leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
43e665287f
linux/net/wireless/nl80211.c

15251 lines
402 KiB
C
Raw Normal View History

[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/*
* This is the new netlink-based wireless configuration interface.
*
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
* Copyright 2006-2010 Johannes Berg <johannes@sipsolutions.net>
cfg80211: add Intel Mobile Communications copyright Our legal structure changed at some point (see wikipedia), but we forgot to immediately switch over to the new copyright notice. For files that we have modified in the time since the change, add the proper copyright notice now. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:24:58 +00:00
* Copyright 2013-2014 Intel Mobile Communications GmbH
nl80211: add HT/VHT capabilities to AP parameters For the benefit of drivers that rebuild IEs in firmware, parse the IEs for HT/VHT capabilities and the respective membership selector in the (extended) supported rates. This avoids duplicating the same code into all drivers that need this information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-07 20:40:44 +00:00
* Copyright 2015-2017 Intel Deutschland GmbH
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
*/
#include <linux/if.h>
#include <linux/module.h>
#include <linux/err.h>
include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo <tj@kernel.org> Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org> Cc: Ingo Molnar <mingo@redhat.com> Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
2010-03-24 08:04:11 +00:00
#include <linux/slab.h>
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
#include <linux/list.h>
#include <linux/if_ether.h>
#include <linux/ieee80211.h>
#include <linux/nl80211.h>
#include <linux/rtnetlink.h>
#include <linux/netlink.h>
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
#include <linux/etherdevice.h>
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
#include <net/net_namespace.h>
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
#include <net/genetlink.h>
#include <net/cfg80211.h>
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
#include <net/sock.h>
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
#include <net/inet_connection_sock.h>
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
#include "core.h"
#include "nl80211.h"
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
#include "reg.h"
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
#include "rdev-ops.h"
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211/cfg80211: Add crypto settings into NEW_BEACON This removes need from drivers to parse the beacon tail/head data to figure out what crypto settings are to be used in AP mode in case the Beacon and Probe Response frames are fully constructed in the driver/firmware. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:54:35 +00:00
static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev,
struct genl_info *info,
struct cfg80211_crypto_settings *settings,
int cipher_limit);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* the netlink family */
genetlink: statically initialize families Instead of providing macros/inline functions to initialize the families, make all users initialize them statically and get rid of the macros. This reduces the kernel code size by about 1.6k on x86-64 (with allyesconfig). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:03 +00:00
static struct genl_family nl80211_fam;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
/* multicast groups */
enum nl80211_multicast_groups {
NL80211_MCGRP_CONFIG,
NL80211_MCGRP_SCAN,
NL80211_MCGRP_REGULATORY,
NL80211_MCGRP_MLME,
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
NL80211_MCGRP_VENDOR,
cfg80211: provide a function to report a match for NAN Provide a function the driver can call to report a match. This will send the event to the user space. If the NAN instance is tied to the owner, the notifications will be sent to the socket that started the NAN interface only. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:17 +00:00
NL80211_MCGRP_NAN,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_TESTMODE /* keep last - ifdef! */
};
static const struct genl_multicast_group nl80211_mcgrps[] = {
nl80211: define multicast group names in header Put the group names into the userspace API header file so that userspace clients can use symbolic names from there instead of hardcoding the actual names. This doesn't really change much, but seems somewhat cleaner. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-23 16:17:38 +00:00
[NL80211_MCGRP_CONFIG] = { .name = NL80211_MULTICAST_GROUP_CONFIG },
[NL80211_MCGRP_SCAN] = { .name = NL80211_MULTICAST_GROUP_SCAN },
[NL80211_MCGRP_REGULATORY] = { .name = NL80211_MULTICAST_GROUP_REG },
[NL80211_MCGRP_MLME] = { .name = NL80211_MULTICAST_GROUP_MLME },
[NL80211_MCGRP_VENDOR] = { .name = NL80211_MULTICAST_GROUP_VENDOR },
cfg80211: provide a function to report a match for NAN Provide a function the driver can call to report a match. This will send the event to the user space. If the NAN instance is tied to the owner, the notifications will be sent to the socket that started the NAN interface only. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:17 +00:00
[NL80211_MCGRP_NAN] = { .name = NL80211_MULTICAST_GROUP_NAN },
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
#ifdef CONFIG_NL80211_TESTMODE
nl80211: define multicast group names in header Put the group names into the userspace API header file so that userspace clients can use symbolic names from there instead of hardcoding the actual names. This doesn't really change much, but seems somewhat cleaner. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-23 16:17:38 +00:00
[NL80211_MCGRP_TESTMODE] = { .name = NL80211_MULTICAST_GROUP_TESTMODE }
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
#endif
};
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
/* returns ERR_PTR values */
static struct wireless_dev *
__cfg80211_wdev_from_attrs(struct net *netns, struct nlattr **attrs)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
struct cfg80211_registered_device *rdev;
struct wireless_dev *result = NULL;
bool have_ifidx = attrs[NL80211_ATTR_IFINDEX];
bool have_wdev_id = attrs[NL80211_ATTR_WDEV];
u64 wdev_id;
int wiphy_idx = -1;
int ifidx = -1;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
ASSERT_RTNL();
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
if (!have_ifidx && !have_wdev_id)
return ERR_PTR(-EINVAL);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
if (have_ifidx)
ifidx = nla_get_u32(attrs[NL80211_ATTR_IFINDEX]);
if (have_wdev_id) {
wdev_id = nla_get_u64(attrs[NL80211_ATTR_WDEV]);
wiphy_idx = wdev_id >> 32;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
list_for_each_entry(rdev, &cfg80211_rdev_list, list) {
struct wireless_dev *wdev;
if (wiphy_net(&rdev->wiphy) != netns)
continue;
if (have_wdev_id && rdev->wiphy_idx != wiphy_idx)
continue;
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) {
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
if (have_ifidx && wdev->netdev &&
wdev->netdev->ifindex == ifidx) {
result = wdev;
break;
}
if (have_wdev_id && wdev->identifier == (u32)wdev_id) {
result = wdev;
break;
}
}
if (result)
break;
}
if (result)
return result;
return ERR_PTR(-ENODEV);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
static struct cfg80211_registered_device *
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
__cfg80211_rdev_from_attrs(struct net *netns, struct nlattr **attrs)
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
{
nl80211: refactor __cfg80211_rdev_from_info Refactor the function to make it easier to extend. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:09:58 +00:00
struct cfg80211_registered_device *rdev = NULL, *tmp;
struct net_device *netdev;
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
ASSERT_RTNL();
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
if (!attrs[NL80211_ATTR_WIPHY] &&
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
!attrs[NL80211_ATTR_IFINDEX] &&
!attrs[NL80211_ATTR_WDEV])
nl80211: refactor __cfg80211_rdev_from_info Refactor the function to make it easier to extend. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:09:58 +00:00
return ERR_PTR(-EINVAL);
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
if (attrs[NL80211_ATTR_WIPHY])
nl80211: refactor __cfg80211_rdev_from_info Refactor the function to make it easier to extend. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:09:58 +00:00
rdev = cfg80211_rdev_by_wiphy_idx(
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
nla_get_u32(attrs[NL80211_ATTR_WIPHY]));
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
if (attrs[NL80211_ATTR_WDEV]) {
u64 wdev_id = nla_get_u64(attrs[NL80211_ATTR_WDEV]);
struct wireless_dev *wdev;
bool found = false;
tmp = cfg80211_rdev_by_wiphy_idx(wdev_id >> 32);
if (tmp) {
/* make sure wdev exists */
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(wdev, &tmp->wiphy.wdev_list, list) {
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
if (wdev->identifier != (u32)wdev_id)
continue;
found = true;
break;
}
if (!found)
tmp = NULL;
if (rdev && tmp != rdev)
return ERR_PTR(-EINVAL);
rdev = tmp;
}
}
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
if (attrs[NL80211_ATTR_IFINDEX]) {
int ifindex = nla_get_u32(attrs[NL80211_ATTR_IFINDEX]);
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
netdev = __dev_get_by_index(netns, ifindex);
nl80211: refactor __cfg80211_rdev_from_info Refactor the function to make it easier to extend. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:09:58 +00:00
if (netdev) {
if (netdev->ieee80211_ptr)
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
tmp = wiphy_to_rdev(
netdev->ieee80211_ptr->wiphy);
nl80211: refactor __cfg80211_rdev_from_info Refactor the function to make it easier to extend. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:09:58 +00:00
else
tmp = NULL;
/* not wireless device -- return error */
if (!tmp)
return ERR_PTR(-EINVAL);
/* mismatch -- return error */
if (rdev && tmp != rdev)
return ERR_PTR(-EINVAL);
rdev = tmp;
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
}
}
nl80211: fix netns separation There are currently a few ways to "escape" the network namespace and access a wiphy that belongs to another namespace. Add a netns argument to the relevant functions to fix this. One remaining issue with testmode will be fixed in a follow-up patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:14:22 +00:00
if (!rdev)
return ERR_PTR(-ENODEV);
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
nl80211: fix netns separation There are currently a few ways to "escape" the network namespace and access a wiphy that belongs to another namespace. Add a netns argument to the relevant functions to fix this. One remaining issue with testmode will be fixed in a follow-up patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:14:22 +00:00
if (netns != wiphy_net(&rdev->wiphy))
return ERR_PTR(-ENODEV);
return rdev;
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
}
/*
* This function returns a pointer to the driver
* that the genl_info item that is passed refers to.
*
* The result of this can be a PTR_ERR and hence must
* be checked with IS_ERR() for errors.
*/
static struct cfg80211_registered_device *
nl80211: fix netns separation There are currently a few ways to "escape" the network namespace and access a wiphy that belongs to another namespace. Add a netns argument to the relevant functions to fix this. One remaining issue with testmode will be fixed in a follow-up patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:14:22 +00:00
cfg80211_get_dev_from_info(struct net *netns, struct genl_info *info)
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
{
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
return __cfg80211_rdev_from_attrs(netns, info->attrs);
cfg80211: make some functions static Some of the functions to retrieve a device can be static as they're used only in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 11:32:49 +00:00
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* policy for the attributes */
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
static const struct nla_policy nl80211_policy[NUM_NL80211_ATTR] = {
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
[NL80211_ATTR_WIPHY] = { .type = NLA_U32 },
[NL80211_ATTR_WIPHY_NAME] = { .type = NLA_NUL_STRING,
nl80211: Eliminate reference to BUS_ID_SIZE. It's going away. Just leave the constant "20" here so that behavior doesn't change. Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-27 04:15:00 +00:00
.len = 20-1 },
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
[NL80211_ATTR_WIPHY_TXQ_PARAMS] = { .type = NLA_NESTED },
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
nl80211: Add frequency configuration (including HT40) This patch adds new NL80211_CMD_SET_WIPHY attributes NL80211_ATTR_WIPHY_FREQ and NL80211_ATTR_WIPHY_SEC_CHAN_OFFSET to allow userspace to set the operating channel (e.g., hostapd for AP mode). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-26 14:15:24 +00:00
[NL80211_ATTR_WIPHY_FREQ] = { .type = NLA_U32 },
mac80211: Fix HT channel selection HT management is done differently for AP and STA modes, unify to just the ->config() callback since HT is fundamentally a PHY property and cannot be per-BSS. Rename enum nl80211_sec_chan_offset as nl80211_channel_type to denote the channel type ( NO_HT, HT20, HT40+, HT40- ). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Sujith <Sujith.Manoharan@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-12-12 06:27:43 +00:00
[NL80211_ATTR_WIPHY_CHANNEL_TYPE] = { .type = NLA_U32 },
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
[NL80211_ATTR_CHANNEL_WIDTH] = { .type = NLA_U32 },
[NL80211_ATTR_CENTER_FREQ1] = { .type = NLA_U32 },
[NL80211_ATTR_CENTER_FREQ2] = { .type = NLA_U32 },
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
[NL80211_ATTR_WIPHY_RETRY_SHORT] = { .type = NLA_U8 },
[NL80211_ATTR_WIPHY_RETRY_LONG] = { .type = NLA_U8 },
[NL80211_ATTR_WIPHY_FRAG_THRESHOLD] = { .type = NLA_U32 },
[NL80211_ATTR_WIPHY_RTS_THRESHOLD] = { .type = NLA_U32 },
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
[NL80211_ATTR_WIPHY_COVERAGE_CLASS] = { .type = NLA_U8 },
cfg80211: enable dynack through nl80211 Enable ACK timeout estimation algorithm (dynack) using mac80211 set_coverage_class API. Dynack is activated passing coverage class equals to -1 to lower drivers and it is automatically disabled setting valid value for coverage class. Define NL80211_ATTR_WIPHY_DYN_ACK flag attribute to enable dynack from userspace. In order to activate dynack NL80211_FEATURE_ACKTO_ESTIMATION feature flag must be set by lower drivers to indicate dynack capability. Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-04 21:57:40 +00:00
[NL80211_ATTR_WIPHY_DYN_ACK] = { .type = NLA_FLAG },
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
[NL80211_ATTR_IFTYPE] = { .type = NLA_U32 },
[NL80211_ATTR_IFINDEX] = { .type = NLA_U32 },
[NL80211_ATTR_IFNAME] = { .type = NLA_NUL_STRING, .len = IFNAMSIZ-1 },
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: fix MAC address validation MAC addresses have a fixed length. The current policy allows passing < ETH_ALEN bytes, which might result in reading beyond the buffer. Cc: stable@vger.kernel.org Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-24 16:13:56 +00:00
[NL80211_ATTR_MAC] = { .len = ETH_ALEN },
[NL80211_ATTR_PREV_BSSID] = { .len = ETH_ALEN },
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
[NL80211_ATTR_KEY] = { .type = NLA_NESTED, },
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
[NL80211_ATTR_KEY_DATA] = { .type = NLA_BINARY,
.len = WLAN_MAX_KEY_LEN },
[NL80211_ATTR_KEY_IDX] = { .type = NLA_U8 },
[NL80211_ATTR_KEY_CIPHER] = { .type = NLA_U32 },
[NL80211_ATTR_KEY_DEFAULT] = { .type = NLA_FLAG },
nl80211: Increase maximum NL80211_ATTR_KEY_SEQ length to 16 WPI-SMS4 uses 16-octet PN field, so we need to allow longer key sequence values to be configured. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-02 21:36:31 +00:00
[NL80211_ATTR_KEY_SEQ] = { .type = NLA_BINARY, .len = 16 },
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
[NL80211_ATTR_KEY_TYPE] = { .type = NLA_U32 },
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
[NL80211_ATTR_BEACON_INTERVAL] = { .type = NLA_U32 },
[NL80211_ATTR_DTIM_PERIOD] = { .type = NLA_U32 },
[NL80211_ATTR_BEACON_HEAD] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
[NL80211_ATTR_BEACON_TAIL] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
[NL80211_ATTR_STA_AID] = { .type = NLA_U16 },
[NL80211_ATTR_STA_FLAGS] = { .type = NLA_NESTED },
[NL80211_ATTR_STA_LISTEN_INTERVAL] = { .type = NLA_U16 },
[NL80211_ATTR_STA_SUPPORTED_RATES] = { .type = NLA_BINARY,
.len = NL80211_MAX_SUPP_RATES },
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
[NL80211_ATTR_STA_PLINK_ACTION] = { .type = NLA_U8 },
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
[NL80211_ATTR_STA_VLAN] = { .type = NLA_U32 },
nl80211: fix monitor flags NLA_NESTED attributes cannot be empty, but we want to be able to specify "no flags" (empty attribute) vs. "no change" (no attribute). Therefore, remove the NLA_NESTED policy so it can work as an empty attribute. I guess I should have used a u32 for these flags instead, but we're stuck with it now. Haven't noticed earlier because of a bug in iw... Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-15 09:54:04 +00:00
[NL80211_ATTR_MNTR_FLAGS] = { /* NLA_NESTED can't be empty */ },
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
[NL80211_ATTR_MESH_ID] = { .type = NLA_BINARY,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
.len = IEEE80211_MAX_MESH_ID_LEN },
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
[NL80211_ATTR_MPATH_NEXT_HOP] = { .type = NLA_U32 },
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
[NL80211_ATTR_REG_ALPHA2] = { .type = NLA_STRING, .len = 2 },
[NL80211_ATTR_REG_RULES] = { .type = NLA_NESTED },
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
[NL80211_ATTR_BSS_CTS_PROT] = { .type = NLA_U8 },
[NL80211_ATTR_BSS_SHORT_PREAMBLE] = { .type = NLA_U8 },
[NL80211_ATTR_BSS_SHORT_SLOT_TIME] = { .type = NLA_U8 },
nl80211: Add basic rate configuration for AP mode Add a new attribute, NL80211_ATTR_BSS_BASIC_RATES, that can be used with NL80211_CMD_SET_BSS for userspace (e.g., hostapd) to set which rates are in the basic rate set. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:22 +00:00
[NL80211_ATTR_BSS_BASIC_RATES] = { .type = NLA_BINARY,
.len = NL80211_MAX_SUPP_RATES },
cfg80211: Add new BSS attribute ht_opmode Add a new BSS attribute to allow hostapd to set the current HT opmode. Otherwise drivers won't be able to set up protection for HT rates in AP mode. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-19 11:40:25 +00:00
[NL80211_ATTR_BSS_HT_OPMODE] = { .type = NLA_U16 },
mac80211/cfg80211: HT capabilities for NEW_STA Allow userspace (e.g., hostapd) to set HT capabilities for associated STAs. This is based on a patch from Zhu Yi <yi.zhu@intel.com> (only the NL80211_ATTR_HT_CAPABILITY for NEW_STA part is included here). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-25 08:58:58 +00:00
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
[NL80211_ATTR_MESH_CONFIG] = { .type = NLA_NESTED },
cfg80211/nl80211: Add userspace authentication flag to mesh setup During mesh setup, use NL80211_MESH_SETUP_USERSPACE_AUTH flag to create a secure mesh and route management frames to userspace. Also, NL80211_CMD_GET_WIPHY now returns a flag NL80211_SUPPORT_MESH_AUTH if the wiphy's mesh implementation supports routing of mesh auth frames to userspace. This is useful for forward compatibility between old kernels and new userspace tools. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:28 +00:00
[NL80211_ATTR_SUPPORT_MESH_AUTH] = { .type = NLA_FLAG },
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
nl80211: fix HT capability attribute validation Since the NL80211_ATTR_HT_CAPABILITY attribute is used as a struct, it needs a minimum, not maximum length. Enforce that properly. Not doing so could potentially lead to reading after the buffer. Cc: stable@vger.kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-03 08:27:01 +00:00
[NL80211_ATTR_HT_CAPABILITY] = { .len = NL80211_HT_CAPABILITY_LEN },
nl80211: New command for adding extra IE(s) into management frames A new nl80211 command, NL80211_CMD_SET_MGMT_EXTRA_IE, can be used to add arbitrary IE data into the end of management frames. The interface allows extra IEs to be configured for each management frame subtype, but only some of them (ProbeReq, ProbeResp, Auth, (Re)AssocReq, Deauth, Disassoc) are currently accepted in mac80211 implementation. This makes it easier to implement IEEE 802.11 extensions like WPS and FT that add IE(s) into some management frames. In addition, this can be useful for testing and experimentation purposes. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-13 14:03:29 +00:00
[NL80211_ATTR_MGMT_SUBTYPE] = { .type = NLA_U8 },
[NL80211_ATTR_IE] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
[NL80211_ATTR_SCAN_FREQUENCIES] = { .type = NLA_NESTED },
[NL80211_ATTR_SCAN_SSIDS] = { .type = NLA_NESTED },
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
[NL80211_ATTR_SSID] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_SSID_LEN },
[NL80211_ATTR_AUTH_TYPE] = { .type = NLA_U32 },
[NL80211_ATTR_REASON_CODE] = { .type = NLA_U16 },
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
[NL80211_ATTR_FREQ_FIXED] = { .type = NLA_FLAG },
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
[NL80211_ATTR_TIMED_OUT] = { .type = NLA_FLAG },
nl80211 : Add support for configuring MFP NL80211_CMD_ASSOCIATE request must be able to indicate whether management frame protection (IEEE 802.11w) is being used. mac80211 was able to use MFP in client mode only with WEXT, but the new NL80211_ATTR_USE_MFP attribute will allow this to be done with nl80211, too. Since we are currently using nl80211 for MFP only with drivers that use user space SME, only MFP disabled and required values are used. However, the NL80211_ATTR_USE_MFP attribute is an enum that can be extended with MFP optional in the future, if that is needed with some drivers (e.g., if the RSN IE is generated by the driver). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-06 19:09:37 +00:00
[NL80211_ATTR_USE_MFP] = { .type = NLA_U32 },
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
[NL80211_ATTR_STA_FLAGS2] = {
.len = sizeof(struct nl80211_sta_flag_update),
},
nl80211: Add IEEE 802.1X PAE control for station mode Add a new NL80211_ATTR_CONTROL_PORT flag for NL80211_CMD_ASSOCIATE to allow user space to indicate that it will control the IEEE 802.1X port in station mode. Previously, mac80211 was always marking the port authorized in station mode. This was enough when drop_unencrypted flag was set. However, drop_unencrypted can currently be controlled only with WEXT and the current nl80211 design does not allow fully secure configuration. Fix this by providing a mechanism for user space to control the IEEE 802.1X port in station mode (i.e., do the same that we are already doing in AP mode). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:57 +00:00
[NL80211_ATTR_CONTROL_PORT] = { .type = NLA_FLAG },
cfg80211: allow changing port control protocol Some vendor specified mechanisms for 802.1X-style functionality use a different protocol than EAP (even if EAP is vendor-extensible). Allow setting the ethertype for the protocol when a driver has support for this. The default if unspecified is EAP, of course. Note: This is suitable only for station mode, not for AP implementation. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-27 11:26:53 +00:00
[NL80211_ATTR_CONTROL_PORT_ETHERTYPE] = { .type = NLA_U16 },
[NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT] = { .type = NLA_FLAG },
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
[NL80211_ATTR_PRIVACY] = { .type = NLA_FLAG },
[NL80211_ATTR_CIPHER_SUITE_GROUP] = { .type = NLA_U32 },
[NL80211_ATTR_WPA_VERSIONS] = { .type = NLA_U32 },
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
[NL80211_ATTR_PID] = { .type = NLA_U32 },
nl80211: add a parameter for using 4-address frames on virtual interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-10 17:53:10 +00:00
[NL80211_ATTR_4ADDR] = { .type = NLA_U8 },
cfg80211: Check if PMKID attribute is of expected size nla policy checks for only maximum length of the attribute data when the attribute type is NLA_BINARY. If userspace sends less data than specified, the wireless drivers may access illegal memory. When type is NLA_UNSPEC, nla policy check ensures that userspace sends minimum specified length number of bytes. Remove type assignment to NLA_BINARY from nla_policy of NL80211_ATTR_PMKID to make this NLA_UNSPEC and to make sure minimum WLAN_PMKID_LEN bytes are received from userspace with NL80211_ATTR_PMKID. Fixes: 67fbb16be69d ("nl80211: PMKSA caching support") Cc: stable@vger.kernel.org Signed-off-by: Srinivas Dasari <dasaris@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-07-06 22:43:39 +00:00
[NL80211_ATTR_PMKID] = { .len = WLAN_PMKID_LEN },
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
[NL80211_ATTR_DURATION] = { .type = NLA_U32 },
[NL80211_ATTR_COOKIE] = { .type = NLA_U64 },
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
[NL80211_ATTR_TX_RATES] = { .type = NLA_NESTED },
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
[NL80211_ATTR_FRAME] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
[NL80211_ATTR_FRAME_MATCH] = { .type = NLA_BINARY, },
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
[NL80211_ATTR_PS_STATE] = { .type = NLA_U32 },
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
[NL80211_ATTR_CQM] = { .type = NLA_NESTED, },
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
[NL80211_ATTR_LOCAL_STATE_CHANGE] = { .type = NLA_FLAG },
cfg80211: add ap isolation support This is used to configure APs to not bridge traffic between connected stations. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-26 23:23:35 +00:00
[NL80211_ATTR_AP_ISOLATE] = { .type = NLA_U8 },
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
[NL80211_ATTR_WIPHY_TX_POWER_SETTING] = { .type = NLA_U32 },
[NL80211_ATTR_WIPHY_TX_POWER_LEVEL] = { .type = NLA_U32 },
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
[NL80211_ATTR_FRAME_TYPE] = { .type = NLA_U16 },
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
[NL80211_ATTR_WIPHY_ANTENNA_TX] = { .type = NLA_U32 },
[NL80211_ATTR_WIPHY_ANTENNA_RX] = { .type = NLA_U32 },
cfg80211: add support for setting the ad-hoc multicast rate Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-11 14:07:22 +00:00
[NL80211_ATTR_MCAST_RATE] = { .type = NLA_U32 },
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
[NL80211_ATTR_OFFCHANNEL_TX_OK] = { .type = NLA_FLAG },
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
[NL80211_ATTR_KEY_DEFAULT_TYPES] = { .type = NLA_NESTED },
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
[NL80211_ATTR_WOWLAN_TRIGGERS] = { .type = NLA_NESTED },
nl80211: Let userspace drive the peer link management states. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:11 +00:00
[NL80211_ATTR_STA_PLINK_STATE] = { .type = NLA_U8 },
cfg80211/nl80211: add interval attribute for scheduled scans Introduce NL80211_ATTR_SCHED_SCAN_INTERVAL as a required attribute for NL80211_CMD_START_SCHED_SCAN. This value informs the driver at which intervals the scheduled scan cycles should be executed. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:37 +00:00
[NL80211_ATTR_SCHED_SCAN_INTERVAL] = { .type = NLA_U32 },
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
[NL80211_ATTR_REKEY_DATA] = { .type = NLA_NESTED },
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
[NL80211_ATTR_SCAN_SUPP_RATES] = { .type = NLA_NESTED },
nl80211/cfg80211: Allow SSID to be specified in new beacon command This makes it easier for drivers that generate Beacon and Probe Response frames internally (in firmware most likely) in AP mode. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:53:31 +00:00
[NL80211_ATTR_HIDDEN_SSID] = { .type = NLA_U32 },
nl80211/cfg80211: Add extra IE configuration to AP mode setup The NL80211_CMD_NEW_BEACON command is, in practice, requesting AP mode operations to be started. Add new attributes to provide extra IEs (e.g., WPS IE, P2P IE) for drivers that build Beacon, Probe Response, and (Re)Association Response frames internally (likely in firmware). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:55:56 +00:00
[NL80211_ATTR_IE_PROBE_RESP] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
[NL80211_ATTR_IE_ASSOC_RESP] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
cfg80211/nl80211: Indicate roaming feature capability to userspace. When the rssi of the current AP drops, both wpa_supplicant and the firmware may do a background scan to find a better AP and try to associate. Since firmware based roaming is faster, inform wpa_supplicant to avoid roaming and let the firmware decide to roam if necessary. For fullmac drivers like ath6kl, it is just enough to provide the ESSID and the firmware will decide on the BSSID. Since it is not possible to do pre-auth during roaming for fullmac drivers, the wpa_supplicant needs to completely disconnect with the old AP and reconnect with the new AP. This consumes lot of time and it is better to leave the roaming decision to the firmware. Signed-off-by: Vivek Natarajan <nataraja@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-29 08:53:03 +00:00
[NL80211_ATTR_ROAM_SUPPORT] = { .type = NLA_FLAG },
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
[NL80211_ATTR_SCHED_SCAN_MATCH] = { .type = NLA_NESTED },
nl80211/cfg80211: Add support to disable CCK rate for management frame Add a new nl80211 attribute to specify whether to send the management frames in CCK rate or not. As of now the wpa_supplicant is disabling CCK rate at P2P init itself. So this patch helps to send P2P probe request/probe response/action frames being sent at non CCK rate in 2GHz without disabling 11b rates. This attribute is used with NL80211_CMD_TRIGGER_SCAN and NL80211_CMD_FRAME commands to disable CCK rate for management frame transmission. Cc: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-25 09:23:30 +00:00
[NL80211_ATTR_TX_NO_CCK_RATE] = { .type = NLA_FLAG },
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
[NL80211_ATTR_TDLS_ACTION] = { .type = NLA_U8 },
[NL80211_ATTR_TDLS_DIALOG_TOKEN] = { .type = NLA_U8 },
[NL80211_ATTR_TDLS_OPERATION] = { .type = NLA_U8 },
[NL80211_ATTR_TDLS_SUPPORT] = { .type = NLA_FLAG },
[NL80211_ATTR_TDLS_EXTERNAL_SETUP] = { .type = NLA_FLAG },
cfg80211: pass TDLS initiator in tdls_mgmt operations The TDLS initiator is set once during link setup. If determines the address ordering in the link identifier IE. Fix dependent drivers - mwifiex and mac80211. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-11 14:18:21 +00:00
[NL80211_ATTR_TDLS_INITIATOR] = { .type = NLA_FLAG },
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
[NL80211_ATTR_DONT_WAIT_FOR_ACK] = { .type = NLA_FLAG },
nl80211: Pass probe response data to drivers Pass probe-response data from usermode via beacon parameters. Signed-off-by: Guy Eilam <guy@wizery.com> Signed-off-by: Arik Nemtsov <arik@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-10 09:28:56 +00:00
[NL80211_ATTR_PROBE_RESP] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
cfg80211: process regulatory DFS region for countries The wireless-regdb now has support for mapping a country to one DFS region. CRDA sends this to us now so process it so we can provide that hint to drivers. This will later be used by code for processing DFS in a way that meets the criteria for the DFS region the country belongs to. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-11 17:59:02 +00:00
[NL80211_ATTR_DFS_REGION] = { .type = NLA_U8 },
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
[NL80211_ATTR_DISABLE_HT] = { .type = NLA_FLAG },
[NL80211_ATTR_HT_CAPABILITY_MASK] = {
.len = NL80211_HT_CAPABILITY_LEN
},
wireless: Add NoAck per tid support This patch contains the configuration changes in nl80211/cfg80211. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 13:20:43 +00:00
[NL80211_ATTR_NOACK_MAP] = { .type = NLA_U16 },
cfg80211: Add an attribute to set inactivity timeout in AP mode This patch adds an attribute, NL80211_ATTR_INACTIVITY_TIMEOUT, to set the inactivity timeout which can be used to remove the station in AP mode. This can be passed in NL80211_CMD_START_AP and used by the drivers which have AP MLME in firmware but don't support get_station() properly. To disable inactivity timer in userspace, wpa_s for example, there is a new flag, NL80211_FEATURE_INACTIVITY_TIMER, in nl80211_feature_flags through which drivers can register their capability to use the inactivity timeout to free the stations. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-02 10:20:02 +00:00
[NL80211_ATTR_INACTIVITY_TIMEOUT] = { .type = NLA_U16 },
cfg80211: Add background scan period attribute. Receive background scan period as part of connect command and pass the same to the driver. Signed-off-by: Bala Shanmugam <bkamatch@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-07 11:57:12 +00:00
[NL80211_ATTR_BG_SCAN_PERIOD] = { .type = NLA_U16 },
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
[NL80211_ATTR_WDEV] = { .type = NLA_U64 },
cfg80211: add cellular base station regulatory hint support Cellular base stations can provide hints to cfg80211 about where they think we are. This can be done for example on a cell phone. To enable these hints we simply allow them through as user regulatory hints but we allow userspace to clasify the hint as either coming directly from the user or coming from a cellular base station. This option is only available when you enable CONFIG_CFG80211_CERTIFICATION_ONUS. The base station hints themselves will not be processed by the core unless at least one device on the system supports this feature. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 18:49:18 +00:00
[NL80211_ATTR_USER_REG_HINT_TYPE] = { .type = NLA_U32 },
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
[NL80211_ATTR_AUTH_DATA] = { .type = NLA_BINARY, },
{nl,cfg}80211: Peer STA VHT caps To save STAs VHT caps in AP mode Signed-off-by: Mahesh Palivela <maheshp@posedge.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-11 08:04:52 +00:00
[NL80211_ATTR_VHT_CAPABILITY] = { .len = NL80211_VHT_CAPABILITY_LEN },
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
[NL80211_ATTR_SCAN_FLAGS] = { .type = NLA_U32 },
nl80211: support P2P GO powersave configuration If a driver supports P2P GO powersave, allow it to set the new feature flags for it and allow userspace to configure the parameters for it. This can be done at GO startup and later changed with SET_BSS. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-14 14:17:28 +00:00
[NL80211_ATTR_P2P_CTWINDOW] = { .type = NLA_U8 },
[NL80211_ATTR_P2P_OPPPS] = { .type = NLA_U8 },
cfg80211: Define nla_policy for NL80211_ATTR_LOCAL_MESH_POWER_MODE Buffer overread may happen as nl80211_set_station() reads 4 bytes from the attribute NL80211_ATTR_LOCAL_MESH_POWER_MODE without validating the size of data received when userspace sends less than 4 bytes of data with NL80211_ATTR_LOCAL_MESH_POWER_MODE. Define nla_policy for NL80211_ATTR_LOCAL_MESH_POWER_MODE to avoid the buffer overread. Fixes: 3b1c5a5307f ("{cfg,nl}80211: mesh power mode primitives and userspace access") Cc: stable@vger.kernel.org Signed-off-by: Srinivas Dasari <dasaris@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-07-06 22:43:41 +00:00
[NL80211_ATTR_LOCAL_MESH_POWER_MODE] = {. type = NLA_U32 },
cfg80211/nl80211: add API for MAC address ACLs Add API to enable drivers to implement MAC address based access control in AP/P2P GO mode. Capable drivers advertise this capability by setting the maximum number of MAC addresses in such a list in wiphy->max_acl_mac_addrs. An initial ACL may be given to the NL80211_CMD_START_AP command and/or changed later with NL80211_CMD_SET_MAC_ACL. Black- and whitelists are supported, but not simultaneously. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> [rewrite commit log, many cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-18 05:48:45 +00:00
[NL80211_ATTR_ACL_POLICY] = {. type = NLA_U32 },
[NL80211_ATTR_MAC_ADDRS] = { .type = NLA_NESTED },
cfg80211: Pass station (extended) capability info to kernel The information of the peer's capabilities and extended capabilities are required for the driver to perform TDLS Peer UAPSD operations and off channel operations. This information of the peer is passed from user space using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass the capability information of the peer to the driver. Similarly, there may be need for capability information for other modes, so allow this to be provided with both add_station and change_station. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:13 +00:00
[NL80211_ATTR_STA_CAPABILITY] = { .type = NLA_U16 },
[NL80211_ATTR_STA_EXT_CAPABILITY] = { .type = NLA_BINARY, },
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
[NL80211_ATTR_SPLIT_WIPHY_DUMP] = { .type = NLA_FLAG, },
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
[NL80211_ATTR_DISABLE_VHT] = { .type = NLA_FLAG },
[NL80211_ATTR_VHT_CAPABILITY_MASK] = {
.len = NL80211_VHT_CAPABILITY_LEN,
},
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
[NL80211_ATTR_MDID] = { .type = NLA_U16 },
[NL80211_ATTR_IE_RIC] = { .type = NLA_BINARY,
.len = IEEE80211_MAX_DATA_LEN },
cfg80211: Allow TDLS peer AID to be configured for VHT VHT uses peer AID in the PARTIAL_AID field in TDLS frames. The current design for TDLS is to first add a dummy STA entry before completing TDLS Setup and then update information on this STA entry based on what was received from the peer during the setup exchange. In theory, this could use NL80211_ATTR_STA_AID to set the peer AID just like this is used in AP mode to set the AID of an association station. However, existing cfg80211 validation rules prevent this attribute from being used with set_station operation. To avoid interoperability issues between different kernel and user space version combinations, introduce a new nl80211 attribute for the purpose of setting TDLS peer AID. This attribute can be used in both the new_station and set_station operations. It is not supposed to be allowed to change the AID value during the lifetime of the STA entry, but that validation is left for drivers to do in the change_station callback. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 17:11:08 +00:00
[NL80211_ATTR_PEER_AID] = { .type = NLA_U16 },
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
[NL80211_ATTR_CH_SWITCH_COUNT] = { .type = NLA_U32 },
[NL80211_ATTR_CH_SWITCH_BLOCK_TX] = { .type = NLA_FLAG },
[NL80211_ATTR_CSA_IES] = { .type = NLA_NESTED },
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
[NL80211_ATTR_CSA_C_OFF_BEACON] = { .type = NLA_BINARY },
[NL80211_ATTR_CSA_C_OFF_PRESP] = { .type = NLA_BINARY },
cfg80211: pass station supported channel and oper class info The information of the peer's supported channels and supported operating classes are required for the driver to perform TDLS off channel operations. This commit enhances the function nl80211_(new)set_station to pass this information of the peer to the driver. Signed-off-by: Sunil Dutt <c_duttus@qti.qualcomm.com> [return errors for malformed tuples] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-09 15:15:21 +00:00
[NL80211_ATTR_STA_SUPPORTED_CHANNELS] = { .type = NLA_BINARY },
[NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES] = { .type = NLA_BINARY },
nl80211/cfg80211: enable DFS for IBSS mode To use DFS in IBSS mode, userspace is required to react to radar events. It can inform nl80211 that it is capable of doing so by adding a NL80211_ATTR_HANDLE_DFS attribute when joining the IBSS. This attribute is supplied to let the kernelspace know that the userspace application can and will handle radar events, e.g. by intiating channel switches to a valid channel. DFS channels may only be used if this attribute is supplied and the driver supports it. Driver support will be checked even if a channel without DFS will be initially joined, as a DFS channel may be chosen later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix attribute name in commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-07 16:41:05 +00:00
[NL80211_ATTR_HANDLE_DFS] = { .type = NLA_FLAG },
nl80211/cfg80211: Set Operating Mode Notification This attribute is needed for setting Operating Mode Notification in AP mode from User Space. This functionality is required when User Space received Assoc Request contains Operation Mode Notification element. Signed-off-by: Marek Kwaczynski <marek.kwaczynski@tieto.com> [fix typos, nl80211 documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-03 09:04:59 +00:00
[NL80211_ATTR_OPMODE_NOTIF] = { .type = NLA_U8 },
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
[NL80211_ATTR_VENDOR_ID] = { .type = NLA_U32 },
[NL80211_ATTR_VENDOR_SUBCMD] = { .type = NLA_U32 },
[NL80211_ATTR_VENDOR_DATA] = { .type = NLA_BINARY },
cfg80211: Add support for QoS mapping This allows QoS mapping from external networks to be implemented as defined in IEEE Std 802.11-2012, 10.24.9. APs can use this to advertise DSCP ranges and exceptions for mapping frames to a specific UP over Wi-Fi. The payload of the QoS Map Set element (IEEE Std 802.11-2012, 8.4.2.97) is sent to the driver through the new NL80211_ATTR_QOS_MAP attribute to configure the local behavior either on the AP (based on local configuration) or on a station (based on information received from the AP). Signed-off-by: Kyeyoon Park <kyeyoonp@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-17 07:01:30 +00:00
[NL80211_ATTR_QOS_MAP] = { .type = NLA_BINARY,
.len = IEEE80211_QOS_MAP_LEN_MAX },
cfg80211: Allow BSS hint to be provided for connect This clarifies the expected driver behavior on the older NL80211_ATTR_MAC and NL80211_ATTR_WIPHY_FREQ attributes and adds a new set of similar attributes with _HINT postfix to enable use of a recommendation of the initial BSS to choose. This can be helpful for some drivers that can avoid an additional full scan on connection request if the information is provided to them (user space tools like wpa_supplicant already has that information available based on earlier scans). In addition, this can be used to get more expected behavior for cases where a specific BSS should be picked first based on operations like Interworking network selection or WPS. These cases were already easily addressed with drivers that leave BSS selection to user space, but there was no convenient way to do this with drivers that take care of BSS selection internally without using the NL80211_ATTR_MAC which is not really desired since it is needed for other purposes to force the association to remain with the same BSS. Signed-off-by: Jouni Malinen <j@w1.fi> [add const, fix policy] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:00:47 +00:00
[NL80211_ATTR_MAC_HINT] = { .len = ETH_ALEN },
[NL80211_ATTR_WIPHY_FREQ_HINT] = { .type = NLA_U32 },
cfg80211: Pass TDLS peer capability information in tdls_mgmt While framing the TDLS Setup Confirmation frame, the driver needs to know if the TDLS peer is VHT/HT/WMM capable and thus shall construct the VHT/HT operation / WMM parameter elements accordingly. Supplicant determines if the TDLS peer is VHT/HT/WMM capable based on the presence of the respective IEs in the received TDLS Setup Response frame. The host driver should not need to parse the received TDLS Response frame and thus, should be able to rely on the supplicant to indicate the capability of the peer through additional flags while transmitting the TDLS Setup Confirmation frame through tdls_mgmt operations. Signed-off-by: Sunil Dutt Undekari <usdutt@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-20 10:52:09 +00:00
[NL80211_ATTR_TDLS_PEER_CAPABILITY] = { .type = NLA_U32 },
nl80211: Replace interface socket owner attribute with more generic one Replace NL80211_ATTR_IFACE_SOCKET_OWNER attribute with more generic NL80211_ATTR_SOCKET_OWNER that can be used with other commands that interface creation. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-13 15:25:14 +00:00
[NL80211_ATTR_SOCKET_OWNER] = { .type = NLA_FLAG },
cfg80211: Add API to update CSA counters in mgmt frames Add NL80211_ATTR_CSA_C_OFFSETS_TX which holds an array of offsets to the CSA counters which should be updated when sending a management frames with NL80211_CMD_FRAME. This API should be used by the drivers that wish to keep the CSA counter updated in probe responses, but do not implement probe response offloading and so, do not use ieee80211_proberesp_get function. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:44 +00:00
[NL80211_ATTR_CSA_C_OFFSETS_TX] = { .type = NLA_BINARY },
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
[NL80211_ATTR_USE_RRM] = { .type = NLA_FLAG },
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
[NL80211_ATTR_TSID] = { .type = NLA_U8 },
[NL80211_ATTR_USER_PRIO] = { .type = NLA_U8 },
[NL80211_ATTR_ADMITTED_TIME] = { .type = NLA_U16 },
cfg80211: allow requesting SMPS mode on ap start Add feature bits to indicate device support for static-smps and dynamic-smps modes. Add a new NL80211_ATTR_SMPS_MODE attribue to allow configuring the smps mode to be used by the ap (e.g. configuring to ap to dynamic smps mode will reduce power consumption while having minor effect on throughput) Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 11:07:34 +00:00
[NL80211_ATTR_SMPS_MODE] = { .type = NLA_U8 },
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
[NL80211_ATTR_MAC_MASK] = { .len = ETH_ALEN },
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
[NL80211_ATTR_WIPHY_SELF_MANAGED_REG] = { .type = NLA_FLAG },
nl80211: Allow set network namespace by fd Added new NL80211_ATTR_NETNS_FD which allows to set namespace via nl80211 by fd. Signed-off-by: Vadim Kochan <vadim4j@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-12 14:34:05 +00:00
[NL80211_ATTR_NETNS_FD] = { .type = NLA_U32 },
nl80211: add an attribute to allow delaying the first scheduled scan cycle The userspace may want to delay the the first scheduled scan or net-detect cycle. Add an optional attribute to the scheduled scan configuration to pass the delay to be (optionally) used by the driver. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> [add the attribute to the policy to validate it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-16 14:04:09 +00:00
[NL80211_ATTR_SCHED_SCAN_DELAY] = { .type = NLA_U32 },
cfg80211: Add API to change the indoor regulatory setting Previously, the indoor setting configuration assumed that as long as a station interface is connected, the indoor environment setting does not change. However, this assumption is problematic as: - It is possible that a station interface is connected to a mobile AP, e.g., softAP or a P2P GO, where it is possible that both the station and the mobile AP move out of the indoor environment making the indoor setting invalid. In such a case, user space has no way to invalidate the setting. - A station interface disconnection does not necessarily imply that the device is no longer operating in an indoor environment, e.g., it is possible that the station interface is roaming but is still stays indoor. To handle the above, extend the indoor configuration API to allow user space to indicate a change of indoor settings, and allow it to indicate weather it controls the indoor setting, such that: 1. If the user space process explicitly indicates that it is going to control the indoor setting, do not clear the indoor setting internally, unless the socket is released. The user space process should use the NL80211_ATTR_SOCKET_OWNER attribute in the command to state that it is going to control the indoor setting. 2. Reset the indoor setting when restoring the regulatory settings in case it is not owned by a user space process. Based on the above, a user space tool that continuously monitors the indoor settings, i.e., tracking power setting, location etc., can indicate environment changes to the regulatory core. It should be noted that currently user space is the only provided mechanism used to hint to the regulatory core over the indoor/outdoor environment -- while the country IEs do have an environment setting this has been completely ignored by the regulatory core by design for a while now since country IEs typically can contain bogus data. Acked-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: ArikX Nemtsov <arik@wizery.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-04 05:32:06 +00:00
[NL80211_ATTR_REG_INDOOR] = { .type = NLA_FLAG },
cfg80211: basic support for PBSS network type PBSS (Personal Basic Service Set) is a new BSS type for DMG networks. It is similar to infrastructure BSS, having an AP-like entity called PCP (PBSS Control Point), but it has few differences. PBSS support is mandatory for 11ad devices. Add support for PBSS by introducing a new PBSS flag attribute. The PBSS flag is used in the START_AP command to request starting a PCP instead of an AP, and in the CONNECT command to request connecting to a PCP instead of an AP. Signed-off-by: Lior David <liord@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-01-28 08:58:25 +00:00
[NL80211_ATTR_PBSS] = { .type = NLA_FLAG },
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
[NL80211_ATTR_BSS_SELECT] = { .type = NLA_NESTED },
cfg80211: allow userspace to specify client P2P PS support Legacy clients don't support P2P power save mechanisms, and thus if a P2P GO has a legacy client connected to it, it has to make some changes in the PS behavior. To handle this, add an attribute to specify whether a station supports P2P PS or not. If the attribute was not specified cfg80211 will assume that station supports it for P2P GO interface, and does NOT support it for AP interface, matching the current assumptions in the code. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-17 13:41:38 +00:00
[NL80211_ATTR_STA_SUPPORT_P2P_PS] = { .type = NLA_U8 },
nl80211: Add API to support VHT MU-MIMO air sniffer add API to support VHT MU-MIMO air sniffer. in MU-MIMO there are parallel frames on the air while the HW has only one RX. add the capability to sniff one of the MU-MIMO parallel frames by giving the sniffer additional information so it'll know which of the parallel frames it shall follow. Add attribute - NL80211_ATTR_MU_MIMO_GROUP_DATA - for getting a MU-MIMO groupID in order to monitor packets from that group using VHT MU-MIMO. And add attribute -NL80211_ATTR_MU_MIMO_FOLLOW_ADDR - for passing MAC address to monitor mode. that option will be used by VHT MU-MIMO air sniffer to follow a station according to it's MAC address using VHT MU-MIMO. Signed-off-by: Aviya Erenfeld <aviya.erenfeld@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-05 12:23:08 +00:00
[NL80211_ATTR_MU_MIMO_GROUP_DATA] = {
.len = VHT_MUMIMO_GROUPS_DATA_LEN
},
[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR] = { .len = ETH_ALEN },
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
[NL80211_ATTR_NAN_MASTER_PREF] = { .type = NLA_U8 },
cfg80211: fix NAN bands definition The nl80211_nan_dual_band_conf enumeration doesn't make much sense. The default value is assigned to a bit, which makes it weird if the default bit and other bits are set at the same time. To improve this, get rid of NL80211_NAN_BAND_DEFAULT and add a wiphy configuration to let the drivers define which bands are supported. This is exposed to the userspace, which then can make a decision on which band(s) to use. Additionally, rename all "dual_band" elements to "bands", to make things clearer. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-08 13:00:34 +00:00
[NL80211_ATTR_BANDS] = { .type = NLA_U32 },
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
[NL80211_ATTR_NAN_FUNC] = { .type = NLA_NESTED },
cfg80211: Add KEK/nonces for FILS association frames The new nl80211 attributes can be used to provide KEK and nonces to allow the driver to encrypt and decrypt FILS (Re)Association Request/Response frames in station mode. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:03 +00:00
[NL80211_ATTR_FILS_KEK] = { .type = NLA_BINARY,
.len = FILS_MAX_KEK_LEN },
[NL80211_ATTR_FILS_NONCES] = { .len = 2 * FILS_NONCE_LEN },
cfg80211: configure multicast to unicast for AP interfaces Add the ability to configure if an AP (and associated VLANs) will do multicast-to-unicast conversion for ARP, IPv4 and IPv6 frames (possibly within 802.1Q). If enabled, such frames are to be sent to each station separately, with the DA replaced by their own MAC address rather than the group address. Note that this may break certain expectations of the receiver, such as the ability to drop unicast IP packets received within multicast L2 frames, or the ability to not send ICMP destination unreachable messages for packets received in L2 multicast (which is required, but the receiver can't tell the difference if this new option is enabled.) This also doesn't implement the 802.11 DMS (directed multicast service). Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [fix disabling, add better documentation & commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-10 17:12:22 +00:00
[NL80211_ATTR_MULTICAST_TO_UNICAST_ENABLED] = { .type = NLA_FLAG, },
nl80211: Use different attrs for BSSID and random MAC addr in scan req NL80211_ATTR_MAC was used to set both the specific BSSID to be scanned and the random MAC address to be used when privacy is enabled. When both the features are enabled, both the BSSID and the local MAC address were getting same value causing Probe Request frames to go with unintended DA. Hence, this has been fixed by using a different NL80211_ATTR_BSSID attribute to set the specific BSSID (which was the more recent addition in cfg80211) for a scan. Backwards compatibility with old userspace software is maintained to some extent by allowing NL80211_ATTR_MAC to be used to set the specific BSSID when scanning without enabling random MAC address use. Scanning with random source MAC address was introduced by commit ad2b26abc157 ("cfg80211: allow drivers to support random MAC addresses for scan") and the issue was introduced with the addition of the second user for the same attribute in commit 818965d39177 ("cfg80211: Allow a scan request for a specific BSSID"). Fixes: 818965d39177 ("cfg80211: Allow a scan request for a specific BSSID") Signed-off-by: Vamsi Krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-02 21:59:08 +00:00
[NL80211_ATTR_BSSID] = { .len = ETH_ALEN },
cfg80211: Add support to sched scan to report better BSSs Enhance sched scan to support option of finding a better BSS while in connected state. Firmware scans the medium and reports when it finds a known BSS which has better RSSI than the current connected BSS. New attributes to specify the relative RSSI (compared to the current BSS) are added to the sched scan to implement this. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:20 +00:00
[NL80211_ATTR_SCHED_SCAN_RELATIVE_RSSI] = { .type = NLA_S8 },
[NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST] = {
.len = sizeof(struct nl80211_bss_select_rssi_adjust)
},
cfg80211: Specify the reason for connect timeout This enhances the connect timeout API to also carry the reason for the timeout. These reason codes for the connect time out are represented by enum nl80211_timeout_reason and are passed to user space through a new attribute NL80211_ATTR_TIMEOUT_REASON (u32). Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [keep gfp_t argument last] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:21 +00:00
[NL80211_ATTR_TIMEOUT_REASON] = { .type = NLA_U32 },
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
[NL80211_ATTR_FILS_ERP_USERNAME] = { .type = NLA_BINARY,
.len = FILS_ERP_MAX_USERNAME_LEN },
[NL80211_ATTR_FILS_ERP_REALM] = { .type = NLA_BINARY,
.len = FILS_ERP_MAX_REALM_LEN },
[NL80211_ATTR_FILS_ERP_NEXT_SEQ_NUM] = { .type = NLA_U16 },
[NL80211_ATTR_FILS_ERP_RRK] = { .type = NLA_BINARY,
.len = FILS_ERP_MAX_RRK_LEN },
[NL80211_ATTR_FILS_CACHE_ID] = { .len = 2 },
[NL80211_ATTR_PMK] = { .type = NLA_BINARY, .len = PMK_MAX_LEN },
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
[NL80211_ATTR_SCHED_SCAN_MULTI] = { .type = NLA_FLAG },
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
};
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
/* policy for the key attributes */
const: struct nla_policy Make remaining netlink policies as const. Fixup coding style where needed. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-18 08:14:31 +00:00
static const struct nla_policy nl80211_key_policy[NL80211_KEY_MAX + 1] = {
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
[NL80211_KEY_DATA] = { .type = NLA_BINARY, .len = WLAN_MAX_KEY_LEN },
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
[NL80211_KEY_IDX] = { .type = NLA_U8 },
[NL80211_KEY_CIPHER] = { .type = NLA_U32 },
nl80211: Increase maximum NL80211_ATTR_KEY_SEQ length to 16 WPI-SMS4 uses 16-octet PN field, so we need to allow longer key sequence values to be configured. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-02 21:36:31 +00:00
[NL80211_KEY_SEQ] = { .type = NLA_BINARY, .len = 16 },
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
[NL80211_KEY_DEFAULT] = { .type = NLA_FLAG },
[NL80211_KEY_DEFAULT_MGMT] = { .type = NLA_FLAG },
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
[NL80211_KEY_TYPE] = { .type = NLA_U32 },
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
[NL80211_KEY_DEFAULT_TYPES] = { .type = NLA_NESTED },
};
/* policy for the key default flags */
static const struct nla_policy
nl80211_key_default_policy[NUM_NL80211_KEY_DEFAULT_TYPES] = {
[NL80211_KEY_DEFAULT_TYPE_UNICAST] = { .type = NLA_FLAG },
[NL80211_KEY_DEFAULT_TYPE_MULTICAST] = { .type = NLA_FLAG },
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
};
nl80211: ifdef WoWLAN related policies To avoid unused variable warnings when CONFIG_PM isn't set, add the appropriate ifdef to the policies that are only used for WoWLAN, which can only be invoked when CONFIG_PM is set. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-17 06:04:07 +00:00
#ifdef CONFIG_PM
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
/* policy for WoWLAN attributes */
static const struct nla_policy
nl80211_wowlan_policy[NUM_NL80211_WOWLAN_TRIG] = {
[NL80211_WOWLAN_TRIG_ANY] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_DISCONNECT] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_MAGIC_PKT] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_PKT_PATTERN] = { .type = NLA_NESTED },
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
[NL80211_WOWLAN_TRIG_GTK_REKEY_FAILURE] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_EAP_IDENT_REQUEST] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_4WAY_HANDSHAKE] = { .type = NLA_FLAG },
[NL80211_WOWLAN_TRIG_RFKILL_RELEASE] = { .type = NLA_FLAG },
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
[NL80211_WOWLAN_TRIG_TCP_CONNECTION] = { .type = NLA_NESTED },
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
[NL80211_WOWLAN_TRIG_NET_DETECT] = { .type = NLA_NESTED },
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
};
static const struct nla_policy
nl80211_wowlan_tcp_policy[NUM_NL80211_WOWLAN_TCP] = {
[NL80211_WOWLAN_TCP_SRC_IPV4] = { .type = NLA_U32 },
[NL80211_WOWLAN_TCP_DST_IPV4] = { .type = NLA_U32 },
[NL80211_WOWLAN_TCP_DST_MAC] = { .len = ETH_ALEN },
[NL80211_WOWLAN_TCP_SRC_PORT] = { .type = NLA_U16 },
[NL80211_WOWLAN_TCP_DST_PORT] = { .type = NLA_U16 },
[NL80211_WOWLAN_TCP_DATA_PAYLOAD] = { .len = 1 },
[NL80211_WOWLAN_TCP_DATA_PAYLOAD_SEQ] = {
.len = sizeof(struct nl80211_wowlan_tcp_data_seq)
},
[NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN] = {
.len = sizeof(struct nl80211_wowlan_tcp_data_token)
},
[NL80211_WOWLAN_TCP_DATA_INTERVAL] = { .type = NLA_U32 },
[NL80211_WOWLAN_TCP_WAKE_PAYLOAD] = { .len = 1 },
[NL80211_WOWLAN_TCP_WAKE_MASK] = { .len = 1 },
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
};
nl80211: ifdef WoWLAN related policies To avoid unused variable warnings when CONFIG_PM isn't set, add the appropriate ifdef to the policies that are only used for WoWLAN, which can only be invoked when CONFIG_PM is set. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-17 06:04:07 +00:00
#endif /* CONFIG_PM */
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
/* policy for coalesce rule attributes */
static const struct nla_policy
nl80211_coalesce_policy[NUM_NL80211_ATTR_COALESCE_RULE] = {
[NL80211_ATTR_COALESCE_RULE_DELAY] = { .type = NLA_U32 },
[NL80211_ATTR_COALESCE_RULE_CONDITION] = { .type = NLA_U32 },
[NL80211_ATTR_COALESCE_RULE_PKT_PATTERN] = { .type = NLA_NESTED },
};
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
/* policy for GTK rekey offload attributes */
static const struct nla_policy
nl80211_rekey_policy[NUM_NL80211_REKEY_DATA] = {
[NL80211_REKEY_DATA_KEK] = { .len = NL80211_KEK_LEN },
[NL80211_REKEY_DATA_KCK] = { .len = NL80211_KCK_LEN },
[NL80211_REKEY_DATA_REPLAY_CTR] = { .len = NL80211_REPLAY_CTR_LEN },
};
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
static const struct nla_policy
nl80211_match_policy[NL80211_SCHED_SCAN_MATCH_ATTR_MAX + 1] = {
nl80211: fix sched scan match attribute name It should be NL80211_SCHED_SCAN_MATCH_ATTR_SSID as documented, not NL80211_ATTR_SCHED_SCAN_MATCH_SSID. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 09:17:11 +00:00
[NL80211_SCHED_SCAN_MATCH_ATTR_SSID] = { .type = NLA_BINARY,
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
.len = IEEE80211_MAX_SSID_LEN },
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
[NL80211_SCHED_SCAN_MATCH_ATTR_BSSID] = { .len = ETH_ALEN },
nl80211: specify RSSI threshold in scheduled scan Support configuring an RSSI threshold in dBm (s32) when requesting scheduled scan, below which a BSS won't be reported by the cfg80211 driver. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-21 18:09:54 +00:00
[NL80211_SCHED_SCAN_MATCH_ATTR_RSSI] = { .type = NLA_U32 },
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
};
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
static const struct nla_policy
nl80211_plan_policy[NL80211_SCHED_SCAN_PLAN_MAX + 1] = {
[NL80211_SCHED_SCAN_PLAN_INTERVAL] = { .type = NLA_U32 },
[NL80211_SCHED_SCAN_PLAN_ITERATIONS] = { .type = NLA_U32 },
};
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
static const struct nla_policy
nl80211_bss_select_policy[NL80211_BSS_SELECT_ATTR_MAX + 1] = {
[NL80211_BSS_SELECT_ATTR_RSSI] = { .type = NLA_FLAG },
[NL80211_BSS_SELECT_ATTR_BAND_PREF] = { .type = NLA_U32 },
[NL80211_BSS_SELECT_ATTR_RSSI_ADJUST] = {
.len = sizeof(struct nl80211_bss_select_rssi_adjust)
},
};
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
/* policy for NAN function attributes */
static const struct nla_policy
nl80211_nan_func_policy[NL80211_NAN_FUNC_ATTR_MAX + 1] = {
[NL80211_NAN_FUNC_TYPE] = { .type = NLA_U8 },
cfg80211: Check if NAN service ID is of expected size nla policy checks for only maximum length of the attribute data when the attribute type is NLA_BINARY. If userspace sends less data than specified, cfg80211 may access illegal memory. When type is NLA_UNSPEC, nla policy check ensures that userspace sends minimum specified length number of bytes. Remove type assignment to NLA_BINARY from nla_policy of NL80211_NAN_FUNC_SERVICE_ID to make these NLA_UNSPEC and to make sure minimum NL80211_NAN_FUNC_SERVICE_ID_LEN bytes are received from userspace with NL80211_NAN_FUNC_SERVICE_ID. Fixes: a442b761b24 ("cfg80211: add add_nan_func / del_nan_func") Cc: stable@vger.kernel.org Signed-off-by: Srinivas Dasari <dasaris@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-07-06 22:43:40 +00:00
[NL80211_NAN_FUNC_SERVICE_ID] = {
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
.len = NL80211_NAN_FUNC_SERVICE_ID_LEN },
[NL80211_NAN_FUNC_PUBLISH_TYPE] = { .type = NLA_U8 },
[NL80211_NAN_FUNC_PUBLISH_BCAST] = { .type = NLA_FLAG },
[NL80211_NAN_FUNC_SUBSCRIBE_ACTIVE] = { .type = NLA_FLAG },
[NL80211_NAN_FUNC_FOLLOW_UP_ID] = { .type = NLA_U8 },
[NL80211_NAN_FUNC_FOLLOW_UP_REQ_ID] = { .type = NLA_U8 },
[NL80211_NAN_FUNC_FOLLOW_UP_DEST] = { .len = ETH_ALEN },
[NL80211_NAN_FUNC_CLOSE_RANGE] = { .type = NLA_FLAG },
[NL80211_NAN_FUNC_TTL] = { .type = NLA_U32 },
[NL80211_NAN_FUNC_SERVICE_INFO] = { .type = NLA_BINARY,
.len = NL80211_NAN_FUNC_SERVICE_SPEC_INFO_MAX_LEN },
[NL80211_NAN_FUNC_SRF] = { .type = NLA_NESTED },
[NL80211_NAN_FUNC_RX_MATCH_FILTER] = { .type = NLA_NESTED },
[NL80211_NAN_FUNC_TX_MATCH_FILTER] = { .type = NLA_NESTED },
[NL80211_NAN_FUNC_INSTANCE_ID] = { .type = NLA_U8 },
[NL80211_NAN_FUNC_TERM_REASON] = { .type = NLA_U8 },
};
/* policy for Service Response Filter attributes */
static const struct nla_policy
nl80211_nan_srf_policy[NL80211_NAN_SRF_ATTR_MAX + 1] = {
[NL80211_NAN_SRF_INCLUDE] = { .type = NLA_FLAG },
[NL80211_NAN_SRF_BF] = { .type = NLA_BINARY,
.len = NL80211_NAN_FUNC_SRF_MAX_LEN },
[NL80211_NAN_SRF_BF_IDX] = { .type = NLA_U8 },
[NL80211_NAN_SRF_MAC_ADDRS] = { .type = NLA_NESTED },
};
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
static int nl80211_prepare_wdev_dump(struct sk_buff *skb,
struct netlink_callback *cb,
struct cfg80211_registered_device **rdev,
struct wireless_dev **wdev)
cfg80211: introduce nl80211_get_ifidx() ... which get's rid of three indentical cut-n-paste sections. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 10:30:02 +00:00
{
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
int err;
cfg80211: introduce nl80211_get_ifidx() ... which get's rid of three indentical cut-n-paste sections. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 10:30:02 +00:00
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (!cb->args[0]) {
err = nlmsg_parse(cb->nlh, GENL_HDRLEN + nl80211_fam.hdrsize,
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
genl_family_attrbuf(&nl80211_fam),
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_fam.maxattr, nl80211_policy, NULL);
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
return err;
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
*wdev = __cfg80211_wdev_from_attrs(
sock_net(skb->sk),
genl_family_attrbuf(&nl80211_fam));
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (IS_ERR(*wdev))
return PTR_ERR(*wdev);
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
*rdev = wiphy_to_rdev((*wdev)->wiphy);
nl80211: fix another nl80211_fam.attrbuf race This is similar to the race Linus had reported, but in this case it's an older bug: nl80211_prepare_wdev_dump() uses the wiphy index in cb->args[0] as it is and thus parses the message over and over again instead of just once because 0 is the first valid wiphy index. Similar code in nl80211_testmode_dump() correctly offsets the wiphy_index by 1, do that here as well. Cc: stable@vger.kernel.org Reported-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-30 20:34:28 +00:00
/* 0 is the first index - add 1 to parse only once */
cb->args[0] = (*rdev)->wiphy_idx + 1;
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
cb->args[1] = (*wdev)->identifier;
} else {
nl80211: fix another nl80211_fam.attrbuf race This is similar to the race Linus had reported, but in this case it's an older bug: nl80211_prepare_wdev_dump() uses the wiphy index in cb->args[0] as it is and thus parses the message over and over again instead of just once because 0 is the first valid wiphy index. Similar code in nl80211_testmode_dump() correctly offsets the wiphy_index by 1, do that here as well. Cc: stable@vger.kernel.org Reported-by: Ben Hutchings <ben@decadent.org.uk> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-30 20:34:28 +00:00
/* subtract the 1 again here */
struct wiphy *wiphy = wiphy_idx_to_wiphy(cb->args[0] - 1);
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
struct wireless_dev *tmp;
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!wiphy)
return -ENODEV;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
*rdev = wiphy_to_rdev(wiphy);
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
*wdev = NULL;
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(tmp, &(*rdev)->wiphy.wdev_list, list) {
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (tmp->identifier == cb->args[1]) {
*wdev = tmp;
break;
}
}
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!*wdev)
return -ENODEV;
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
}
return 0;
}
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
/* IE validation */
static bool is_valid_ie_attr(const struct nlattr *attr)
{
const u8 *pos;
int len;
if (!attr)
return true;
pos = nla_data(attr);
len = nla_len(attr);
while (len) {
u8 elemlen;
if (len < 2)
return false;
len -= 2;
elemlen = pos[1];
if (elemlen > len)
return false;
len -= elemlen;
pos += 2 + elemlen;
}
return true;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* message building helper */
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
static inline void *nl80211hdr_put(struct sk_buff *skb, u32 portid, u32 seq,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
int flags, u8 cmd)
{
/* since there is no private header just add the generic one */
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
return genlmsg_put(skb, portid, seq, &nl80211_fam, flags, cmd);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
static int nl80211_msg_put_channel(struct sk_buff *msg,
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
struct ieee80211_channel *chan,
bool large)
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
{
cfg80211: Add attributes describing prohibited channel bandwidth Since there are frequency bands (e.g. 5.9GHz) allowing channels with only 10 or 5 MHz bandwidth, this patch adds attributes that allow keeping track about this information. When channel attributes are reported to user-space, make sure to not break old tools, i.e. if the 'split wiphy dump' is enabled, report the extra attributes (if present) describing the bandwidth restrictions. If the 'split wiphy dump' is not enabled, completely omit those channels that have flags set to either IEEE80211_CHAN_NO_10MHZ or IEEE80211_CHAN_NO_20MHZ. Add the check for new bandwidth restriction flags in cfg80211_chandef_usable() to comply with the restrictions. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-15 12:37:55 +00:00
/* Some channels must be completely excluded from the
* list to protect old user-space tools from breaking
*/
if (!large && chan->flags &
(IEEE80211_CHAN_NO_10MHZ | IEEE80211_CHAN_NO_20MHZ))
return 0;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_FREQ,
chan->center_freq))
goto nla_put_failure;
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((chan->flags & IEEE80211_CHAN_DISABLED) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_DISABLED))
goto nla_put_failure;
cfg80211: consolidate passive-scan and no-ibss flags These two flags are used for the same purpose, just combine them into a no-ir flag to annotate no initiating radiation is allowed. Old userspace sending either flag will have it treated as the no-ir flag. To be considerate to older userspace we also send both the no-ir flag and the old no-ibss flags. Newer userspace will have to be aware of older kernels. Update all places in the tree using these flags with the following semantic patch: @@ @@ -NL80211_RRF_PASSIVE_SCAN +NL80211_RRF_NO_IR @@ @@ -NL80211_RRF_NO_IBSS +NL80211_RRF_NO_IR @@ @@ -IEEE80211_CHAN_PASSIVE_SCAN +IEEE80211_CHAN_NO_IR @@ @@ -IEEE80211_CHAN_NO_IBSS +IEEE80211_CHAN_NO_IR @@ @@ -NL80211_RRF_NO_IR | NL80211_RRF_NO_IR +NL80211_RRF_NO_IR @@ @@ -IEEE80211_CHAN_NO_IR | IEEE80211_CHAN_NO_IR +IEEE80211_CHAN_NO_IR @@ @@ -(NL80211_RRF_NO_IR) +NL80211_RRF_NO_IR @@ @@ -(IEEE80211_CHAN_NO_IR) +IEEE80211_CHAN_NO_IR Along with some hand-optimisations in documentation, to remove duplicates and to fix some indentation. Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> [do all the driver updates in one go] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-21 17:22:25 +00:00
if (chan->flags & IEEE80211_CHAN_NO_IR) {
if (nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_IR))
goto nla_put_failure;
if (nla_put_flag(msg, __NL80211_FREQUENCY_ATTR_NO_IBSS))
goto nla_put_failure;
}
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
if (chan->flags & IEEE80211_CHAN_RADAR) {
if (nla_put_flag(msg, NL80211_FREQUENCY_ATTR_RADAR))
goto nla_put_failure;
if (large) {
u32 time;
time = elapsed_jiffies_msecs(chan->dfs_state_entered);
if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_DFS_STATE,
chan->dfs_state))
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_DFS_TIME,
time))
goto nla_put_failure;
cfg80211: regulatory: allow getting DFS CAC time from userspace Introduce DFS CAC time as a regd param, configured per REG_RULE and set per channel in cfg80211. DFS CAC time is close connected with regulatory database configuration. Instead of using hardcoded values, get DFS CAC time form regulatory database. Pass DFS CAC time to user mode (mainly for iw reg get, iw list, iw info). Allow setting DFS CAC time via CRDA. Add support for internal regulatory database. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> [rewrap commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-21 18:46:12 +00:00
if (nla_put_u32(msg,
NL80211_FREQUENCY_ATTR_DFS_CAC_TIME,
chan->dfs_cac_ms))
goto nla_put_failure;
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
}
}
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
if (large) {
if ((chan->flags & IEEE80211_CHAN_NO_HT40MINUS) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_HT40_MINUS))
goto nla_put_failure;
if ((chan->flags & IEEE80211_CHAN_NO_HT40PLUS) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_HT40_PLUS))
goto nla_put_failure;
if ((chan->flags & IEEE80211_CHAN_NO_80MHZ) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_80MHZ))
goto nla_put_failure;
if ((chan->flags & IEEE80211_CHAN_NO_160MHZ) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_160MHZ))
goto nla_put_failure;
cfg80211: Add indoor only and GO concurrent channel attributes The FCC are clarifying some soft configuration requirements, which among other include the following: 1. Indoor operation, where a device can use channels requiring indoor operation, subject to that it can guarantee indoor operation, i.e., the device is connected to AC Power or the device is under the control of a local master that is acting as an AP and is connected to AC Power. 2. Concurrent GO operation, where devices may instantiate a P2P GO while they are under the guidance of an authorized master. For example, on a channel on which a BSS is connected to an authorized master, i.e., with DFS and radar detection capability in the UNII band. See https://apps.fcc.gov/eas/comments/GetPublishedDocument.html?id=327&tn=528122 Add support for advertising Indoor-only and GO-Concurrent channel properties. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-23 07:12:59 +00:00
if ((chan->flags & IEEE80211_CHAN_INDOOR_ONLY) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_INDOOR_ONLY))
goto nla_put_failure;
cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA The GO_CONCURRENT regulatory definition can be extended to station interfaces requesting to IR as part of TDLS off-channel operations. Rename the GO_CONCURRENT flag to IR_CONCURRENT and allow the added use-case. Change internal users of GO_CONCURRENT to use the new definition. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-05-06 13:28:31 +00:00
if ((chan->flags & IEEE80211_CHAN_IR_CONCURRENT) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_IR_CONCURRENT))
cfg80211: Add indoor only and GO concurrent channel attributes The FCC are clarifying some soft configuration requirements, which among other include the following: 1. Indoor operation, where a device can use channels requiring indoor operation, subject to that it can guarantee indoor operation, i.e., the device is connected to AC Power or the device is under the control of a local master that is acting as an AP and is connected to AC Power. 2. Concurrent GO operation, where devices may instantiate a P2P GO while they are under the guidance of an authorized master. For example, on a channel on which a BSS is connected to an authorized master, i.e., with DFS and radar detection capability in the UNII band. See https://apps.fcc.gov/eas/comments/GetPublishedDocument.html?id=327&tn=528122 Add support for advertising Indoor-only and GO-Concurrent channel properties. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-23 07:12:59 +00:00
goto nla_put_failure;
cfg80211: Add attributes describing prohibited channel bandwidth Since there are frequency bands (e.g. 5.9GHz) allowing channels with only 10 or 5 MHz bandwidth, this patch adds attributes that allow keeping track about this information. When channel attributes are reported to user-space, make sure to not break old tools, i.e. if the 'split wiphy dump' is enabled, report the extra attributes (if present) describing the bandwidth restrictions. If the 'split wiphy dump' is not enabled, completely omit those channels that have flags set to either IEEE80211_CHAN_NO_10MHZ or IEEE80211_CHAN_NO_20MHZ. Add the check for new bandwidth restriction flags in cfg80211_chandef_usable() to comply with the restrictions. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-15 12:37:55 +00:00
if ((chan->flags & IEEE80211_CHAN_NO_20MHZ) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_20MHZ))
goto nla_put_failure;
if ((chan->flags & IEEE80211_CHAN_NO_10MHZ) &&
nla_put_flag(msg, NL80211_FREQUENCY_ATTR_NO_10MHZ))
goto nla_put_failure;
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_FREQUENCY_ATTR_MAX_TX_POWER,
DBM_TO_MBM(chan->max_power)))
goto nla_put_failure;
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
return 0;
nla_put_failure:
return -ENOBUFS;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* netlink command implementations */
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
struct key_parse {
struct key_params p;
int idx;
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
int type;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
bool def, defmgmt;
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
bool def_uni, def_multi;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
};
static int nl80211_parse_key_new(struct nlattr *key, struct key_parse *k)
{
struct nlattr *tb[NL80211_KEY_MAX + 1];
int err = nla_parse_nested(tb, NL80211_KEY_MAX, key,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_key_policy, NULL);
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (err)
return err;
k->def = !!tb[NL80211_KEY_DEFAULT];
k->defmgmt = !!tb[NL80211_KEY_DEFAULT_MGMT];
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (k->def) {
k->def_uni = true;
k->def_multi = true;
}
if (k->defmgmt)
k->def_multi = true;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (tb[NL80211_KEY_IDX])
k->idx = nla_get_u8(tb[NL80211_KEY_IDX]);
if (tb[NL80211_KEY_DATA]) {
k->p.key = nla_data(tb[NL80211_KEY_DATA]);
k->p.key_len = nla_len(tb[NL80211_KEY_DATA]);
}
if (tb[NL80211_KEY_SEQ]) {
k->p.seq = nla_data(tb[NL80211_KEY_SEQ]);
k->p.seq_len = nla_len(tb[NL80211_KEY_SEQ]);
}
if (tb[NL80211_KEY_CIPHER])
k->p.cipher = nla_get_u32(tb[NL80211_KEY_CIPHER]);
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (tb[NL80211_KEY_TYPE]) {
k->type = nla_get_u32(tb[NL80211_KEY_TYPE]);
if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES)
return -EINVAL;
}
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (tb[NL80211_KEY_DEFAULT_TYPES]) {
struct nlattr *kdt[NUM_NL80211_KEY_DEFAULT_TYPES];
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211: fix a few -Wshadow warnings It seems that -Wshadow is no longer default in sparse runs, but let's fix the warnings anyway. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-20 12:52:37 +00:00
err = nla_parse_nested(kdt, NUM_NL80211_KEY_DEFAULT_TYPES - 1,
tb[NL80211_KEY_DEFAULT_TYPES],
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_key_default_policy, NULL);
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (err)
return err;
k->def_uni = kdt[NL80211_KEY_DEFAULT_TYPE_UNICAST];
k->def_multi = kdt[NL80211_KEY_DEFAULT_TYPE_MULTICAST];
}
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
return 0;
}
static int nl80211_parse_key_old(struct genl_info *info, struct key_parse *k)
{
if (info->attrs[NL80211_ATTR_KEY_DATA]) {
k->p.key = nla_data(info->attrs[NL80211_ATTR_KEY_DATA]);
k->p.key_len = nla_len(info->attrs[NL80211_ATTR_KEY_DATA]);
}
if (info->attrs[NL80211_ATTR_KEY_SEQ]) {
k->p.seq = nla_data(info->attrs[NL80211_ATTR_KEY_SEQ]);
k->p.seq_len = nla_len(info->attrs[NL80211_ATTR_KEY_SEQ]);
}
if (info->attrs[NL80211_ATTR_KEY_IDX])
k->idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
if (info->attrs[NL80211_ATTR_KEY_CIPHER])
k->p.cipher = nla_get_u32(info->attrs[NL80211_ATTR_KEY_CIPHER]);
k->def = !!info->attrs[NL80211_ATTR_KEY_DEFAULT];
k->defmgmt = !!info->attrs[NL80211_ATTR_KEY_DEFAULT_MGMT];
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (k->def) {
k->def_uni = true;
k->def_multi = true;
}
if (k->defmgmt)
k->def_multi = true;
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (info->attrs[NL80211_ATTR_KEY_TYPE]) {
k->type = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]);
if (k->type < 0 || k->type >= NUM_NL80211_KEYTYPES)
return -EINVAL;
}
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (info->attrs[NL80211_ATTR_KEY_DEFAULT_TYPES]) {
struct nlattr *kdt[NUM_NL80211_KEY_DEFAULT_TYPES];
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
int err = nla_parse_nested(kdt,
NUM_NL80211_KEY_DEFAULT_TYPES - 1,
info->attrs[NL80211_ATTR_KEY_DEFAULT_TYPES],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_key_default_policy,
info->extack);
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (err)
return err;
k->def_uni = kdt[NL80211_KEY_DEFAULT_TYPE_UNICAST];
k->def_multi = kdt[NL80211_KEY_DEFAULT_TYPE_MULTICAST];
}
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
return 0;
}
static int nl80211_parse_key(struct genl_info *info, struct key_parse *k)
{
int err;
memset(k, 0, sizeof(*k));
k->idx = -1;
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
k->type = -1;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (info->attrs[NL80211_ATTR_KEY])
err = nl80211_parse_key_new(info->attrs[NL80211_ATTR_KEY], k);
else
err = nl80211_parse_key_old(info, k);
if (err)
return err;
if (k->def && k->defmgmt)
return -EINVAL;
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (k->defmgmt) {
if (k->def_uni || !k->def_multi)
return -EINVAL;
}
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (k->idx != -1) {
if (k->defmgmt) {
if (k->idx < 4 || k->idx > 5)
return -EINVAL;
} else if (k->def) {
if (k->idx < 0 || k->idx > 3)
return -EINVAL;
} else {
if (k->idx < 0 || k->idx > 5)
return -EINVAL;
}
}
return 0;
}
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
static struct cfg80211_cached_keys *
nl80211_parse_connkeys(struct cfg80211_registered_device *rdev,
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
struct nlattr *keys, bool *no_ht)
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
{
struct key_parse parse;
struct nlattr *key;
struct cfg80211_cached_keys *result;
int rem, err, def = 0;
cfg80211: allow connect keys only with default (TX) key There's no point in allowing connect keys when one of them isn't also configured as the TX key, it would just confuse drivers and probably cause them to pick something for TX. Disallow this confusing and erroneous configuration. As wpa_supplicant will always send NL80211_ATTR_KEYS, even when there are no keys inside, allow that and treat it as though the attribute isn't present at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 15:08:23 +00:00
bool have_key = false;
nla_for_each_nested(key, keys, rem) {
have_key = true;
break;
}
if (!have_key)
return NULL;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
result = kzalloc(sizeof(*result), GFP_KERNEL);
if (!result)
return ERR_PTR(-ENOMEM);
result->def = -1;
nla_for_each_nested(key, keys, rem) {
memset(&parse, 0, sizeof(parse));
parse.idx = -1;
err = nl80211_parse_key_new(key, &parse);
if (err)
goto error;
err = -EINVAL;
if (!parse.p.key)
goto error;
nl80211: fix connect keys range check Only key index 0-3 should be accepted, 4/5 are for IGTKs and cannot be used as connect keys. Fix the range checking to not allow such erroneous configurations. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 13:51:03 +00:00
if (parse.idx < 0 || parse.idx > 3)
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
goto error;
if (parse.def) {
if (def)
goto error;
def = 1;
result->def = parse.idx;
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (!parse.def_uni || !parse.def_multi)
goto error;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
} else if (parse.defmgmt)
goto error;
err = cfg80211_validate_key_settings(rdev, &parse.p,
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
parse.idx, false, NULL);
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (err)
goto error;
nl80211: only allow WEP keys during connect command This was already documented that way in nl80211.h, but the parsing code still accepted other key types. Change it to really only accept WEP keys as documented. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 14:10:02 +00:00
if (parse.p.cipher != WLAN_CIPHER_SUITE_WEP40 &&
parse.p.cipher != WLAN_CIPHER_SUITE_WEP104) {
err = -EINVAL;
goto error;
}
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
result->params[parse.idx].cipher = parse.p.cipher;
result->params[parse.idx].key_len = parse.p.key_len;
result->params[parse.idx].key = result->data[parse.idx];
memcpy(result->data[parse.idx], parse.p.key, parse.p.key_len);
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
nl80211: only allow WEP keys during connect command This was already documented that way in nl80211.h, but the parsing code still accepted other key types. Change it to really only accept WEP keys as documented. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 14:10:02 +00:00
/* must be WEP key if we got here */
if (no_ht)
*no_ht = true;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
}
cfg80211: allow connect keys only with default (TX) key There's no point in allowing connect keys when one of them isn't also configured as the TX key, it would just confuse drivers and probably cause them to pick something for TX. Disallow this confusing and erroneous configuration. As wpa_supplicant will always send NL80211_ATTR_KEYS, even when there are no keys inside, allow that and treat it as though the attribute isn't present at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 15:08:23 +00:00
if (result->def < 0) {
err = -EINVAL;
goto error;
}
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
return result;
error:
kfree(result);
return ERR_PTR(err);
}
static int nl80211_key_allowed(struct wireless_dev *wdev)
{
ASSERT_WDEV_LOCK(wdev);
switch (wdev->iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
case NL80211_IFTYPE_P2P_GO:
nl80211: allow installing keys for a meshif Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:12 +00:00
case NL80211_IFTYPE_MESH_POINT:
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
break;
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_STATION:
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
case NL80211_IFTYPE_P2P_CLIENT:
cfg80211: separate internal SME implementation The current internal SME implementation in cfg80211 is very mixed up with the MLME handling, which has been causing issues for a long time. There are three things that the implementation has to provide: * a basic SME implementation for nl80211's connect() call (for drivers implementing auth/assoc, which is really just mac80211) and wireless extensions * MLME events for the userspace SME * SME events (connected, disconnected etc.) for all different SME implementation possibilities (driver, cfg80211 and userspace) To achieve these goals it isn't necessary to track the software SME's connection status outside of it's state (which is the part that caused many issues.) Instead, track it only in the SME data (wdev->conn) and in the general case only track whether the wdev is connected or not (via wdev->current_bss.) Also separate the internal implementation to not have callbacks from the SME events, but rather call it from the API functions that the driver (or rather mac80211) calls. This separates the code better. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:45 +00:00
if (!wdev->current_bss)
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
return -ENOLINK;
break;
cfg80211: avoid using default in interface type switch Most code avoids having a default case in interface type switch statements already, to make it easier to find places that need to be extended. Change the code in the __cfg80211_leave() and nl80211_key_allowed() functions to not have a default case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-31 13:16:12 +00:00
case NL80211_IFTYPE_UNSPECIFIED:
cfg80211: 802.11p OCB mode handling This patch adds new iface type (NL80211_IFTYPE_OCB) representing the OCB (Outside the Context of a BSS) mode. When establishing a connection to the network a cfg80211_join_ocb function is called (particular nl80211_command is added as well). A mandatory parameters during the ocb_join operation are 'center frequency' and 'channel width (5/10 MHz)'. Changes done in mac80211 are minimal possible required to avoid many warnings (warning: enumeration value 'NL80211_IFTYPE_OCB' not handled in switch) during compilation. Full functionality (where needed) is added in the following patch. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-03 09:33:18 +00:00
case NL80211_IFTYPE_OCB:
cfg80211: avoid using default in interface type switch Most code avoids having a default case in interface type switch statements already, to make it easier to find places that need to be extended. Change the code in the __cfg80211_leave() and nl80211_key_allowed() functions to not have a default case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-31 13:16:12 +00:00
case NL80211_IFTYPE_MONITOR:
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
case NL80211_IFTYPE_NAN:
cfg80211: avoid using default in interface type switch Most code avoids having a default case in interface type switch statements already, to make it easier to find places that need to be extended. Change the code in the __cfg80211_leave() and nl80211_key_allowed() functions to not have a default case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-31 13:16:12 +00:00
case NL80211_IFTYPE_P2P_DEVICE:
case NL80211_IFTYPE_WDS:
case NUM_NL80211_IFTYPES:
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
return -EINVAL;
}
return 0;
}
cfg80211: Clean up connect params and channel fetching Addition of the frequency hints showed up couple of places in cfg80211 where pointers could be marked const and a shared function could be used to fetch a valid channel. Signed-off-by: Jouni Malinen <j@w1.fi> [fix mwifiex] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:44 +00:00
static struct ieee80211_channel *nl80211_get_valid_chan(struct wiphy *wiphy,
struct nlattr *tb)
{
struct ieee80211_channel *chan;
if (tb == NULL)
return NULL;
chan = ieee80211_get_channel(wiphy, nla_get_u32(tb));
if (!chan || chan->flags & IEEE80211_CHAN_DISABLED)
return NULL;
return chan;
}
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
static int nl80211_put_iftypes(struct sk_buff *msg, u32 attr, u16 ifmodes)
{
struct nlattr *nl_modes = nla_nest_start(msg, attr);
int i;
if (!nl_modes)
goto nla_put_failure;
i = 0;
while (ifmodes) {
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((ifmodes & 1) && nla_put_flag(msg, i))
goto nla_put_failure;
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
ifmodes >>= 1;
i++;
}
nla_nest_end(msg, nl_modes);
return 0;
nla_put_failure:
return -ENOBUFS;
}
static int nl80211_put_iface_combinations(struct wiphy *wiphy,
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
struct sk_buff *msg,
bool large)
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
{
struct nlattr *nl_combis;
int i, j;
nl_combis = nla_nest_start(msg,
NL80211_ATTR_INTERFACE_COMBINATIONS);
if (!nl_combis)
goto nla_put_failure;
for (i = 0; i < wiphy->n_iface_combinations; i++) {
const struct ieee80211_iface_combination *c;
struct nlattr *nl_combi, *nl_limits;
c = &wiphy->iface_combinations[i];
nl_combi = nla_nest_start(msg, i + 1);
if (!nl_combi)
goto nla_put_failure;
nl_limits = nla_nest_start(msg, NL80211_IFACE_COMB_LIMITS);
if (!nl_limits)
goto nla_put_failure;
for (j = 0; j < c->n_limits; j++) {
struct nlattr *nl_limit;
nl_limit = nla_nest_start(msg, j + 1);
if (!nl_limit)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_IFACE_LIMIT_MAX,
c->limits[j].max))
goto nla_put_failure;
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
if (nl80211_put_iftypes(msg, NL80211_IFACE_LIMIT_TYPES,
c->limits[j].types))
goto nla_put_failure;
nla_nest_end(msg, nl_limit);
}
nla_nest_end(msg, nl_limits);
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (c->beacon_int_infra_match &&
nla_put_flag(msg, NL80211_IFACE_COMB_STA_AP_BI_MATCH))
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_IFACE_COMB_NUM_CHANNELS,
c->num_different_channels) ||
nla_put_u32(msg, NL80211_IFACE_COMB_MAXNUM,
c->max_interfaces))
goto nla_put_failure;
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
if (large &&
cfg80211: allow restricting supported dfs regions At the moment, the ath9k/ath10k DFS module only supports detecting ETSI radar patterns. Add a bitmap in the interface combinations, indicating which DFS regions are supported by the detector. If unset, support for all regions is assumed. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-05 09:48:40 +00:00
(nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_WIDTHS,
c->radar_detect_widths) ||
nla_put_u32(msg, NL80211_IFACE_COMB_RADAR_DETECT_REGIONS,
c->radar_detect_regions)))
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
goto nla_put_failure;
cfg80211: support virtual interfaces with different beacon intervals This commit provides a mechanism for the host drivers to advertise the support for different beacon intervals among the respective interface combinations in a group, through NL80211_IFACE_COMB_BI_MIN_GCD (u32). This value will be compared against GCD of all beaconing interfaces of matching combinations. If the driver doesn't advertise this value, the old behaviour where all beacon intervals must be identical is retained. If it is specified, then any beacon interval for an interface in the interface combination as well as the GCD of all active beacon intervals in the combination must be greater or equal to this value. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [change commit message, some variable names, small other things] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-12 12:56:51 +00:00
if (c->beacon_int_min_gcd &&
nla_put_u32(msg, NL80211_IFACE_COMB_BI_MIN_GCD,
c->beacon_int_min_gcd))
goto nla_put_failure;
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
nla_nest_end(msg, nl_combi);
}
nla_nest_end(msg, nl_combis);
return 0;
nla_put_failure:
return -ENOBUFS;
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
#ifdef CONFIG_PM
nl80211: conditionally add back TCP WoWLAN information Add back the previously removed TCP WoWLAN information, but only if userspace is prepared to deal with large wiphy capability data dumps. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-20 00:02:38 +00:00
static int nl80211_send_wowlan_tcp_caps(struct cfg80211_registered_device *rdev,
struct sk_buff *msg)
{
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
const struct wiphy_wowlan_tcp_support *tcp = rdev->wiphy.wowlan->tcp;
nl80211: conditionally add back TCP WoWLAN information Add back the previously removed TCP WoWLAN information, but only if userspace is prepared to deal with large wiphy capability data dumps. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-20 00:02:38 +00:00
struct nlattr *nl_tcp;
if (!tcp)
return 0;
nl_tcp = nla_nest_start(msg, NL80211_WOWLAN_TRIG_TCP_CONNECTION);
if (!nl_tcp)
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD,
tcp->data_payload_max))
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD,
tcp->data_payload_max))
return -ENOBUFS;
if (tcp->seq && nla_put_flag(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD_SEQ))
return -ENOBUFS;
if (tcp->tok && nla_put(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN,
sizeof(*tcp->tok), tcp->tok))
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_WOWLAN_TCP_DATA_INTERVAL,
tcp->data_interval_max))
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_WOWLAN_TCP_WAKE_PAYLOAD,
tcp->wake_payload_max))
return -ENOBUFS;
nla_nest_end(msg, nl_tcp);
return 0;
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
static int nl80211_send_wowlan(struct sk_buff *msg,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev,
nl80211: conditionally add back TCP WoWLAN information Add back the previously removed TCP WoWLAN information, but only if userspace is prepared to deal with large wiphy capability data dumps. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-20 00:02:38 +00:00
bool large)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
struct nlattr *nl_wowlan;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (!rdev->wiphy.wowlan)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
return 0;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nl_wowlan = nla_nest_start(msg, NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED);
if (!nl_wowlan)
return -ENOBUFS;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_ANY) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_ANY)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_DISCONNECT) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_DISCONNECT)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_MAGIC_PKT) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_MAGIC_PKT)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_SUPPORTS_GTK_REKEY) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_GTK_REKEY_SUPPORTED)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_GTK_REKEY_FAILURE) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_GTK_REKEY_FAILURE)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_EAP_IDENTITY_REQ) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_EAP_IDENT_REQUEST)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_4WAY_HANDSHAKE) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_4WAY_HANDSHAKE)) ||
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_RFKILL_RELEASE) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_RFKILL_RELEASE)))
return -ENOBUFS;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.wowlan->n_patterns) {
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
struct nl80211_pattern_support pat = {
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
.max_patterns = rdev->wiphy.wowlan->n_patterns,
.min_pattern_len = rdev->wiphy.wowlan->pattern_min_len,
.max_pattern_len = rdev->wiphy.wowlan->pattern_max_len,
.max_pkt_offset = rdev->wiphy.wowlan->max_pkt_offset,
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
};
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (nla_put(msg, NL80211_WOWLAN_TRIG_PKT_PATTERN,
sizeof(pat), &pat))
return -ENOBUFS;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
if ((rdev->wiphy.wowlan->flags & WIPHY_WOWLAN_NET_DETECT) &&
nla_put_u32(msg, NL80211_WOWLAN_TRIG_NET_DETECT,
rdev->wiphy.wowlan->max_nd_match_sets))
return -ENOBUFS;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (large && nl80211_send_wowlan_tcp_caps(rdev, msg))
nl80211: conditionally add back TCP WoWLAN information Add back the previously removed TCP WoWLAN information, but only if userspace is prepared to deal with large wiphy capability data dumps. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-20 00:02:38 +00:00
return -ENOBUFS;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_wowlan);
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
return 0;
}
#endif
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
static int nl80211_send_coalesce(struct sk_buff *msg,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev)
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
{
struct nl80211_coalesce_rule_support rule;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (!rdev->wiphy.coalesce)
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
return 0;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rule.max_rules = rdev->wiphy.coalesce->n_rules;
rule.max_delay = rdev->wiphy.coalesce->max_delay;
rule.pat.max_patterns = rdev->wiphy.coalesce->n_patterns;
rule.pat.min_pattern_len = rdev->wiphy.coalesce->pattern_min_len;
rule.pat.max_pattern_len = rdev->wiphy.coalesce->pattern_max_len;
rule.pat.max_pkt_offset = rdev->wiphy.coalesce->max_pkt_offset;
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
if (nla_put(msg, NL80211_ATTR_COALESCE_RULE, sizeof(rule), &rule))
return -ENOBUFS;
return 0;
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
static int nl80211_send_band_rateinfo(struct sk_buff *msg,
struct ieee80211_supported_band *sband)
{
struct nlattr *nl_rates, *nl_rate;
struct ieee80211_rate *rate;
int i;
nl80211: Add probe response offload attribute Notify user-space about probe-response offloading support in the driver. A wiphy flag is used to indicate support and a bitmap of protocols determines which protocols are supported. Signed-off-by: Guy Eilam <guy@wizery.com> Signed-off-by: Arik Nemtsov <arik@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-10 09:28:55 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* add HT info */
if (sband->ht_cap.ht_supported &&
(nla_put(msg, NL80211_BAND_ATTR_HT_MCS_SET,
sizeof(sband->ht_cap.mcs),
&sband->ht_cap.mcs) ||
nla_put_u16(msg, NL80211_BAND_ATTR_HT_CAPA,
sband->ht_cap.cap) ||
nla_put_u8(msg, NL80211_BAND_ATTR_HT_AMPDU_FACTOR,
sband->ht_cap.ampdu_factor) ||
nla_put_u8(msg, NL80211_BAND_ATTR_HT_AMPDU_DENSITY,
sband->ht_cap.ampdu_density)))
return -ENOBUFS;
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* add VHT info */
if (sband->vht_cap.vht_supported &&
(nla_put(msg, NL80211_BAND_ATTR_VHT_MCS_SET,
sizeof(sband->vht_cap.vht_mcs),
&sband->vht_cap.vht_mcs) ||
nla_put_u32(msg, NL80211_BAND_ATTR_VHT_CAPA,
sband->vht_cap.cap)))
return -ENOBUFS;
cfg80211: keep track of supported interface modes It is obviously good for userspace to know up front which interface modes a given piece of hardware might support (even if adding such an interface might fail later because of concurrency issues), so let's make cfg80211 aware of that. For good measure, disallow adding interfaces in all other modes so drivers don't forget to announce support for one mode when they add it. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Stephen Blackheath <tramp.enshrine.stephen@blacksapphire.com> Signed-off-by: Ivo van Doorn <IvDoorn@gmail.com> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-29 23:26:43 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* add bitrates */
nl_rates = nla_nest_start(msg, NL80211_BAND_ATTR_RATES);
if (!nl_rates)
return -ENOBUFS;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
for (i = 0; i < sband->n_bitrates; i++) {
nl_rate = nla_nest_start(msg, i);
if (!nl_rate)
return -ENOBUFS;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
rate = &sband->bitrates[i];
if (nla_put_u32(msg, NL80211_BITRATE_ATTR_RATE,
rate->bitrate))
return -ENOBUFS;
if ((rate->flags & IEEE80211_RATE_SHORT_PREAMBLE) &&
nla_put_flag(msg,
NL80211_BITRATE_ATTR_2GHZ_SHORTPREAMBLE))
return -ENOBUFS;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_rate);
}
nl80211: export HT capabilities This exports the local HT capabilities in nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-09 10:20:13 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_rates);
cfg80211: allow advertising VHT capabilities Allow drivers to advertise their VHT capabilities and export them to userspace via nl80211. Signed-off-by: Mahesh Palivela <maheshp@posedge.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-22 07:27:46 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
return 0;
}
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
static int
nl80211_send_mgmt_stypes(struct sk_buff *msg,
const struct ieee80211_txrx_stypes *mgmt_stypes)
{
u16 stypes;
struct nlattr *nl_ftypes, *nl_ifs;
enum nl80211_iftype ift;
int i;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (!mgmt_stypes)
return 0;
cfg80211: add netlink channel put helper This adds a netlink channel put helper, nl80211_msg_put_channel(), which we will also make use of later for the beacon hints events. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:08 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nl_ifs = nla_nest_start(msg, NL80211_ATTR_TX_FRAME_TYPES);
if (!nl_ifs)
return -ENOBUFS;
nl80211: Report max TX power in NL80211_BAND_ATTR_FREQS This is useful information to provide for userspace (e.g., hostapd needs this to generate Country IE). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-21 17:01:30 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) {
nl_ftypes = nla_nest_start(msg, ift);
if (!nl_ftypes)
return -ENOBUFS;
i = 0;
stypes = mgmt_stypes[ift].tx;
while (stypes) {
if ((stypes & 1) &&
nla_put_u16(msg, NL80211_ATTR_FRAME_TYPE,
(i << 4) | IEEE80211_FTYPE_MGMT))
return -ENOBUFS;
stypes >>= 1;
i++;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_ftypes);
}
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_ifs);
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nl_ifs = nla_nest_start(msg, NL80211_ATTR_RX_FRAME_TYPES);
if (!nl_ifs)
return -ENOBUFS;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
for (ift = 0; ift < NUM_NL80211_IFTYPES; ift++) {
nl_ftypes = nla_nest_start(msg, ift);
if (!nl_ftypes)
return -ENOBUFS;
i = 0;
stypes = mgmt_stypes[ift].rx;
while (stypes) {
if ((stypes & 1) &&
nla_put_u16(msg, NL80211_ATTR_FRAME_TYPE,
(i << 4) | IEEE80211_FTYPE_MGMT))
return -ENOBUFS;
stypes >>= 1;
i++;
}
nla_nest_end(msg, nl_ftypes);
}
nla_nest_end(msg, nl_ifs);
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
return 0;
}
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: move unsplit command advertising to a separate function When we split the wiphy dump because it got too large, I added a comment and asked that all new command advertising be done only for userspace clients capable of receiving split data, in order to not break older ones (which can't use the new commands anyway) This mostly worked, and we haven't added many new commands, but I occasionally get patches that modify the wrong place. Make this easier to detect and understand by splitting out the old commands to a separate function that makes it more clear it should never be modified again. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 09:42:04 +00:00
#define CMD(op, n) \
do { \
if (rdev->ops->op) { \
i++; \
if (nla_put_u32(msg, i, NL80211_CMD_ ## n)) \
goto nla_put_failure; \
} \
} while (0)
static int nl80211_add_commands_unsplit(struct cfg80211_registered_device *rdev,
struct sk_buff *msg)
{
int i = 0;
/*
* do *NOT* add anything into this function, new things need to be
* advertised only to new versions of userspace that can deal with
* the split (and they can't possibly care about new features...
*/
CMD(add_virtual_intf, NEW_INTERFACE);
CMD(change_virtual_intf, SET_INTERFACE);
CMD(add_key, NEW_KEY);
CMD(start_ap, START_AP);
CMD(add_station, NEW_STATION);
CMD(add_mpath, NEW_MPATH);
CMD(update_mesh_config, SET_MESH_CONFIG);
CMD(change_bss, SET_BSS);
CMD(auth, AUTHENTICATE);
CMD(assoc, ASSOCIATE);
CMD(deauth, DEAUTHENTICATE);
CMD(disassoc, DISASSOCIATE);
CMD(join_ibss, JOIN_IBSS);
CMD(join_mesh, JOIN_MESH);
CMD(set_pmksa, SET_PMKSA);
CMD(del_pmksa, DEL_PMKSA);
CMD(flush_pmksa, FLUSH_PMKSA);
if (rdev->wiphy.flags & WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL)
CMD(remain_on_channel, REMAIN_ON_CHANNEL);
CMD(set_bitrate_mask, SET_TX_BITRATE_MASK);
CMD(mgmt_tx, FRAME);
CMD(mgmt_tx_cancel_wait, FRAME_WAIT_CANCEL);
if (rdev->wiphy.flags & WIPHY_FLAG_NETNS_OK) {
i++;
if (nla_put_u32(msg, i, NL80211_CMD_SET_WIPHY_NETNS))
goto nla_put_failure;
}
if (rdev->ops->set_monitor_channel || rdev->ops->start_ap ||
rdev->ops->join_mesh) {
i++;
if (nla_put_u32(msg, i, NL80211_CMD_SET_CHANNEL))
goto nla_put_failure;
}
CMD(set_wds_peer, SET_WDS_PEER);
if (rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_TDLS) {
CMD(tdls_mgmt, TDLS_MGMT);
CMD(tdls_oper, TDLS_OPER);
}
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
if (rdev->wiphy.max_sched_scan_reqs)
nl80211: move unsplit command advertising to a separate function When we split the wiphy dump because it got too large, I added a comment and asked that all new command advertising be done only for userspace clients capable of receiving split data, in order to not break older ones (which can't use the new commands anyway) This mostly worked, and we haven't added many new commands, but I occasionally get patches that modify the wrong place. Make this easier to detect and understand by splitting out the old commands to a separate function that makes it more clear it should never be modified again. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 09:42:04 +00:00
CMD(sched_scan_start, START_SCHED_SCAN);
CMD(probe_client, PROBE_CLIENT);
CMD(set_noack_map, SET_NOACK_MAP);
if (rdev->wiphy.flags & WIPHY_FLAG_REPORTS_OBSS) {
i++;
if (nla_put_u32(msg, i, NL80211_CMD_REGISTER_BEACONS))
goto nla_put_failure;
}
CMD(start_p2p_device, START_P2P_DEVICE);
CMD(set_mcast_rate, SET_MCAST_RATE);
#ifdef CONFIG_NL80211_TESTMODE
CMD(testmode_cmd, TESTMODE);
#endif
if (rdev->ops->connect || rdev->ops->auth) {
i++;
if (nla_put_u32(msg, i, NL80211_CMD_CONNECT))
goto nla_put_failure;
}
if (rdev->ops->disconnect || rdev->ops->deauth) {
i++;
if (nla_put_u32(msg, i, NL80211_CMD_DISCONNECT))
goto nla_put_failure;
}
return i;
nla_put_failure:
return -ENOBUFS;
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
struct nl80211_dump_wiphy_state {
s64 filter_wiphy;
long start;
cfg80211: Advertise extended capabilities per interface type to userspace The driver extended capabilities may differ for different interface types which the userspace needs to know (for example the fine timing measurement initiator and responder bits might differ for a station and AP). Add a new nl80211 attribute to provide extended capabilities per interface type to userspace. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Reviewed-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-16 05:11:04 +00:00
long split_start, band_start, chan_start, capa_start;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
bool split;
};
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
static int nl80211_send_wiphy(struct cfg80211_registered_device *rdev,
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
enum nl80211_commands cmd,
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
struct sk_buff *msg, u32 portid, u32 seq,
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
int flags, struct nl80211_dump_wiphy_state *state)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
{
void *hdr;
struct nlattr *nl_bands, *nl_band;
struct nlattr *nl_freqs, *nl_freq;
struct nlattr *nl_cmds;
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
enum nl80211_band band;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
struct ieee80211_channel *chan;
int i;
const struct ieee80211_txrx_stypes *mgmt_stypes =
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.mgmt_stypes;
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
u32 features;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
hdr = nl80211hdr_put(msg, portid, seq, flags, cmd);
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (!hdr)
return -ENOBUFS;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (WARN_ON(!state))
return -EINVAL;
nl80211: export hardware bitrate/channel capabilities This makes nl80211 export the hardware bitrate/channel capabilities as registered in a wiphy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-24 18:38:39 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_string(msg, NL80211_ATTR_WIPHY_NAME,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
wiphy_name(&rdev->wiphy)) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_GENERATION,
cfg80211_rdev_list_generation))
nl80211: export supported commands This makes nl80211 export the supported commands (command groups) per wiphy so userspace has an idea what it can do -- this will be required reading for userspace when we introduce auth/assoc /or/ connect for older hardware that cannot separate auth and assoc. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-14 08:34:01 +00:00
goto nla_put_failure;
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
if (cmd != NL80211_CMD_NEW_WIPHY)
goto finish;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
switch (state->split_start) {
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
case 0:
if (nla_put_u8(msg, NL80211_ATTR_WIPHY_RETRY_SHORT,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.retry_short) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u8(msg, NL80211_ATTR_WIPHY_RETRY_LONG,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.retry_long) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_FRAG_THRESHOLD,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.frag_threshold) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_RTS_THRESHOLD,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.rts_threshold) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u8(msg, NL80211_ATTR_WIPHY_COVERAGE_CLASS,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.coverage_class) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u8(msg, NL80211_ATTR_MAX_NUM_SCAN_SSIDS,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_scan_ssids) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u8(msg, NL80211_ATTR_MAX_NUM_SCHED_SCAN_SSIDS,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_sched_scan_ssids) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u16(msg, NL80211_ATTR_MAX_SCAN_IE_LEN,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_scan_ie_len) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u16(msg, NL80211_ATTR_MAX_SCHED_SCAN_IE_LEN,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_sched_scan_ie_len) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u8(msg, NL80211_ATTR_MAX_MATCH_SETS,
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
rdev->wiphy.max_match_sets) ||
nla_put_u32(msg, NL80211_ATTR_MAX_NUM_SCHED_SCAN_PLANS,
rdev->wiphy.max_sched_scan_plans) ||
nla_put_u32(msg, NL80211_ATTR_MAX_SCAN_PLAN_INTERVAL,
rdev->wiphy.max_sched_scan_plan_interval) ||
nla_put_u32(msg, NL80211_ATTR_MAX_SCAN_PLAN_ITERATIONS,
rdev->wiphy.max_sched_scan_plan_iterations))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_SUPPORT_IBSS_RSN))
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_MESH_AUTH) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_SUPPORT_MESH_AUTH))
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_AP_UAPSD) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_SUPPORT_AP_UAPSD))
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_FW_ROAM) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_ROAM_SUPPORT))
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_TDLS) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_TDLS_SUPPORT))
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_TDLS_EXTERNAL_SETUP) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_TDLS_EXTERNAL_SETUP))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 1:
if (nla_put(msg, NL80211_ATTR_CIPHER_SUITES,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
sizeof(u32) * rdev->wiphy.n_cipher_suites,
rdev->wiphy.cipher_suites))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: add testmode to the list of supported commands User space might want to test if driver supports testmode. Adding testmode to the list of supported commands makes this easier. I omitted testmode_dump() in purpose. I assume all drivers implementing testmode_dump() will also implement testmode_cmd(). Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-17 17:06:10 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (nla_put_u8(msg, NL80211_ATTR_MAX_NUM_PMKIDS,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_num_pmkids))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_CONTROL_PORT_ETHERTYPE))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY_ANTENNA_AVAIL_TX,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.available_antennas_tx) ||
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_ANTENNA_AVAIL_RX,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.available_antennas_rx))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_AP_PROBE_RESP_OFFLOAD) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_PROBE_RESP_OFFLOAD,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.probe_resp_offload))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: export supported commands This makes nl80211 export the supported commands (command groups) per wiphy so userspace has an idea what it can do -- this will be required reading for userspace when we introduce auth/assoc /or/ connect for older hardware that cannot separate auth and assoc. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-14 08:34:01 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.available_antennas_tx ||
rdev->wiphy.available_antennas_rx) &&
rdev->ops->get_antenna) {
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
u32 tx_ant = 0, rx_ant = 0;
int res;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
res = rdev_get_antenna(rdev, &tx_ant, &rx_ant);
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (!res) {
if (nla_put_u32(msg,
NL80211_ATTR_WIPHY_ANTENNA_TX,
tx_ant) ||
nla_put_u32(msg,
NL80211_ATTR_WIPHY_ANTENNA_RX,
rx_ant))
goto nla_put_failure;
}
}
nl80211: advertise maximum remain-on-channel duration With the upcoming hardware offload implementation, some devices will have a different maximum duration for the remain-on-channel command. Advertise the maximum duration in mac80211, and make mac80211 set it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-14 16:54:28 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 2:
if (nl80211_put_iftypes(msg, NL80211_ATTR_SUPPORTED_IFTYPES,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.interface_modes))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 3:
nl_bands = nla_nest_start(msg, NL80211_ATTR_WIPHY_BANDS);
if (!nl_bands)
goto nla_put_failure;
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
for (band = state->band_start;
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
band < NUM_NL80211_BANDS; band++) {
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
struct ieee80211_supported_band *sband;
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
sband = rdev->wiphy.bands[band];
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (!sband)
continue;
nl_band = nla_nest_start(msg, band);
if (!nl_band)
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
goto nla_put_failure;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
switch (state->chan_start) {
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
case 0:
if (nl80211_send_band_rateinfo(msg, sband))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->chan_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
default:
/* add frequencies */
nl_freqs = nla_nest_start(
msg, NL80211_BAND_ATTR_FREQS);
if (!nl_freqs)
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
for (i = state->chan_start - 1;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
i < sband->n_channels;
i++) {
nl_freq = nla_nest_start(msg, i);
if (!nl_freq)
goto nla_put_failure;
chan = &sband->channels[i];
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (nl80211_msg_put_channel(
msg, chan,
state->split))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nla_nest_end(msg, nl_freq);
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
}
if (i < sband->n_channels)
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->chan_start = i + 2;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
else
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->chan_start = 0;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_freqs);
}
nla_nest_end(msg, nl_band);
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->split) {
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* start again here */
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->chan_start)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
band--;
break;
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
}
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_bands);
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
if (band < NUM_NL80211_BANDS)
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->band_start = band + 1;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
else
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->band_start = 0;
nl80211: fix missing nesting commit 95a6ccbb46c70cff376684c752831c014c87029d Author: Johannes Berg <johannes.berg@intel.com> Date: Thu Aug 12 15:38:38 2010 +0200 cfg80211/mac80211: extensible frame processing introduced a netlink bug that caused parsing errors in userspace because it forgot to close a nesting, which would advertise a nesting length of zero to userspace, which then completely threw off parsing and led to Illegal nla->nla_type == 0 being printed by libnl. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-24 10:15:53 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* if bands & channels are done, continue outside */
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->band_start == 0 && state->chan_start == 0)
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 4:
nl_cmds = nla_nest_start(msg, NL80211_ATTR_SUPPORTED_COMMANDS);
if (!nl_cmds)
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
goto nla_put_failure;
nl80211: move unsplit command advertising to a separate function When we split the wiphy dump because it got too large, I added a comment and asked that all new command advertising be done only for userspace clients capable of receiving split data, in order to not break older ones (which can't use the new commands anyway) This mostly worked, and we haven't added many new commands, but I occasionally get patches that modify the wrong place. Make this easier to detect and understand by splitting out the old commands to a separate function that makes it more clear it should never be modified again. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 09:42:04 +00:00
i = nl80211_add_commands_unsplit(rdev, msg);
if (i < 0)
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->split) {
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
CMD(crit_proto_start, CRIT_PROTOCOL_START);
CMD(crit_proto_stop, CRIT_PROTOCOL_STOP);
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.flags & WIPHY_FLAG_HAS_CHANNEL_SWITCH)
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
CMD(channel_switch, CHANNEL_SWITCH);
nl80211: move set_qos_map command into split state The non-split wiphy state shouldn't be increased in size so move the new set_qos_map command into the split if statement. Cc: stable@vger.kernel.org (3.14+) Fixes: fa9ffc745610 ("cfg80211: Add support for QoS mapping") Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-10 12:06:25 +00:00
CMD(set_qos_map, SET_QOS_MAP);
cfg80211: make WMM TSPEC support flag an nl80211 feature flag During the review of the corresponding wpa_supplicant patches we noticed that the only way for it to detect that this functionality is supported currently is to check for the command support. This can be misleading though, as the command was also designed to, in the future, support pure 802.11 TSPECs. Expose the WMM-TSPEC feature flag to nl80211 so later we can also expose an 802.11-TSPEC feature flag (if needed) to differentiate the two cases. Note: this change isn't needed in 3.18 as there's no driver there yet that supports the functionality at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-22 07:25:06 +00:00
if (rdev->wiphy.features &
NL80211_FEATURE_SUPPORTS_WMM_ADMISSION)
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
CMD(add_tx_ts, ADD_TX_TS);
cfg80211: configure multicast to unicast for AP interfaces Add the ability to configure if an AP (and associated VLANs) will do multicast-to-unicast conversion for ARP, IPv4 and IPv6 frames (possibly within 802.1Q). If enabled, such frames are to be sent to each station separately, with the DA replaced by their own MAC address rather than the group address. Note that this may break certain expectations of the receiver, such as the ability to drop unicast IP packets received within multicast L2 frames, or the ability to not send ICMP destination unreachable messages for packets received in L2 multicast (which is required, but the receiver can't tell the difference if this new option is enabled.) This also doesn't implement the 802.11 DMS (directed multicast service). Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [fix disabling, add better documentation & commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-10 17:12:22 +00:00
CMD(set_multicast_to_unicast, SET_MULTICAST_TO_UNICAST);
cfg80211: Add support to update connection parameters Add functionality to update the connection parameters when in connected state, so that driver/firmware uses the updated parameters for subsequent roaming. This is for drivers that support internal BSS selection and roaming. The new command does not change the current association state, i.e., it can be used to update IE contents for future (re)associations without causing an immediate disassociation or reassociation with the current BSS. This commit implements the required functionality for updating IEs for (Re)Association Request frame only. Other parameters can be added in future when required. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-27 13:51:11 +00:00
CMD(update_connect_params, UPDATE_CONNECT_PARAMS);
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
}
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
#undef CMD
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_nest_end(msg, nl_cmds);
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 5:
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->ops->remain_on_channel &&
(rdev->wiphy.flags & WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg,
NL80211_ATTR_MAX_REMAIN_ON_CHANNEL_DURATION,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_remain_on_channel_duration))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_OFFCHAN_TX) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_flag(msg, NL80211_ATTR_OFFCHANNEL_TX_OK))
goto nla_put_failure;
if (nl80211_send_mgmt_stypes(msg, mgmt_stypes))
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 6:
#ifdef CONFIG_PM
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (nl80211_send_wowlan(msg, rdev, state->split))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
#else
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
cfg80211: don't allow WoWLAN support without CONFIG_PM When CONFIG_PM is disabled, no device can possibly support WoWLAN since it can't go to sleep to start with. Due to this, mac80211 had even rejected the hardware registration. By making all the code and data for WoWLAN depend on CONFIG_PM we can promote this runtime error to a compile-time error. Add #ifdef around all WoWLAN code to remove it in systems that don't need it as they never suspend. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Acked-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 07:23:48 +00:00
#endif
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
case 7:
if (nl80211_put_iftypes(msg, NL80211_ATTR_SOFTWARE_IFTYPES,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.software_iftypes))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (nl80211_put_iface_combinations(&rdev->wiphy, msg,
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
if (state->split)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 8:
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_HAVE_AP_SME) &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_DEVICE_AP_SME,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.ap_sme_capa))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
cfg80211: advertise possible interface combinations Add the ability to advertise interface combinations in nl80211. This allows the driver to indicate what the combinations are that it supports. "Combinations" of just a single interface are implicit, as previously. Note that cfg80211 will enforce that the restrictions are met, but not for all drivers yet (once all drivers are updated, we can remove the flag and enforce for all). When no combinations are actually supported, an empty list will be exported so that userspace can know if the kernel exported this info or not (although it isn't clear to me what tools using the info should do if the kernel didn't export it). Since some interface types are purely virtual/software and don't fit the restrictions, those are exposed in a new list of pure SW types, not subject to restrictions. This mainly exists to handle AP-VLAN and monitor interfaces in mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-13 08:58:57 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
features = rdev->wiphy.features;
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
/*
* We can only add the per-channel limit information if the
* dump is split, otherwise it makes it too big. Therefore
* only advertise it in that case.
*/
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->split)
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
features |= NL80211_FEATURE_ADVERTISE_CHAN_LIMITS;
if (nla_put_u32(msg, NL80211_ATTR_FEATURE_FLAGS, features))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: advertise device AP SME Add the ability to advertise that the device contains the AP SME and what features it can support. There are currently no features in the bitmap -- probe response offload will be advertised by a few patches Arik is working on now (who took over from Guy Eilam) and a device with AP SME will typically implement and require response offload. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-07 11:39:33 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.ht_capa_mod_mask &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put(msg, NL80211_ATTR_HT_CAPABILITY_MASK,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
sizeof(*rdev->wiphy.ht_capa_mod_mask),
rdev->wiphy.ht_capa_mod_mask))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
nl80211: advertise socket TX status capability The new wifi socket TX capability should be supported by wifi drivers, let them advertise whether they do or not. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-06 13:13:33 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.flags & WIPHY_FLAG_HAVE_AP_SME &&
rdev->wiphy.max_acl_mac_addrs &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
nla_put_u32(msg, NL80211_ATTR_MAC_ACL_MAX,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_acl_mac_addrs))
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
goto nla_put_failure;
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/*
* Any information below this point is only available to
* applications that can deal with it being split. This
* helps ensure that newly added capabilities don't break
* older tools by overrunning their buffers.
*
* We still increment split_start so that in the split
* case we'll continue with more data in the next round,
* but break unconditionally so unsplit data stops here.
*/
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start++;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
case 9:
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.extended_capabilities &&
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
(nla_put(msg, NL80211_ATTR_EXT_CAPA,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.extended_capabilities_len,
rdev->wiphy.extended_capabilities) ||
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
nla_put(msg, NL80211_ATTR_EXT_CAPA_MASK,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.extended_capabilities_len,
rdev->wiphy.extended_capabilities_mask)))
nl80211: re-add channel width and extended capa advertising Add back the channel width and extended capability data to wiphy information if split information is supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 14:39:45 +00:00
goto nla_put_failure;
cfg80211: advertise extended capabilities to userspace In many cases, userspace may need to know which of the 802.11 extended capabilities ("Extended Capabilities element") are implemented in the driver or device, to include them e.g. in beacons, assoc request/response or other frames. Add a new nl80211 attribute to hold the extended capabilities bitmap for this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-11 13:20:05 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.vht_capa_mod_mask &&
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
nla_put(msg, NL80211_ATTR_VHT_CAPABILITY_MASK,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
sizeof(*rdev->wiphy.vht_capa_mod_mask),
rdev->wiphy.vht_capa_mod_mask))
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
goto nla_put_failure;
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
state->split_start++;
break;
case 10:
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (nl80211_send_coalesce(msg, rdev))
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if ((rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_5_10_MHZ) &&
cfg80211: fix reporting 5/10 MHz support to user space nla_put_flag needs a real nl80211 attribute id, not a wiphy flag bit. While at it, split 5 and 10 MHz capability flags in case we ever need to support hardware that can only do one of the two. Also move the flag settings to the split-only information so we don't increase the space needed for old userspace. Signed-off-by: Felix Fietkau <nbd@openwrt.org> [change location of flag setting] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-09 13:57:54 +00:00
(nla_put_flag(msg, NL80211_ATTR_SUPPORT_5_MHZ) ||
nla_put_flag(msg, NL80211_ATTR_SUPPORT_10_MHZ)))
goto nla_put_failure;
cfg80211: Advertise maximum associated STAs in AP mode This allows drivers to advertise the maximum number of associated stations they support in AP mode (including P2P GO). User space applications can use this for cleaner way of handling the limit (e.g., hostapd rejecting IEEE 802.11 authentication without manual configuration of the limit) or to figure out what type of use cases can be executed with multiple devices before trying and failing. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:08 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.max_ap_assoc_sta &&
cfg80211: Advertise maximum associated STAs in AP mode This allows drivers to advertise the maximum number of associated stations they support in AP mode (including P2P GO). User space applications can use this for cleaner way of handling the limit (e.g., hostapd rejecting IEEE 802.11 authentication without manual configuration of the limit) or to figure out what type of use cases can be executed with multiple devices before trying and failing. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:08 +00:00
nla_put_u32(msg, NL80211_ATTR_MAX_AP_ASSOC_STA,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev->wiphy.max_ap_assoc_sta))
cfg80211: Advertise maximum associated STAs in AP mode This allows drivers to advertise the maximum number of associated stations they support in AP mode (including P2P GO). User space applications can use this for cleaner way of handling the limit (e.g., hostapd rejecting IEEE 802.11 authentication without manual configuration of the limit) or to figure out what type of use cases can be executed with multiple devices before trying and failing. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:08 +00:00
goto nla_put_failure;
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
state->split_start++;
break;
case 11:
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.n_vendor_commands) {
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
const struct nl80211_vendor_cmd_info *info;
struct nlattr *nested;
nested = nla_nest_start(msg, NL80211_ATTR_VENDOR_DATA);
if (!nested)
goto nla_put_failure;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
for (i = 0; i < rdev->wiphy.n_vendor_commands; i++) {
info = &rdev->wiphy.vendor_commands[i].info;
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
if (nla_put(msg, i + 1, sizeof(*info), info))
goto nla_put_failure;
}
nla_nest_end(msg, nested);
}
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (rdev->wiphy.n_vendor_events) {
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
const struct nl80211_vendor_cmd_info *info;
struct nlattr *nested;
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
nested = nla_nest_start(msg,
NL80211_ATTR_VENDOR_EVENTS);
if (!nested)
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
goto nla_put_failure;
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
for (i = 0; i < rdev->wiphy.n_vendor_events; i++) {
info = &rdev->wiphy.vendor_events[i];
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
if (nla_put(msg, i + 1, sizeof(*info), info))
goto nla_put_failure;
}
nla_nest_end(msg, nested);
}
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
state->split_start++;
break;
case 12:
if (rdev->wiphy.flags & WIPHY_FLAG_HAS_CHANNEL_SWITCH &&
nla_put_u8(msg, NL80211_ATTR_MAX_CSA_COUNTERS,
rdev->wiphy.max_num_csa_counters))
goto nla_put_failure;
cfg80211: fix reporting 5/10 MHz support to user space nla_put_flag needs a real nl80211 attribute id, not a wiphy flag bit. While at it, split 5 and 10 MHz capability flags in case we ever need to support hardware that can only do one of the two. Also move the flag settings to the split-only information so we don't increase the space needed for old userspace. Signed-off-by: Felix Fietkau <nbd@openwrt.org> [change location of flag setting] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-09 13:57:54 +00:00
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
if (rdev->wiphy.regulatory_flags & REGULATORY_WIPHY_SELF_MANAGED &&
nla_put_flag(msg, NL80211_ATTR_WIPHY_SELF_MANAGED_REG))
goto nla_put_failure;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
if (rdev->wiphy.max_sched_scan_reqs &&
nla_put_u32(msg, NL80211_ATTR_SCHED_SCAN_MAX_REQS,
rdev->wiphy.max_sched_scan_reqs))
goto nla_put_failure;
cfg80211: add extensible feature flag attribute With the wiphy::features flag being used up this patch adds a new field wiphy::ext_features. Considering extensibility this new field is declared as a byte array. This extensible flag is exposed to user-space by NL80211_ATTR_EXT_FEATURES. Cc: Avinash Patil <patila@marvell.com> Signed-off-by: Gautam (Gautam Kumar) Shukla <gautams@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-23 15:55:19 +00:00
if (nla_put(msg, NL80211_ATTR_EXT_FEATURES,
sizeof(rdev->wiphy.ext_features),
rdev->wiphy.ext_features))
goto nla_put_failure;
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
if (rdev->wiphy.bss_select_support) {
struct nlattr *nested;
u32 bss_select_support = rdev->wiphy.bss_select_support;
nested = nla_nest_start(msg, NL80211_ATTR_BSS_SELECT);
if (!nested)
goto nla_put_failure;
i = 0;
while (bss_select_support) {
if ((bss_select_support & 1) &&
nla_put_flag(msg, i))
goto nla_put_failure;
i++;
bss_select_support >>= 1;
}
nla_nest_end(msg, nested);
}
cfg80211: Advertise extended capabilities per interface type to userspace The driver extended capabilities may differ for different interface types which the userspace needs to know (for example the fine timing measurement initiator and responder bits might differ for a station and AP). Add a new nl80211 attribute to provide extended capabilities per interface type to userspace. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Reviewed-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-16 05:11:04 +00:00
state->split_start++;
break;
case 13:
if (rdev->wiphy.num_iftype_ext_capab &&
rdev->wiphy.iftype_ext_capab) {
struct nlattr *nested_ext_capab, *nested;
nested = nla_nest_start(msg,
NL80211_ATTR_IFTYPE_EXT_CAPA);
if (!nested)
goto nla_put_failure;
for (i = state->capa_start;
i < rdev->wiphy.num_iftype_ext_capab; i++) {
const struct wiphy_iftype_ext_capab *capab;
capab = &rdev->wiphy.iftype_ext_capab[i];
nested_ext_capab = nla_nest_start(msg, i);
if (!nested_ext_capab ||
nla_put_u32(msg, NL80211_ATTR_IFTYPE,
capab->iftype) ||
nla_put(msg, NL80211_ATTR_EXT_CAPA,
capab->extended_capabilities_len,
capab->extended_capabilities) ||
nla_put(msg, NL80211_ATTR_EXT_CAPA_MASK,
capab->extended_capabilities_len,
capab->extended_capabilities_mask))
goto nla_put_failure;
nla_nest_end(msg, nested_ext_capab);
if (state->split)
break;
}
nla_nest_end(msg, nested);
if (i < rdev->wiphy.num_iftype_ext_capab) {
state->capa_start = i + 1;
break;
}
}
cfg80211: fix NAN bands definition The nl80211_nan_dual_band_conf enumeration doesn't make much sense. The default value is assigned to a bit, which makes it weird if the default bit and other bits are set at the same time. To improve this, get rid of NL80211_NAN_BAND_DEFAULT and add a wiphy configuration to let the drivers define which bands are supported. This is exposed to the userspace, which then can make a decision on which band(s) to use. Additionally, rename all "dual_band" elements to "bands", to make things clearer. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-08 13:00:34 +00:00
if (nla_put_u32(msg, NL80211_ATTR_BANDS,
rdev->wiphy.nan_supported_bands))
goto nla_put_failure;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
/* done */
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->split_start = 0;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
}
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
finish:
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nla_put_failure:
netlink: Improve returned error codes Make nlmsg_trim(), nlmsg_cancel(), genlmsg_cancel(), and nla_nest_cancel() void functions. Return -EMSGSIZE instead of -1 if the provided message buffer is not big enough. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-03 23:36:54 +00:00
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
static int nl80211_dump_wiphy_parse(struct sk_buff *skb,
struct netlink_callback *cb,
struct nl80211_dump_wiphy_state *state)
{
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
struct nlattr **tb = genl_family_attrbuf(&nl80211_fam);
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
int ret = nlmsg_parse(cb->nlh, GENL_HDRLEN + nl80211_fam.hdrsize, tb,
nl80211_fam.maxattr, nl80211_policy, NULL);
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
/* ignore parse errors for backward compatibility */
if (ret)
return 0;
state->split = tb[NL80211_ATTR_SPLIT_WIPHY_DUMP];
if (tb[NL80211_ATTR_WIPHY])
state->filter_wiphy = nla_get_u32(tb[NL80211_ATTR_WIPHY]);
if (tb[NL80211_ATTR_WDEV])
state->filter_wiphy = nla_get_u64(tb[NL80211_ATTR_WDEV]) >> 32;
if (tb[NL80211_ATTR_IFINDEX]) {
struct net_device *netdev;
struct cfg80211_registered_device *rdev;
int ifidx = nla_get_u32(tb[NL80211_ATTR_IFINDEX]);
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
netdev = __dev_get_by_index(sock_net(skb->sk), ifidx);
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (!netdev)
return -ENODEV;
if (netdev->ieee80211_ptr) {
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
rdev = wiphy_to_rdev(
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
netdev->ieee80211_ptr->wiphy);
state->filter_wiphy = rdev->wiphy_idx;
}
}
return 0;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
static int nl80211_dump_wiphy(struct sk_buff *skb, struct netlink_callback *cb)
{
nl80211: increase wiphy dump size dynamically Given a device with many channels capabilities the wiphy information can still overflow even though its size in 3.9 was reduced to 3.8 levels. For new userspace and kernel 3.10 we're going to implement a new "split dump" protocol that can use multiple messages per wiphy. For now though, add a workaround to be able to send more information to userspace. Since generic netlink doesn't have a way to set the minimum dump size globally, and we wouldn't really want to set it globally anyway, increase the size only when needed, as described in the comments. As userspace might not be prepared for large buffers, we can only use 4k. Also increase the size for the get_wiphy command. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-01 13:03:49 +00:00
int idx = 0, ret;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
struct nl80211_dump_wiphy_state *state = (void *)cb->args[0];
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev;
nl80211: fix attrbuf access race by allocating a separate one Since my commit 3713b4e364 ("nl80211: allow splitting wiphy information in dumps"), nl80211_dump_wiphy() uses the global nl80211_fam.attrbuf for parsing the incoming data. This wouldn't be a problem if it only did so on the first dump iteration which is locked against other commands in generic netlink, but due to space constraints in cb->args (the needed state doesn't fit) I decided to always parse the original message. That's racy though since nl80211_fam.attrbuf could be used by some other parsing in generic netlink concurrently. For now, fix this by allocating a separate parse buffer (it's a bit too big for the stack, currently 1448 bytes on 64-bit). For -next, I'll change the code to parse into the global buffer in the first round only and then allocate a smaller buffer to keep the data in cb->args. Reported-by: Linus Torvalds <torvalds@linux-foundation.org> Acked-by: David S. Miller <davem@davemloft.net> Acked-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:09:57 +00:00
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
rtnl_lock();
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (!state) {
state = kzalloc(sizeof(*state), GFP_KERNEL);
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-next into for-davem Conflicts: net/wireless/nl80211.c
2013-06-28 17:18:21 +00:00
if (!state) {
rtnl_unlock();
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
return -ENOMEM;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->filter_wiphy = -1;
ret = nl80211_dump_wiphy_parse(skb, cb, state);
if (ret) {
kfree(state);
rtnl_unlock();
return ret;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
cb->args[0] = (long)state;
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
}
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
list_for_each_entry(rdev, &cfg80211_rdev_list, list) {
if (!net_eq(wiphy_net(&rdev->wiphy), sock_net(skb->sk)))
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
continue;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (++idx <= state->start)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
continue;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
if (state->filter_wiphy != -1 &&
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
state->filter_wiphy != rdev->wiphy_idx)
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
continue;
/* attempt to fit multiple wiphy data chunks into the skb */
do {
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
ret = nl80211_send_wiphy(rdev, NL80211_CMD_NEW_WIPHY,
skb,
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq,
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
NLM_F_MULTI, state);
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
if (ret < 0) {
/*
* If sending the wiphy data didn't fit (ENOBUFS
* or EMSGSIZE returned), this SKB is still
* empty (so it's not too big because another
* wiphy dataset is already in the skb) and
* we've not tried to adjust the dump allocation
* yet ... then adjust the alloc size to be
* bigger, and return 1 but with the empty skb.
* This results in an empty message being RX'ed
* in userspace, but that is ignored.
*
* We can then retry with the larger buffer.
*/
if ((ret == -ENOBUFS || ret == -EMSGSIZE) &&
nl80211: Reset split_start when netlink skb is exhausted When the netlink skb is exhausted split_start is left set. In the subsequent retry, with a larger buffer, the dump is continued from the failing point instead of from the beginning. This was causing my rt28xx based USB dongle to now show up when running "iw list" with an old iw version without split dump support. Cc: stable@vger.kernel.org Fixes: 3713b4e364ef ("nl80211: allow splitting wiphy information in dumps") Signed-off-by: Pontus Fuchs <pontus.fuchs@gmail.com> [avoid the entire workaround when state->split is set] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-16 14:00:40 +00:00
!skb->len && !state->split &&
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
cb->min_dump_alloc < 4096) {
cb->min_dump_alloc = 4096;
nl80211: Reset split_start when netlink skb is exhausted When the netlink skb is exhausted split_start is left set. In the subsequent retry, with a larger buffer, the dump is continued from the failing point instead of from the beginning. This was causing my rt28xx based USB dongle to now show up when running "iw list" with an old iw version without split dump support. Cc: stable@vger.kernel.org Fixes: 3713b4e364ef ("nl80211: allow splitting wiphy information in dumps") Signed-off-by: Pontus Fuchs <pontus.fuchs@gmail.com> [avoid the entire workaround when state->split is set] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-16 14:00:40 +00:00
state->split_start = 0;
Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net Conflicts: drivers/net/wireless/ath/ath9k/Kconfig drivers/net/xen-netback/netback.c net/batman-adv/bat_iv_ogm.c net/wireless/nl80211.c The ath9k Kconfig conflict was a change of a Kconfig option name right next to the deletion of another option. The xen-netback conflict was overlapping changes involving the handling of the notify list in xen_netbk_rx_action(). Batman conflict resolution provided by Antonio Quartulli, basically keep everything in both conflict hunks. The nl80211 conflict is a little more involved. In 'net' we added a dynamic memory allocation to nl80211_dump_wiphy() to fix a race that Linus reported. Meanwhile in 'net-next' the handlers were converted to use pre and post doit handlers which use a flag to determine whether to hold the RTNL mutex around the operation. However, the dump handlers to not use this logic. Instead they have to explicitly do the locking. There were apparent bugs in the conversion of nl80211_dump_wiphy() in that we were not dropping the RTNL mutex in all the return paths, and it seems we very much should be doing so. So I fixed that whilst handling the overlapping changes. To simplify the initial returns, I take the RTNL mutex after we try to allocate 'tb'. Signed-off-by: David S. Miller <davem@davemloft.net>
2013-06-19 23:49:39 +00:00
rtnl_unlock();
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
return 1;
}
idx--;
break;
nl80211: increase wiphy dump size dynamically Given a device with many channels capabilities the wiphy information can still overflow even though its size in 3.9 was reduced to 3.8 levels. For new userspace and kernel 3.10 we're going to implement a new "split dump" protocol that can use multiple messages per wiphy. For now though, add a workaround to be able to send more information to userspace. Since generic netlink doesn't have a way to set the minimum dump size globally, and we wouldn't really want to set it globally anyway, increase the size only when needed, as described in the comments. As userspace might not be prepared for large buffers, we can only use 4k. Also increase the size for the get_wiphy command. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-01 13:03:49 +00:00
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
} while (state->split_start > 0);
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
break;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
rtnl_unlock();
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
state->start = idx;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
return skb->len;
}
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
static int nl80211_dump_wiphy_done(struct netlink_callback *cb)
{
kfree((void *)cb->args[0]);
return 0;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
static int nl80211_get_wiphy(struct sk_buff *skb, struct genl_info *info)
{
struct sk_buff *msg;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
struct nl80211_dump_wiphy_state state = {};
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: increase wiphy dump size dynamically Given a device with many channels capabilities the wiphy information can still overflow even though its size in 3.9 was reduced to 3.8 levels. For new userspace and kernel 3.10 we're going to implement a new "split dump" protocol that can use multiple messages per wiphy. For now though, add a workaround to be able to send more information to userspace. Since generic netlink doesn't have a way to set the minimum dump size globally, and we wouldn't really want to set it globally anyway, increase the size only when needed, as described in the comments. As userspace might not be prepared for large buffers, we can only use 4k. Also increase the size for the get_wiphy command. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-01 13:03:49 +00:00
msg = nlmsg_new(4096, GFP_KERNEL);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (!msg)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -ENOMEM;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
if (nl80211_send_wiphy(rdev, NL80211_CMD_NEW_WIPHY, msg,
info->snd_portid, info->snd_seq, 0,
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
&state) < 0) {
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
nlmsg_free(msg);
return -ENOBUFS;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
genetlink: make netns aware This makes generic netlink network namespace aware. No generic netlink families except for the controller family are made namespace aware, they need to be checked one by one and then set the family->netnsok member to true. A new function genlmsg_multicast_netns() is introduced to allow sending a multicast message in a given namespace, for example when it applies to an object that lives in that namespace, a new function genlmsg_multicast_allns() to send a message to all network namespaces (for objects that do not have an associated netns). The function genlmsg_multicast() is changed to multicast the message in just init_net, which is currently correct for all generic netlink families since they only work in init_net right now. Some will later want to work in all net namespaces because they do not care about the netns at all -- those will have to be converted to use one of the new functions genlmsg_multicast_allns() or genlmsg_multicast_netns() whenever they are made netns aware in some way. After this patch families can easily decide whether or not they should be available in all net namespaces. Many genl families us it for objects not related to networking and should therefore be available in all namespaces, but that will have to be done on a per family basis. Note that this doesn't touch on the checkpoint/restart problem where network namespaces could be used, genl families and multicast groups are numbered globally and I see no easy way of changing that, especially since it must be possible to multicast to all network namespaces for those families that do not care about netns. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-10 09:51:34 +00:00
return genlmsg_reply(msg, info);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
static const struct nla_policy txq_params_policy[NL80211_TXQ_ATTR_MAX + 1] = {
[NL80211_TXQ_ATTR_QUEUE] = { .type = NLA_U8 },
[NL80211_TXQ_ATTR_TXOP] = { .type = NLA_U16 },
[NL80211_TXQ_ATTR_CWMIN] = { .type = NLA_U16 },
[NL80211_TXQ_ATTR_CWMAX] = { .type = NLA_U16 },
[NL80211_TXQ_ATTR_AIFS] = { .type = NLA_U8 },
};
static int parse_txq_params(struct nlattr *tb[],
struct ieee80211_txq_params *txq_params)
{
cfg80211/nl80211: clarify TX queue API With the plan to change mac80211's queue API to not map ACs to queues 1:1, it seems necessary to clarify some APIs that act on ACs rather than on queues to spell that out explicitly. Do this. Also verify that the AC number given is valid. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-28 09:04:24 +00:00
if (!tb[NL80211_TXQ_ATTR_AC] || !tb[NL80211_TXQ_ATTR_TXOP] ||
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
!tb[NL80211_TXQ_ATTR_CWMIN] || !tb[NL80211_TXQ_ATTR_CWMAX] ||
!tb[NL80211_TXQ_ATTR_AIFS])
return -EINVAL;
cfg80211/nl80211: clarify TX queue API With the plan to change mac80211's queue API to not map ACs to queues 1:1, it seems necessary to clarify some APIs that act on ACs rather than on queues to spell that out explicitly. Do this. Also verify that the AC number given is valid. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-28 09:04:24 +00:00
txq_params->ac = nla_get_u8(tb[NL80211_TXQ_ATTR_AC]);
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
txq_params->txop = nla_get_u16(tb[NL80211_TXQ_ATTR_TXOP]);
txq_params->cwmin = nla_get_u16(tb[NL80211_TXQ_ATTR_CWMIN]);
txq_params->cwmax = nla_get_u16(tb[NL80211_TXQ_ATTR_CWMAX]);
txq_params->aifs = nla_get_u8(tb[NL80211_TXQ_ATTR_AIFS]);
cfg80211/nl80211: clarify TX queue API With the plan to change mac80211's queue API to not map ACs to queues 1:1, it seems necessary to clarify some APIs that act on ACs rather than on queues to spell that out explicitly. Do this. Also verify that the AC number given is valid. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-28 09:04:24 +00:00
if (txq_params->ac >= NL80211_NUM_ACS)
return -EINVAL;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
return 0;
}
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
static bool nl80211_can_set_dev_channel(struct wireless_dev *wdev)
{
/*
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
* You can only set the channel explicitly for WDS interfaces,
* all others have their channel managed via their respective
* "establish a connection" command (connect, join, ...)
*
* For AP/GO and mesh mode, the channel can be set with the
* channel userspace API, but is only stored and passed to the
* low-level driver when the AP starts or the mesh is joined.
* This is for backward compatibility, userspace can also give
* the channel in the start-ap or join-mesh commands instead.
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
*
* Monitors are special as they are normally slaved to
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
* whatever else is going on, so they have their own special
* operation to set the monitor channel if possible.
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
*/
return !wdev ||
wdev->iftype == NL80211_IFTYPE_AP ||
wdev->iftype == NL80211_IFTYPE_MESH_POINT ||
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
wdev->iftype == NL80211_IFTYPE_MONITOR ||
wdev->iftype == NL80211_IFTYPE_P2P_GO;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
}
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
static int nl80211_parse_chandef(struct cfg80211_registered_device *rdev,
struct genl_info *info,
struct cfg80211_chan_def *chandef)
{
cfg80211: Remove unused VHT chan code Cleanup of unused VHT channel config related code. Signed-off-by: Mahesh Palivela <maheshp@posedge.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 08:41:07 +00:00
u32 control_freq;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (!info->attrs[NL80211_ATTR_WIPHY_FREQ])
return -EINVAL;
control_freq = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_FREQ]);
chandef->chan = ieee80211_get_channel(&rdev->wiphy, control_freq);
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
chandef->width = NL80211_CHAN_WIDTH_20_NOHT;
chandef->center_freq1 = control_freq;
chandef->center_freq2 = 0;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
/* Primary channel not allowed */
if (!chandef->chan || chandef->chan->flags & IEEE80211_CHAN_DISABLED)
return -EINVAL;
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]) {
enum nl80211_channel_type chantype;
chantype = nla_get_u32(
info->attrs[NL80211_ATTR_WIPHY_CHANNEL_TYPE]);
switch (chantype) {
case NL80211_CHAN_NO_HT:
case NL80211_CHAN_HT20:
case NL80211_CHAN_HT40PLUS:
case NL80211_CHAN_HT40MINUS:
cfg80211_chandef_create(chandef, chandef->chan,
chantype);
break;
default:
return -EINVAL;
}
} else if (info->attrs[NL80211_ATTR_CHANNEL_WIDTH]) {
chandef->width =
nla_get_u32(info->attrs[NL80211_ATTR_CHANNEL_WIDTH]);
if (info->attrs[NL80211_ATTR_CENTER_FREQ1])
chandef->center_freq1 =
nla_get_u32(
info->attrs[NL80211_ATTR_CENTER_FREQ1]);
if (info->attrs[NL80211_ATTR_CENTER_FREQ2])
chandef->center_freq2 =
nla_get_u32(
info->attrs[NL80211_ATTR_CENTER_FREQ2]);
}
cfg80211: rework chandef checking and export it Some of the chandef checking that we do in cfg80211 to check if a channel is supported or not is also needed in mac80211, so rework that a bit and export the functions that are needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-22 15:59:45 +00:00
if (!cfg80211_chandef_valid(chandef))
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
return -EINVAL;
cfg80211: rework chandef checking and export it Some of the chandef checking that we do in cfg80211 to check if a channel is supported or not is also needed in mac80211, so rework that a bit and export the functions that are needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-22 15:59:45 +00:00
if (!cfg80211_chandef_usable(&rdev->wiphy, chandef,
IEEE80211_CHAN_DISABLED))
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
return -EINVAL;
nl80211/cfg80211: add 5 and 10 MHz defines and wiphy flag Add defines for 5 and 10 MHz channel width and fix channel handling functions accordingly. Also check for and report the WIPHY_FLAG_SUPPORTS_5_10_MHZ capability. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix spelling in comment] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 11:00:28 +00:00
if ((chandef->width == NL80211_CHAN_WIDTH_5 ||
chandef->width == NL80211_CHAN_WIDTH_10) &&
!(rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_5_10_MHZ))
return -EINVAL;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
return 0;
}
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
static int __nl80211_set_channel(struct cfg80211_registered_device *rdev,
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
struct net_device *dev,
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
struct genl_info *info)
{
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
struct cfg80211_chan_def chandef;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
int result;
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
enum nl80211_iftype iftype = NL80211_IFTYPE_MONITOR;
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
struct wireless_dev *wdev = NULL;
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
if (dev)
wdev = dev->ieee80211_ptr;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
if (!nl80211_can_set_dev_channel(wdev))
return -EOPNOTSUPP;
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
if (wdev)
iftype = wdev->iftype;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
result = nl80211_parse_chandef(rdev, info, &chandef);
if (result)
return result;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
switch (iftype) {
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
cfg80211: use RTNL locked reg_can_beacon for IR-relaxation The RTNL is required to check for IR-relaxation conditions that allow more channels to beacon. Export an RTNL locked version of reg_can_beacon and use it where possible in AP/STA interface type flows, where IR-relaxation may be applicable. Fixes: 06f207fc5418 ("cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA") Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-07-08 12:41:44 +00:00
if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &chandef,
iftype)) {
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
result = -EINVAL;
break;
}
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
if (wdev->beacon_interval) {
if (!dev || !rdev->ops->set_ap_chanwidth ||
!(rdev->wiphy.features &
NL80211_FEATURE_AP_MODE_CHAN_WIDTH_CHANGE)) {
result = -EBUSY;
break;
}
/* Only allow dynamic channel width changes */
if (chandef.chan != wdev->preset_chandef.chan) {
result = -EBUSY;
break;
}
result = rdev_set_ap_chanwidth(rdev, dev, &chandef);
if (result)
break;
}
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
wdev->preset_chandef = chandef;
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
result = 0;
break;
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
case NL80211_IFTYPE_MESH_POINT:
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
result = cfg80211_set_mesh_channel(rdev, wdev, &chandef);
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
break;
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
case NL80211_IFTYPE_MONITOR:
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
result = cfg80211_set_monitor_channel(rdev, &chandef);
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
break;
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
default:
cfg80211: clarify set_channel APIs Now that we've removed all uses of the set_channel API except for the monitor channel and in libertas, clarify this. Split the libertas mesh use into a new libertas_set_mesh_channel() operation, just to keep backward compatibility, and rename the normal set_channel() to set_monitor_channel(). Also describe the desired set_monitor_channel() semantics more clearly. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-06 06:18:22 +00:00
result = -EINVAL;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
}
return result;
}
static int nl80211_set_channel(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *netdev = info->user_ptr[1];
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
return __nl80211_set_channel(rdev, netdev, info);
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
}
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
static int nl80211_set_wds_peer(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use new genl helpers for WDS Bill Jordan's patch to allow setting the WDS peer crossed with my patch removing all the boilerplate code in nl80211, and consequently he didn't make use of it yet. Fix that. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:10:30 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: constify WDS address There's no need for the WDS peer address to not be const, so make it const. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:11:09 +00:00
const u8 *bssid;
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
nl80211: use new genl helpers for WDS Bill Jordan's patch to allow setting the WDS peer crossed with my patch removing all the boilerplate code in nl80211, and consequently he didn't make use of it yet. Fix that. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:10:30 +00:00
if (netif_running(dev))
return -EBUSY;
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
nl80211: use new genl helpers for WDS Bill Jordan's patch to allow setting the WDS peer crossed with my patch removing all the boilerplate code in nl80211, and consequently he didn't make use of it yet. Fix that. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:10:30 +00:00
if (!rdev->ops->set_wds_peer)
return -EOPNOTSUPP;
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
nl80211: use new genl helpers for WDS Bill Jordan's patch to allow setting the WDS peer crossed with my patch removing all the boilerplate code in nl80211, and consequently he didn't make use of it yet. Fix that. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:10:30 +00:00
if (wdev->iftype != NL80211_IFTYPE_WDS)
return -EOPNOTSUPP;
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_set_wds_peer(rdev, dev, bssid);
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
static int nl80211_set_wiphy(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev;
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
struct net_device *netdev = NULL;
struct wireless_dev *wdev;
nl80211: Uninitialized variable There is a path in nl80211_set_wiphy where result is tested but uninitialized. I am hitting this path when I attempt: sh# iw dev wlan0 set channel 10 command failed: Unknown error 1069727332 (-1069727332) Signed-off-by: William Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-10 15:22:32 +00:00
int result = 0, rem_txq_params = 0;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
struct nlattr *nl_txq_params;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
u32 changed;
u8 retry_short = 0, retry_long = 0;
u32 frag_threshold = 0, rts_threshold = 0;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
u8 coverage_class = 0;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
ASSERT_RTNL();
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
/*
* Try to find the wiphy and netdev. Normally this
* function shouldn't need the netdev, but this is
* done for backward compatibility -- previously
* setting the channel was done per wiphy, but now
* it is per netdev. Previous userland like hostapd
* also passed a netdev to set_wiphy, so that it is
* possible to let that go to the right netdev!
*/
cfg80211: fix locking in nl80211_set_wiphy Luis reports that there's a circular locking dependency; this is because cfg80211_dev_rename() will acquire the cfg80211_mutex while the device mutex is held, while this normally is done the other way around. The solution is to open-code the device-getting in nl80211_set_wiphy and require holding the mutex around cfg80211_dev_rename rather than acquiring it within. Also fix a bug -- rtnl locking is expected by drivers so we need to provide it. Reported-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-24 08:35:46 +00:00
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
if (info->attrs[NL80211_ATTR_IFINDEX]) {
int ifindex = nla_get_u32(info->attrs[NL80211_ATTR_IFINDEX]);
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
netdev = __dev_get_by_index(genl_info_net(info), ifindex);
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
if (netdev && netdev->ieee80211_ptr)
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
rdev = wiphy_to_rdev(netdev->ieee80211_ptr->wiphy);
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
else
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
netdev = NULL;
cfg80211: fix locking in nl80211_set_wiphy Luis reports that there's a circular locking dependency; this is because cfg80211_dev_rename() will acquire the cfg80211_mutex while the device mutex is held, while this normally is done the other way around. The solution is to open-code the device-getting in nl80211_set_wiphy and require holding the mutex around cfg80211_dev_rename rather than acquiring it within. Also fix a bug -- rtnl locking is expected by drivers so we need to provide it. Reported-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-24 08:35:46 +00:00
}
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
if (!netdev) {
nl80211: change __cfg80211_rdev_from_info Change the function to __cfg80211_rdev_from_attrs to take attributes instead of the info struct to make it usable from dump callbacks for testmode. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:18:32 +00:00
rdev = __cfg80211_rdev_from_attrs(genl_info_net(info),
info->attrs);
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
if (IS_ERR(rdev))
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return PTR_ERR(rdev);
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
wdev = NULL;
netdev = NULL;
result = 0;
nl80211: move "can set channel" check Setting the wdev to NULL when the channel can't be set for that interface type (to treat the channel setting for the wiphy/monitor) currently works, but is confusing in the code if netdev/wdev aren't both set/unset in the same way. Move the check whether the channel can be set to where it's needed so that wdev and netdev are always both assigned or NULL. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-24 08:04:58 +00:00
} else
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
wdev = netdev->ieee80211_ptr;
/*
* end workaround code, by now the rdev is available
* and locked, and wdev may or may not be NULL.
*/
cfg80211: fix locking in nl80211_set_wiphy Luis reports that there's a circular locking dependency; this is because cfg80211_dev_rename() will acquire the cfg80211_mutex while the device mutex is held, while this normally is done the other way around. The solution is to open-code the device-getting in nl80211_set_wiphy and require holding the mutex around cfg80211_dev_rename rather than acquiring it within. Also fix a bug -- rtnl locking is expected by drivers so we need to provide it. Reported-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-24 08:35:46 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_NAME])
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
result = cfg80211_dev_rename(
rdev, nla_data(info->attrs[NL80211_ATTR_WIPHY_NAME]));
cfg80211: fix locking in nl80211_set_wiphy Luis reports that there's a circular locking dependency; this is because cfg80211_dev_rename() will acquire the cfg80211_mutex while the device mutex is held, while this normally is done the other way around. The solution is to open-code the device-getting in nl80211_set_wiphy and require holding the mutex around cfg80211_dev_rename rather than acquiring it within. Also fix a bug -- rtnl locking is expected by drivers so we need to provide it. Reported-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-24 08:35:46 +00:00
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_TXQ_PARAMS]) {
struct ieee80211_txq_params txq_params;
struct nlattr *tb[NL80211_TXQ_ATTR_MAX + 1];
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (!rdev->ops->set_txq_params)
return -EOPNOTSUPP;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (!netdev)
return -EINVAL;
cfg80211/mac80211: add netdev param to set_txq_params() tx params are currently configured per hw, although they should be configured per interface. Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-25 17:06:53 +00:00
cfg80211: allow setting TXQ parameters only in AP mode In other modes the parameters should not be set. Right now, mac80211 will set them, even if the user asked for setting them on VLANs which the driver doesn't know about, causing all kinds of trouble. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-03 13:50:13 +00:00
if (netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
netdev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
cfg80211: allow setting TXQ parameters only in AP mode In other modes the parameters should not be set. Right now, mac80211 will set them, even if the user asked for setting them on VLANs which the driver doesn't know about, causing all kinds of trouble. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-03 13:50:13 +00:00
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (!netif_running(netdev))
return -ENETDOWN;
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
nla_for_each_nested(nl_txq_params,
info->attrs[NL80211_ATTR_WIPHY_TXQ_PARAMS],
rem_txq_params) {
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
result = nla_parse_nested(tb, NL80211_TXQ_ATTR_MAX,
nl_txq_params,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
txq_params_policy,
info->extack);
nl80211: check nla_parse() return values If there's a policy, then nla_parse() return values must be checked, otherwise the policy is useless and there's nothing that ensures the attributes are actually what we expect them to be. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:17:47 +00:00
if (result)
return result;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
result = parse_txq_params(tb, &txq_params);
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
result = rdev_set_txq_params(rdev, netdev,
&txq_params);
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
nl80211: Add TX queue parameter configuration Add a new attribute, NL80211_ATTR_WIPHY_TXQ_PARAMS, that can be used with NL80211_CMD_SET_WIPHY for userspace (e.g., hostapd) to set TX queue parameters (txop, cwmin, cwmax, aifs). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:24 +00:00
}
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: Add frequency configuration (including HT40) This patch adds new NL80211_CMD_SET_WIPHY attributes NL80211_ATTR_WIPHY_FREQ and NL80211_ATTR_WIPHY_SEC_CHAN_OFFSET to allow userspace to set the operating channel (e.g., hostapd for AP mode). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-26 14:15:24 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) {
cfg80211: Dynamic channel bandwidth changes in AP mode This extends NL80211_CMD_SET_CHANNEL to allow dynamic channel bandwidth changes in AP mode (including P2P GO) during a lifetime of the BSS. This can be used to implement, e.g., HT 20/40 MHz co-existence rules on the 2.4 GHz band. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-28 08:22:08 +00:00
result = __nl80211_set_channel(
rdev,
nl80211_can_set_dev_channel(wdev) ? netdev : NULL,
info);
nl80211: Add frequency configuration (including HT40) This patch adds new NL80211_CMD_SET_WIPHY attributes NL80211_ATTR_WIPHY_FREQ and NL80211_ATTR_WIPHY_SEC_CHAN_OFFSET to allow userspace to set the operating channel (e.g., hostapd for AP mode). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-26 14:15:24 +00:00
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
nl80211: Add frequency configuration (including HT40) This patch adds new NL80211_CMD_SET_WIPHY attributes NL80211_ATTR_WIPHY_FREQ and NL80211_ATTR_WIPHY_SEC_CHAN_OFFSET to allow userspace to set the operating channel (e.g., hostapd for AP mode). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-11-26 14:15:24 +00:00
}
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_TX_POWER_SETTING]) {
cfg80211: allow per interface TX power setting The TX power setting is currently per wiphy (hardware device) but with multi-channel capabilities that doesn't make much sense any more. Allow drivers (and mac80211) to advertise support for per-interface TX power configuration. When the TX power is configured for the wiphy, the wdev will be NULL and the driver can still handle that, but when a wdev is given the TX power can be set only for that wdev now. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-24 08:17:18 +00:00
struct wireless_dev *txp_wdev = wdev;
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
enum nl80211_tx_power_setting type;
int idx, mbm = 0;
cfg80211: allow per interface TX power setting The TX power setting is currently per wiphy (hardware device) but with multi-channel capabilities that doesn't make much sense any more. Allow drivers (and mac80211) to advertise support for per-interface TX power configuration. When the TX power is configured for the wiphy, the wdev will be NULL and the driver can still handle that, but when a wdev is given the TX power can be set only for that wdev now. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-24 08:17:18 +00:00
if (!(rdev->wiphy.features & NL80211_FEATURE_VIF_TXPOWER))
txp_wdev = NULL;
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (!rdev->ops->set_tx_power)
return -EOPNOTSUPP;
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
idx = NL80211_ATTR_WIPHY_TX_POWER_SETTING;
type = nla_get_u32(info->attrs[idx]);
if (!info->attrs[NL80211_ATTR_WIPHY_TX_POWER_LEVEL] &&
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
(type != NL80211_TX_POWER_AUTOMATIC))
return -EINVAL;
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
if (type != NL80211_TX_POWER_AUTOMATIC) {
idx = NL80211_ATTR_WIPHY_TX_POWER_LEVEL;
mbm = nla_get_u32(info->attrs[idx]);
}
cfg80211: allow per interface TX power setting The TX power setting is currently per wiphy (hardware device) but with multi-channel capabilities that doesn't make much sense any more. Allow drivers (and mac80211) to advertise support for per-interface TX power configuration. When the TX power is configured for the wiphy, the wdev will be NULL and the driver can still handle that, but when a wdev is given the TX power can be set only for that wdev now. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-24 08:17:18 +00:00
result = rdev_set_tx_power(rdev, txp_wdev, type, mbm);
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
nl80211: Add option to adjust transmit power This patch adds transmit power setting type and transmit power level attributes to NL80211_CMD_SET_WIPHY in order to facilitate adjusting of the transmit power level of the device. The added attributes allow selection of automatic, limited or fixed transmit power level, with the level definable in signed mBm format. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-23 09:12:38 +00:00
}
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_ANTENNA_TX] &&
info->attrs[NL80211_ATTR_WIPHY_ANTENNA_RX]) {
u32 tx_ant, rx_ant;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211: Separate available antennas for RX and TX As has been pointed out by Daniel Halperin some devices (e.g. Intel IWL5100) can only TX from a subset of RX antennas, so use separate availability masks for RX and TX. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-16 02:30:22 +00:00
if ((!rdev->wiphy.available_antennas_tx &&
!rdev->wiphy.available_antennas_rx) ||
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
!rdev->ops->set_antenna)
return -EOPNOTSUPP;
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
tx_ant = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_ANTENNA_TX]);
rx_ant = nla_get_u32(info->attrs[NL80211_ATTR_WIPHY_ANTENNA_RX]);
cfg80211: Add antenna availability information Add a field to wiphy for the hardware to report the availble antennas for configuration. Only if this is set to something bigger than zero, will the anntenna configuration ops be executed. Allthough this could be a simple number of antennas, I defined it as a bitmap of antennas which are available for configuration, since it's more consistent with the rest of the antenna API and there could be cases where the hardware allows only configuration of certain antennas. As it does not make much of a difference in size or normal usage, I think it's better to be able to support this, in case the need arises. The antenna configuration is now also checked against the availabe antennas and rejected if it does not match. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v3: always apply available antenna mask (for "all" antennas case). v2: reject antenna configurations which don't match the available antennas Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-08 04:59:24 +00:00
/* reject antenna configurations which don't match the
cfg80211: Separate available antennas for RX and TX As has been pointed out by Daniel Halperin some devices (e.g. Intel IWL5100) can only TX from a subset of RX antennas, so use separate availability masks for RX and TX. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-16 02:30:22 +00:00
* available antenna masks, except for the "all" mask */
if ((~tx_ant && (tx_ant & ~rdev->wiphy.available_antennas_tx)) ||
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
(~rx_ant && (rx_ant & ~rdev->wiphy.available_antennas_rx)))
return -EINVAL;
cfg80211: Add antenna availability information Add a field to wiphy for the hardware to report the availble antennas for configuration. Only if this is set to something bigger than zero, will the anntenna configuration ops be executed. Allthough this could be a simple number of antennas, I defined it as a bitmap of antennas which are available for configuration, since it's more consistent with the rest of the antenna API and there could be cases where the hardware allows only configuration of certain antennas. As it does not make much of a difference in size or normal usage, I think it's better to be able to support this, in case the need arises. The antenna configuration is now also checked against the availabe antennas and rejected if it does not match. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v3: always apply available antenna mask (for "all" antennas case). v2: reject antenna configurations which don't match the available antennas Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-08 04:59:24 +00:00
cfg80211: Separate available antennas for RX and TX As has been pointed out by Daniel Halperin some devices (e.g. Intel IWL5100) can only TX from a subset of RX antennas, so use separate availability masks for RX and TX. Signed-off-by: Bruno Randolf <br1@einfach.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-16 02:30:22 +00:00
tx_ant = tx_ant & rdev->wiphy.available_antennas_tx;
rx_ant = rx_ant & rdev->wiphy.available_antennas_rx;
cfg80211: Add antenna availability information Add a field to wiphy for the hardware to report the availble antennas for configuration. Only if this is set to something bigger than zero, will the anntenna configuration ops be executed. Allthough this could be a simple number of antennas, I defined it as a bitmap of antennas which are available for configuration, since it's more consistent with the rest of the antenna API and there could be cases where the hardware allows only configuration of certain antennas. As it does not make much of a difference in size or normal usage, I think it's better to be able to support this, in case the need arises. The antenna configuration is now also checked against the availabe antennas and rejected if it does not match. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v3: always apply available antenna mask (for "all" antennas case). v2: reject antenna configurations which don't match the available antennas Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-08 04:59:24 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
result = rdev_set_antenna(rdev, tx_ant, rx_ant);
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
if (result)
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return result;
cfg80211: Add nl80211 antenna configuration Allow setting of TX and RX antennas configuration via nl80211. The antenna configuration is defined as a bitmap of allowed antennas to use. This API can be used to mask out antennas which are not attached or should not be used for other reasons like regulatory concerns or special setups. Separate bitmaps are used for RX and TX to allow configuring different antennas for receiving and transmitting. Each bitmap is 32 bit long, each bit representing one antenna, starting with antenna 1 at the first bit. If an antenna bit is set, this means the driver is allowed to use this antenna for RX or TX respectively; if the bit is not set the hardware is not allowed to use this antenna. Using bitmaps has the benefit of allowing for a flexible configuration interface which can support many different configurations and which can be used for 802.11n as well as non-802.11n devices. Instead of relying on some hardware specific assumptions, drivers can use this information to know which antennas are actually attached to the system and derive their capabilities based on that. 802.11n devices should enable or disable chains, based on which antennas are present (If all antennas belonging to a particular chain are disabled, the entire chain should be disabled). HT capabilities (like STBC, TX Beamforming, Antenna selection) should be calculated based on the available chains after applying the antenna masks. Should a 802.11n device have diversity antennas attached to one of their chains, diversity can be enabled or disabled based on the antenna information. Non-802.11n drivers can use the antenna masks to select RX and TX antennas and to enable or disable antenna diversity. While covering chainmasks for 802.11n and the standard "legacy" modes "fixed antenna 1", "fixed antenna 2" and "diversity" this API also allows more rare, but useful configurations as follows: 1) Send on antenna 1, receive on antenna 2 (or vice versa). This can be used to have a low gain antenna for TX in order to keep within the regulatory constraints and a high gain antenna for RX in order to receive weaker signals ("speak softly, but listen harder"). This can be useful for building long-shot outdoor links. Another usage of this setup is having a low-noise pre-amplifier on antenna 1 and a power amplifier on the other antenna. This way transmit noise is mostly kept out of the low noise receive channel. (This would be bitmaps: tx 1 rx 2). 2) Another similar setup is: Use RX diversity on both antennas, but always send on antenna 1. Again that would allow us to benefit from a higher gain RX antenna, while staying within the legal limits. (This would be: tx 0 rx 3). 3) And finally there can be special experimental setups in research and development even with pre 802.11n hardware where more than 2 antennas are available. It's good to keep the API simple, yet flexible. Signed-off-by: Bruno Randolf <br1@einfach.org> -- v7: Made bitmasks 32 bit wide and rebased to latest wireless-testing. Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-10 03:50:50 +00:00
}
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
changed = 0;
if (info->attrs[NL80211_ATTR_WIPHY_RETRY_SHORT]) {
retry_short = nla_get_u8(
info->attrs[NL80211_ATTR_WIPHY_RETRY_SHORT]);
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (retry_short == 0)
return -EINVAL;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
changed |= WIPHY_PARAM_RETRY_SHORT;
}
if (info->attrs[NL80211_ATTR_WIPHY_RETRY_LONG]) {
retry_long = nla_get_u8(
info->attrs[NL80211_ATTR_WIPHY_RETRY_LONG]);
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (retry_long == 0)
return -EINVAL;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
changed |= WIPHY_PARAM_RETRY_LONG;
}
if (info->attrs[NL80211_ATTR_WIPHY_FRAG_THRESHOLD]) {
frag_threshold = nla_get_u32(
info->attrs[NL80211_ATTR_WIPHY_FRAG_THRESHOLD]);
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (frag_threshold < 256)
return -EINVAL;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
if (frag_threshold != (u32) -1) {
/*
* Fragments (apart from the last one) are required to
* have even length. Make the fragmentation code
* simpler by stripping LSB should someone try to use
* odd threshold value.
*/
frag_threshold &= ~0x1;
}
changed |= WIPHY_PARAM_FRAG_THRESHOLD;
}
if (info->attrs[NL80211_ATTR_WIPHY_RTS_THRESHOLD]) {
rts_threshold = nla_get_u32(
info->attrs[NL80211_ATTR_WIPHY_RTS_THRESHOLD]);
changed |= WIPHY_PARAM_RTS_THRESHOLD;
}
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]) {
cfg80211: enable dynack through nl80211 Enable ACK timeout estimation algorithm (dynack) using mac80211 set_coverage_class API. Dynack is activated passing coverage class equals to -1 to lower drivers and it is automatically disabled setting valid value for coverage class. Define NL80211_ATTR_WIPHY_DYN_ACK flag attribute to enable dynack from userspace. In order to activate dynack NL80211_FEATURE_ACKTO_ESTIMATION feature flag must be set by lower drivers to indicate dynack capability. Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-04 21:57:40 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_DYN_ACK])
return -EINVAL;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
coverage_class = nla_get_u8(
info->attrs[NL80211_ATTR_WIPHY_COVERAGE_CLASS]);
changed |= WIPHY_PARAM_COVERAGE_CLASS;
}
cfg80211: enable dynack through nl80211 Enable ACK timeout estimation algorithm (dynack) using mac80211 set_coverage_class API. Dynack is activated passing coverage class equals to -1 to lower drivers and it is automatically disabled setting valid value for coverage class. Define NL80211_ATTR_WIPHY_DYN_ACK flag attribute to enable dynack from userspace. In order to activate dynack NL80211_FEATURE_ACKTO_ESTIMATION feature flag must be set by lower drivers to indicate dynack capability. Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi83@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-04 21:57:40 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_DYN_ACK]) {
if (!(rdev->wiphy.features & NL80211_FEATURE_ACKTO_ESTIMATION))
return -EOPNOTSUPP;
changed |= WIPHY_PARAM_DYN_ACK;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
}
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
if (changed) {
u8 old_retry_short, old_retry_long;
u32 old_frag_threshold, old_rts_threshold;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
u8 old_coverage_class;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
if (!rdev->ops->set_wiphy_params)
return -EOPNOTSUPP;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
old_retry_short = rdev->wiphy.retry_short;
old_retry_long = rdev->wiphy.retry_long;
old_frag_threshold = rdev->wiphy.frag_threshold;
old_rts_threshold = rdev->wiphy.rts_threshold;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
old_coverage_class = rdev->wiphy.coverage_class;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
if (changed & WIPHY_PARAM_RETRY_SHORT)
rdev->wiphy.retry_short = retry_short;
if (changed & WIPHY_PARAM_RETRY_LONG)
rdev->wiphy.retry_long = retry_long;
if (changed & WIPHY_PARAM_FRAG_THRESHOLD)
rdev->wiphy.frag_threshold = frag_threshold;
if (changed & WIPHY_PARAM_RTS_THRESHOLD)
rdev->wiphy.rts_threshold = rts_threshold;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
if (changed & WIPHY_PARAM_COVERAGE_CLASS)
rdev->wiphy.coverage_class = coverage_class;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
result = rdev_set_wiphy_params(rdev, changed);
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
if (result) {
rdev->wiphy.retry_short = old_retry_short;
rdev->wiphy.retry_long = old_retry_long;
rdev->wiphy.frag_threshold = old_frag_threshold;
rdev->wiphy.rts_threshold = old_rts_threshold;
nl80211: Add new WIPHY attribute COVERAGE_CLASS The new attribute NL80211_ATTR_WIPHY_COVERAGE_CLASS sets IEEE 802.11 Coverage Class, which depends on maximum distance of nodes in a wireless network. It's required for long distance links (more than a few hundred meters). The attribute is now ignored by two non-mac80211 drivers, rndis and iwmc3200wifi, together with WIPHY_PARAM_RETRY_SHORT and WIPHY_PARAM_RETRY_LONG. If it turns out to be a problem, we could split set_wiphy_params callback or add new capability bits. Signed-off-by: Lukas Turek <8an@praha12.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-21 21:50:47 +00:00
rdev->wiphy.coverage_class = old_coverage_class;
cfg80211: propagate set_wiphy failure to userspace If driver failed to setup wiphy params (e.g. rts threshold, fragmentation treshold) userspace wasn't properly notified about this. This could lead to user confusion who would think the command succeeded even if that wasn't the case. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-03 08:55:24 +00:00
return result;
nl80211: Add set/get for frag/rts threshold and retry limits Add new nl80211 attributes that can be used with NL80211_CMD_SET_WIPHY and NL80211_CMD_GET_WIPHY to manage fragmentation/RTS threshold and retry limits. Since these values are stored in struct wiphy, remove the local copy from mac80211 where feasible (frag & rts threshold). The retry limits are currently needed in struct ieee80211_conf, but these could be eventually removed since the driver should have access to the values in struct wiphy. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-20 16:39:05 +00:00
}
}
net: nl80211: __dev_get_by_index instead of dev_get_by_index to find interface As __cfg80211_rdev_from_attrs(), nl80211_dump_wiphy_parse() and nl80211_set_wiphy() are all under rtnl_lock protection, __dev_get_by_index() instead of dev_get_by_index() should be used to find interface handler in them allowing us to avoid to change interface reference counter. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Ying Xue <ying.xue@windriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2014-01-15 02:23:45 +00:00
return 0;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
static inline u64 wdev_id(struct wireless_dev *wdev)
{
return (u64)wdev->identifier |
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
((u64)wiphy_to_rdev(wdev->wiphy)->wiphy_idx << 32);
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
static int nl80211_send_chandef(struct sk_buff *msg,
cfg80211/mac80211: DFS setup chandef for cac event To report channel width correctly we have to send correct channel parameters from mac80211 when calling cfg80211_cac_event(). This is required in case of using channel width higher than 20MHz and we have to set correct dfs channel state after CAC (NL80211_DFS_AVAILABLE). Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Reviewed-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-06 12:55:51 +00:00
const struct cfg80211_chan_def *chandef)
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
{
nl80211: don't crash sending invalid chandef One of the cases for an invalid channel definition is that the channel pointer is NULL, in which case the warning is a bit late since we'll dereference the pointer. Bail out of the function upon warning about this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-27 16:26:56 +00:00
if (WARN_ON(!cfg80211_chandef_valid(chandef)))
return -EINVAL;
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY_FREQ,
chandef->chan->center_freq))
return -ENOBUFS;
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
switch (chandef->width) {
case NL80211_CHAN_WIDTH_20_NOHT:
case NL80211_CHAN_WIDTH_20:
case NL80211_CHAN_WIDTH_40:
if (nla_put_u32(msg, NL80211_ATTR_WIPHY_CHANNEL_TYPE,
cfg80211_get_chandef_type(chandef)))
return -ENOBUFS;
break;
default:
break;
}
if (nla_put_u32(msg, NL80211_ATTR_CHANNEL_WIDTH, chandef->width))
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_ATTR_CENTER_FREQ1, chandef->center_freq1))
return -ENOBUFS;
if (chandef->center_freq2 &&
nla_put_u32(msg, NL80211_ATTR_CENTER_FREQ2, chandef->center_freq2))
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
return -ENOBUFS;
return 0;
}
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
static int nl80211_send_iface(struct sk_buff *msg, u32 portid, u32 seq, int flags,
nl80211: send wiphy along with netdev When listing all wireless netdevs in the system this is useful to print which wiphy they belong to. Just add the attribute, any program that doesn't care will just ignore it. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 14:23:20 +00:00
struct cfg80211_registered_device *rdev,
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
struct wireless_dev *wdev, bool removal)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
nl80211: retrieve interface data by wdev Since soon there will be virtual interfaces that don't have a netdev, use the wdev identifier for the API to retrieve interface data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 15:52:47 +00:00
struct net_device *dev = wdev->netdev;
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
u8 cmd = NL80211_CMD_NEW_INTERFACE;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
void *hdr;
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
if (removal)
cmd = NL80211_CMD_DEL_INTERFACE;
hdr = nl80211hdr_put(msg, portid, seq, flags, cmd);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (!hdr)
return -1;
nl80211: retrieve interface data by wdev Since soon there will be virtual interfaces that don't have a netdev, use the wdev identifier for the API to retrieve interface data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 15:52:47 +00:00
if (dev &&
(nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
nla_put_string(msg, NL80211_ATTR_IFNAME, dev->name)))
nl80211: retrieve interface data by wdev Since soon there will be virtual interfaces that don't have a netdev, use the wdev identifier for the API to retrieve interface data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 15:52:47 +00:00
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFTYPE, wdev->iftype) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD) ||
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, wdev_address(wdev)) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_ATTR_GENERATION,
rdev->devlist_generation ^
(cfg80211_rdev_list_generation << 2)))
goto nla_put_failure;
nl80211: add generation number to all dumps In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 14:17:38 +00:00
cfg80211/mac80211: re-add get_channel operation This essentially reverts commit 2e165b818456 but introduces the get_channel operation with a new wireless_dev argument so that you can retrieve the channel per interface. This is necessary as even though we can track all interface channels (except monitor) we can't track the channel type used. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 17:45:08 +00:00
if (rdev->ops->get_channel) {
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
int ret;
struct cfg80211_chan_def chandef;
ret = rdev_get_channel(rdev, wdev, &chandef);
if (ret == 0) {
if (nl80211_send_chandef(msg, &chandef))
goto nla_put_failure;
}
cfg80211: Add channel information to NL80211_CMD_GET_INTERFACE If the current channel is known, add frequency and channel type to NL80211_CMD_GET_INTERFACE. Signed-off-by: Pontus Fuchs <pontus.fuchs@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-03 14:39:58 +00:00
}
nl80211: put current TX power in interface info Many drivers implement reading current TX power (using either cfg80211 or ieee80211 op) but userspace can't get it using nl80211. Right now the only way to access it is to call some wext ioctl. Let's put TX power in interface info reply (callback is wdev specific) just like we do with current channel. To be consistent (e.g. NL80211_CMD_SET_WIPHY) let's use mBm as na unit. Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-31 20:59:38 +00:00
if (rdev->ops->get_tx_power) {
int dbm, ret;
ret = rdev_get_tx_power(rdev, wdev, &dbm);
if (ret == 0 &&
nla_put_u32(msg, NL80211_ATTR_WIPHY_TX_POWER_LEVEL,
DBM_TO_MBM(dbm)))
goto nla_put_failure;
}
nl80211: send the NL80211_ATTR_SSID in nl80211_send_iface() The userspace may want to know what is the current ssid that a given interface is using. This patch enables nl80211 to send the NL80211_ATTR_SSID attribute in nl80211_send_iface(). Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-07 11:52:20 +00:00
if (wdev->ssid_len) {
if (nla_put(msg, NL80211_ATTR_SSID, wdev->ssid_len, wdev->ssid))
goto nla_put_failure;
}
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nla_put_failure:
netlink: Improve returned error codes Make nlmsg_trim(), nlmsg_cancel(), genlmsg_cancel(), and nla_nest_cancel() void functions. Return -EMSGSIZE instead of -1 if the provided message buffer is not big enough. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-03 23:36:54 +00:00
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
static int nl80211_dump_interface(struct sk_buff *skb, struct netlink_callback *cb)
{
int wp_idx = 0;
int if_idx = 0;
int wp_start = cb->args[0];
int if_start = cb->args[1];
nl80211: Allow GET_INTERFACE dumps to be filtered This patch allows GET_INTERFACE dumps to be filtered based on NL80211_ATTR_WIPHY or NL80211_ATTR_WDEV. The documentation for GET_INTERFACE mentions that this is possible: "Request an interface's configuration; either a dump request on a %NL80211_ATTR_WIPHY or ..." However, this behavior has not been implemented until now. Johannes: rewrite most of the patch: * use nl80211_dump_wiphy_parse() to also allow passing an interface to be able to dump its siblings * fix locking (must hold rtnl around using nl80211_fam.attrbuf) * make init self-contained instead of relying on other cb->args Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 22:02:15 +00:00
int filter_wiphy = -1;
nl80211: add generation number to all dumps In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 14:17:38 +00:00
struct cfg80211_registered_device *rdev;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
struct wireless_dev *wdev;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
int ret;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
rtnl_lock();
nl80211: Allow GET_INTERFACE dumps to be filtered This patch allows GET_INTERFACE dumps to be filtered based on NL80211_ATTR_WIPHY or NL80211_ATTR_WDEV. The documentation for GET_INTERFACE mentions that this is possible: "Request an interface's configuration; either a dump request on a %NL80211_ATTR_WIPHY or ..." However, this behavior has not been implemented until now. Johannes: rewrite most of the patch: * use nl80211_dump_wiphy_parse() to also allow passing an interface to be able to dump its siblings * fix locking (must hold rtnl around using nl80211_fam.attrbuf) * make init self-contained instead of relying on other cb->args Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 22:02:15 +00:00
if (!cb->args[2]) {
struct nl80211_dump_wiphy_state state = {
.filter_wiphy = -1,
};
ret = nl80211_dump_wiphy_parse(skb, cb, &state);
if (ret)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out_unlock;
nl80211: Allow GET_INTERFACE dumps to be filtered This patch allows GET_INTERFACE dumps to be filtered based on NL80211_ATTR_WIPHY or NL80211_ATTR_WDEV. The documentation for GET_INTERFACE mentions that this is possible: "Request an interface's configuration; either a dump request on a %NL80211_ATTR_WIPHY or ..." However, this behavior has not been implemented until now. Johannes: rewrite most of the patch: * use nl80211_dump_wiphy_parse() to also allow passing an interface to be able to dump its siblings * fix locking (must hold rtnl around using nl80211_fam.attrbuf) * make init self-contained instead of relying on other cb->args Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 22:02:15 +00:00
filter_wiphy = state.filter_wiphy;
/*
* if filtering, set cb->args[2] to +1 since 0 is the default
* value needed to determine that parsing is necessary.
*/
if (filter_wiphy >= 0)
cb->args[2] = filter_wiphy + 1;
else
cb->args[2] = -1;
} else if (cb->args[2] > 0) {
filter_wiphy = cb->args[2] - 1;
}
nl80211: add generation number to all dumps In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 14:17:38 +00:00
list_for_each_entry(rdev, &cfg80211_rdev_list, list) {
if (!net_eq(wiphy_net(&rdev->wiphy), sock_net(skb->sk)))
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
continue;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
if (wp_idx < wp_start) {
wp_idx++;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
continue;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
}
nl80211: Allow GET_INTERFACE dumps to be filtered This patch allows GET_INTERFACE dumps to be filtered based on NL80211_ATTR_WIPHY or NL80211_ATTR_WDEV. The documentation for GET_INTERFACE mentions that this is possible: "Request an interface's configuration; either a dump request on a %NL80211_ATTR_WIPHY or ..." However, this behavior has not been implemented until now. Johannes: rewrite most of the patch: * use nl80211_dump_wiphy_parse() to also allow passing an interface to be able to dump its siblings * fix locking (must hold rtnl around using nl80211_fam.attrbuf) * make init self-contained instead of relying on other cb->args Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 22:02:15 +00:00
if (filter_wiphy >= 0 && filter_wiphy != rdev->wiphy_idx)
continue;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if_idx = 0;
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) {
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
if (if_idx < if_start) {
if_idx++;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
continue;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
}
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (nl80211_send_iface(skb, NETLINK_CB(cb->skb).portid,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
rdev, wdev, false) < 0) {
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
goto out;
}
if_idx++;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
wp_idx++;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
out:
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cb->args[0] = wp_idx;
cb->args[1] = if_idx;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
ret = skb->len;
out_unlock:
rtnl_unlock();
return ret;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
static int nl80211_get_interface(struct sk_buff *skb, struct genl_info *info)
{
struct sk_buff *msg;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: retrieve interface data by wdev Since soon there will be virtual interfaces that don't have a netdev, use the wdev identifier for the API to retrieve interface data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 15:52:47 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (!msg)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -ENOMEM;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (nl80211_send_iface(msg, info->snd_portid, info->snd_seq, 0,
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
rdev, wdev, false) < 0) {
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
nlmsg_free(msg);
return -ENOBUFS;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
genetlink: make netns aware This makes generic netlink network namespace aware. No generic netlink families except for the controller family are made namespace aware, they need to be checked one by one and then set the family->netnsok member to true. A new function genlmsg_multicast_netns() is introduced to allow sending a multicast message in a given namespace, for example when it applies to an object that lives in that namespace, a new function genlmsg_multicast_allns() to send a message to all network namespaces (for objects that do not have an associated netns). The function genlmsg_multicast() is changed to multicast the message in just init_net, which is currently correct for all generic netlink families since they only work in init_net right now. Some will later want to work in all net namespaces because they do not care about the netns at all -- those will have to be converted to use one of the new functions genlmsg_multicast_allns() or genlmsg_multicast_netns() whenever they are made netns aware in some way. After this patch families can easily decide whether or not they should be available in all net namespaces. Many genl families us it for objects not related to networking and should therefore be available in all namespaces, but that will have to be done on a per family basis. Note that this doesn't touch on the checkpoint/restart problem where network namespaces could be used, genl families and multicast groups are numbered globally and I see no easy way of changing that, especially since it must be possible to multicast to all network namespaces for those families that do not care about netns. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-07-10 09:51:34 +00:00
return genlmsg_reply(msg, info);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
nl80211: Add monitor interface configuration flags This allows precise control over what a monitor interface shows. Signed-off-by: Michael Wu <flamingice@sourmilk.net> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-31 18:48:22 +00:00
static const struct nla_policy mntr_flags_policy[NL80211_MNTR_FLAG_MAX + 1] = {
[NL80211_MNTR_FLAG_FCSFAIL] = { .type = NLA_FLAG },
[NL80211_MNTR_FLAG_PLCPFAIL] = { .type = NLA_FLAG },
[NL80211_MNTR_FLAG_CONTROL] = { .type = NLA_FLAG },
[NL80211_MNTR_FLAG_OTHER_BSS] = { .type = NLA_FLAG },
[NL80211_MNTR_FLAG_COOK_FRAMES] = { .type = NLA_FLAG },
cfg80211: support an active monitor interface flag An active monitor interface is one that is used for communication (via injection). It is expected to ACK incoming unicast packets. This is useful for running various 802.11 testing utilities that associate to an AP via injection and manage the state in user space. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-28 11:01:52 +00:00
[NL80211_MNTR_FLAG_ACTIVE] = { .type = NLA_FLAG },
nl80211: Add monitor interface configuration flags This allows precise control over what a monitor interface shows. Signed-off-by: Michael Wu <flamingice@sourmilk.net> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-31 18:48:22 +00:00
};
static int parse_monitor_flags(struct nlattr *nla, u32 *mntrflags)
{
struct nlattr *flags[NL80211_MNTR_FLAG_MAX + 1];
int flag;
*mntrflags = 0;
if (!nla)
return -EINVAL;
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
if (nla_parse_nested(flags, NL80211_MNTR_FLAG_MAX, nla,
mntr_flags_policy, NULL))
nl80211: Add monitor interface configuration flags This allows precise control over what a monitor interface shows. Signed-off-by: Michael Wu <flamingice@sourmilk.net> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-31 18:48:22 +00:00
return -EINVAL;
for (flag = 1; flag <= NL80211_MNTR_FLAG_MAX; flag++)
if (flags[flag])
*mntrflags |= (1<<flag);
cfg80211: move add/change interface monitor flags into params Instead passing both flags, which can be NULL, and vif_params, which are never NULL, move the flags into the vif_params and use BIT(0), which is invalid from userspace, to indicate that the flags were changed. While updating all drivers, fix a small bug in wil6210 where it was setting the flags to 0 instead of leaving them unchanged. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:23:28 +00:00
*mntrflags |= MONITOR_FLAG_CHANGED;
nl80211: Add monitor interface configuration flags This allows precise control over what a monitor interface shows. Signed-off-by: Michael Wu <flamingice@sourmilk.net> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-01-31 18:48:22 +00:00
return 0;
}
cfg80211: refactor nl80211 monitor option parsing Refactor the parsing of monitor flags and the MU-MIMO options. This will allow adding more things cleanly in the future and also allows setting the latter already when creating a monitor interface. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:36:31 +00:00
static int nl80211_parse_mon_options(struct cfg80211_registered_device *rdev,
enum nl80211_iftype type,
struct genl_info *info,
struct vif_params *params)
{
bool change = false;
int err;
if (info->attrs[NL80211_ATTR_MNTR_FLAGS]) {
if (type != NL80211_IFTYPE_MONITOR)
return -EINVAL;
err = parse_monitor_flags(info->attrs[NL80211_ATTR_MNTR_FLAGS],
&params->flags);
if (err)
return err;
change = true;
}
if (params->flags & MONITOR_FLAG_ACTIVE &&
!(rdev->wiphy.features & NL80211_FEATURE_ACTIVE_MONITOR))
return -EOPNOTSUPP;
if (info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]) {
const u8 *mumimo_groups;
u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER;
if (type != NL80211_IFTYPE_MONITOR)
return -EINVAL;
if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag))
return -EOPNOTSUPP;
mumimo_groups =
nla_data(info->attrs[NL80211_ATTR_MU_MIMO_GROUP_DATA]);
/* bits 0 and 63 are reserved and must be zero */
nl80211: correctly validate MU-MIMO groups Since groups 0 and 63 are invalid, we should check for those bits. Note that the 802.11 spec specifies the *bit* order, but the CPU doesn't care about bit order since it can't address bits, so it's always treating BIT(0) as the lowest bit within a byte. Reported-by: Jan Fuchs <jan.fuchs@lancom.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-27 07:13:38 +00:00
if ((mumimo_groups[0] & BIT(0)) ||
(mumimo_groups[VHT_MUMIMO_GROUPS_DATA_LEN - 1] & BIT(7)))
cfg80211: refactor nl80211 monitor option parsing Refactor the parsing of monitor flags and the MU-MIMO options. This will allow adding more things cleanly in the future and also allows setting the latter already when creating a monitor interface. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:36:31 +00:00
return -EINVAL;
params->vht_mumimo_groups = mumimo_groups;
change = true;
}
if (info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR]) {
u32 cap_flag = NL80211_EXT_FEATURE_MU_MIMO_AIR_SNIFFER;
if (type != NL80211_IFTYPE_MONITOR)
return -EINVAL;
if (!wiphy_ext_feature_isset(&rdev->wiphy, cap_flag))
return -EOPNOTSUPP;
params->vht_mumimo_follow_addr =
nla_data(info->attrs[NL80211_ATTR_MU_MIMO_FOLLOW_MAC_ADDR]);
change = true;
}
return change ? 1 : 0;
}
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
static int nl80211_valid_4addr(struct cfg80211_registered_device *rdev,
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
struct net_device *netdev, u8 use_4addr,
enum nl80211_iftype iftype)
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
{
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
if (!use_4addr) {
bridge: use rx_handler_data pointer to store net_bridge_port pointer Register net_bridge_port pointer as rx_handler data pointer. As br_port is removed from struct net_device, another netdev priv_flag is added to indicate the device serves as a bridge port. Also rcuized pointers are now correctly dereferenced in br_fdb.c and in netfilter parts. Signed-off-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-06-15 06:50:45 +00:00
if (netdev && (netdev->priv_flags & IFF_BRIDGE_PORT))
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
return -EBUSY;
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
return 0;
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
}
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
switch (iftype) {
case NL80211_IFTYPE_AP_VLAN:
if (rdev->wiphy.flags & WIPHY_FLAG_4ADDR_AP)
return 0;
break;
case NL80211_IFTYPE_STATION:
if (rdev->wiphy.flags & WIPHY_FLAG_4ADDR_STATION)
return 0;
break;
default:
break;
}
return -EOPNOTSUPP;
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
static int nl80211_set_interface(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
struct vif_params params;
cfg80211: pass netdev to change_virtual_intf If there was a reason I'm passing the ifidx I cannot remember it any more and don't see one now, so let's just pass the pointer itself. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-09 19:04:43 +00:00
int err;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
enum nl80211_iftype otype, ntype;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211: accept no-op interface mode changes When somebody tries to set the interface mode to the existing mode, don't ask the driver but silently accept the setting. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-21 16:07:59 +00:00
bool change = false;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
memset(&params, 0, sizeof(params));
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
otype = ntype = dev->ieee80211_ptr->iftype;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: allow set_interface without type Which then causes no type change. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-16 18:22:09 +00:00
if (info->attrs[NL80211_ATTR_IFTYPE]) {
cfg80211: accept no-op interface mode changes When somebody tries to set the interface mode to the existing mode, don't ask the driver but silently accept the setting. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-21 16:07:59 +00:00
ntype = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
if (otype != ntype)
cfg80211: accept no-op interface mode changes When somebody tries to set the interface mode to the existing mode, don't ask the driver but silently accept the setting. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-21 16:07:59 +00:00
change = true;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (ntype > NL80211_IFTYPE_MAX)
return -EINVAL;
cfg80211: allow set_interface without type Which then causes no type change. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-16 18:22:09 +00:00
}
cfg80211: reject invalid configuration items Reject configuring mesh-id for non-mesh, monitor flags for non-monitor. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-16 18:39:36 +00:00
if (info->attrs[NL80211_ATTR_MESH_ID]) {
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (ntype != NL80211_IFTYPE_MESH_POINT)
return -EINVAL;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
if (netif_running(dev))
return -EBUSY;
wdev_lock(wdev);
BUILD_BUG_ON(IEEE80211_MAX_SSID_LEN !=
IEEE80211_MAX_MESH_ID_LEN);
wdev->mesh_id_up_len =
nla_len(info->attrs[NL80211_ATTR_MESH_ID]);
memcpy(wdev->ssid, nla_data(info->attrs[NL80211_ATTR_MESH_ID]),
wdev->mesh_id_up_len);
wdev_unlock(wdev);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
nl80211: add a parameter for using 4-address frames on virtual interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-10 17:53:10 +00:00
if (info->attrs[NL80211_ATTR_4ADDR]) {
params.use_4addr = !!nla_get_u8(info->attrs[NL80211_ATTR_4ADDR]);
change = true;
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
err = nl80211_valid_4addr(rdev, dev, params.use_4addr, ntype);
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
if (err)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return err;
nl80211: add a parameter for using 4-address frames on virtual interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-10 17:53:10 +00:00
} else {
params.use_4addr = -1;
}
cfg80211: refactor nl80211 monitor option parsing Refactor the parsing of monitor flags and the MU-MIMO options. This will allow adding more things cleanly in the future and also allows setting the latter already when creating a monitor interface. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:36:31 +00:00
err = nl80211_parse_mon_options(rdev, ntype, info, &params);
if (err < 0)
return err;
if (err > 0)
nl80211: Add API to support VHT MU-MIMO air sniffer add API to support VHT MU-MIMO air sniffer. in MU-MIMO there are parallel frames on the air while the HW has only one RX. add the capability to sniff one of the MU-MIMO parallel frames by giving the sniffer additional information so it'll know which of the parallel frames it shall follow. Add attribute - NL80211_ATTR_MU_MIMO_GROUP_DATA - for getting a MU-MIMO groupID in order to monitor packets from that group using VHT MU-MIMO. And add attribute -NL80211_ATTR_MU_MIMO_FOLLOW_ADDR - for passing MAC address to monitor mode. that option will be used by VHT MU-MIMO air sniffer to follow a station according to it's MAC address using VHT MU-MIMO. Signed-off-by: Aviya Erenfeld <aviya.erenfeld@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-05 12:23:08 +00:00
change = true;
cfg80211: support an active monitor interface flag An active monitor interface is one that is used for communication (via injection). It is expected to ACK incoming unicast packets. This is useful for running various 802.11 testing utilities that associate to an AP via injection and manage the state in user space. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-28 11:01:52 +00:00
cfg80211: accept no-op interface mode changes When somebody tries to set the interface mode to the existing mode, don't ask the driver but silently accept the setting. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-21 16:07:59 +00:00
if (change)
cfg80211: move add/change interface monitor flags into params Instead passing both flags, which can be NULL, and vif_params, which are never NULL, move the flags into the vif_params and use BIT(0), which is invalid from userspace, to indicate that the flags were changed. While updating all drivers, fix a small bug in wil6210 where it was setting the flags to 0 instead of leaving them unchanged. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:23:28 +00:00
err = cfg80211_change_iface(rdev, dev, ntype, &params);
cfg80211: accept no-op interface mode changes When somebody tries to set the interface mode to the existing mode, don't ask the driver but silently accept the setting. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-21 16:07:59 +00:00
else
err = 0;
cfg80211: show interface type This patch makes cfg80211 show the interface in the nl80211 information about a specific interface. API users are required to keep the type updated (everything else is fairly complicated) but you will get a warning if you fail to keep it updated. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-16 12:55:09 +00:00
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
if (!err && params.use_4addr != -1)
dev->ieee80211_ptr->use_4addr = params.use_4addr;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
return err;
}
static int nl80211_new_interface(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
struct vif_params params;
cfg80211: use wireless_dev for interface management In order to be able to create P2P Device wdevs, move the virtual interface management over to wireless_dev structures. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:00:26 +00:00
struct wireless_dev *wdev;
cfg80211: always notify userspace of new wireless netdevs This change alters the semantics of NL80211_CMD_NEW_INTERFACE events by always sending this event whenever a new net_device object associated with a wdev is registered. Prior to this change, this event was only sent as a result of NL80211_CMD_NEW_INTERFACE command sent from userspace. This allows userspace to reliably detect new wireless interfaces (e.g. due to hardware hot-plug events, etc). For wdevs created without an associated net_device object (e.g. NL80211_IFTYPE_P2P_DEVICE), the NL80211_CMD_NEW_INTERFACE event is still generated inside the relevant nl80211 command handler. Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 21:58:33 +00:00
struct sk_buff *msg;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
int err;
enum nl80211_iftype type = NL80211_IFTYPE_UNSPECIFIED;
cfg80211: allow userspace to take ownership of interfaces When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-03-24 16:57:27 +00:00
/* to avoid failing a new interface creation due to pending removal */
cfg80211_destroy_ifaces(rdev);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
memset(&params, 0, sizeof(params));
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (!info->attrs[NL80211_ATTR_IFNAME])
return -EINVAL;
if (info->attrs[NL80211_ATTR_IFTYPE]) {
type = nla_get_u32(info->attrs[NL80211_ATTR_IFTYPE]);
if (type > NL80211_IFTYPE_MAX)
return -EINVAL;
}
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
if (!rdev->ops->add_virtual_intf ||
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
!(rdev->wiphy.interface_modes & (1 << type)))
return -EOPNOTSUPP;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
if ((type == NL80211_IFTYPE_P2P_DEVICE || type == NL80211_IFTYPE_NAN ||
cfg80211: support configuring vif mac addr on create This is useful when creating virtual interfaces. Keeps udev from mucking with things it shouldn't, since the default MAC is never seen by udev when specified on the cmd-line during creation. Signed-off-by: Ben Greear <greearb@candelatech.com> [check for feature flag in nl80211 to force drivers to set it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-22 19:23:05 +00:00
rdev->wiphy.features & NL80211_FEATURE_MAC_ON_CREATE) &&
info->attrs[NL80211_ATTR_MAC]) {
nl80211: allow user-space to set address for P2P_DEVICE As per email discussion Jouni Malinen pointed out that: "P2P message exchanges can be executed on the current operating channel of any operation (both P2P and non-P2P station). These can be on 5 GHz and even on 60 GHz (so yes, you _can_ do GO Negotiation on 60 GHz). As an example, it would be possible to receive a GO Negotiation Request frame on a 5 GHz only radio and then to complete GO Negotiation on that band. This can happen both when connected to a P2P group (through client discoverability mechanism) and when connected to a legacy AP (assuming the station receive Probe Request frame from full scan in the beginning of P2P device discovery)." This means that P2P messages can be sent over different radio devices. However, these should use the same P2P device address so it should be able to provision this from user-space. This patch adds a parameter for this to struct vif_params which should only be used during creation of the P2P device interface. Cc: Jouni Malinen <j@w1.fi> Cc: Greg Goldman <ggoldman@broadcom.com> Cc: Jithu Jance <jithu@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [add error checking] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-08 09:17:27 +00:00
nla_memcpy(params.macaddr, info->attrs[NL80211_ATTR_MAC],
ETH_ALEN);
if (!is_valid_ether_addr(params.macaddr))
return -EADDRNOTAVAIL;
}
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
if (info->attrs[NL80211_ATTR_4ADDR]) {
nl80211: add a parameter for using 4-address frames on virtual interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-10 17:53:10 +00:00
params.use_4addr = !!nla_get_u8(info->attrs[NL80211_ATTR_4ADDR]);
cfg80211: disallow bridging managed/adhoc interfaces A number of people have tried to add a wireless interface (in managed mode) to a bridge and then complained that it doesn't work. It cannot work, however, because in 802.11 networks all packets need to be acknowledged and as such need to be sent to the right address. Promiscuous doesn't help here. The wireless address format used for these links has only space for three addresses, the * transmitter, which must be equal to the sender (origin) * receiver (on the wireless medium), which is the AP in the case of managed mode * the recipient (destination), which is on the APs local network segment In an IBSS, it is similar, but the receiver and recipient must match and the third address is used as the BSSID. To avoid such mistakes in the future, disallow adding a wireless interface to a bridge. Felix has recently added a four-address mode to the AP and client side that can be used (after negotiating that it is possible, which must happen out-of-band by setting up both sides) for bridging, so allow that case. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-18 23:56:30 +00:00
err = nl80211_valid_4addr(rdev, NULL, params.use_4addr, type);
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
if (err)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return err;
cfg80211: introduce capability for 4addr mode It's very likely that not many devices will support four-address mode in station or AP mode so introduce capability bits for both modes, set them in mac80211 and check them when userspace tries to use the mode. Also, keep track of 4addr in cfg80211 (wireless_dev) and not in mac80211 any more. mac80211 can also be improved for the VLAN case by not looking at the 4addr flag but maintaining the station pointer for it correctly. However, keep track of use_4addr for station mode in mac80211 to avoid all the derefs. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-19 10:55:19 +00:00
}
nl80211: add a parameter for using 4-address frames on virtual interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-10 17:53:10 +00:00
cfg80211: refactor nl80211 monitor option parsing Refactor the parsing of monitor flags and the MU-MIMO options. This will allow adding more things cleanly in the future and also allows setting the latter already when creating a monitor interface. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:36:31 +00:00
err = nl80211_parse_mon_options(rdev, type, info, &params);
if (err < 0)
return err;
cfg80211: support an active monitor interface flag An active monitor interface is one that is used for communication (via injection). It is expected to ACK incoming unicast packets. This is useful for running various 802.11 testing utilities that associate to an AP via injection and manage the state in user space. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-28 11:01:52 +00:00
nl80211: fix memory leak in monitor flags parsing If monitor flags parsing results in active monitor but that isn't supported, the already allocated message is leaked. Fix this by moving the allocation after this check. Reported-by: Christian Engelmayer <cengelma@gmx.at> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-24 09:56:42 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
wdev = rdev_add_virtual_intf(rdev,
nla_data(info->attrs[NL80211_ATTR_IFNAME]),
cfg80211: move add/change interface monitor flags into params Instead passing both flags, which can be NULL, and vif_params, which are never NULL, move the flags into the vif_params and use BIT(0), which is invalid from userspace, to indicate that the flags were changed. While updating all drivers, fix a small bug in wil6210 where it was setting the flags to 0 instead of leaving them unchanged. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-12 09:23:28 +00:00
NET_NAME_USER, type, &params);
cfg80211: protect fools returning NULL in add_virtual_intf Callback add_virtual_intf is supposed to return ERR_PTR and trying to return NULL results in some "Unable to handle kernel paging request", etc. As it may be complicated to debug & trace, let's catch it (WARN). Signed-off-by: Rafał Miłecki <zajec5@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 17:43:28 +00:00
if (WARN_ON(!wdev)) {
nlmsg_free(msg);
return -EPROTO;
} else if (IS_ERR(wdev)) {
nl80211: send interface after creation After a new virtual interface is created, reply to userspace with a message detailing it so it knows the new wdev identifier. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:05:37 +00:00
nlmsg_free(msg);
cfg80211: use wireless_dev for interface management In order to be able to create P2P Device wdevs, move the virtual interface management over to wireless_dev structures. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:00:26 +00:00
return PTR_ERR(wdev);
nl80211: send interface after creation After a new virtual interface is created, reply to userspace with a message detailing it so it knows the new wdev identifier. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:05:37 +00:00
}
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nl80211: Replace interface socket owner attribute with more generic one Replace NL80211_ATTR_IFACE_SOCKET_OWNER attribute with more generic NL80211_ATTR_SOCKET_OWNER that can be used with other commands that interface creation. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-13 15:25:14 +00:00
if (info->attrs[NL80211_ATTR_SOCKET_OWNER])
cfg80211: allow userspace to take ownership of interfaces When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-03-24 16:57:27 +00:00
wdev->owner_nlportid = info->snd_portid;
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
switch (type) {
case NL80211_IFTYPE_MESH_POINT:
if (!info->attrs[NL80211_ATTR_MESH_ID])
break;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
wdev_lock(wdev);
BUILD_BUG_ON(IEEE80211_MAX_SSID_LEN !=
IEEE80211_MAX_MESH_ID_LEN);
wdev->mesh_id_up_len =
nla_len(info->attrs[NL80211_ATTR_MESH_ID]);
memcpy(wdev->ssid, nla_data(info->attrs[NL80211_ATTR_MESH_ID]),
wdev->mesh_id_up_len);
wdev_unlock(wdev);
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
break;
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
case NL80211_IFTYPE_NAN:
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
case NL80211_IFTYPE_P2P_DEVICE:
/*
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
* P2P Device and NAN do not have a netdev, so don't go
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
* through the netdev notifier and must be added here
*/
mutex_init(&wdev->mtx);
INIT_LIST_HEAD(&wdev->event_list);
spin_lock_init(&wdev->event_lock);
INIT_LIST_HEAD(&wdev->mgmt_registrations);
spin_lock_init(&wdev->mgmt_registrations_lock);
wdev->identifier = ++rdev->wdev_id;
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_add_rcu(&wdev->list, &rdev->wiphy.wdev_list);
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
rdev->devlist_generation++;
break;
default:
break;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
}
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (nl80211_send_iface(msg, info->snd_portid, info->snd_seq, 0,
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
rdev, wdev, false) < 0) {
nl80211: send interface after creation After a new virtual interface is created, reply to userspace with a message detailing it so it knows the new wdev identifier. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:05:37 +00:00
nlmsg_free(msg);
return -ENOBUFS;
}
cfg80211: always notify userspace of new wireless netdevs This change alters the semantics of NL80211_CMD_NEW_INTERFACE events by always sending this event whenever a new net_device object associated with a wdev is registered. Prior to this change, this event was only sent as a result of NL80211_CMD_NEW_INTERFACE command sent from userspace. This allows userspace to reliably detect new wireless interfaces (e.g. due to hardware hot-plug events, etc). For wdevs created without an associated net_device object (e.g. NL80211_IFTYPE_P2P_DEVICE), the NL80211_CMD_NEW_INTERFACE event is still generated inside the relevant nl80211 command handler. Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 21:58:33 +00:00
/*
* For wdevs which have no associated netdev object (e.g. of type
* NL80211_IFTYPE_P2P_DEVICE), emit the NEW_INTERFACE event here.
* For all other types, the event will be generated from the
* netdev notifier
*/
if (!wdev->netdev)
nl80211_notify_iface(rdev, wdev, NL80211_CMD_NEW_INTERFACE);
nl80211: Broadcast CMD_NEW_INTERFACE and CMD_DEL_INTERFACE Let the other listeners being notified when a new or del interface command has been issued, thus reducing later necessary request to be in sync with current context. Signed-off-by: Tomasz Bursztyka <tomasz.bursztyka@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-12 14:26:45 +00:00
nl80211: send interface after creation After a new virtual interface is created, reply to userspace with a message detailing it so it knows the new wdev identifier. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:05:37 +00:00
return genlmsg_reply(msg, info);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
static int nl80211_del_interface(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wireless_dev for interface management In order to be able to create P2P Device wdevs, move the virtual interface management over to wireless_dev structures. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:00:26 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->del_virtual_intf)
return -EOPNOTSUPP;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: use wireless_dev for interface management In order to be able to create P2P Device wdevs, move the virtual interface management over to wireless_dev structures. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:00:26 +00:00
/*
* If we remove a wireless device without a netdev then clear
* user_ptr[1] so that nl80211_post_doit won't dereference it
* to check if it needs to do dev_put(). Otherwise it crashes
* since the wdev has been freed, unlike with a netdev where
* we need the dev_put() for the netdev to really be freed.
*/
if (!wdev->netdev)
info->user_ptr[1] = NULL;
cfg80211: always notify userspace when wireless netdev is removed This change alters the semantics of NL80211_CMD_DEL_INTERFACE events by always sending this event whenever a net_device object associated with a wdev is destroyed. Prior to this change, this event was only emitted as a result of NL80211_CMD_DEL_INTERFACE command sent from userspace. This allows userspace to reliably detect when wireless interfaces have been removed, e.g. due to USB removal events, etc. For wireless device objects without an associated net_device (e.g. NL80211_IFTYPE_P2P_DEVICE), the NL80211_CMD_DEL_INTERFACE event is now generated inside cfg80211_unregister_wdev. Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 21:58:35 +00:00
return rdev_del_virtual_intf(rdev, wdev);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
wireless: Add NoAck per tid support This patch contains the configuration changes in nl80211/cfg80211. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 13:20:43 +00:00
static int nl80211_set_noack_map(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
u16 noack_map;
if (!info->attrs[NL80211_ATTR_NOACK_MAP])
return -EINVAL;
if (!rdev->ops->set_noack_map)
return -EOPNOTSUPP;
noack_map = nla_get_u16(info->attrs[NL80211_ATTR_NOACK_MAP]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_set_noack_map(rdev, dev, noack_map);
wireless: Add NoAck per tid support This patch contains the configuration changes in nl80211/cfg80211. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 13:20:43 +00:00
}
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
struct get_key_cookie {
struct sk_buff *msg;
int error;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
int idx;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
};
static void get_key_callback(void *c, struct key_params *params)
{
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
struct nlattr *key;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
struct get_key_cookie *cookie = c;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((params->key &&
nla_put(cookie->msg, NL80211_ATTR_KEY_DATA,
params->key_len, params->key)) ||
(params->seq &&
nla_put(cookie->msg, NL80211_ATTR_KEY_SEQ,
params->seq_len, params->seq)) ||
(params->cipher &&
nla_put_u32(cookie->msg, NL80211_ATTR_KEY_CIPHER,
params->cipher)))
goto nla_put_failure;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
key = nla_nest_start(cookie->msg, NL80211_ATTR_KEY);
if (!key)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((params->key &&
nla_put(cookie->msg, NL80211_KEY_DATA,
params->key_len, params->key)) ||
(params->seq &&
nla_put(cookie->msg, NL80211_KEY_SEQ,
params->seq_len, params->seq)) ||
(params->cipher &&
nla_put_u32(cookie->msg, NL80211_KEY_CIPHER,
params->cipher)))
goto nla_put_failure;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u8(cookie->msg, NL80211_ATTR_KEY_IDX, cookie->idx))
goto nla_put_failure;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
nla_nest_end(cookie->msg, key);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return;
nla_put_failure:
cookie->error = 1;
}
static int nl80211_get_key(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
u8 key_idx = 0;
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
const u8 *mac_addr = NULL;
bool pairwise;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
struct get_key_cookie cookie = {
.error = 0,
};
void *hdr;
struct sk_buff *msg;
if (info->attrs[NL80211_ATTR_KEY_IDX])
key_idx = nla_get_u8(info->attrs[NL80211_ATTR_KEY_IDX]);
mac80211: 802.11w - Use BIP (AES-128-CMAC) Add mechanism for managing BIP keys (IGTK) and integrate BIP into the TX/RX paths. Signed-off-by: Jouni Malinen <j@w1.fi> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-08 11:32:02 +00:00
if (key_idx > 5)
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return -EINVAL;
if (info->attrs[NL80211_ATTR_MAC])
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
pairwise = !!mac_addr;
if (info->attrs[NL80211_ATTR_KEY_TYPE]) {
u32 kt = nla_get_u32(info->attrs[NL80211_ATTR_KEY_TYPE]);
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (kt >= NUM_NL80211_KEYTYPES)
return -EINVAL;
if (kt != NL80211_KEYTYPE_GROUP &&
kt != NL80211_KEYTYPE_PAIRWISE)
return -EINVAL;
pairwise = kt == NL80211_KEYTYPE_PAIRWISE;
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->get_key)
return -EOPNOTSUPP;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: fix per-station group key get/del and memory leak In case userspace attempts to obtain key information for or delete a unicast key, this is currently erroneously rejected unless the driver sets the WIPHY_FLAG_IBSS_RSN flag. Apparently enough drivers do so it was never noticed. Fix that, and while at it fix a potential memory leak: the error path in the get_key() function was placed after allocating a message but didn't free it - move it to a better place. Luckily admin permissions are needed to call this operation. Cc: stable@vger.kernel.org Fixes: e31b82136d1ad ("cfg80211/mac80211: allow per-station GTKs") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:10:12 +00:00
if (!pairwise && mac_addr && !(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
return -ENOENT;
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!msg)
return -ENOMEM;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
NL80211_CMD_NEW_KEY);
nl80211: nl80211hdr_put() doesn't return an ERR_PTR There are a few places which check nl80211hdr_put() for an ERR_PTR but actually it returns NULL on error and never error values. In nl80211_testmode_dump() the return wasn't checked at all so I have added one. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> [some whitespace changes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-14 11:50:01 +00:00
if (!hdr)
nl80211: fix error path in nl80211_get_key() Coverity pointed out that in the (practically impossible) error case we leak the message - fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:15:12 +00:00
goto nla_put_failure;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
cookie.msg = msg;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
cookie.idx = key_idx;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put_u8(msg, NL80211_ATTR_KEY_IDX, key_idx))
goto nla_put_failure;
if (mac_addr &&
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr))
goto nla_put_failure;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_get_key(rdev, dev, key_idx, pairwise, mac_addr, &cookie,
get_key_callback);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
if (err)
nl80211: Memory leak fixed Potential memory leak via msg pointer in nl80211_get_key() function. Signed-off-by: Niko Jokinen <ext-niko.k.jokinen@nokia.com> Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-15 08:00:53 +00:00
goto free_msg;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
if (cookie.error)
goto nla_put_failure;
genlmsg_end(msg, hdr);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nla_put_failure:
err = -ENOBUFS;
nl80211: Memory leak fixed Potential memory leak via msg pointer in nl80211_get_key() function. Signed-off-by: Niko Jokinen <ext-niko.k.jokinen@nokia.com> Signed-off-by: Luciano Coelho <luciano.coelho@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-15 08:00:53 +00:00
free_msg:
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nlmsg_free(msg);
return err;
}
static int nl80211_set_key(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
struct key_parse key;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
err = nl80211_parse_key(info, &key);
if (err)
return err;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (key.idx < 0)
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return -EINVAL;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
/* only support setting default key */
if (!key.def && !key.defmgmt)
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return -EINVAL;
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
wdev_lock(dev->ieee80211_ptr);
mac80211: 802.11w - Use BIP (AES-128-CMAC) Add mechanism for managing BIP keys (IGTK) and integrate BIP into the TX/RX paths. Signed-off-by: Jouni Malinen <j@w1.fi> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-08 11:32:02 +00:00
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (key.def) {
if (!rdev->ops->set_default_key) {
err = -EOPNOTSUPP;
goto out;
}
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
err = nl80211_key_allowed(dev->ieee80211_ptr);
if (err)
goto out;
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_set_default_key(rdev, dev, key.idx,
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
key.def_uni, key.def_multi);
if (err)
goto out;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wext: refactor Refactor wext to * split out iwpriv handling * split out iwspy handling * split out procfs support * allow cfg80211 to have wireless extensions compat code w/o CONFIG_WIRELESS_EXT After this, drivers need to - select WIRELESS_EXT - for wext support - select WEXT_PRIV - for iwpriv support - select WEXT_SPY - for iwspy support except cfg80211 -- which gets new hooks in wext-core.c and can then get wext handlers without CONFIG_WIRELESS_EXT. Wireless extensions procfs support is auto-selected based on PROC_FS and anything that requires the wext core (i.e. WIRELESS_EXT or CFG80211_WEXT). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-09-29 21:27:28 +00:00
#ifdef CONFIG_CFG80211_WEXT
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
dev->ieee80211_ptr->wext.default_key = key.idx;
#endif
} else {
if (key.def_uni || !key.def_multi) {
err = -EINVAL;
goto out;
}
if (!rdev->ops->set_default_mgmt_key) {
err = -EOPNOTSUPP;
goto out;
}
err = nl80211_key_allowed(dev->ieee80211_ptr);
if (err)
goto out;
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_set_default_mgmt_key(rdev, dev, key.idx);
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
if (err)
goto out;
#ifdef CONFIG_CFG80211_WEXT
dev->ieee80211_ptr->wext.default_mgmt_key = key.idx;
cfg80211: implement wext key handling Move key handling wireless extension ioctls from mac80211 to cfg80211 so that all drivers that implement the cfg80211 operations get wext compatibility. Note that this drops the SIOCGIWENCODE ioctl support for getting IW_ENCODE_RESTRICTED/IW_ENCODE_OPEN. This means that iwconfig will no longer report "Security mode:open" or "Security mode:restricted" for mac80211. However, what we displayed there (the authentication algo used) was actually wrong -- linux/wireless.h states that this setting is meant to differentiate between "Refuse non-encoded packets" and "Accept non-encoded packets". (Combined with "cfg80211: fix a couple of bugs with key ioctls". -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 11:54:58 +00:00
#endif
cfg80211/nl80211: separate unicast/multicast default TX keys Allow userspace to specify that a given key is default only for unicast and/or multicast transmissions. Only WEP keys are for both, WPA/RSN keys set here are GTKs for multicast only. For more future flexibility, allow to specify all combiations. Wireless extensions can only set both so use nl80211; WEP keys (connect keys) must be set as default for both (but 802.1X WEP is still possible). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-09 18:58:59 +00:00
}
out:
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wdev_unlock(dev->ieee80211_ptr);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return err;
}
static int nl80211_new_key(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
struct key_parse key;
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
const u8 *mac_addr = NULL;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
err = nl80211_parse_key(info, &key);
if (err)
return err;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (!key.p.key)
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return -EINVAL;
if (info->attrs[NL80211_ATTR_MAC])
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (key.type == -1) {
if (mac_addr)
key.type = NL80211_KEYTYPE_PAIRWISE;
else
key.type = NL80211_KEYTYPE_GROUP;
}
/* for now */
if (key.type != NL80211_KEYTYPE_PAIRWISE &&
key.type != NL80211_KEYTYPE_GROUP)
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->add_key)
return -EOPNOTSUPP;
cfg80211: add cipher capabilities This adds the necessary code and fields to let drivers specify their cipher capabilities and exports them to userspace. Also update mac80211 to export the ciphers it has. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:14:06 +00:00
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (cfg80211_validate_key_settings(rdev, &key.p, key.idx,
key.type == NL80211_KEYTYPE_PAIRWISE,
mac_addr))
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wdev_lock(dev->ieee80211_ptr);
err = nl80211_key_allowed(dev->ieee80211_ptr);
if (!err)
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_add_key(rdev, dev, key.idx,
key.type == NL80211_KEYTYPE_PAIRWISE,
mac_addr, &key.p);
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wdev_unlock(dev->ieee80211_ptr);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return err;
}
static int nl80211_del_key(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
u8 *mac_addr = NULL;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
struct key_parse key;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
err = nl80211_parse_key(info, &key);
if (err)
return err;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
if (info->attrs[NL80211_ATTR_MAC])
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (key.type == -1) {
if (mac_addr)
key.type = NL80211_KEYTYPE_PAIRWISE;
else
key.type = NL80211_KEYTYPE_GROUP;
}
/* for now */
if (key.type != NL80211_KEYTYPE_PAIRWISE &&
key.type != NL80211_KEYTYPE_GROUP)
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->del_key)
return -EOPNOTSUPP;
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wdev_lock(dev->ieee80211_ptr);
err = nl80211_key_allowed(dev->ieee80211_ptr);
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
nl80211: fix per-station group key get/del and memory leak In case userspace attempts to obtain key information for or delete a unicast key, this is currently erroneously rejected unless the driver sets the WIPHY_FLAG_IBSS_RSN flag. Apparently enough drivers do so it was never noticed. Fix that, and while at it fix a potential memory leak: the error path in the get_key() function was placed after allocating a message but didn't free it - move it to a better place. Luckily admin permissions are needed to call this operation. Cc: stable@vger.kernel.org Fixes: e31b82136d1ad ("cfg80211/mac80211: allow per-station GTKs") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:10:12 +00:00
if (key.type == NL80211_KEYTYPE_GROUP && mac_addr &&
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
!(rdev->wiphy.flags & WIPHY_FLAG_IBSS_RSN))
err = -ENOENT;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (!err)
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_del_key(rdev, dev, key.idx,
key.type == NL80211_KEYTYPE_PAIRWISE,
mac_addr);
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
wext: refactor Refactor wext to * split out iwpriv handling * split out iwspy handling * split out procfs support * allow cfg80211 to have wireless extensions compat code w/o CONFIG_WIRELESS_EXT After this, drivers need to - select WIRELESS_EXT - for wext support - select WEXT_PRIV - for iwpriv support - select WEXT_SPY - for iwspy support except cfg80211 -- which gets new hooks in wext-core.c and can then get wext handlers without CONFIG_WIRELESS_EXT. Wireless extensions procfs support is auto-selected based on PROC_FS and anything that requires the wext core (i.e. WIRELESS_EXT or CFG80211_WEXT). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-09-29 21:27:28 +00:00
#ifdef CONFIG_CFG80211_WEXT
cfg80211: implement wext key handling Move key handling wireless extension ioctls from mac80211 to cfg80211 so that all drivers that implement the cfg80211 operations get wext compatibility. Note that this drops the SIOCGIWENCODE ioctl support for getting IW_ENCODE_RESTRICTED/IW_ENCODE_OPEN. This means that iwconfig will no longer report "Security mode:open" or "Security mode:restricted" for mac80211. However, what we displayed there (the authentication algo used) was actually wrong -- linux/wireless.h states that this setting is meant to differentiate between "Refuse non-encoded packets" and "Accept non-encoded packets". (Combined with "cfg80211: fix a couple of bugs with key ioctls". -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 11:54:58 +00:00
if (!err) {
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
if (key.idx == dev->ieee80211_ptr->wext.default_key)
cfg80211: implement wext key handling Move key handling wireless extension ioctls from mac80211 to cfg80211 so that all drivers that implement the cfg80211 operations get wext compatibility. Note that this drops the SIOCGIWENCODE ioctl support for getting IW_ENCODE_RESTRICTED/IW_ENCODE_OPEN. This means that iwconfig will no longer report "Security mode:open" or "Security mode:restricted" for mac80211. However, what we displayed there (the authentication algo used) was actually wrong -- linux/wireless.h states that this setting is meant to differentiate between "Refuse non-encoded packets" and "Accept non-encoded packets". (Combined with "cfg80211: fix a couple of bugs with key ioctls". -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 11:54:58 +00:00
dev->ieee80211_ptr->wext.default_key = -1;
nl80211: introduce new key attributes We will soon want to nest key attributes into some new attribute for configuring static WEP keys at connect() and ibss_join() time, so we need nested attributes for that. However, key attributes right now are 'global'. This patch thus introduces new nested attributes for the key settings and functions for parsing them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 11:29:08 +00:00
else if (key.idx == dev->ieee80211_ptr->wext.default_mgmt_key)
cfg80211: implement wext key handling Move key handling wireless extension ioctls from mac80211 to cfg80211 so that all drivers that implement the cfg80211 operations get wext compatibility. Note that this drops the SIOCGIWENCODE ioctl support for getting IW_ENCODE_RESTRICTED/IW_ENCODE_OPEN. This means that iwconfig will no longer report "Security mode:open" or "Security mode:restricted" for mac80211. However, what we displayed there (the authentication algo used) was actually wrong -- linux/wireless.h states that this setting is meant to differentiate between "Refuse non-encoded packets" and "Accept non-encoded packets". (Combined with "cfg80211: fix a couple of bugs with key ioctls". -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 11:54:58 +00:00
dev->ieee80211_ptr->wext.default_mgmt_key = -1;
}
#endif
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
wdev_unlock(dev->ieee80211_ptr);
cfg80211: implement wext key handling Move key handling wireless extension ioctls from mac80211 to cfg80211 so that all drivers that implement the cfg80211 operations get wext compatibility. Note that this drops the SIOCGIWENCODE ioctl support for getting IW_ENCODE_RESTRICTED/IW_ENCODE_OPEN. This means that iwconfig will no longer report "Security mode:open" or "Security mode:restricted" for mac80211. However, what we displayed there (the authentication algo used) was actually wrong -- linux/wireless.h states that this setting is meant to differentiate between "Refuse non-encoded packets" and "Accept non-encoded packets". (Combined with "cfg80211: fix a couple of bugs with key ioctls". -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 11:54:58 +00:00
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
return err;
}
cfg80211/nl80211: add API for MAC address ACLs Add API to enable drivers to implement MAC address based access control in AP/P2P GO mode. Capable drivers advertise this capability by setting the maximum number of MAC addresses in such a list in wiphy->max_acl_mac_addrs. An initial ACL may be given to the NL80211_CMD_START_AP command and/or changed later with NL80211_CMD_SET_MAC_ACL. Black- and whitelists are supported, but not simultaneously. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> [rewrite commit log, many cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-18 05:48:45 +00:00
/* This function returns an error or the number of nested attributes */
static int validate_acl_mac_addrs(struct nlattr *nl_attr)
{
struct nlattr *attr;
int n_entries = 0, tmp;
nla_for_each_nested(attr, nl_attr, tmp) {
if (nla_len(attr) != ETH_ALEN)
return -EINVAL;
n_entries++;
}
return n_entries;
}
/*
* This function parses ACL information and allocates memory for ACL data.
* On successful return, the calling function is responsible to free the
* ACL buffer returned by this function.
*/
static struct cfg80211_acl_data *parse_acl_data(struct wiphy *wiphy,
struct genl_info *info)
{
enum nl80211_acl_policy acl_policy;
struct nlattr *attr;
struct cfg80211_acl_data *acl;
int i = 0, n_entries, tmp;
if (!wiphy->max_acl_mac_addrs)
return ERR_PTR(-EOPNOTSUPP);
if (!info->attrs[NL80211_ATTR_ACL_POLICY])
return ERR_PTR(-EINVAL);
acl_policy = nla_get_u32(info->attrs[NL80211_ATTR_ACL_POLICY]);
if (acl_policy != NL80211_ACL_POLICY_ACCEPT_UNLESS_LISTED &&
acl_policy != NL80211_ACL_POLICY_DENY_UNLESS_LISTED)
return ERR_PTR(-EINVAL);
if (!info->attrs[NL80211_ATTR_MAC_ADDRS])
return ERR_PTR(-EINVAL);
n_entries = validate_acl_mac_addrs(info->attrs[NL80211_ATTR_MAC_ADDRS]);
if (n_entries < 0)
return ERR_PTR(n_entries);
if (n_entries > wiphy->max_acl_mac_addrs)
return ERR_PTR(-ENOTSUPP);
acl = kzalloc(sizeof(*acl) + (sizeof(struct mac_address) * n_entries),
GFP_KERNEL);
if (!acl)
return ERR_PTR(-ENOMEM);
nla_for_each_nested(attr, info->attrs[NL80211_ATTR_MAC_ADDRS], tmp) {
memcpy(acl->mac_addrs[i].addr, nla_data(attr), ETH_ALEN);
i++;
}
acl->n_acl_entries = n_entries;
acl->acl_policy = acl_policy;
return acl;
}
static int nl80211_set_mac_acl(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct cfg80211_acl_data *acl;
int err;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
if (!dev->ieee80211_ptr->beacon_interval)
return -EINVAL;
acl = parse_acl_data(&rdev->wiphy, info);
if (IS_ERR(acl))
return PTR_ERR(acl);
err = rdev_set_mac_acl(rdev, dev, acl);
kfree(acl);
return err;
}
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
static u32 rateset_to_mask(struct ieee80211_supported_band *sband,
u8 *rates, u8 rates_len)
{
u8 i;
u32 mask = 0;
for (i = 0; i < rates_len; i++) {
int rate = (rates[i] & 0x7f) * 5;
int ridx;
for (ridx = 0; ridx < sband->n_bitrates; ridx++) {
struct ieee80211_rate *srate =
&sband->bitrates[ridx];
if (rate == srate->bitrate) {
mask |= 1 << ridx;
break;
}
}
if (ridx == sband->n_bitrates)
return 0; /* rate not found */
}
return mask;
}
static bool ht_rateset_to_mask(struct ieee80211_supported_band *sband,
u8 *rates, u8 rates_len,
u8 mcs[IEEE80211_HT_MCS_MASK_LEN])
{
u8 i;
memset(mcs, 0, IEEE80211_HT_MCS_MASK_LEN);
for (i = 0; i < rates_len; i++) {
int ridx, rbit;
ridx = rates[i] / 8;
rbit = BIT(rates[i] % 8);
/* check validity */
if ((ridx < 0) || (ridx >= IEEE80211_HT_MCS_MASK_LEN))
return false;
/* check availability */
if (sband->ht_cap.mcs.rx_mask[ridx] & rbit)
mcs[ridx] |= rbit;
else
return false;
}
return true;
}
static u16 vht_mcs_map_to_mcs_mask(u8 vht_mcs_map)
{
u16 mcs_mask = 0;
switch (vht_mcs_map) {
case IEEE80211_VHT_MCS_NOT_SUPPORTED:
break;
case IEEE80211_VHT_MCS_SUPPORT_0_7:
mcs_mask = 0x00FF;
break;
case IEEE80211_VHT_MCS_SUPPORT_0_8:
mcs_mask = 0x01FF;
break;
case IEEE80211_VHT_MCS_SUPPORT_0_9:
mcs_mask = 0x03FF;
break;
default:
break;
}
return mcs_mask;
}
static void vht_build_mcs_mask(u16 vht_mcs_map,
u16 vht_mcs_mask[NL80211_VHT_NSS_MAX])
{
u8 nss;
for (nss = 0; nss < NL80211_VHT_NSS_MAX; nss++) {
vht_mcs_mask[nss] = vht_mcs_map_to_mcs_mask(vht_mcs_map & 0x03);
vht_mcs_map >>= 2;
}
}
static bool vht_set_mcs_mask(struct ieee80211_supported_band *sband,
struct nl80211_txrate_vht *txrate,
u16 mcs[NL80211_VHT_NSS_MAX])
{
u16 tx_mcs_map = le16_to_cpu(sband->vht_cap.vht_mcs.tx_mcs_map);
u16 tx_mcs_mask[NL80211_VHT_NSS_MAX] = {};
u8 i;
if (!sband->vht_cap.vht_supported)
return false;
memset(mcs, 0, sizeof(u16) * NL80211_VHT_NSS_MAX);
/* Build vht_mcs_mask from VHT capabilities */
vht_build_mcs_mask(tx_mcs_map, tx_mcs_mask);
for (i = 0; i < NL80211_VHT_NSS_MAX; i++) {
if ((tx_mcs_mask[i] & txrate->mcs[i]) == txrate->mcs[i])
mcs[i] = txrate->mcs[i];
else
return false;
}
return true;
}
static const struct nla_policy nl80211_txattr_policy[NL80211_TXRATE_MAX + 1] = {
[NL80211_TXRATE_LEGACY] = { .type = NLA_BINARY,
.len = NL80211_MAX_SUPP_RATES },
[NL80211_TXRATE_HT] = { .type = NLA_BINARY,
.len = NL80211_MAX_SUPP_HT_RATES },
[NL80211_TXRATE_VHT] = { .len = sizeof(struct nl80211_txrate_vht)},
[NL80211_TXRATE_GI] = { .type = NLA_U8 },
};
static int nl80211_parse_tx_bitrate_mask(struct genl_info *info,
struct cfg80211_bitrate_mask *mask)
{
struct nlattr *tb[NL80211_TXRATE_MAX + 1];
struct cfg80211_registered_device *rdev = info->user_ptr[0];
int rem, i;
struct nlattr *tx_rates;
struct ieee80211_supported_band *sband;
u16 vht_tx_mcs_map;
memset(mask, 0, sizeof(*mask));
/* Default to all rates enabled */
for (i = 0; i < NUM_NL80211_BANDS; i++) {
sband = rdev->wiphy.bands[i];
if (!sband)
continue;
mask->control[i].legacy = (1 << sband->n_bitrates) - 1;
memcpy(mask->control[i].ht_mcs,
sband->ht_cap.mcs.rx_mask,
sizeof(mask->control[i].ht_mcs));
if (!sband->vht_cap.vht_supported)
continue;
vht_tx_mcs_map = le16_to_cpu(sband->vht_cap.vht_mcs.tx_mcs_map);
vht_build_mcs_mask(vht_tx_mcs_map, mask->control[i].vht_mcs);
}
/* if no rates are given set it back to the defaults */
if (!info->attrs[NL80211_ATTR_TX_RATES])
goto out;
/* The nested attribute uses enum nl80211_band as the index. This maps
* directly to the enum nl80211_band values used in cfg80211.
*/
BUILD_BUG_ON(NL80211_MAX_SUPP_HT_RATES > IEEE80211_HT_MCS_MASK_LEN * 8);
nla_for_each_nested(tx_rates, info->attrs[NL80211_ATTR_TX_RATES], rem) {
enum nl80211_band band = nla_type(tx_rates);
int err;
if (band < 0 || band >= NUM_NL80211_BANDS)
return -EINVAL;
sband = rdev->wiphy.bands[band];
if (sband == NULL)
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, NL80211_TXRATE_MAX, tx_rates,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_txattr_policy, info->extack);
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
if (err)
return err;
if (tb[NL80211_TXRATE_LEGACY]) {
mask->control[band].legacy = rateset_to_mask(
sband,
nla_data(tb[NL80211_TXRATE_LEGACY]),
nla_len(tb[NL80211_TXRATE_LEGACY]));
if ((mask->control[band].legacy == 0) &&
nla_len(tb[NL80211_TXRATE_LEGACY]))
return -EINVAL;
}
if (tb[NL80211_TXRATE_HT]) {
if (!ht_rateset_to_mask(
sband,
nla_data(tb[NL80211_TXRATE_HT]),
nla_len(tb[NL80211_TXRATE_HT]),
mask->control[band].ht_mcs))
return -EINVAL;
}
if (tb[NL80211_TXRATE_VHT]) {
if (!vht_set_mcs_mask(
sband,
nla_data(tb[NL80211_TXRATE_VHT]),
mask->control[band].vht_mcs))
return -EINVAL;
}
if (tb[NL80211_TXRATE_GI]) {
mask->control[band].gi =
nla_get_u8(tb[NL80211_TXRATE_GI]);
if (mask->control[band].gi > NL80211_TXRATE_FORCE_LGI)
return -EINVAL;
}
if (mask->control[band].legacy == 0) {
/* don't allow empty legacy rates if HT or VHT
* are not even supported.
*/
if (!(rdev->wiphy.bands[band]->ht_cap.ht_supported ||
rdev->wiphy.bands[band]->vht_cap.vht_supported))
return -EINVAL;
for (i = 0; i < IEEE80211_HT_MCS_MASK_LEN; i++)
if (mask->control[band].ht_mcs[i])
goto out;
for (i = 0; i < NL80211_VHT_NSS_MAX; i++)
if (mask->control[band].vht_mcs[i])
goto out;
/* legacy and mcs rates may not be both empty */
return -EINVAL;
}
}
out:
return 0;
}
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
static int validate_beacon_tx_rate(struct cfg80211_registered_device *rdev,
enum nl80211_band band,
struct cfg80211_bitrate_mask *beacon_rate)
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
{
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
u32 count_ht, count_vht, i;
u32 rate = beacon_rate->control[band].legacy;
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
/* Allow only one rate */
if (hweight32(rate) > 1)
return -EINVAL;
count_ht = 0;
for (i = 0; i < IEEE80211_HT_MCS_MASK_LEN; i++) {
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
if (hweight8(beacon_rate->control[band].ht_mcs[i]) > 1) {
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
return -EINVAL;
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
} else if (beacon_rate->control[band].ht_mcs[i]) {
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
count_ht++;
if (count_ht > 1)
return -EINVAL;
}
if (count_ht && rate)
return -EINVAL;
}
count_vht = 0;
for (i = 0; i < NL80211_VHT_NSS_MAX; i++) {
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
if (hweight16(beacon_rate->control[band].vht_mcs[i]) > 1) {
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
return -EINVAL;
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
} else if (beacon_rate->control[band].vht_mcs[i]) {
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
count_vht++;
if (count_vht > 1)
return -EINVAL;
}
if (count_vht && rate)
return -EINVAL;
}
if ((count_ht && count_vht) || (!rate && !count_ht && !count_vht))
return -EINVAL;
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
if (rate &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_BEACON_RATE_LEGACY))
return -EINVAL;
if (count_ht &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_BEACON_RATE_HT))
return -EINVAL;
if (count_vht &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_BEACON_RATE_VHT))
return -EINVAL;
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
return 0;
}
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
static int nl80211_parse_beacon(struct nlattr *attrs[],
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
struct cfg80211_beacon_data *bcn)
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
{
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
bool haveinfo = false;
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (!is_valid_ie_attr(attrs[NL80211_ATTR_BEACON_TAIL]) ||
!is_valid_ie_attr(attrs[NL80211_ATTR_IE]) ||
!is_valid_ie_attr(attrs[NL80211_ATTR_IE_PROBE_RESP]) ||
!is_valid_ie_attr(attrs[NL80211_ATTR_IE_ASSOC_RESP]))
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
return -EINVAL;
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
memset(bcn, 0, sizeof(*bcn));
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_BEACON_HEAD]) {
bcn->head = nla_data(attrs[NL80211_ATTR_BEACON_HEAD]);
bcn->head_len = nla_len(attrs[NL80211_ATTR_BEACON_HEAD]);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
if (!bcn->head_len)
return -EINVAL;
haveinfo = true;
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
}
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_BEACON_TAIL]) {
bcn->tail = nla_data(attrs[NL80211_ATTR_BEACON_TAIL]);
bcn->tail_len = nla_len(attrs[NL80211_ATTR_BEACON_TAIL]);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
haveinfo = true;
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!haveinfo)
return -EINVAL;
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_IE]) {
bcn->beacon_ies = nla_data(attrs[NL80211_ATTR_IE]);
bcn->beacon_ies_len = nla_len(attrs[NL80211_ATTR_IE]);
nl80211/cfg80211: Add extra IE configuration to AP mode setup The NL80211_CMD_NEW_BEACON command is, in practice, requesting AP mode operations to be started. Add new attributes to provide extra IEs (e.g., WPS IE, P2P IE) for drivers that build Beacon, Probe Response, and (Re)Association Response frames internally (likely in firmware). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:55:56 +00:00
}
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_IE_PROBE_RESP]) {
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
bcn->proberesp_ies =
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
nla_data(attrs[NL80211_ATTR_IE_PROBE_RESP]);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
bcn->proberesp_ies_len =
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
nla_len(attrs[NL80211_ATTR_IE_PROBE_RESP]);
nl80211/cfg80211: Add extra IE configuration to AP mode setup The NL80211_CMD_NEW_BEACON command is, in practice, requesting AP mode operations to be started. Add new attributes to provide extra IEs (e.g., WPS IE, P2P IE) for drivers that build Beacon, Probe Response, and (Re)Association Response frames internally (likely in firmware). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:55:56 +00:00
}
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_IE_ASSOC_RESP]) {
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
bcn->assocresp_ies =
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
nla_data(attrs[NL80211_ATTR_IE_ASSOC_RESP]);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
bcn->assocresp_ies_len =
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
nla_len(attrs[NL80211_ATTR_IE_ASSOC_RESP]);
nl80211/cfg80211: Add extra IE configuration to AP mode setup The NL80211_CMD_NEW_BEACON command is, in practice, requesting AP mode operations to be started. Add new attributes to provide extra IEs (e.g., WPS IE, P2P IE) for drivers that build Beacon, Probe Response, and (Re)Association Response frames internally (likely in firmware). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-10 20:55:56 +00:00
}
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
if (attrs[NL80211_ATTR_PROBE_RESP]) {
bcn->probe_resp = nla_data(attrs[NL80211_ATTR_PROBE_RESP]);
bcn->probe_resp_len = nla_len(attrs[NL80211_ATTR_PROBE_RESP]);
nl80211: Pass probe response data to drivers Pass probe-response data from usermode via beacon parameters. Signed-off-by: Guy Eilam <guy@wizery.com> Signed-off-by: Arik Nemtsov <arik@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-10 09:28:56 +00:00
}
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
return 0;
}
nl80211: add HT/VHT capabilities to AP parameters For the benefit of drivers that rebuild IEs in firmware, parse the IEs for HT/VHT capabilities and the respective membership selector in the (extended) supported rates. This avoids duplicating the same code into all drivers that need this information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-07 20:40:44 +00:00
static void nl80211_check_ap_rate_selectors(struct cfg80211_ap_settings *params,
const u8 *rates)
{
int i;
if (!rates)
return;
for (i = 0; i < rates[1]; i++) {
if (rates[2 + i] == BSS_MEMBERSHIP_SELECTOR_HT_PHY)
params->ht_required = true;
if (rates[2 + i] == BSS_MEMBERSHIP_SELECTOR_VHT_PHY)
params->vht_required = true;
}
}
/*
* Since the nl80211 API didn't include, from the beginning, attributes about
* HT/VHT requirements/capabilities, we parse them out of the IEs for the
* benefit of drivers that rebuild IEs in the firmware.
*/
static void nl80211_calculate_ap_params(struct cfg80211_ap_settings *params)
{
const struct cfg80211_beacon_data *bcn = &params->beacon;
size_t ies_len = bcn->beacon_ies_len;
const u8 *ies = bcn->beacon_ies;
const u8 *rates;
const u8 *cap;
rates = cfg80211_find_ie(WLAN_EID_SUPP_RATES, ies, ies_len);
nl80211_check_ap_rate_selectors(params, rates);
rates = cfg80211_find_ie(WLAN_EID_EXT_SUPP_RATES, ies, ies_len);
nl80211_check_ap_rate_selectors(params, rates);
cap = cfg80211_find_ie(WLAN_EID_HT_CAPABILITY, ies, ies_len);
if (cap && cap[1] >= sizeof(*params->ht_cap))
params->ht_cap = (void *)(cap + 2);
cap = cfg80211_find_ie(WLAN_EID_VHT_CAPABILITY, ies, ies_len);
if (cap && cap[1] >= sizeof(*params->vht_cap))
params->vht_cap = (void *)(cap + 2);
}
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
static bool nl80211_get_ap_channel(struct cfg80211_registered_device *rdev,
struct cfg80211_ap_settings *params)
{
struct wireless_dev *wdev;
bool ret = false;
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(wdev, &rdev->wiphy.wdev_list, list) {
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
if (wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO)
continue;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (!wdev->preset_chandef.chan)
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
continue;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
params->chandef = wdev->preset_chandef;
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
ret = true;
break;
}
return ret;
}
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
static bool nl80211_valid_auth_type(struct cfg80211_registered_device *rdev,
enum nl80211_auth_type auth_type,
enum nl80211_commands cmd)
{
if (auth_type > NL80211_AUTHTYPE_MAX)
return false;
switch (cmd) {
case NL80211_CMD_AUTHENTICATE:
if (!(rdev->wiphy.features & NL80211_FEATURE_SAE) &&
auth_type == NL80211_AUTHTYPE_SAE)
return false;
cfg80211: Add Fast Initial Link Setup (FILS) auth algs This defines authentication algorithms for FILS (IEEE 802.11ai). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:02 +00:00
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_FILS_STA) &&
(auth_type == NL80211_AUTHTYPE_FILS_SK ||
auth_type == NL80211_AUTHTYPE_FILS_SK_PFS ||
auth_type == NL80211_AUTHTYPE_FILS_PK))
return false;
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
return true;
case NL80211_CMD_CONNECT:
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
/* SAE not supported yet */
if (auth_type == NL80211_AUTHTYPE_SAE)
return false;
/* FILS with SK PFS or PK not supported yet */
if (auth_type == NL80211_AUTHTYPE_FILS_SK_PFS ||
auth_type == NL80211_AUTHTYPE_FILS_PK)
return false;
if (!wiphy_ext_feature_isset(
&rdev->wiphy,
NL80211_EXT_FEATURE_FILS_SK_OFFLOAD) &&
auth_type == NL80211_AUTHTYPE_FILS_SK)
return false;
return true;
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
case NL80211_CMD_START_AP:
/* SAE not supported yet */
if (auth_type == NL80211_AUTHTYPE_SAE)
return false;
cfg80211: Add Fast Initial Link Setup (FILS) auth algs This defines authentication algorithms for FILS (IEEE 802.11ai). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:02 +00:00
/* FILS not supported yet */
if (auth_type == NL80211_AUTHTYPE_FILS_SK ||
auth_type == NL80211_AUTHTYPE_FILS_SK_PFS ||
auth_type == NL80211_AUTHTYPE_FILS_PK)
return false;
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
return true;
default:
return false;
}
}
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
static int nl80211_start_ap(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_ap_settings params;
int err;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
if (!rdev->ops->start_ap)
return -EOPNOTSUPP;
if (wdev->beacon_interval)
return -EALREADY;
memset(&params, 0, sizeof(params));
/* these are required for START_AP */
if (!info->attrs[NL80211_ATTR_BEACON_INTERVAL] ||
!info->attrs[NL80211_ATTR_DTIM_PERIOD] ||
!info->attrs[NL80211_ATTR_BEACON_HEAD])
return -EINVAL;
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
err = nl80211_parse_beacon(info->attrs, &params.beacon);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
if (err)
return err;
params.beacon_interval =
nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]);
params.dtim_period =
nla_get_u32(info->attrs[NL80211_ATTR_DTIM_PERIOD]);
cfg80211: support virtual interfaces with different beacon intervals This commit provides a mechanism for the host drivers to advertise the support for different beacon intervals among the respective interface combinations in a group, through NL80211_IFACE_COMB_BI_MIN_GCD (u32). This value will be compared against GCD of all beaconing interfaces of matching combinations. If the driver doesn't advertise this value, the old behaviour where all beacon intervals must be identical is retained. If it is specified, then any beacon interval for an interface in the interface combination as well as the GCD of all active beacon intervals in the combination must be greater or equal to this value. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [change commit message, some variable names, small other things] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-12 12:56:51 +00:00
err = cfg80211_validate_beacon_int(rdev, dev->ieee80211_ptr->iftype,
params.beacon_interval);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
if (err)
return err;
/*
* In theory, some of these attributes should be required here
* but since they were not used when the command was originally
* added, keep them optional for old user space programs to let
* them continue to work with drivers that do not need the
* additional information -- drivers must check!
*/
if (info->attrs[NL80211_ATTR_SSID]) {
params.ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
params.ssid_len =
nla_len(info->attrs[NL80211_ATTR_SSID]);
if (params.ssid_len == 0 ||
params.ssid_len > IEEE80211_MAX_SSID_LEN)
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_HIDDEN_SSID]) {
params.hidden_ssid = nla_get_u32(
info->attrs[NL80211_ATTR_HIDDEN_SSID]);
if (params.hidden_ssid != NL80211_HIDDEN_SSID_NOT_IN_USE &&
params.hidden_ssid != NL80211_HIDDEN_SSID_ZERO_LEN &&
params.hidden_ssid != NL80211_HIDDEN_SSID_ZERO_CONTENTS)
return -EINVAL;
}
params.privacy = !!info->attrs[NL80211_ATTR_PRIVACY];
if (info->attrs[NL80211_ATTR_AUTH_TYPE]) {
params.auth_type = nla_get_u32(
info->attrs[NL80211_ATTR_AUTH_TYPE]);
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
if (!nl80211_valid_auth_type(rdev, params.auth_type,
NL80211_CMD_START_AP))
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
return -EINVAL;
} else
params.auth_type = NL80211_AUTHTYPE_AUTOMATIC;
err = nl80211_crypto_settings(rdev, info, &params.crypto,
NL80211_MAX_NR_CIPHER_SUITES);
if (err)
return err;
cfg80211: Add an attribute to set inactivity timeout in AP mode This patch adds an attribute, NL80211_ATTR_INACTIVITY_TIMEOUT, to set the inactivity timeout which can be used to remove the station in AP mode. This can be passed in NL80211_CMD_START_AP and used by the drivers which have AP MLME in firmware but don't support get_station() properly. To disable inactivity timer in userspace, wpa_s for example, there is a new flag, NL80211_FEATURE_INACTIVITY_TIMER, in nl80211_feature_flags through which drivers can register their capability to use the inactivity timeout to free the stations. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-02 10:20:02 +00:00
if (info->attrs[NL80211_ATTR_INACTIVITY_TIMEOUT]) {
if (!(rdev->wiphy.features & NL80211_FEATURE_INACTIVITY_TIMER))
return -EOPNOTSUPP;
params.inactivity_timeout = nla_get_u16(
info->attrs[NL80211_ATTR_INACTIVITY_TIMEOUT]);
}
nl80211: support P2P GO powersave configuration If a driver supports P2P GO powersave, allow it to set the new feature flags for it and allow userspace to configure the parameters for it. This can be done at GO startup and later changed with SET_BSS. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-14 14:17:28 +00:00
if (info->attrs[NL80211_ATTR_P2P_CTWINDOW]) {
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
params.p2p_ctwindow =
nla_get_u8(info->attrs[NL80211_ATTR_P2P_CTWINDOW]);
if (params.p2p_ctwindow > 127)
return -EINVAL;
if (params.p2p_ctwindow != 0 &&
!(rdev->wiphy.features & NL80211_FEATURE_P2P_GO_CTWIN))
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_P2P_OPPPS]) {
u8 tmp;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
tmp = nla_get_u8(info->attrs[NL80211_ATTR_P2P_OPPPS]);
if (tmp > 1)
return -EINVAL;
params.p2p_opp_ps = tmp;
if (params.p2p_opp_ps != 0 &&
!(rdev->wiphy.features & NL80211_FEATURE_P2P_GO_OPPPS))
return -EINVAL;
}
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) {
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
err = nl80211_parse_chandef(rdev, info, &params.chandef);
if (err)
return err;
} else if (wdev->preset_chandef.chan) {
params.chandef = wdev->preset_chandef;
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
} else if (!nl80211_get_ap_channel(rdev, &params))
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
return -EINVAL;
cfg80211: use RTNL locked reg_can_beacon for IR-relaxation The RTNL is required to check for IR-relaxation conditions that allow more channels to beacon. Export an RTNL locked version of reg_can_beacon and use it where possible in AP/STA interface type flows, where IR-relaxation may be applicable. Fixes: 06f207fc5418 ("cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA") Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-07-08 12:41:44 +00:00
if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &params.chandef,
wdev->iftype))
cfg80211: provide channel to start_ap function Instead of setting the channel first and then starting the AP, let cfg80211 store the channel and provide it as one of the AP settings. This means that now you have to set the channel before you can start an AP interface, but since hostapd/wpa_supplicant always do that we're OK with this change. Alternatively, it's now possible to give the channel as an attribute to the start-ap nl80211 command, overriding any preset channel. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:18 +00:00
return -EINVAL;
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
if (info->attrs[NL80211_ATTR_TX_RATES]) {
err = nl80211_parse_tx_bitrate_mask(info, &params.beacon_rate);
if (err)
return err;
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
err = validate_beacon_tx_rate(rdev, params.chandef.chan->band,
&params.beacon_rate);
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
if (err)
return err;
}
cfg80211: allow requesting SMPS mode on ap start Add feature bits to indicate device support for static-smps and dynamic-smps modes. Add a new NL80211_ATTR_SMPS_MODE attribue to allow configuring the smps mode to be used by the ap (e.g. configuring to ap to dynamic smps mode will reduce power consumption while having minor effect on throughput) Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 11:07:34 +00:00
if (info->attrs[NL80211_ATTR_SMPS_MODE]) {
params.smps_mode =
nla_get_u8(info->attrs[NL80211_ATTR_SMPS_MODE]);
switch (params.smps_mode) {
case NL80211_SMPS_OFF:
break;
case NL80211_SMPS_STATIC:
if (!(rdev->wiphy.features &
NL80211_FEATURE_STATIC_SMPS))
return -EINVAL;
break;
case NL80211_SMPS_DYNAMIC:
if (!(rdev->wiphy.features &
NL80211_FEATURE_DYNAMIC_SMPS))
return -EINVAL;
break;
default:
return -EINVAL;
}
} else {
params.smps_mode = NL80211_SMPS_OFF;
}
nl80211: Move ACL parsing later to avoid a possible memory leak No support for pbss results in a memory leak for the acl_data (if parse_acl_data succeeds). Fix this by moving the ACL parsing later. Cc: stable@vger.kernel.org Fixes: 34d505193bd10 ("cfg80211: basic support for PBSS network type") Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-07-05 08:14:51 +00:00
params.pbss = nla_get_flag(info->attrs[NL80211_ATTR_PBSS]);
if (params.pbss && !rdev->wiphy.bands[NL80211_BAND_60GHZ])
return -EOPNOTSUPP;
nl80211: Fix potential memory leak from parse_acl_data If parse_acl_data succeeds but the subsequent parsing of smps attributes fails, there will be a memory leak due to early returns. Fix that by moving the ACL parsing later. Cc: stable@vger.kernel.org Fixes: 18998c381b19b ("cfg80211: allow requesting SMPS mode on ap start") Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-29 06:04:58 +00:00
if (info->attrs[NL80211_ATTR_ACL_POLICY]) {
params.acl = parse_acl_data(&rdev->wiphy, info);
if (IS_ERR(params.acl))
return PTR_ERR(params.acl);
}
nl80211: add HT/VHT capabilities to AP parameters For the benefit of drivers that rebuild IEs in firmware, parse the IEs for HT/VHT capabilities and the respective membership selector in the (extended) supported rates. This avoids duplicating the same code into all drivers that need this information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-07 20:40:44 +00:00
nl80211_calculate_ap_params(&params);
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
wdev_lock(wdev);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_start_ap(rdev, dev, &params);
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
if (!err) {
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
wdev->preset_chandef = params.chandef;
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
wdev->beacon_interval = params.beacon_interval;
cfg80211: consider existing DFS interfaces It was possible to break interface combinations in the following way: combo 1: iftype = AP, num_ifaces = 2, num_chans = 2, combo 2: iftype = AP, num_ifaces = 1, num_chans = 1, radar = HT20 With the above interface combinations it was possible to: step 1. start AP on DFS channel by matching combo 2 step 2. start AP on non-DFS channel by matching combo 1 This was possible beacuse (step 2) did not consider if other interfaces require radar detection. The patch changes how cfg80211 tracks channels - instead of channel itself now a complete chandef is stored. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-29 13:22:27 +00:00
wdev->chandef = params.chandef;
cfg80211: store the ssid into wirless_dev in AP mode Store the configured ssid in wdev->ssid when starting an AP Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-07 11:52:19 +00:00
wdev->ssid_len = params.ssid_len;
memcpy(wdev->ssid, params.ssid, wdev->ssid_len);
cfg80211: fix regression in multi-vif AP start Commit "cfg80211: provide channel to start_ap function" assumes that the channel is always passed to the NL80211_CMD_START_AP command, however in case of multi-BSSID, hostapd only passes the channel for the first vif. This makes starting beaconing on secondary vifs fail with -EINVAL. Fix this by storing the channel provided to .start_ap in wdev->preset_chan and picking the first AP vif's channel for secondary vifs if not provided. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-19 00:50:57 +00:00
}
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
wdev_unlock(wdev);
cfg80211/nl80211: add API for MAC address ACLs Add API to enable drivers to implement MAC address based access control in AP/P2P GO mode. Capable drivers advertise this capability by setting the maximum number of MAC addresses in such a list in wiphy->max_acl_mac_addrs. An initial ACL may be given to the NL80211_CMD_START_AP command and/or changed later with NL80211_CMD_SET_MAC_ACL. Black- and whitelists are supported, but not simultaneously. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> [rewrite commit log, many cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-18 05:48:45 +00:00
kfree(params.acl);
cfg80211: restrict AP beacon intervals Multiple virtual AP interfaces can currently try to use different beacon intervals, but that just leads to problems since it won't actually be done that way by drivers. Return an error in this case to make sure it won't be done wrong. Also, ignore attempts to change the DTIM period or beacon interval during the lifetime of the BSS. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-09 16:41:15 +00:00
return err;
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
}
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
static int nl80211_set_beacon(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_beacon_data params;
int err;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
if (!rdev->ops->change_beacon)
return -EOPNOTSUPP;
if (!wdev->beacon_interval)
return -EINVAL;
nl80211: use attributes to parse beacons only the attributes are required and not the whole netlink info, as the function accesses the attributes only anyway. This makes it easier to parse nested beacon IEs later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-14 12:15:19 +00:00
err = nl80211_parse_beacon(info->attrs, &params);
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
if (err)
return err;
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
wdev_lock(wdev);
err = rdev_change_beacon(rdev, dev, &params);
wdev_unlock(wdev);
return err;
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
}
static int nl80211_stop_ap(struct sk_buff *skb, struct genl_info *info)
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
cfg80211: send stop AP event only due to internal reason Commit "nl80211: send event when AP operation is stopped" added an event to notify user space that an AP interface has been stopped, to handle cases such as suspend etc. The event is sent regardless if the stop AP flow was triggered by user space or due to internal state change. This might cause issues with wpa_supplicant/hostapd flows that consider stop AP flow as a synchronous one, e.g., AP/GO channel change in the absence of CSA support. In such cases, the flow will restart the AP immediately after the stop AP flow is done, and only handle the stop AP event after the current flow is done, and as a result stop the AP again. Change the current implementation to only send the event in case the stop AP was triggered due to an internal reason. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-25 13:33:38 +00:00
return cfg80211_stop_ap(rdev, dev, false);
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
}
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
static const struct nla_policy sta_flags_policy[NL80211_STA_FLAG_MAX + 1] = {
[NL80211_STA_FLAG_AUTHORIZED] = { .type = NLA_FLAG },
[NL80211_STA_FLAG_SHORT_PREAMBLE] = { .type = NLA_FLAG },
[NL80211_STA_FLAG_WME] = { .type = NLA_FLAG },
nl80211: Validate MFP flag type when parsing STA flags NL80211_STA_FLAG_MFP was forgotten from sta_flags_policy. The previous version added the flag due to the loop used in parse_station_flags, but the proper behavior would be to allow nla_parse_nested() to go through the policy for all flags. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:55 +00:00
[NL80211_STA_FLAG_MFP] = { .type = NLA_FLAG },
nl80211/mac80211: let userspace authenticate stations Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:30 +00:00
[NL80211_STA_FLAG_AUTHENTICATED] = { .type = NLA_FLAG },
nl80211: add TDLS peer flag to policy This was evidently missed in the TDLS patch (07ba55d7). Cc: Arik Nemtsov <arik@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 08:29:15 +00:00
[NL80211_STA_FLAG_TDLS_PEER] = { .type = NLA_FLAG },
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
};
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
static int parse_station_flags(struct genl_info *info,
nl80211: fix old station flags compatibility My patch to validate station flags broke compatibility with the old station flags setting where all flags are always set at once since it always set the mask as all possible flags which ended up being rejected later in the station add/modify code. Fix by parsing only the current flags in the old flags attribute -- new applications and new flags should use (and will now require) the new flags attribute where the mask is given by the application. Reported-and-tested-by: Thomas Hilber <ath9k-dev@toh.cx> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-02 12:30:03 +00:00
enum nl80211_iftype iftype,
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
struct station_parameters *params)
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
{
struct nlattr *flags[NL80211_STA_FLAG_MAX + 1];
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
struct nlattr *nla;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
int flag;
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
/*
* Try parsing the new attribute first so userspace
* can specify both for older kernels.
*/
nla = info->attrs[NL80211_ATTR_STA_FLAGS2];
if (nla) {
struct nl80211_sta_flag_update *sta_flags;
sta_flags = nla_data(nla);
params->sta_flags_mask = sta_flags->mask;
params->sta_flags_set = sta_flags->set;
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
params->sta_flags_set &= params->sta_flags_mask;
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
if ((params->sta_flags_mask |
params->sta_flags_set) & BIT(__NL80211_STA_FLAG_INVALID))
return -EINVAL;
return 0;
}
/* if present, parse the old attribute */
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
nla = info->attrs[NL80211_ATTR_STA_FLAGS];
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (!nla)
return 0;
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
if (nla_parse_nested(flags, NL80211_STA_FLAG_MAX, nla,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
sta_flags_policy, info->extack))
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
return -EINVAL;
nl80211: fix old station flags compatibility My patch to validate station flags broke compatibility with the old station flags setting where all flags are always set at once since it always set the mask as all possible flags which ended up being rejected later in the station add/modify code. Fix by parsing only the current flags in the old flags attribute -- new applications and new flags should use (and will now require) the new flags attribute where the mask is given by the application. Reported-and-tested-by: Thomas Hilber <ath9k-dev@toh.cx> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-02 12:30:03 +00:00
/*
* Only allow certain flags for interface types so that
* other attributes are silently ignored. Remember that
* this is backward compatibility code with old userspace
* and shouldn't be hit in other cases anyway.
*/
switch (iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_P2P_GO:
params->sta_flags_mask = BIT(NL80211_STA_FLAG_AUTHORIZED) |
BIT(NL80211_STA_FLAG_SHORT_PREAMBLE) |
BIT(NL80211_STA_FLAG_WME) |
BIT(NL80211_STA_FLAG_MFP);
break;
case NL80211_IFTYPE_P2P_CLIENT:
case NL80211_IFTYPE_STATION:
params->sta_flags_mask = BIT(NL80211_STA_FLAG_AUTHORIZED) |
BIT(NL80211_STA_FLAG_TDLS_PEER);
break;
case NL80211_IFTYPE_MESH_POINT:
params->sta_flags_mask = BIT(NL80211_STA_FLAG_AUTHENTICATED) |
BIT(NL80211_STA_FLAG_MFP) |
BIT(NL80211_STA_FLAG_AUTHORIZED);
default:
return -EINVAL;
}
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
nl80211: prevent additions to old station flags API We don't really want/need to maintain the old station flags API any more, so refuse changes to new (not yet defined) flags from the old flags API. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-10 18:14:43 +00:00
for (flag = 1; flag <= NL80211_STA_FLAG_MAX; flag++) {
if (flags[flag]) {
nl80211: improve station flags handling It is currently not possible to modify station flags, but that capability would be very useful. This patch introduces a new nl80211 attribute that contains a set/mask for station flags, and updates the internal API (and mac80211) to mirror that. The new attribute is parsed before falling back to the old so that userspace can specify both (if it can) to work on all kernels. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 18:57:56 +00:00
params->sta_flags_set |= (1<<flag);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
nl80211: prevent additions to old station flags API We don't really want/need to maintain the old station flags API any more, so refuse changes to new (not yet defined) flags from the old flags API. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-10 18:14:43 +00:00
/* no longer support new API additions in old API */
if (flag > NL80211_STA_FLAG_MAX_OLD_API)
return -EINVAL;
}
}
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
return 0;
}
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
static bool nl80211_put_sta_rate(struct sk_buff *msg, struct rate_info *info,
int attr)
{
struct nlattr *rate;
{nl,cfg}80211: support high bitrates Until now, a u16 value was used to represent bitrate value. With VHT bitrates this becomes too small. Introduce a new 32-bit bitrate attribute. nl80211 will report both the new and the old attribute, unless the bitrate doesn't fit into the old u16 attribute in which case only the new one will be reported. User space tools encouraged to prefer the 32-bit attribute, if available (since it won't be available on older kernels.) Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [reword commit message and comments a bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-05 11:25:49 +00:00
u32 bitrate;
u16 bitrate_compat;
nl80211: Fix enum type of variable in nl80211_put_sta_rate() rate_flg is of type 'enum nl80211_attrs', however it is assigned with 'enum nl80211_rate_info' values. Change the type of rate_flg accordingly. Signed-off-by: Matthias Kaehlcke <mka@chromium.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-17 22:59:52 +00:00
enum nl80211_rate_info rate_flg;
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
rate = nla_nest_start(msg, attr);
if (!rate)
nl80211/cfg80211: add VHT MCS support Add support for reporting and calculating VHT MCSes. Note that I'm not completely sure that the bitrate calculations are correct, nor that they can't be simplified. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-09 13:56:41 +00:00
return false;
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
/* cfg80211_calculate_bitrate will return 0 for mcs >= 32 */
bitrate = cfg80211_calculate_bitrate(info);
{nl,cfg}80211: support high bitrates Until now, a u16 value was used to represent bitrate value. With VHT bitrates this becomes too small. Introduce a new 32-bit bitrate attribute. nl80211 will report both the new and the old attribute, unless the bitrate doesn't fit into the old u16 attribute in which case only the new one will be reported. User space tools encouraged to prefer the 32-bit attribute, if available (since it won't be available on older kernels.) Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [reword commit message and comments a bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-05 11:25:49 +00:00
/* report 16-bit bitrate only if we can */
bitrate_compat = bitrate < (1UL << 16) ? bitrate : 0;
nl80211/cfg80211: add VHT MCS support Add support for reporting and calculating VHT MCSes. Note that I'm not completely sure that the bitrate calculations are correct, nor that they can't be simplified. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-09 13:56:41 +00:00
if (bitrate > 0 &&
nla_put_u32(msg, NL80211_RATE_INFO_BITRATE32, bitrate))
return false;
if (bitrate_compat > 0 &&
nla_put_u16(msg, NL80211_RATE_INFO_BITRATE, bitrate_compat))
return false;
cfg80211: change bandwidth reporting to explicit field For some reason, we made the bandwidth separate flags, which is rather confusing - a single rate cannot have different bandwidths at the same time. Change this to no longer be flags but use a separate field for the bandwidth ('bw') instead. While at it, add support for 5 and 10 MHz rates - these are reported as regular legacy rates with their real bitrate, but tagged as 5/10 now to make it easier to distinguish them. In the nl80211 API, the flags are preserved, but the code now can also clearly only set a single one of the flags. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-15 15:14:02 +00:00
switch (info->bw) {
case RATE_INFO_BW_5:
rate_flg = NL80211_RATE_INFO_5_MHZ_WIDTH;
break;
case RATE_INFO_BW_10:
rate_flg = NL80211_RATE_INFO_10_MHZ_WIDTH;
break;
default:
WARN_ON(1);
/* fall through */
case RATE_INFO_BW_20:
rate_flg = 0;
break;
case RATE_INFO_BW_40:
rate_flg = NL80211_RATE_INFO_40_MHZ_WIDTH;
break;
case RATE_INFO_BW_80:
rate_flg = NL80211_RATE_INFO_80_MHZ_WIDTH;
break;
case RATE_INFO_BW_160:
rate_flg = NL80211_RATE_INFO_160_MHZ_WIDTH;
break;
}
if (rate_flg && nla_put_flag(msg, rate_flg))
return false;
nl80211/cfg80211: add VHT MCS support Add support for reporting and calculating VHT MCSes. Note that I'm not completely sure that the bitrate calculations are correct, nor that they can't be simplified. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-09 13:56:41 +00:00
if (info->flags & RATE_INFO_FLAGS_MCS) {
if (nla_put_u8(msg, NL80211_RATE_INFO_MCS, info->mcs))
return false;
if (info->flags & RATE_INFO_FLAGS_SHORT_GI &&
nla_put_flag(msg, NL80211_RATE_INFO_SHORT_GI))
return false;
} else if (info->flags & RATE_INFO_FLAGS_VHT_MCS) {
if (nla_put_u8(msg, NL80211_RATE_INFO_VHT_MCS, info->mcs))
return false;
if (nla_put_u8(msg, NL80211_RATE_INFO_VHT_NSS, info->nss))
return false;
if (info->flags & RATE_INFO_FLAGS_SHORT_GI &&
nla_put_flag(msg, NL80211_RATE_INFO_SHORT_GI))
return false;
}
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
nla_nest_end(msg, rate);
return true;
}
cfg80211: add support for per-chain signal strength reporting Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-22 14:29:30 +00:00
static bool nl80211_put_signal(struct sk_buff *msg, u8 mask, s8 *signal,
int id)
{
void *attr;
int i = 0;
if (!mask)
return true;
attr = nla_nest_start(msg, id);
if (!attr)
return false;
for (i = 0; i < IEEE80211_MAX_CHAINS; i++) {
if (!(mask & BIT(i)))
continue;
if (nla_put_u8(msg, i, signal[i]))
return false;
}
nla_nest_end(msg, attr);
return true;
}
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
static int nl80211_send_station(struct sk_buff *msg, u32 cmd, u32 portid,
u32 seq, int flags,
cfg80211: allow CFG80211_SIGNAL_TYPE_UNSPEC in station_info The station_info struct had demanded dBm signal values, but the cfg80211 wireless extensions implementation was also accepting "unspecified" (i.e. RSSI) unit values while the nl80211 code was completely unaware of them. Resolve this by formally allowing the "unspecified" units while making nl80211 ignore them. Signed-off-by: John W. Linville <linville@tuxdriver.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
2012-03-15 17:25:41 +00:00
struct cfg80211_registered_device *rdev,
struct net_device *dev,
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
const u8 *mac_addr, struct station_info *sinfo)
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
{
void *hdr;
nl80211: Add BSS parameters to station This allows user-space monitoring of BSS parameters for the associated station. This is useful for debugging and verifying that the paramaters are as expected. [Exactly the same as before but bundled into a single message] Signed-off-by: Paul Stewart <pstew@chromium.org> Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-31 16:25:41 +00:00
struct nlattr *sinfoattr, *bss_param;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
hdr = nl80211hdr_put(msg, portid, seq, flags, cmd);
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
if (!hdr)
return -1;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr) ||
nla_put_u32(msg, NL80211_ATTR_GENERATION, sinfo->generation))
goto nla_put_failure;
nl80211: add generation number to all dumps In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 14:17:38 +00:00
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
sinfoattr = nla_nest_start(msg, NL80211_ATTR_STA_INFO);
if (!sinfoattr)
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
goto nla_put_failure;
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
#define PUT_SINFO(attr, memb, type) do { \
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
BUILD_BUG_ON(sizeof(type) == sizeof(u64)); \
cfg80211/nl80211: Add support for NL80211_STA_INFO_RX_DURATION Add support for the a station statistics netlink attribute: NL80211_STA_INFO_RX_DURATION. If present, this attribute contains the aggregate PPDU duration (in microseconds) for all the frames from the peer. This is useful to help understand the total time spent transmitting to us by all of the connected peers. Signed-off-by: Mohammed Shafi Shajakhan <mohammed@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-07 14:29:34 +00:00
if (sinfo->filled & (1ULL << NL80211_STA_INFO_ ## attr) && \
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
nla_put_ ## type(msg, NL80211_STA_INFO_ ## attr, \
sinfo->memb)) \
goto nla_put_failure; \
} while (0)
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
#define PUT_SINFO_U64(attr, memb) do { \
if (sinfo->filled & (1ULL << NL80211_STA_INFO_ ## attr) && \
nla_put_u64_64bit(msg, NL80211_STA_INFO_ ## attr, \
sinfo->memb, NL80211_STA_INFO_PAD)) \
goto nla_put_failure; \
} while (0)
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
PUT_SINFO(CONNECTED_TIME, connected_time, u32);
PUT_SINFO(INACTIVE_TIME, inactive_time, u32);
if (sinfo->filled & (BIT(NL80211_STA_INFO_RX_BYTES) |
BIT(NL80211_STA_INFO_RX_BYTES64)) &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_STA_INFO_RX_BYTES,
cfg80211: expand per-station byte counters to 64bit In per-station statistics, present 32bit counters are too small for practical purposes - with gigabit speeds, it get overlapped every few seconds. Expand counters in the struct station_info to be 64-bit. Driver can still fill only 32-bit and indicate in @filled only bits like STATION_INFO_[TR]X_BYTES; in case driver provides full 64-bit counter, it should also set in @filled bit STATION_INFO_[TR]RX_BYTES64 Netlink sends both 32-bit and 64-bit counters, if present, to not break userspace. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [change to also have 32-bit counters if driver advertises 64-bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-04 11:53:11 +00:00
(u32)sinfo->rx_bytes))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->filled & (BIT(NL80211_STA_INFO_TX_BYTES) |
BIT(NL80211_STA_INFO_TX_BYTES64)) &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_STA_INFO_TX_BYTES,
cfg80211: expand per-station byte counters to 64bit In per-station statistics, present 32bit counters are too small for practical purposes - with gigabit speeds, it get overlapped every few seconds. Expand counters in the struct station_info to be 64-bit. Driver can still fill only 32-bit and indicate in @filled only bits like STATION_INFO_[TR]X_BYTES; in case driver provides full 64-bit counter, it should also set in @filled bit STATION_INFO_[TR]RX_BYTES64 Netlink sends both 32-bit and 64-bit counters, if present, to not break userspace. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [change to also have 32-bit counters if driver advertises 64-bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-04 11:53:11 +00:00
(u32)sinfo->tx_bytes))
goto nla_put_failure;
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
PUT_SINFO_U64(RX_BYTES64, rx_bytes);
PUT_SINFO_U64(TX_BYTES64, tx_bytes);
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
PUT_SINFO(LLID, llid, u16);
PUT_SINFO(PLID, plid, u16);
PUT_SINFO(PLINK_STATE, plink_state, u8);
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
PUT_SINFO_U64(RX_DURATION, rx_duration);
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
cfg80211: allow CFG80211_SIGNAL_TYPE_UNSPEC in station_info The station_info struct had demanded dBm signal values, but the cfg80211 wireless extensions implementation was also accepting "unspecified" (i.e. RSSI) unit values while the nl80211 code was completely unaware of them. Resolve this by formally allowing the "unspecified" units while making nl80211 ignore them. Signed-off-by: John W. Linville <linville@tuxdriver.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
2012-03-15 17:25:41 +00:00
switch (rdev->wiphy.signal_type) {
case CFG80211_SIGNAL_TYPE_MBM:
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
PUT_SINFO(SIGNAL, signal, u8);
PUT_SINFO(SIGNAL_AVG, signal_avg, u8);
cfg80211: allow CFG80211_SIGNAL_TYPE_UNSPEC in station_info The station_info struct had demanded dBm signal values, but the cfg80211 wireless extensions implementation was also accepting "unspecified" (i.e. RSSI) unit values while the nl80211 code was completely unaware of them. Resolve this by formally allowing the "unspecified" units while making nl80211 ignore them. Signed-off-by: John W. Linville <linville@tuxdriver.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
2012-03-15 17:25:41 +00:00
break;
default:
break;
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->filled & BIT(NL80211_STA_INFO_CHAIN_SIGNAL)) {
cfg80211: add support for per-chain signal strength reporting Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-22 14:29:30 +00:00
if (!nl80211_put_signal(msg, sinfo->chains,
sinfo->chain_signal,
NL80211_STA_INFO_CHAIN_SIGNAL))
goto nla_put_failure;
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->filled & BIT(NL80211_STA_INFO_CHAIN_SIGNAL_AVG)) {
cfg80211: add support for per-chain signal strength reporting Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-22 14:29:30 +00:00
if (!nl80211_put_signal(msg, sinfo->chains,
sinfo->chain_signal_avg,
NL80211_STA_INFO_CHAIN_SIGNAL_AVG))
goto nla_put_failure;
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->filled & BIT(NL80211_STA_INFO_TX_BITRATE)) {
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
if (!nl80211_put_sta_rate(msg, &sinfo->txrate,
NL80211_STA_INFO_TX_BITRATE))
goto nla_put_failure;
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->filled & BIT(NL80211_STA_INFO_RX_BITRATE)) {
cfg80211: add a field for the bitrate of the last rx data packet from a station Also fix a typo in the STATION_INFO_TX_BITRATE description Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-02-27 21:08:00 +00:00
if (!nl80211_put_sta_rate(msg, &sinfo->rxrate,
NL80211_STA_INFO_RX_BITRATE))
nl80211: Add signal strength and bandwith to nl80211station info This patch adds signal strength and transmission bitrate to the station_info of nl80211. Signed-off-by: Henning Rogge <rogge@fgan.de> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-12-11 21:04:19 +00:00
goto nla_put_failure;
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
PUT_SINFO(RX_PACKETS, rx_packets, u32);
PUT_SINFO(TX_PACKETS, tx_packets, u32);
PUT_SINFO(TX_RETRIES, tx_retries, u32);
PUT_SINFO(TX_FAILED, tx_failed, u32);
PUT_SINFO(EXPECTED_THROUGHPUT, expected_throughput, u32);
PUT_SINFO(BEACON_LOSS, beacon_loss_count, u32);
PUT_SINFO(LOCAL_PM, local_pm, u32);
PUT_SINFO(PEER_PM, peer_pm, u32);
PUT_SINFO(NONPEER_PM, nonpeer_pm, u32);
if (sinfo->filled & BIT(NL80211_STA_INFO_BSS_PARAM)) {
nl80211: Add BSS parameters to station This allows user-space monitoring of BSS parameters for the associated station. This is useful for debugging and verifying that the paramaters are as expected. [Exactly the same as before but bundled into a single message] Signed-off-by: Paul Stewart <pstew@chromium.org> Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-31 16:25:41 +00:00
bss_param = nla_nest_start(msg, NL80211_STA_INFO_BSS_PARAM);
if (!bss_param)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (((sinfo->bss_param.flags & BSS_PARAM_FLAGS_CTS_PROT) &&
nla_put_flag(msg, NL80211_STA_BSS_PARAM_CTS_PROT)) ||
((sinfo->bss_param.flags & BSS_PARAM_FLAGS_SHORT_PREAMBLE) &&
nla_put_flag(msg, NL80211_STA_BSS_PARAM_SHORT_PREAMBLE)) ||
((sinfo->bss_param.flags & BSS_PARAM_FLAGS_SHORT_SLOT_TIME) &&
nla_put_flag(msg, NL80211_STA_BSS_PARAM_SHORT_SLOT_TIME)) ||
nla_put_u8(msg, NL80211_STA_BSS_PARAM_DTIM_PERIOD,
sinfo->bss_param.dtim_period) ||
nla_put_u16(msg, NL80211_STA_BSS_PARAM_BEACON_INTERVAL,
sinfo->bss_param.beacon_interval))
goto nla_put_failure;
nl80211: Add BSS parameters to station This allows user-space monitoring of BSS parameters for the associated station. This is useful for debugging and verifying that the paramaters are as expected. [Exactly the same as before but bundled into a single message] Signed-off-by: Paul Stewart <pstew@chromium.org> Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-31 16:25:41 +00:00
nla_nest_end(msg, bss_param);
}
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if ((sinfo->filled & BIT(NL80211_STA_INFO_STA_FLAGS)) &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put(msg, NL80211_STA_INFO_STA_FLAGS,
sizeof(struct nl80211_sta_flag_update),
&sinfo->sta_flags))
goto nla_put_failure;
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
PUT_SINFO_U64(T_OFFSET, t_offset);
PUT_SINFO_U64(RX_DROP_MISC, rx_dropped_misc);
PUT_SINFO_U64(BEACON_RX, rx_beacon);
cfg80211: add nl80211 beacon-only statistics Add these two values: * BEACON_RX: number of beacons received from this peer * BEACON_SIGNAL_AVG: signal strength average for beacons only These can then be used for Android Lollipop's statistics request. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:12:22 +00:00
PUT_SINFO(BEACON_SIGNAL_AVG, rx_beacon_signal_avg, u8);
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
#undef PUT_SINFO
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
#undef PUT_SINFO_U64
nl80211: support per-TID station statistics The base for the current statistics is pretty mixed up, support exporting RX/TX statistics for MSDUs per TID. This (currently) covers received MSDUs, transmitted MSDUs and retries/failures thereof. Doing it per TID for MSDUs makes more sense than say only per AC because it's symmetric - we could export per-AC statistics for all frames (which AC we used for transmission can be determined also for management frames) but per TID is better and usually data frames are really the ones we care about. Also, on RX we can't determine the AC - but we do know the TID for any QoS MPDU we received. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-19 11:34:00 +00:00
if (sinfo->filled & BIT(NL80211_STA_INFO_TID_STATS)) {
struct nlattr *tidsattr;
int tid;
tidsattr = nla_nest_start(msg, NL80211_STA_INFO_TID_STATS);
if (!tidsattr)
goto nla_put_failure;
for (tid = 0; tid < IEEE80211_NUM_TIDS + 1; tid++) {
struct cfg80211_tid_stats *tidstats;
struct nlattr *tidattr;
tidstats = &sinfo->pertid[tid];
if (!tidstats->filled)
continue;
tidattr = nla_nest_start(msg, tid + 1);
if (!tidattr)
goto nla_put_failure;
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
#define PUT_TIDVAL_U64(attr, memb) do { \
nl80211: support per-TID station statistics The base for the current statistics is pretty mixed up, support exporting RX/TX statistics for MSDUs per TID. This (currently) covers received MSDUs, transmitted MSDUs and retries/failures thereof. Doing it per TID for MSDUs makes more sense than say only per AC because it's symmetric - we could export per-AC statistics for all frames (which AC we used for transmission can be determined also for management frames) but per TID is better and usually data frames are really the ones we care about. Also, on RX we can't determine the AC - but we do know the TID for any QoS MPDU we received. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-19 11:34:00 +00:00
if (tidstats->filled & BIT(NL80211_TID_STATS_ ## attr) && \
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
nla_put_u64_64bit(msg, NL80211_TID_STATS_ ## attr, \
tidstats->memb, NL80211_TID_STATS_PAD)) \
nl80211: support per-TID station statistics The base for the current statistics is pretty mixed up, support exporting RX/TX statistics for MSDUs per TID. This (currently) covers received MSDUs, transmitted MSDUs and retries/failures thereof. Doing it per TID for MSDUs makes more sense than say only per AC because it's symmetric - we could export per-AC statistics for all frames (which AC we used for transmission can be determined also for management frames) but per TID is better and usually data frames are really the ones we care about. Also, on RX we can't determine the AC - but we do know the TID for any QoS MPDU we received. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-19 11:34:00 +00:00
goto nla_put_failure; \
} while (0)
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
PUT_TIDVAL_U64(RX_MSDU, rx_msdu);
PUT_TIDVAL_U64(TX_MSDU, tx_msdu);
PUT_TIDVAL_U64(TX_MSDU_RETRIES, tx_msdu_retries);
PUT_TIDVAL_U64(TX_MSDU_FAILED, tx_msdu_failed);
nl80211: support per-TID station statistics The base for the current statistics is pretty mixed up, support exporting RX/TX statistics for MSDUs per TID. This (currently) covers received MSDUs, transmitted MSDUs and retries/failures thereof. Doing it per TID for MSDUs makes more sense than say only per AC because it's symmetric - we could export per-AC statistics for all frames (which AC we used for transmission can be determined also for management frames) but per TID is better and usually data frames are really the ones we care about. Also, on RX we can't determine the AC - but we do know the TID for any QoS MPDU we received. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-19 11:34:00 +00:00
nl80211: use nla_put_u64_64bit() for the remaining u64 attributes Nicolas converted most users, but didn't realize some were generated by macros. Convert those over as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-26 07:54:11 +00:00
#undef PUT_TIDVAL_U64
nl80211: support per-TID station statistics The base for the current statistics is pretty mixed up, support exporting RX/TX statistics for MSDUs per TID. This (currently) covers received MSDUs, transmitted MSDUs and retries/failures thereof. Doing it per TID for MSDUs makes more sense than say only per AC because it's symmetric - we could export per-AC statistics for all frames (which AC we used for transmission can be determined also for management frames) but per TID is better and usually data frames are really the ones we care about. Also, on RX we can't determine the AC - but we do know the TID for any QoS MPDU we received. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-19 11:34:00 +00:00
nla_nest_end(msg, tidattr);
}
nla_nest_end(msg, tidsattr);
}
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nla_nest_end(msg, sinfoattr);
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
cfg80211: remove enum station_info_flags This is really just duplicating the list of information that's already available in the nl80211 attribute, so remove the list. Two small changes are needed: * remove STATION_INFO_ASSOC_REQ_IES complete, but the length (assoc_req_ies_len) can be used instead * add NL80211_STA_INFO_RX_DROP_MISC which exists internally but not in nl80211 yet This gets rid of the duplicate maintenance of the two lists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-17 13:08:11 +00:00
if (sinfo->assoc_req_ies_len &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put(msg, NL80211_ATTR_IE, sinfo->assoc_req_ies_len,
sinfo->assoc_req_ies))
goto nla_put_failure;
cfg80211/nl80211: Send AssocReq IEs to user space in AP mode When user space SME/MLME (e.g., hostapd) is not used in AP mode, the IEs from the (Re)Association Request frame that was processed in firmware need to be made available for user space (e.g., RSN IE for hostapd). Allow this to be done with cfg80211_new_sta(). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-08 09:11:52 +00:00
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
nla_put_failure:
netlink: Improve returned error codes Make nlmsg_trim(), nlmsg_cancel(), genlmsg_cancel(), and nla_nest_cancel() void functions. Return -EMSGSIZE instead of -1 if the provided message buffer is not big enough. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-03 23:36:54 +00:00
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
}
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
static int nl80211_dump_station(struct sk_buff *skb,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
struct netlink_callback *cb)
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
{
struct station_info sinfo;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev;
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
struct wireless_dev *wdev;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
u8 mac_addr[ETH_ALEN];
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
int sta_idx = cb->args[2];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
int err;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
err = nl80211_prepare_wdev_dump(skb, cb, &rdev, &wdev);
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out_err;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (!wdev->netdev) {
err = -EINVAL;
goto out_err;
}
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (!rdev->ops->dump_station) {
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
err = -EOPNOTSUPP;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
goto out_err;
}
while (1) {
nl80211/cfg80211: Make addition of new sinfo fields safer Add a comment pointing out the use of enum station_info_flags for all new struct station_info fields. In addition, memset the sinfo buffer to zero before use on all paths in the current tree to avoid leaving uninitialized pointers in the data. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-11 08:46:22 +00:00
memset(&sinfo, 0, sizeof(sinfo));
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
err = rdev_dump_station(rdev, wdev->netdev, sta_idx,
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
mac_addr, &sinfo);
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
if (err == -ENOENT)
break;
if (err)
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
goto out_err;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
if (nl80211_send_station(skb, NL80211_CMD_NEW_STATION,
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
NETLINK_CB(cb->skb).portid,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
rdev, wdev->netdev, mac_addr,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
&sinfo) < 0)
goto out;
sta_idx++;
}
out:
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
cb->args[2] = sta_idx;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
err = skb->len;
out_err:
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_unlock();
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
return err;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
static int nl80211_get_station(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
struct station_info sinfo;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
struct sk_buff *msg;
u8 *mac_addr = NULL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
int err;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
memset(&sinfo, 0, sizeof(sinfo));
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->get_station)
return -EOPNOTSUPP;
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_get_station(rdev, dev, mac_addr, &sinfo);
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
if (err)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return err;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
if (!msg)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -ENOMEM;
cfg80211/nl80211: implement station attribute retrieval After a station is added to the kernel's structures, userspace has to be able to retrieve statistics about that station, especially whether the station was idle and how much bytes were transferred to and from it. This adds the necessary code to nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:36 +00:00
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
if (nl80211_send_station(msg, NL80211_CMD_NEW_STATION,
info->snd_portid, info->snd_seq, 0,
cfg80211: allow CFG80211_SIGNAL_TYPE_UNSPEC in station_info The station_info struct had demanded dBm signal values, but the cfg80211 wireless extensions implementation was also accepting "unspecified" (i.e. RSSI) unit values while the nl80211 code was completely unaware of them. Resolve this by formally allowing the "unspecified" units while making nl80211 ignore them. Signed-off-by: John W. Linville <linville@tuxdriver.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
2012-03-15 17:25:41 +00:00
rdev, dev, mac_addr, &sinfo) < 0) {
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
nlmsg_free(msg);
return -ENOBUFS;
}
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
}
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
int cfg80211_check_station_change(struct wiphy *wiphy,
struct station_parameters *params,
enum cfg80211_station_type statype)
{
cfg80211: allow AID/listen interval changes for unassociated station Currently, cfg80211 rejects updates of AID and listen interval parameters for existing entries. This information is known only at association stage and as a result it's impossible to update entries that were added unassociated. Fix this by allowing updates of these properies for stations that the driver (or mac80211) assigned unassociated state. This then fixes mac80211's use of NL80211_FEATURE_FULL_AP_CLIENT_STATE. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-23 08:20:06 +00:00
if (params->listen_interval != -1 &&
statype != CFG80211_STA_AP_CLIENT_UNASSOC)
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
return -EINVAL;
cfg80211: allow AID/listen interval changes for unassociated station Currently, cfg80211 rejects updates of AID and listen interval parameters for existing entries. This information is known only at association stage and as a result it's impossible to update entries that were added unassociated. Fix this by allowing updates of these properies for stations that the driver (or mac80211) assigned unassociated state. This then fixes mac80211's use of NL80211_FEATURE_FULL_AP_CLIENT_STATE. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-23 08:20:06 +00:00
cfg80211: allow userspace to specify client P2P PS support Legacy clients don't support P2P power save mechanisms, and thus if a P2P GO has a legacy client connected to it, it has to make some changes in the PS behavior. To handle this, add an attribute to specify whether a station supports P2P PS or not. If the attribute was not specified cfg80211 will assume that station supports it for P2P GO interface, and does NOT support it for AP interface, matching the current assumptions in the code. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-17 13:41:38 +00:00
if (params->support_p2p_ps != -1 &&
statype != CFG80211_STA_AP_CLIENT_UNASSOC)
return -EINVAL;
cfg80211: fix TDLS setup with VHT peers Some VHT TDLS peers (Google Nexus 5) include the VHT-AID IE in their TDLS setup request/response. Usermode passes this aid as the station aid, causing it to fail verifiction, since this happens in the "set_station" stage. Make an exception for the TDLS use-case. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-07-17 14:14:29 +00:00
if (params->aid &&
cfg80211: allow AID/listen interval changes for unassociated station Currently, cfg80211 rejects updates of AID and listen interval parameters for existing entries. This information is known only at association stage and as a result it's impossible to update entries that were added unassociated. Fix this by allowing updates of these properies for stations that the driver (or mac80211) assigned unassociated state. This then fixes mac80211's use of NL80211_FEATURE_FULL_AP_CLIENT_STATE. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-23 08:20:06 +00:00
!(params->sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)) &&
statype != CFG80211_STA_AP_CLIENT_UNASSOC)
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
return -EINVAL;
/* When you run into this, adjust the code below for the new flag */
BUILD_BUG_ON(NL80211_STA_FLAG_MAX != 7);
switch (statype) {
cfg80211: rename mesh station types The mesh station types used to refer to whether the station was secure or nonsecure. Really the salient information is whether it is managed by the kernel or userspace Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:11 +00:00
case CFG80211_STA_MESH_PEER_KERNEL:
case CFG80211_STA_MESH_PEER_USER:
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/*
* No ignoring the TDLS flag here -- the userspace mesh
* code doesn't have the bug of including TDLS in the
* mask everywhere.
*/
if (params->sta_flags_mask &
~(BIT(NL80211_STA_FLAG_AUTHENTICATED) |
BIT(NL80211_STA_FLAG_MFP) |
BIT(NL80211_STA_FLAG_AUTHORIZED)))
return -EINVAL;
break;
case CFG80211_STA_TDLS_PEER_SETUP:
case CFG80211_STA_TDLS_PEER_ACTIVE:
if (!(params->sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)))
return -EINVAL;
/* ignore since it can't change */
params->sta_flags_mask &= ~BIT(NL80211_STA_FLAG_TDLS_PEER);
break;
default:
/* disallow mesh-specific things */
if (params->plink_action != NL80211_PLINK_ACTION_NO_ACTION)
return -EINVAL;
if (params->local_pm)
return -EINVAL;
if (params->sta_modify_mask & STATION_PARAM_APPLY_PLINK_STATE)
return -EINVAL;
}
if (statype != CFG80211_STA_TDLS_PEER_SETUP &&
statype != CFG80211_STA_TDLS_PEER_ACTIVE) {
/* TDLS can't be set, ... */
if (params->sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER))
return -EINVAL;
/*
* ... but don't bother the driver with it. This works around
* a hostapd/wpa_supplicant issue -- it always includes the
* TLDS_PEER flag in the mask even for AP mode.
*/
params->sta_flags_mask &= ~BIT(NL80211_STA_FLAG_TDLS_PEER);
}
cfg80211: allow changing station capabilities for unassociated stations Currently, cfg80211 rejects capability updates for existing entries and as a result it's impossible to update entries that were added unassociated, but that is necessary to go through the full station states from userspace, adding a station before authentication etc. Fix this by allowing updates to capabilities for stations that the driver (or mac80211) assigned unassociated state. Drivers setting the full station state support flag must use the new station type for proper operation. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-09-21 12:49:53 +00:00
if (statype != CFG80211_STA_TDLS_PEER_SETUP &&
statype != CFG80211_STA_AP_CLIENT_UNASSOC) {
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* reject other things that can't change */
if (params->sta_modify_mask & STATION_PARAM_APPLY_UAPSD)
return -EINVAL;
if (params->sta_modify_mask & STATION_PARAM_APPLY_CAPABILITY)
return -EINVAL;
if (params->supported_rates)
return -EINVAL;
if (params->ext_capab || params->ht_capa || params->vht_capa)
return -EINVAL;
}
cfg80211: allow changing station capabilities for unassociated stations Currently, cfg80211 rejects capability updates for existing entries and as a result it's impossible to update entries that were added unassociated, but that is necessary to go through the full station states from userspace, adding a station before authentication etc. Fix this by allowing updates to capabilities for stations that the driver (or mac80211) assigned unassociated state. Drivers setting the full station state support flag must use the new station type for proper operation. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-09-21 12:49:53 +00:00
if (statype != CFG80211_STA_AP_CLIENT &&
statype != CFG80211_STA_AP_CLIENT_UNASSOC) {
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
if (params->vlan)
return -EINVAL;
}
switch (statype) {
case CFG80211_STA_AP_MLME_CLIENT:
/* Use this only for authorizing/unauthorizing a station */
if (!(params->sta_flags_mask & BIT(NL80211_STA_FLAG_AUTHORIZED)))
return -EOPNOTSUPP;
break;
case CFG80211_STA_AP_CLIENT:
cfg80211: allow changing station capabilities for unassociated stations Currently, cfg80211 rejects capability updates for existing entries and as a result it's impossible to update entries that were added unassociated, but that is necessary to go through the full station states from userspace, adding a station before authentication etc. Fix this by allowing updates to capabilities for stations that the driver (or mac80211) assigned unassociated state. Drivers setting the full station state support flag must use the new station type for proper operation. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-09-21 12:49:53 +00:00
case CFG80211_STA_AP_CLIENT_UNASSOC:
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* accept only the listed bits */
if (params->sta_flags_mask &
~(BIT(NL80211_STA_FLAG_AUTHORIZED) |
BIT(NL80211_STA_FLAG_AUTHENTICATED) |
BIT(NL80211_STA_FLAG_ASSOCIATED) |
BIT(NL80211_STA_FLAG_SHORT_PREAMBLE) |
BIT(NL80211_STA_FLAG_WME) |
BIT(NL80211_STA_FLAG_MFP)))
return -EINVAL;
/* but authenticated/associated only if driver handles it */
if (!(wiphy->features & NL80211_FEATURE_FULL_AP_CLIENT_STATE) &&
params->sta_flags_mask &
(BIT(NL80211_STA_FLAG_AUTHENTICATED) |
BIT(NL80211_STA_FLAG_ASSOCIATED)))
return -EINVAL;
break;
case CFG80211_STA_IBSS:
case CFG80211_STA_AP_STA:
/* reject any changes other than AUTHORIZED */
if (params->sta_flags_mask & ~BIT(NL80211_STA_FLAG_AUTHORIZED))
return -EINVAL;
break;
case CFG80211_STA_TDLS_PEER_SETUP:
/* reject any changes other than AUTHORIZED or WME */
if (params->sta_flags_mask & ~(BIT(NL80211_STA_FLAG_AUTHORIZED) |
BIT(NL80211_STA_FLAG_WME)))
return -EINVAL;
/* force (at least) rates when authorizing */
if (params->sta_flags_set & BIT(NL80211_STA_FLAG_AUTHORIZED) &&
!params->supported_rates)
return -EINVAL;
break;
case CFG80211_STA_TDLS_PEER_ACTIVE:
/* reject any changes */
return -EINVAL;
cfg80211: rename mesh station types The mesh station types used to refer to whether the station was secure or nonsecure. Really the salient information is whether it is managed by the kernel or userspace Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:11 +00:00
case CFG80211_STA_MESH_PEER_KERNEL:
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
if (params->sta_modify_mask & STATION_PARAM_APPLY_PLINK_STATE)
return -EINVAL;
break;
cfg80211: rename mesh station types The mesh station types used to refer to whether the station was secure or nonsecure. Really the salient information is whether it is managed by the kernel or userspace Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:11 +00:00
case CFG80211_STA_MESH_PEER_USER:
cfg80211: allow the plink state blocking for user managed mesh wpa_supplicant or authsae handles the mesh peering in user space, but the plink state is still managed in kernel space. Currently, there is no implementation by wpa_supplicant or authsae to block the plink state after it is set to ESTAB. By applying this patch, we can use the "iw mesh0 station set <MAC address> plink_action block" to block the peer mesh STA. This is useful for experimenting purposes. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-04-17 17:30:02 +00:00
if (params->plink_action != NL80211_PLINK_ACTION_NO_ACTION &&
params->plink_action != NL80211_PLINK_ACTION_BLOCK)
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
return -EINVAL;
break;
}
cfg80211: consider VHT opmode on station update Currently, this attribute is only fetched on station addition, but not on station change. Since this info is only present in the assoc request, with full station state support in the driver it cannot be present when the station is added. Thus, add support for changing the VHT opmode on station update if done before (or while) the station is marked as associated. After this, ignore it, since it used to be ignored. Signed-off-by: Beni Lev <beni.lev@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-19 16:28:56 +00:00
/*
* Older kernel versions ignored this attribute entirely, so don't
* reject attempts to update it but mark it as unused instead so the
* driver won't look at the data.
*/
if (statype != CFG80211_STA_AP_CLIENT_UNASSOC &&
statype != CFG80211_STA_TDLS_PEER_SETUP)
params->opmode_notif_used = false;
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
return 0;
}
EXPORT_SYMBOL(cfg80211_check_station_change);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
/*
nl80211: only allow adding stations to running vlan interfaces Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 16:23:31 +00:00
* Get vlan interface making sure it is running and on the right wiphy.
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
*/
nl80211: make get_vlan logic more common get_vlan() sets the output parameter even if it returns an error, which is a bit odd. Instead, convert it to use ERR_PTR. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 15:23:01 +00:00
static struct net_device *get_vlan(struct genl_info *info,
struct cfg80211_registered_device *rdev)
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
{
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
struct nlattr *vlanattr = info->attrs[NL80211_ATTR_STA_VLAN];
nl80211: make get_vlan logic more common get_vlan() sets the output parameter even if it returns an error, which is a bit odd. Instead, convert it to use ERR_PTR. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 15:23:01 +00:00
struct net_device *v;
int ret;
if (!vlanattr)
return NULL;
v = dev_get_by_index(genl_info_net(info), nla_get_u32(vlanattr));
if (!v)
return ERR_PTR(-ENODEV);
if (!v->ieee80211_ptr || v->ieee80211_ptr->wiphy != &rdev->wiphy) {
ret = -EINVAL;
goto error;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
}
nl80211: make get_vlan logic more common get_vlan() sets the output parameter even if it returns an error, which is a bit odd. Instead, convert it to use ERR_PTR. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 15:23:01 +00:00
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
if (v->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
v->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
v->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO) {
ret = -EINVAL;
goto error;
}
nl80211: make get_vlan logic more common get_vlan() sets the output parameter even if it returns an error, which is a bit odd. Instead, convert it to use ERR_PTR. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 15:23:01 +00:00
if (!netif_running(v)) {
ret = -ENETDOWN;
goto error;
}
return v;
error:
dev_put(v);
return ERR_PTR(ret);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
}
nl80211: make netlink attribute policies const There's no reason for netlink attribute policies to be __read_mostly, they can just be const. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-20 22:58:15 +00:00
static const struct nla_policy
nl80211_sta_wme_policy[NL80211_STA_WME_MAX + 1] = {
cfg80211: Pass TDLS peer's QoS/HT/VHT information during set_station The information of the peer's capabilities is required for the driver to perform TDLS Peer UAPSD operations. This information of the peer is passed by the supplicant using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass this information of the peer to the driver in case this command is used with the TDLS peer STA. In addition, make the HT/VHT capability configuration handled more consistently for other STA cases (reject both instead of just HT). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:54 +00:00
[NL80211_STA_WME_UAPSD_QUEUES] = { .type = NLA_U8 },
[NL80211_STA_WME_MAX_SP] = { .type = NLA_U8 },
};
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
static int nl80211_parse_sta_wme(struct genl_info *info,
struct station_parameters *params)
cfg80211: Pass TDLS peer's QoS/HT/VHT information during set_station The information of the peer's capabilities is required for the driver to perform TDLS Peer UAPSD operations. This information of the peer is passed by the supplicant using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass this information of the peer to the driver in case this command is used with the TDLS peer STA. In addition, make the HT/VHT capability configuration handled more consistently for other STA cases (reject both instead of just HT). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:54 +00:00
{
struct nlattr *tb[NL80211_STA_WME_MAX + 1];
struct nlattr *nla;
int err;
/* parse WME attributes if present */
if (!info->attrs[NL80211_ATTR_STA_WME])
return 0;
nla = info->attrs[NL80211_ATTR_STA_WME];
err = nla_parse_nested(tb, NL80211_STA_WME_MAX, nla,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_sta_wme_policy, info->extack);
cfg80211: Pass TDLS peer's QoS/HT/VHT information during set_station The information of the peer's capabilities is required for the driver to perform TDLS Peer UAPSD operations. This information of the peer is passed by the supplicant using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass this information of the peer to the driver in case this command is used with the TDLS peer STA. In addition, make the HT/VHT capability configuration handled more consistently for other STA cases (reject both instead of just HT). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:54 +00:00
if (err)
return err;
if (tb[NL80211_STA_WME_UAPSD_QUEUES])
params->uapsd_queues = nla_get_u8(
tb[NL80211_STA_WME_UAPSD_QUEUES]);
if (params->uapsd_queues & ~IEEE80211_WMM_IE_STA_QOSINFO_AC_MASK)
return -EINVAL;
if (tb[NL80211_STA_WME_MAX_SP])
params->max_sp = nla_get_u8(tb[NL80211_STA_WME_MAX_SP]);
if (params->max_sp & ~IEEE80211_WMM_IE_STA_QOSINFO_SP_MASK)
return -EINVAL;
params->sta_modify_mask |= STATION_PARAM_APPLY_UAPSD;
return 0;
}
cfg80211: pass station supported channel and oper class info The information of the peer's supported channels and supported operating classes are required for the driver to perform TDLS off channel operations. This commit enhances the function nl80211_(new)set_station to pass this information of the peer to the driver. Signed-off-by: Sunil Dutt <c_duttus@qti.qualcomm.com> [return errors for malformed tuples] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-09 15:15:21 +00:00
static int nl80211_parse_sta_channel_info(struct genl_info *info,
struct station_parameters *params)
{
if (info->attrs[NL80211_ATTR_STA_SUPPORTED_CHANNELS]) {
params->supported_channels =
nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_CHANNELS]);
params->supported_channels_len =
nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_CHANNELS]);
/*
* Need to include at least one (first channel, number of
* channels) tuple for each subband, and must have proper
* tuples for the rest of the data as well.
*/
if (params->supported_channels_len < 2)
return -EINVAL;
if (params->supported_channels_len % 2)
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES]) {
params->supported_oper_classes =
nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES]);
params->supported_oper_classes_len =
nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_OPER_CLASSES]);
/*
* The value of the Length field of the Supported Operating
* Classes element is between 2 and 253.
*/
if (params->supported_oper_classes_len < 2 ||
params->supported_oper_classes_len > 253)
return -EINVAL;
}
return 0;
}
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
static int nl80211_set_station_tdls(struct genl_info *info,
struct station_parameters *params)
{
cfg80211: pass station supported channel and oper class info The information of the peer's supported channels and supported operating classes are required for the driver to perform TDLS off channel operations. This commit enhances the function nl80211_(new)set_station to pass this information of the peer to the driver. Signed-off-by: Sunil Dutt <c_duttus@qti.qualcomm.com> [return errors for malformed tuples] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-09 15:15:21 +00:00
int err;
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
/* Dummy STA entry gets updated once the peer capabilities are known */
cfg80211: Allow TDLS peer AID to be configured for VHT VHT uses peer AID in the PARTIAL_AID field in TDLS frames. The current design for TDLS is to first add a dummy STA entry before completing TDLS Setup and then update information on this STA entry based on what was received from the peer during the setup exchange. In theory, this could use NL80211_ATTR_STA_AID to set the peer AID just like this is used in AP mode to set the AID of an association station. However, existing cfg80211 validation rules prevent this attribute from being used with set_station operation. To avoid interoperability issues between different kernel and user space version combinations, introduce a new nl80211 attribute for the purpose of setting TDLS peer AID. This attribute can be used in both the new_station and set_station operations. It is not supposed to be allowed to change the AID value during the lifetime of the STA entry, but that validation is left for drivers to do in the change_station callback. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 17:11:08 +00:00
if (info->attrs[NL80211_ATTR_PEER_AID])
params->aid = nla_get_u16(info->attrs[NL80211_ATTR_PEER_AID]);
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
if (info->attrs[NL80211_ATTR_HT_CAPABILITY])
params->ht_capa =
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY]);
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY])
params->vht_capa =
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY]);
cfg80211: pass station supported channel and oper class info The information of the peer's supported channels and supported operating classes are required for the driver to perform TDLS off channel operations. This commit enhances the function nl80211_(new)set_station to pass this information of the peer to the driver. Signed-off-by: Sunil Dutt <c_duttus@qti.qualcomm.com> [return errors for malformed tuples] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-09 15:15:21 +00:00
err = nl80211_parse_sta_channel_info(info, params);
if (err)
return err;
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
return nl80211_parse_sta_wme(info, params);
}
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
static int nl80211_set_station(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
struct station_parameters params;
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
u8 *mac_addr;
int err;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
memset(&params, 0, sizeof(params));
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
if (!rdev->ops->change_station)
return -EOPNOTSUPP;
cfg80211: allow AID/listen interval changes for unassociated station Currently, cfg80211 rejects updates of AID and listen interval parameters for existing entries. This information is known only at association stage and as a result it's impossible to update entries that were added unassociated. Fix this by allowing updates of these properies for stations that the driver (or mac80211) assigned unassociated state. This then fixes mac80211's use of NL80211_FEATURE_FULL_AP_CLIENT_STATE. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-23 08:20:06 +00:00
/*
* AID and listen_interval properties can be set only for unassociated
* station. Include these parameters here and will check them in
* cfg80211_check_station_change().
*/
cfg80211: use NL80211_ATTR_STA_AID in nl82011_set_station Fix nl80211_set_station() to use the value of NL80211_ATTR_STA_AID attribute instead of NL80211_ATTR_PEER_AID attribute. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-11-26 15:26:12 +00:00
if (info->attrs[NL80211_ATTR_STA_AID])
params.aid = nla_get_u16(info->attrs[NL80211_ATTR_STA_AID]);
cfg80211: allow AID/listen interval changes for unassociated station Currently, cfg80211 rejects updates of AID and listen interval parameters for existing entries. This information is known only at association stage and as a result it's impossible to update entries that were added unassociated. Fix this by allowing updates of these properies for stations that the driver (or mac80211) assigned unassociated state. This then fixes mac80211's use of NL80211_FEATURE_FULL_AP_CLIENT_STATE. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-23 08:20:06 +00:00
if (info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL])
params.listen_interval =
nla_get_u16(info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL]);
else
params.listen_interval = -1;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
cfg80211: allow userspace to specify client P2P PS support Legacy clients don't support P2P power save mechanisms, and thus if a P2P GO has a legacy client connected to it, it has to make some changes in the PS behavior. To handle this, add an attribute to specify whether a station supports P2P PS or not. If the attribute was not specified cfg80211 will assume that station supports it for P2P GO interface, and does NOT support it for AP interface, matching the current assumptions in the code. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-17 13:41:38 +00:00
if (info->attrs[NL80211_ATTR_STA_SUPPORT_P2P_PS]) {
u8 tmp;
tmp = nla_get_u8(info->attrs[NL80211_ATTR_STA_SUPPORT_P2P_PS]);
if (tmp >= NUM_NL80211_P2P_PS_STATUS)
return -EINVAL;
params.support_p2p_ps = tmp;
} else {
params.support_p2p_ps = -1;
}
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
if (info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]) {
params.supported_rates =
nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
params.supported_rates_len =
nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
}
cfg80211: Pass station (extended) capability info to kernel The information of the peer's capabilities and extended capabilities are required for the driver to perform TDLS Peer UAPSD operations and off channel operations. This information of the peer is passed from user space using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass the capability information of the peer to the driver. Similarly, there may be need for capability information for other modes, so allow this to be provided with both add_station and change_station. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:13 +00:00
if (info->attrs[NL80211_ATTR_STA_CAPABILITY]) {
params.capability =
nla_get_u16(info->attrs[NL80211_ATTR_STA_CAPABILITY]);
params.sta_modify_mask |= STATION_PARAM_APPLY_CAPABILITY;
}
if (info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]) {
params.ext_capab =
nla_data(info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]);
params.ext_capab_len =
nla_len(info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]);
}
nl80211: fix old station flags compatibility My patch to validate station flags broke compatibility with the old station flags setting where all flags are always set at once since it always set the mask as all possible flags which ended up being rejected later in the station add/modify code. Fix by parsing only the current flags in the old flags attribute -- new applications and new flags should use (and will now require) the new flags attribute where the mask is given by the application. Reported-and-tested-by: Thomas Hilber <ath9k-dev@toh.cx> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-02 12:30:03 +00:00
if (parse_station_flags(info, dev->ieee80211_ptr->iftype, &params))
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
return -EINVAL;
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
if (info->attrs[NL80211_ATTR_STA_PLINK_ACTION]) {
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
params.plink_action =
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_ACTION]);
if (params.plink_action >= NUM_NL80211_PLINK_ACTIONS)
return -EINVAL;
}
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
if (info->attrs[NL80211_ATTR_STA_PLINK_STATE]) {
nl80211: Let userspace drive the peer link management states. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:11 +00:00
params.plink_state =
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_STATE]);
if (params.plink_state >= NUM_NL80211_PLINK_STATES)
return -EINVAL;
cfg80211: Add mesh peer AID setting API Previously, mesh power management functionality works only with kernel MPM. Because user space MPM did not report mesh peer AID to kernel, the kernel could not identify the bit in TIM element. So this patch adds mesh peer AID setting API. Signed-off-by: Masashi Honma <masashi.honma@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-01 01:19:34 +00:00
if (info->attrs[NL80211_ATTR_MESH_PEER_AID]) {
params.peer_aid = nla_get_u16(
info->attrs[NL80211_ATTR_MESH_PEER_AID]);
if (params.peer_aid > IEEE80211_MAX_AID)
return -EINVAL;
}
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
params.sta_modify_mask |= STATION_PARAM_APPLY_PLINK_STATE;
}
nl80211: Let userspace drive the peer link management states. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:11 +00:00
{cfg,nl}80211: mesh power mode primitives and userspace access Add the nl80211_mesh_power_mode enumeration which holds possible values for the mesh power mode. These modes are unknown, active, light sleep and deep sleep. Add power_mode entry to the mesh config structure to hold the user-configured default mesh power mode. This value will be used for new peer links. Add the dot11MeshAwakeWindowDuration value to the mesh config. The awake window is a duration in TU describing how long the STA will stay awake after transmitting its beacon in PS mode. Add access routines to: - get/set local link-specific power mode (STA) - get remote STA's link-specific power mode (STA) - get remote STA's non-peer power mode (STA) - get/set default mesh power mode (mesh config) - get/set mesh awake window duration (mesh config) All config changes may be done at mesh runtime and take effect immediately. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Ivan Bezyazychnyy <ivan.bezyazychnyy@gmail.com> Signed-off-by: Mike Krinkin <krinkin.m.u@gmail.com> [fix commit message line length, error handling in set station] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:52 +00:00
if (info->attrs[NL80211_ATTR_LOCAL_MESH_POWER_MODE]) {
enum nl80211_mesh_power_mode pm = nla_get_u32(
info->attrs[NL80211_ATTR_LOCAL_MESH_POWER_MODE]);
if (pm <= NL80211_MESH_POWER_UNKNOWN ||
pm > NL80211_MESH_POWER_MAX)
return -EINVAL;
params.local_pm = pm;
}
cfg80211: consider VHT opmode on station update Currently, this attribute is only fetched on station addition, but not on station change. Since this info is only present in the assoc request, with full station state support in the driver it cannot be present when the station is added. Thus, add support for changing the VHT opmode on station update if done before (or while) the station is marked as associated. After this, ignore it, since it used to be ignored. Signed-off-by: Beni Lev <beni.lev@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-19 16:28:56 +00:00
if (info->attrs[NL80211_ATTR_OPMODE_NOTIF]) {
params.opmode_notif_used = true;
params.opmode_notif =
nla_get_u8(info->attrs[NL80211_ATTR_OPMODE_NOTIF]);
}
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* Include parameters for TDLS peer (will check later) */
err = nl80211_set_station_tdls(info, &params);
if (err)
return err;
params.vlan = get_vlan(info, rdev);
if (IS_ERR(params.vlan))
return PTR_ERR(params.vlan);
cfg80211: validate station settings When I disallowed interfering with stations on non-AP interfaces, I not only forget mesh but also managed interfaces which need this for the authorized flag. Let's actually validate everything properly. This fixes an nl80211 regression introduced by the interfering, under which wpa_supplicant -Dnl80211 could not properly connect. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-18 15:23:43 +00:00
switch (dev->ieee80211_ptr->iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
case NL80211_IFTYPE_P2P_GO:
case NL80211_IFTYPE_P2P_CLIENT:
cfg80211: validate station settings When I disallowed interfering with stations on non-AP interfaces, I not only forget mesh but also managed interfaces which need this for the authorized flag. Let's actually validate everything properly. This fixes an nl80211 regression introduced by the interfering, under which wpa_supplicant -Dnl80211 could not properly connect. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-18 15:23:43 +00:00
case NL80211_IFTYPE_STATION:
cfg80211/mac80211: userspace peer authorization in IBSS If the IBSS network is RSN-protected, let userspace authorize the stations instead of adding them as AUTHORIZED by default. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-31 19:25:47 +00:00
case NL80211_IFTYPE_ADHOC:
cfg80211: validate station settings When I disallowed interfering with stations on non-AP interfaces, I not only forget mesh but also managed interfaces which need this for the authorized flag. Let's actually validate everything properly. This fixes an nl80211 regression introduced by the interfering, under which wpa_supplicant -Dnl80211 could not properly connect. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-18 15:23:43 +00:00
case NL80211_IFTYPE_MESH_POINT:
break;
default:
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
err = -EOPNOTSUPP;
goto out_put_vlan;
cfg80211: disallow interfering with stations on non-AP (part 2) On non-AP interfaces userspace has no business interfering with the station management, this can confuse mac80211 (and other drivers probably wouldn't support it anyway). Allow adding and removing stations only on AP interfaces. (Reconcile this w/ previous version of patch posted with same subject... -- JWL) Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-27 08:35:29 +00:00
}
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* driver will call cfg80211_check_station_change() */
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_change_station(rdev, dev, mac_addr, &params);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
out_put_vlan:
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (params.vlan)
dev_put(params.vlan);
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
return err;
}
static int nl80211_new_station(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
struct station_parameters params;
u8 *mac_addr = NULL;
cfg80211: handle add_station auth/assoc flag quirks When a new station is added to AP/GO interfaces the default behaviour is for it to be added authenticated and associated, due to backwards compatibility. To prevent that, the driver must be able to do that (setting the NL80211_FEATURE_FULL_AP_CLIENT_STATE feature flag) and userspace must set the flag mask to auth|assoc and clear the set. Handle this quirk in the API entirely in nl80211, and always push the full flags to the drivers. NL80211_FEATURE_FULL_AP_CLIENT_STATE is still required for userspace to be allowed to set the mask including those bits, but after checking that add both flags to the mask and set in case userspace didn't set them otherwise. This obsoletes the mac80211 code handling this difference, no other driver is currently using these flags. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-11-26 15:26:13 +00:00
u32 auth_assoc = BIT(NL80211_STA_FLAG_AUTHENTICATED) |
BIT(NL80211_STA_FLAG_ASSOCIATED);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
memset(&params, 0, sizeof(params));
cfg80211: clean up station WME attribute parsing Parse the attributes first, and then disable the apply flag if needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:43:25 +00:00
if (!rdev->ops->add_station)
return -EOPNOTSUPP;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES])
return -EINVAL;
cfg80211: Allow TDLS peer AID to be configured for VHT VHT uses peer AID in the PARTIAL_AID field in TDLS frames. The current design for TDLS is to first add a dummy STA entry before completing TDLS Setup and then update information on this STA entry based on what was received from the peer during the setup exchange. In theory, this could use NL80211_ATTR_STA_AID to set the peer AID just like this is used in AP mode to set the AID of an association station. However, existing cfg80211 validation rules prevent this attribute from being used with set_station operation. To avoid interoperability issues between different kernel and user space version combinations, introduce a new nl80211 attribute for the purpose of setting TDLS peer AID. This attribute can be used in both the new_station and set_station operations. It is not supposed to be allowed to change the AID value during the lifetime of the STA entry, but that validation is left for drivers to do in the change_station callback. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 17:11:08 +00:00
if (!info->attrs[NL80211_ATTR_STA_AID] &&
!info->attrs[NL80211_ATTR_PEER_AID])
nl80211: does not allow NEW_STATION and DEL_STATION for mesh As discussed in linux-wireless mailing list, adding and removing stations for mesh topologies is not necessary. Since doing it triggers bugs, the sugestion was to simply disable it. Tested using a custom iw command "station new". Works only after using hostapd. "station del" command also works. Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Simon Raffeiner <sturmflut@lieberbiber.de> Cc: Andrey Yurovsky <andrey@cozybit.com> Cc: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-12 14:34:50 +00:00
return -EINVAL;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
mac_addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
params.supported_rates =
nla_data(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
params.supported_rates_len =
nla_len(info->attrs[NL80211_ATTR_STA_SUPPORTED_RATES]);
params.listen_interval =
nla_get_u16(info->attrs[NL80211_ATTR_STA_LISTEN_INTERVAL]);
cfg80211: validate AID of stations being added We have some validation code in mac80211 but said code will force an invalid AID to 0 which isn't a valid AID either; instead require a valid AID (1-2007) to be passed in from userspace in cfg80211 already. Also move the code before the race comment since it can only be executed during STA addition and thus is not racy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:42:30 +00:00
cfg80211: allow userspace to specify client P2P PS support Legacy clients don't support P2P power save mechanisms, and thus if a P2P GO has a legacy client connected to it, it has to make some changes in the PS behavior. To handle this, add an attribute to specify whether a station supports P2P PS or not. If the attribute was not specified cfg80211 will assume that station supports it for P2P GO interface, and does NOT support it for AP interface, matching the current assumptions in the code. Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-17 13:41:38 +00:00
if (info->attrs[NL80211_ATTR_STA_SUPPORT_P2P_PS]) {
u8 tmp;
tmp = nla_get_u8(info->attrs[NL80211_ATTR_STA_SUPPORT_P2P_PS]);
if (tmp >= NUM_NL80211_P2P_PS_STATUS)
return -EINVAL;
params.support_p2p_ps = tmp;
} else {
/*
* if not specified, assume it's supported for P2P GO interface,
* and is NOT supported for AP interface
*/
params.support_p2p_ps =
dev->ieee80211_ptr->iftype == NL80211_IFTYPE_P2P_GO;
}
cfg80211: fix VHT TDLS peer AID verification I (Johannes) accidentally applied the first version of the patch ("Allow TDLS peer AID to be configured for VHT"). Now apply just the changes between v1 and v2 to get the AID verification and prefer the new attribute over the old one. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-27 15:24:02 +00:00
if (info->attrs[NL80211_ATTR_PEER_AID])
cfg80211: Allow TDLS peer AID to be configured for VHT VHT uses peer AID in the PARTIAL_AID field in TDLS frames. The current design for TDLS is to first add a dummy STA entry before completing TDLS Setup and then update information on this STA entry based on what was received from the peer during the setup exchange. In theory, this could use NL80211_ATTR_STA_AID to set the peer AID just like this is used in AP mode to set the AID of an association station. However, existing cfg80211 validation rules prevent this attribute from being used with set_station operation. To avoid interoperability issues between different kernel and user space version combinations, introduce a new nl80211 attribute for the purpose of setting TDLS peer AID. This attribute can be used in both the new_station and set_station operations. It is not supposed to be allowed to change the AID value during the lifetime of the STA entry, but that validation is left for drivers to do in the change_station callback. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 17:11:08 +00:00
params.aid = nla_get_u16(info->attrs[NL80211_ATTR_PEER_AID]);
cfg80211: fix VHT TDLS peer AID verification I (Johannes) accidentally applied the first version of the patch ("Allow TDLS peer AID to be configured for VHT"). Now apply just the changes between v1 and v2 to get the AID verification and prefer the new attribute over the old one. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-27 15:24:02 +00:00
else
params.aid = nla_get_u16(info->attrs[NL80211_ATTR_STA_AID]);
nl80211: does not allow NEW_STATION and DEL_STATION for mesh As discussed in linux-wireless mailing list, adding and removing stations for mesh topologies is not necessary. Since doing it triggers bugs, the sugestion was to simply disable it. Tested using a custom iw command "station new". Works only after using hostapd. "station del" command also works. Signed-off-by: Thadeu Lima de Souza Cascardo <cascardo@holoscopio.com> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Simon Raffeiner <sturmflut@lieberbiber.de> Cc: Andrey Yurovsky <andrey@cozybit.com> Cc: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-12 14:34:50 +00:00
if (!params.aid || params.aid > IEEE80211_MAX_AID)
return -EINVAL;
cfg80211: validate AID of stations being added We have some validation code in mac80211 but said code will force an invalid AID to 0 which isn't a valid AID either; instead require a valid AID (1-2007) to be passed in from userspace in cfg80211 already. Also move the code before the race comment since it can only be executed during STA addition and thus is not racy. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:42:30 +00:00
cfg80211: Pass station (extended) capability info to kernel The information of the peer's capabilities and extended capabilities are required for the driver to perform TDLS Peer UAPSD operations and off channel operations. This information of the peer is passed from user space using NL80211_CMD_SET_STATION command. This commit enhances the function nl80211_set_station to pass the capability information of the peer to the driver. Similarly, there may be need for capability information for other modes, so allow this to be provided with both add_station and change_station. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 19:10:13 +00:00
if (info->attrs[NL80211_ATTR_STA_CAPABILITY]) {
params.capability =
nla_get_u16(info->attrs[NL80211_ATTR_STA_CAPABILITY]);
params.sta_modify_mask |= STATION_PARAM_APPLY_CAPABILITY;
}
if (info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]) {
params.ext_capab =
nla_data(info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]);
params.ext_capab_len =
nla_len(info->attrs[NL80211_ATTR_STA_EXT_CAPABILITY]);
}
mac80211/cfg80211: HT capabilities for NEW_STA Allow userspace (e.g., hostapd) to set HT capabilities for associated STAs. This is based on a patch from Zhu Yi <yi.zhu@intel.com> (only the NL80211_ATTR_HT_CAPABILITY for NEW_STA part is included here). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-25 08:58:58 +00:00
if (info->attrs[NL80211_ATTR_HT_CAPABILITY])
params.ht_capa =
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY]);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
{nl,cfg}80211: Peer STA VHT caps To save STAs VHT caps in AP mode Signed-off-by: Mahesh Palivela <maheshp@posedge.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-11 08:04:52 +00:00
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY])
params.vht_capa =
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY]);
nl80211/cfg80211: Set Operating Mode Notification This attribute is needed for setting Operating Mode Notification in AP mode from User Space. This functionality is required when User Space received Assoc Request contains Operation Mode Notification element. Signed-off-by: Marek Kwaczynski <marek.kwaczynski@tieto.com> [fix typos, nl80211 documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-03 09:04:59 +00:00
if (info->attrs[NL80211_ATTR_OPMODE_NOTIF]) {
params.opmode_notif_used = true;
params.opmode_notif =
nla_get_u8(info->attrs[NL80211_ATTR_OPMODE_NOTIF]);
}
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
if (info->attrs[NL80211_ATTR_STA_PLINK_ACTION]) {
nl80211/mac80211: Perform PLINK_ACTION on new station Modify the NEW_STATION command to accept PLINK_ACTIONS, in case userspace wants to create stations and initiate a peer link right away (for authenticated stations) or create a blocked station (for debugging). Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:33 +00:00
params.plink_action =
cfg80211: clean up mesh plink station change API Make the ability to leave the plink_state unchanged not use a magic -1 variable that isn't in the enum, but an explicit change flag; reject invalid plink states or actions and move the needed constants for plink actions to the right header file. Also reject plink_state changes for non-mesh interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:27:01 +00:00
nla_get_u8(info->attrs[NL80211_ATTR_STA_PLINK_ACTION]);
if (params.plink_action >= NUM_NL80211_PLINK_ACTIONS)
return -EINVAL;
}
nl80211/mac80211: Perform PLINK_ACTION on new station Modify the NEW_STATION command to accept PLINK_ACTIONS, in case userspace wants to create stations and initiate a peer link right away (for authenticated stations) or create a blocked station (for debugging). Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:33 +00:00
cfg80211: pass station supported channel and oper class info The information of the peer's supported channels and supported operating classes are required for the driver to perform TDLS off channel operations. This commit enhances the function nl80211_(new)set_station to pass this information of the peer to the driver. Signed-off-by: Sunil Dutt <c_duttus@qti.qualcomm.com> [return errors for malformed tuples] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-09 15:15:21 +00:00
err = nl80211_parse_sta_channel_info(info, &params);
if (err)
return err;
cfg80211: unify station WME parsing Instead of copying the code, create a new function to parse the station's WME information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:09:01 +00:00
err = nl80211_parse_sta_wme(info, &params);
if (err)
return err;
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
nl80211: fix old station flags compatibility My patch to validate station flags broke compatibility with the old station flags setting where all flags are always set at once since it always set the mask as all possible flags which ended up being rejected later in the station add/modify code. Fix by parsing only the current flags in the old flags attribute -- new applications and new flags should use (and will now require) the new flags attribute where the mask is given by the application. Reported-and-tested-by: Thomas Hilber <ath9k-dev@toh.cx> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-02 12:30:03 +00:00
if (parse_station_flags(info, dev->ieee80211_ptr->iftype, &params))
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
return -EINVAL;
nl80211: ignore HT/VHT capabilities without QoS/WMM As HT/VHT depend heavily on QoS/WMM, it's not a good idea to let userspace add clients that have HT/VHT but not QoS/WMM. Since it does so in certain cases we've observed (client is using HT IEs but not QoS/WMM) just ignore the HT/VHT info at this point and don't pass it down to the drivers which might unconditionally use it. Cc: stable@vger.kernel.org Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-12 06:53:27 +00:00
/* HT/VHT requires QoS, but if we don't have that just ignore HT/VHT
* as userspace might just pass through the capabilities from the IEs
* directly, rather than enforcing this restriction and returning an
* error in this case.
*/
if (!(params.sta_flags_set & BIT(NL80211_STA_FLAG_WME))) {
params.ht_capa = NULL;
params.vht_capa = NULL;
}
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* When you run into this, adjust the code below for the new flag */
BUILD_BUG_ON(NL80211_STA_FLAG_MAX != 7);
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
switch (dev->ieee80211_ptr->iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_P2P_GO:
cfg80211: clean up station WME attribute parsing Parse the attributes first, and then disable the apply flag if needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:43:25 +00:00
/* ignore WME attributes if iface/sta is not capable */
if (!(rdev->wiphy.flags & WIPHY_FLAG_AP_UAPSD) ||
!(params.sta_flags_set & BIT(NL80211_STA_FLAG_WME)))
params.sta_modify_mask &= ~STATION_PARAM_APPLY_UAPSD;
nl80211/cfg80211: add STA WME parameters Add new NL80211_ATTR_STA_WME nested attribute that contains wme params needed by the low-level driver (uapsd_queues and max_sp). Add these params to the station_parameters struct as well. Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-23 11:37:46 +00:00
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* TDLS peers cannot be added */
cfg80211: fix VHT TDLS peer AID verification I (Johannes) accidentally applied the first version of the patch ("Allow TDLS peer AID to be configured for VHT"). Now apply just the changes between v1 and v2 to get the AID verification and prefer the new attribute over the old one. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-27 15:24:02 +00:00
if ((params.sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)) ||
info->attrs[NL80211_ATTR_PEER_AID])
cfg80211: verify format of uAPSD information The format is intended to be like the subfields in the QoS Info field, verify that is the case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-07 09:50:48 +00:00
return -EINVAL;
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* but don't bother the driver with it */
params.sta_flags_mask &= ~BIT(NL80211_STA_FLAG_TDLS_PEER);
cfg80211/mac80211: apply station uAPSD parameters selectively Currently, when hostapd sets the station as authorized we also overwrite its uAPSD parameter. This obviously leads to buggy behaviour (later, with my patches that actually add uAPSD support). To fix this, only apply those parameters if they were actually set in nl80211, and to achieve that add a bitmap of things to apply. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-27 18:56:12 +00:00
nl80211/mac80211: support full station state in AP mode Today, stations are added already associated. That is inefficient if, for example, the driver has no room for stations any more because then the station will go through the entire auth/assoc handshake, only to be kicked out afterwards. To address this a bit better, at least with drivers using the new station state callback, allow hostapd to add stations in unauthenticated mode, just after receiving the AUTH frame, before even replying. Thus if there's no more space at that point, it can send a negative auth frame back. It still needs to handle later state transition errors though, of course. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 15:53:44 +00:00
/* allow authenticated/associated only if driver handles it */
if (!(rdev->wiphy.features &
NL80211_FEATURE_FULL_AP_CLIENT_STATE) &&
cfg80211: handle add_station auth/assoc flag quirks When a new station is added to AP/GO interfaces the default behaviour is for it to be added authenticated and associated, due to backwards compatibility. To prevent that, the driver must be able to do that (setting the NL80211_FEATURE_FULL_AP_CLIENT_STATE feature flag) and userspace must set the flag mask to auth|assoc and clear the set. Handle this quirk in the API entirely in nl80211, and always push the full flags to the drivers. NL80211_FEATURE_FULL_AP_CLIENT_STATE is still required for userspace to be allowed to set the mask including those bits, but after checking that add both flags to the mask and set in case userspace didn't set them otherwise. This obsoletes the mac80211 code handling this difference, no other driver is currently using these flags. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-11-26 15:26:13 +00:00
params.sta_flags_mask & auth_assoc)
nl80211/mac80211: support full station state in AP mode Today, stations are added already associated. That is inefficient if, for example, the driver has no room for stations any more because then the station will go through the entire auth/assoc handshake, only to be kicked out afterwards. To address this a bit better, at least with drivers using the new station state callback, allow hostapd to add stations in unauthenticated mode, just after receiving the AUTH frame, before even replying. Thus if there's no more space at that point, it can send a negative auth frame back. It still needs to handle later state transition errors though, of course. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 15:53:44 +00:00
return -EINVAL;
cfg80211: handle add_station auth/assoc flag quirks When a new station is added to AP/GO interfaces the default behaviour is for it to be added authenticated and associated, due to backwards compatibility. To prevent that, the driver must be able to do that (setting the NL80211_FEATURE_FULL_AP_CLIENT_STATE feature flag) and userspace must set the flag mask to auth|assoc and clear the set. Handle this quirk in the API entirely in nl80211, and always push the full flags to the drivers. NL80211_FEATURE_FULL_AP_CLIENT_STATE is still required for userspace to be allowed to set the mask including those bits, but after checking that add both flags to the mask and set in case userspace didn't set them otherwise. This obsoletes the mac80211 code handling this difference, no other driver is currently using these flags. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-11-26 15:26:13 +00:00
/* Older userspace, or userspace wanting to be compatible with
* !NL80211_FEATURE_FULL_AP_CLIENT_STATE, will not set the auth
* and assoc flags in the mask, but assumes the station will be
* added as associated anyway since this was the required driver
* behaviour before NL80211_FEATURE_FULL_AP_CLIENT_STATE was
* introduced.
* In order to not bother drivers with this quirk in the API
* set the flags in both the mask and set for new stations in
* this case.
*/
if (!(params.sta_flags_mask & auth_assoc)) {
params.sta_flags_mask |= auth_assoc;
params.sta_flags_set |= auth_assoc;
}
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* must be last in here for error handling */
params.vlan = get_vlan(info, rdev);
if (IS_ERR(params.vlan))
return PTR_ERR(params.vlan);
break;
case NL80211_IFTYPE_MESH_POINT:
cfg80211: clean up station WME attribute parsing Parse the attributes first, and then disable the apply flag if needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:43:25 +00:00
/* ignore uAPSD data */
params.sta_modify_mask &= ~STATION_PARAM_APPLY_UAPSD;
nl80211/mac80211: support full station state in AP mode Today, stations are added already associated. That is inefficient if, for example, the driver has no room for stations any more because then the station will go through the entire auth/assoc handshake, only to be kicked out afterwards. To address this a bit better, at least with drivers using the new station state callback, allow hostapd to add stations in unauthenticated mode, just after receiving the AUTH frame, before even replying. Thus if there's no more space at that point, it can send a negative auth frame back. It still needs to handle later state transition errors though, of course. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 15:53:44 +00:00
/* associated is disallowed */
if (params.sta_flags_mask & BIT(NL80211_STA_FLAG_ASSOCIATED))
return -EINVAL;
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* TDLS peers cannot be added */
cfg80211: fix VHT TDLS peer AID verification I (Johannes) accidentally applied the first version of the patch ("Allow TDLS peer AID to be configured for VHT"). Now apply just the changes between v1 and v2 to get the AID verification and prefer the new attribute over the old one. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-27 15:24:02 +00:00
if ((params.sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)) ||
info->attrs[NL80211_ATTR_PEER_AID])
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
return -EINVAL;
break;
case NL80211_IFTYPE_STATION:
cfg80211: enable TDLS on P2P client interfaces There's no reason TDLS should be prevented on P2P client interfaces, and most of the code already handles it, so allow adding stations for it. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 08:29:46 +00:00
case NL80211_IFTYPE_P2P_CLIENT:
cfg80211: clean up station WME attribute parsing Parse the attributes first, and then disable the apply flag if needed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 22:43:25 +00:00
/* ignore uAPSD data */
params.sta_modify_mask &= ~STATION_PARAM_APPLY_UAPSD;
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/* these are disallowed */
if (params.sta_flags_mask &
(BIT(NL80211_STA_FLAG_ASSOCIATED) |
BIT(NL80211_STA_FLAG_AUTHENTICATED)))
nl80211/mac80211: support full station state in AP mode Today, stations are added already associated. That is inefficient if, for example, the driver has no room for stations any more because then the station will go through the entire auth/assoc handshake, only to be kicked out afterwards. To address this a bit better, at least with drivers using the new station state callback, allow hostapd to add stations in unauthenticated mode, just after receiving the AUTH frame, before even replying. Thus if there's no more space at that point, it can send a negative auth frame back. It still needs to handle later state transition errors though, of course. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 15:53:44 +00:00
return -EINVAL;
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* Only TDLS peers can be added */
if (!(params.sta_flags_set & BIT(NL80211_STA_FLAG_TDLS_PEER)))
return -EINVAL;
/* Can only add if TDLS ... */
if (!(rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_TDLS))
return -EOPNOTSUPP;
/* ... with external setup is supported */
if (!(rdev->wiphy.flags & WIPHY_FLAG_TDLS_EXTERNAL_SETUP))
return -EOPNOTSUPP;
cfg80211: comprehensively check station changes The station change API isn't being checked properly before drivers are called, and as a result it is difficult to see what should be allowed and what not. In order to comprehensively check the API parameters parse everything first, and then have the driver call a function (cfg80211_check_station_change()) with the additionally information about the kind of station that is being changed; this allows the function to make better decisions than the old code could. While at it, also add a few checks, particularly in mesh and clarify the TDLS station lifetime in documentation. To be able to reduce a few checks, ignore any flag set bits when the mask isn't set, they shouldn't be applied then. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 23:48:33 +00:00
/*
* Older wpa_supplicant versions always mark the TDLS peer
* as authorized, but it shouldn't yet be.
*/
params.sta_flags_mask &= ~BIT(NL80211_STA_FLAG_AUTHORIZED);
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
break;
default:
return -EOPNOTSUPP;
nl80211/cfg80211: add STA WME parameters Add new NL80211_ATTR_STA_WME nested attribute that contains wme params needed by the low-level driver (uapsd_queues and max_sp). Add these params to the station_parameters struct as well. Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-23 11:37:46 +00:00
}
cfg80211: validate nl80211 station handling better The nl80211 station handling code is a bit messy and doesn't do a lot of validation. It seems like this could be an issue for drivers that don't use mac80211 to validate everything. As cfg80211 doesn't keep station state, move the validation of allowing supported_rates to change for TDLS only in station mode to mac80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-14 11:20:27 +00:00
/* be aware of params.vlan when changing code here */
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_add_station(rdev, dev, mac_addr, &params);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (params.vlan)
dev_put(params.vlan);
return err;
}
static int nl80211_del_station(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: Convert del_station() callback to use a param struct This makes it easier to add new parameters for the del_station calls without having to modify all drivers that use this. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-10 17:52:40 +00:00
struct station_del_parameters params;
memset(&params, 0, sizeof(params));
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
if (info->attrs[NL80211_ATTR_MAC])
cfg80211: Convert del_station() callback to use a param struct This makes it easier to add new parameters for the del_station calls without having to modify all drivers that use this. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-10 17:52:40 +00:00
params.mac = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
cfg80211: disallow interfering with stations on non-AP On non-AP interfaces userspace has no business interfering with the station management, this can confuse mac80211 (and other drivers probably wouldn't support it anyway). Allow adding and removing stations only on AP interfaces. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-11 12:43:13 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
nl80211: reenable station del for mesh iw dev <devname> station del <MAC address> is quiet useful in mesh mode and should be possible. Signed-off-by: Marco Porsch <marco.porsch@siemens.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-30 08:00:16 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP_VLAN &&
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->del_station)
return -EOPNOTSUPP;
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211: Specify frame and reason code for NL80211_CMD_DEL_STATION The optional NL80211_ATTR_MGMT_SUBTYPE and NL80211_ATTR_REASON_CODE attributes can now be included in NL80211_CMD_DEL_STATION to indicate to the driver which frame (Deauthentication/Disassociation) and reason code in that frame should be used to indicate removal to the specific station. This is used by drivers that implement AP SME and generate those frames internally. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-20 10:20:45 +00:00
if (info->attrs[NL80211_ATTR_MGMT_SUBTYPE]) {
params.subtype =
nla_get_u8(info->attrs[NL80211_ATTR_MGMT_SUBTYPE]);
if (params.subtype != IEEE80211_STYPE_DISASSOC >> 4 &&
params.subtype != IEEE80211_STYPE_DEAUTH >> 4)
return -EINVAL;
} else {
/* Default to Deauthentication frame */
params.subtype = IEEE80211_STYPE_DEAUTH >> 4;
}
if (info->attrs[NL80211_ATTR_REASON_CODE]) {
params.reason_code =
nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]);
if (params.reason_code == 0)
return -EINVAL; /* 0 is reserved */
} else {
/* Default to reason code 2 */
params.reason_code = WLAN_REASON_PREV_AUTH_NOT_VALID;
}
cfg80211: Convert del_station() callback to use a param struct This makes it easier to add new parameters for the del_station calls without having to modify all drivers that use this. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-10 17:52:40 +00:00
return rdev_del_station(rdev, dev, &params);
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
}
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
static int nl80211_send_mpath(struct sk_buff *msg, u32 portid, u32 seq,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
int flags, struct net_device *dev,
u8 *dst, u8 *next_hop,
struct mpath_info *pinfo)
{
void *hdr;
struct nlattr *pinfoattr;
cfg80211: fix nl80211 cmd id in nl80211_send_mpath() Netlink command for nl80211_send_mpath() should be NL80211_CMD_NEW_MPATH. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-04 15:14:58 +00:00
hdr = nl80211hdr_put(msg, portid, seq, flags, NL80211_CMD_NEW_MPATH);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
if (!hdr)
return -1;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, dst) ||
nla_put(msg, NL80211_ATTR_MPATH_NEXT_HOP, ETH_ALEN, next_hop) ||
nla_put_u32(msg, NL80211_ATTR_GENERATION, pinfo->generation))
goto nla_put_failure;
nl80211: add generation number to all dumps In order for userspace to be able to figure out whether it obtained a consistent snapshot of data or not when using netlink dumps, we need to have a generation number in each dump message that indicates whether the list has changed or not -- its value is arbitrary. This patch adds such a number to all dumps, this needs some mac80211 involvement to keep track of a generation number to start with when adding/removing mesh paths or stations. The wiphy and netdev lists can be fully handled within cfg80211, of course, but generation numbers need to be stored there as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 14:17:38 +00:00
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
pinfoattr = nla_nest_start(msg, NL80211_ATTR_MPATH_INFO);
if (!pinfoattr)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((pinfo->filled & MPATH_INFO_FRAME_QLEN) &&
nla_put_u32(msg, NL80211_MPATH_INFO_FRAME_QLEN,
pinfo->frame_qlen))
goto nla_put_failure;
if (((pinfo->filled & MPATH_INFO_SN) &&
nla_put_u32(msg, NL80211_MPATH_INFO_SN, pinfo->sn)) ||
((pinfo->filled & MPATH_INFO_METRIC) &&
nla_put_u32(msg, NL80211_MPATH_INFO_METRIC,
pinfo->metric)) ||
((pinfo->filled & MPATH_INFO_EXPTIME) &&
nla_put_u32(msg, NL80211_MPATH_INFO_EXPTIME,
pinfo->exptime)) ||
((pinfo->filled & MPATH_INFO_FLAGS) &&
nla_put_u8(msg, NL80211_MPATH_INFO_FLAGS,
pinfo->flags)) ||
((pinfo->filled & MPATH_INFO_DISCOVERY_TIMEOUT) &&
nla_put_u32(msg, NL80211_MPATH_INFO_DISCOVERY_TIMEOUT,
pinfo->discovery_timeout)) ||
((pinfo->filled & MPATH_INFO_DISCOVERY_RETRIES) &&
nla_put_u8(msg, NL80211_MPATH_INFO_DISCOVERY_RETRIES,
pinfo->discovery_retries)))
goto nla_put_failure;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nla_nest_end(msg, pinfoattr);
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nla_put_failure:
netlink: Improve returned error codes Make nlmsg_trim(), nlmsg_cancel(), genlmsg_cancel(), and nla_nest_cancel() void functions. Return -EMSGSIZE instead of -1 if the provided message buffer is not big enough. Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2008-06-03 23:36:54 +00:00
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
static int nl80211_dump_mpath(struct sk_buff *skb,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
struct netlink_callback *cb)
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
{
struct mpath_info pinfo;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev;
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
struct wireless_dev *wdev;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
u8 dst[ETH_ALEN];
u8 next_hop[ETH_ALEN];
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
int path_idx = cb->args[2];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
int err;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
err = nl80211_prepare_wdev_dump(skb, cb, &rdev, &wdev);
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out_err;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (!rdev->ops->dump_mpath) {
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
err = -EOPNOTSUPP;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
goto out_err;
}
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (wdev->iftype != NL80211_IFTYPE_MESH_POINT) {
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
err = -EOPNOTSUPP;
nl80211: jump to out_err upon unsupported iftype Jump to out_err when the iftype is not supported. Signed-off-by: Roel Kluin <roel.kluin@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-22 19:15:49 +00:00
goto out_err;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
}
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
while (1) {
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
err = rdev_dump_mpath(rdev, wdev->netdev, path_idx, dst,
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
next_hop, &pinfo);
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
if (err == -ENOENT)
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
break;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
if (err)
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
goto out_err;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (nl80211_send_mpath(skb, NETLINK_CB(cb->skb).portid,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
wdev->netdev, dst, next_hop,
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
&pinfo) < 0)
goto out;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
path_idx++;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
out:
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
cb->args[2] = path_idx;
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
err = skb->len;
out_err:
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_unlock();
nl80211: fix dump callbacks Julius Volz pointed out that the dump callbacks in nl80211 were broken and fixed one of them. This patch fixes the other three and also addresses the TODOs there. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: Julius Volz <juliusv@google.com> Cc: Thomas Graf <tgraf@suug.ch> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-07-29 11:22:51 +00:00
return err;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
static int nl80211_get_mpath(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
struct mpath_info pinfo;
struct sk_buff *msg;
u8 *dst = NULL;
u8 next_hop[ETH_ALEN];
memset(&pinfo, 0, sizeof(pinfo));
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->get_mpath)
return -EOPNOTSUPP;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_get_mpath(rdev, dev, dst, next_hop, &pinfo);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
if (err)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return err;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
if (!msg)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -ENOMEM;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (nl80211_send_mpath(msg, info->snd_portid, info->snd_seq, 0,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev, dst, next_hop, &pinfo) < 0) {
nlmsg_free(msg);
return -ENOBUFS;
}
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
static int nl80211_set_mpath(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
u8 *dst = NULL;
u8 *next_hop = NULL;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP])
return -EINVAL;
dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->change_mpath)
return -EOPNOTSUPP;
nl80211: Check that netif_runnin is true in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this or to be prepared to handle these commands when the interface is down. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:18 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_change_mpath(rdev, dev, dst, next_hop);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
static int nl80211_new_mpath(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
u8 *dst = NULL;
u8 *next_hop = NULL;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MPATH_NEXT_HOP])
return -EINVAL;
dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
next_hop = nla_data(info->attrs[NL80211_ATTR_MPATH_NEXT_HOP]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->add_mpath)
return -EOPNOTSUPP;
nl80211: Check that netif_runnin is true in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this or to be prepared to handle these commands when the interface is down. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:18 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_add_mpath(rdev, dev, dst, next_hop);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
static int nl80211_del_mpath(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
u8 *dst = NULL;
if (info->attrs[NL80211_ATTR_MAC])
dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->del_mpath)
return -EOPNOTSUPP;
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_del_mpath(rdev, dev, dst);
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
}
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
static int nl80211_get_mpp(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
int err;
struct net_device *dev = info->user_ptr[1];
struct mpath_info pinfo;
struct sk_buff *msg;
u8 *dst = NULL;
u8 mpp[ETH_ALEN];
memset(&pinfo, 0, sizeof(pinfo));
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
dst = nla_data(info->attrs[NL80211_ATTR_MAC]);
if (!rdev->ops->get_mpp)
return -EOPNOTSUPP;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
err = rdev_get_mpp(rdev, dev, dst, mpp, &pinfo);
if (err)
return err;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
if (nl80211_send_mpath(msg, info->snd_portid, info->snd_seq, 0,
dev, dst, mpp, &pinfo) < 0) {
nlmsg_free(msg);
return -ENOBUFS;
}
return genlmsg_reply(msg, info);
}
static int nl80211_dump_mpp(struct sk_buff *skb,
struct netlink_callback *cb)
{
struct mpath_info pinfo;
struct cfg80211_registered_device *rdev;
struct wireless_dev *wdev;
u8 dst[ETH_ALEN];
u8 mpp[ETH_ALEN];
int path_idx = cb->args[2];
int err;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
err = nl80211_prepare_wdev_dump(skb, cb, &rdev, &wdev);
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out_err;
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
if (!rdev->ops->dump_mpp) {
err = -EOPNOTSUPP;
goto out_err;
}
if (wdev->iftype != NL80211_IFTYPE_MESH_POINT) {
err = -EOPNOTSUPP;
goto out_err;
}
while (1) {
err = rdev_dump_mpp(rdev, wdev->netdev, path_idx, dst,
mpp, &pinfo);
if (err == -ENOENT)
break;
if (err)
goto out_err;
if (nl80211_send_mpath(skb, NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, NLM_F_MULTI,
wdev->netdev, dst, mpp,
&pinfo) < 0)
goto out;
path_idx++;
}
out:
cb->args[2] = path_idx;
err = skb->len;
out_err:
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_unlock();
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
return err;
}
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
static int nl80211_set_bss(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
struct bss_parameters params;
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
int err;
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
memset(&params, 0, sizeof(params));
/* default to not changing parameters */
params.use_cts_prot = -1;
params.use_short_preamble = -1;
params.use_short_slot_time = -1;
cfg80211: add ap isolation support This is used to configure APs to not bridge traffic between connected stations. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-26 23:23:35 +00:00
params.ap_isolate = -1;
cfg80211: Add new BSS attribute ht_opmode Add a new BSS attribute to allow hostapd to set the current HT opmode. Otherwise drivers won't be able to set up protection for HT rates in AP mode. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-19 11:40:25 +00:00
params.ht_opmode = -1;
nl80211: support P2P GO powersave configuration If a driver supports P2P GO powersave, allow it to set the new feature flags for it and allow userspace to configure the parameters for it. This can be done at GO startup and later changed with SET_BSS. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-14 14:17:28 +00:00
params.p2p_ctwindow = -1;
params.p2p_opp_ps = -1;
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
if (info->attrs[NL80211_ATTR_BSS_CTS_PROT])
params.use_cts_prot =
nla_get_u8(info->attrs[NL80211_ATTR_BSS_CTS_PROT]);
if (info->attrs[NL80211_ATTR_BSS_SHORT_PREAMBLE])
params.use_short_preamble =
nla_get_u8(info->attrs[NL80211_ATTR_BSS_SHORT_PREAMBLE]);
if (info->attrs[NL80211_ATTR_BSS_SHORT_SLOT_TIME])
params.use_short_slot_time =
nla_get_u8(info->attrs[NL80211_ATTR_BSS_SHORT_SLOT_TIME]);
nl80211: Add basic rate configuration for AP mode Add a new attribute, NL80211_ATTR_BSS_BASIC_RATES, that can be used with NL80211_CMD_SET_BSS for userspace (e.g., hostapd) to set which rates are in the basic rate set. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-30 14:59:22 +00:00
if (info->attrs[NL80211_ATTR_BSS_BASIC_RATES]) {
params.basic_rates =
nla_data(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
params.basic_rates_len =
nla_len(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
}
cfg80211: add ap isolation support This is used to configure APs to not bridge traffic between connected stations. Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-26 23:23:35 +00:00
if (info->attrs[NL80211_ATTR_AP_ISOLATE])
params.ap_isolate = !!nla_get_u8(info->attrs[NL80211_ATTR_AP_ISOLATE]);
cfg80211: Add new BSS attribute ht_opmode Add a new BSS attribute to allow hostapd to set the current HT opmode. Otherwise drivers won't be able to set up protection for HT rates in AP mode. Cc: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Helmut Schaa <helmut.schaa@googlemail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-19 11:40:25 +00:00
if (info->attrs[NL80211_ATTR_BSS_HT_OPMODE])
params.ht_opmode =
nla_get_u16(info->attrs[NL80211_ATTR_BSS_HT_OPMODE]);
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
nl80211: support P2P GO powersave configuration If a driver supports P2P GO powersave, allow it to set the new feature flags for it and allow userspace to configure the parameters for it. This can be done at GO startup and later changed with SET_BSS. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-14 14:17:28 +00:00
if (info->attrs[NL80211_ATTR_P2P_CTWINDOW]) {
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
params.p2p_ctwindow =
nla_get_s8(info->attrs[NL80211_ATTR_P2P_CTWINDOW]);
if (params.p2p_ctwindow < 0)
return -EINVAL;
if (params.p2p_ctwindow != 0 &&
!(rdev->wiphy.features & NL80211_FEATURE_P2P_GO_CTWIN))
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_P2P_OPPPS]) {
u8 tmp;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
tmp = nla_get_u8(info->attrs[NL80211_ATTR_P2P_OPPPS]);
if (tmp > 1)
return -EINVAL;
params.p2p_opp_ps = tmp;
if (params.p2p_opp_ps &&
!(rdev->wiphy.features & NL80211_FEATURE_P2P_GO_OPPPS))
return -EINVAL;
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->change_bss)
return -EOPNOTSUPP;
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_AP &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
wdev_lock(wdev);
err = rdev_change_bss(rdev, dev, &params);
wdev_unlock(wdev);
return err;
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
}
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
static int nl80211_req_set_reg(struct sk_buff *skb, struct genl_info *info)
{
char *data = NULL;
cfg80211: Add API to change the indoor regulatory setting Previously, the indoor setting configuration assumed that as long as a station interface is connected, the indoor environment setting does not change. However, this assumption is problematic as: - It is possible that a station interface is connected to a mobile AP, e.g., softAP or a P2P GO, where it is possible that both the station and the mobile AP move out of the indoor environment making the indoor setting invalid. In such a case, user space has no way to invalidate the setting. - A station interface disconnection does not necessarily imply that the device is no longer operating in an indoor environment, e.g., it is possible that the station interface is roaming but is still stays indoor. To handle the above, extend the indoor configuration API to allow user space to indicate a change of indoor settings, and allow it to indicate weather it controls the indoor setting, such that: 1. If the user space process explicitly indicates that it is going to control the indoor setting, do not clear the indoor setting internally, unless the socket is released. The user space process should use the NL80211_ATTR_SOCKET_OWNER attribute in the command to state that it is going to control the indoor setting. 2. Reset the indoor setting when restoring the regulatory settings in case it is not owned by a user space process. Based on the above, a user space tool that continuously monitors the indoor settings, i.e., tracking power setting, location etc., can indicate environment changes to the regulatory core. It should be noted that currently user space is the only provided mechanism used to hint to the regulatory core over the indoor/outdoor environment -- while the country IEs do have an environment setting this has been completely ignored by the regulatory core by design for a while now since country IEs typically can contain bogus data. Acked-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: ArikX Nemtsov <arik@wizery.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-04 05:32:06 +00:00
bool is_indoor;
cfg80211: add cellular base station regulatory hint support Cellular base stations can provide hints to cfg80211 about where they think we are. This can be done for example on a cell phone. To enable these hints we simply allow them through as user regulatory hints but we allow userspace to clasify the hint as either coming directly from the user or coming from a cellular base station. This option is only available when you enable CONFIG_CFG80211_CERTIFICATION_ONUS. The base station hints themselves will not be processed by the core unless at least one device on the system supports this feature. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 18:49:18 +00:00
enum nl80211_user_reg_hint_type user_reg_hint_type;
cfg80211: Add API to change the indoor regulatory setting Previously, the indoor setting configuration assumed that as long as a station interface is connected, the indoor environment setting does not change. However, this assumption is problematic as: - It is possible that a station interface is connected to a mobile AP, e.g., softAP or a P2P GO, where it is possible that both the station and the mobile AP move out of the indoor environment making the indoor setting invalid. In such a case, user space has no way to invalidate the setting. - A station interface disconnection does not necessarily imply that the device is no longer operating in an indoor environment, e.g., it is possible that the station interface is roaming but is still stays indoor. To handle the above, extend the indoor configuration API to allow user space to indicate a change of indoor settings, and allow it to indicate weather it controls the indoor setting, such that: 1. If the user space process explicitly indicates that it is going to control the indoor setting, do not clear the indoor setting internally, unless the socket is released. The user space process should use the NL80211_ATTR_SOCKET_OWNER attribute in the command to state that it is going to control the indoor setting. 2. Reset the indoor setting when restoring the regulatory settings in case it is not owned by a user space process. Based on the above, a user space tool that continuously monitors the indoor settings, i.e., tracking power setting, location etc., can indicate environment changes to the regulatory core. It should be noted that currently user space is the only provided mechanism used to hint to the regulatory core over the indoor/outdoor environment -- while the country IEs do have an environment setting this has been completely ignored by the regulatory core by design for a while now since country IEs typically can contain bogus data. Acked-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: ArikX Nemtsov <arik@wizery.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-04 05:32:06 +00:00
u32 owner_nlportid;
nl80211: disallow user requests prior to regulatory_init() If cfg80211 is built into the kernel there is perhaps a small time window betwen nl80211_init() and regulatory_init() where cfg80211_regdomain hasn't yet been initialized to let the wireless core do its work. During that rare case and time frame (if its even possible) we don't allow user regulatory changes as cfg80211 is working on enabling its first regulatory domain. To check for cfg80211_regdomain we now contend the entire operation using the cfg80211_mutex. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-21 05:04:22 +00:00
/*
* You should only get this when cfg80211 hasn't yet initialized
* completely when built-in to the kernel right between the time
* window between nl80211_init() and regulatory_init(), if that is
* even possible.
*/
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
if (unlikely(!rcu_access_pointer(cfg80211_regdomain)))
cfg80211: move all regulatory hints to workqueue All regulatory hints (core, driver, userspace and 11d) are now processed in a workqueue. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-21 05:04:30 +00:00
return -EINPROGRESS;
nl80211: disallow user requests prior to regulatory_init() If cfg80211 is built into the kernel there is perhaps a small time window betwen nl80211_init() and regulatory_init() where cfg80211_regdomain hasn't yet been initialized to let the wireless core do its work. During that rare case and time frame (if its even possible) we don't allow user regulatory changes as cfg80211 is working on enabling its first regulatory domain. To check for cfg80211_regdomain we now contend the entire operation using the cfg80211_mutex. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-21 05:04:22 +00:00
cfg80211: add cellular base station regulatory hint support Cellular base stations can provide hints to cfg80211 about where they think we are. This can be done for example on a cell phone. To enable these hints we simply allow them through as user regulatory hints but we allow userspace to clasify the hint as either coming directly from the user or coming from a cellular base station. This option is only available when you enable CONFIG_CFG80211_CERTIFICATION_ONUS. The base station hints themselves will not be processed by the core unless at least one device on the system supports this feature. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 18:49:18 +00:00
if (info->attrs[NL80211_ATTR_USER_REG_HINT_TYPE])
user_reg_hint_type =
nla_get_u32(info->attrs[NL80211_ATTR_USER_REG_HINT_TYPE]);
else
user_reg_hint_type = NL80211_USER_REG_HINT_USER;
switch (user_reg_hint_type) {
case NL80211_USER_REG_HINT_USER:
case NL80211_USER_REG_HINT_CELL_BASE:
cfg80211: Add an option to hint indoor operation Add the option to hint the wireless core that it is operating in an indoor environment. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-25 14:26:00 +00:00
if (!info->attrs[NL80211_ATTR_REG_ALPHA2])
return -EINVAL;
data = nla_data(info->attrs[NL80211_ATTR_REG_ALPHA2]);
return regulatory_hint_user(data, user_reg_hint_type);
case NL80211_USER_REG_HINT_INDOOR:
cfg80211: Add API to change the indoor regulatory setting Previously, the indoor setting configuration assumed that as long as a station interface is connected, the indoor environment setting does not change. However, this assumption is problematic as: - It is possible that a station interface is connected to a mobile AP, e.g., softAP or a P2P GO, where it is possible that both the station and the mobile AP move out of the indoor environment making the indoor setting invalid. In such a case, user space has no way to invalidate the setting. - A station interface disconnection does not necessarily imply that the device is no longer operating in an indoor environment, e.g., it is possible that the station interface is roaming but is still stays indoor. To handle the above, extend the indoor configuration API to allow user space to indicate a change of indoor settings, and allow it to indicate weather it controls the indoor setting, such that: 1. If the user space process explicitly indicates that it is going to control the indoor setting, do not clear the indoor setting internally, unless the socket is released. The user space process should use the NL80211_ATTR_SOCKET_OWNER attribute in the command to state that it is going to control the indoor setting. 2. Reset the indoor setting when restoring the regulatory settings in case it is not owned by a user space process. Based on the above, a user space tool that continuously monitors the indoor settings, i.e., tracking power setting, location etc., can indicate environment changes to the regulatory core. It should be noted that currently user space is the only provided mechanism used to hint to the regulatory core over the indoor/outdoor environment -- while the country IEs do have an environment setting this has been completely ignored by the regulatory core by design for a while now since country IEs typically can contain bogus data. Acked-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: ArikX Nemtsov <arik@wizery.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-04 05:32:06 +00:00
if (info->attrs[NL80211_ATTR_SOCKET_OWNER]) {
owner_nlportid = info->snd_portid;
is_indoor = !!info->attrs[NL80211_ATTR_REG_INDOOR];
} else {
owner_nlportid = 0;
is_indoor = true;
}
return regulatory_hint_indoor(is_indoor, owner_nlportid);
cfg80211: add cellular base station regulatory hint support Cellular base stations can provide hints to cfg80211 about where they think we are. This can be done for example on a cell phone. To enable these hints we simply allow them through as user regulatory hints but we allow userspace to clasify the hint as either coming directly from the user or coming from a cellular base station. This option is only available when you enable CONFIG_CFG80211_CERTIFICATION_ONUS. The base station hints themselves will not be processed by the core unless at least one device on the system supports this feature. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 18:49:18 +00:00
default:
return -EINVAL;
}
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
}
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
static int nl80211_get_mesh_config(struct sk_buff *skb,
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
struct genl_info *info)
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct mesh_config cur_params;
int err = 0;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
void *hdr;
struct nlattr *pinfoattr;
struct sk_buff *msg;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
if (wdev->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
if (!rdev->ops->get_mesh_config)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EOPNOTSUPP;
nl80211: Check that function pointer != NULL before using it NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS handlers did not verify whether a function pointer is NULL (not supported by the driver) before trying to call the function. The former nl80211 command is available for unprivileged users, too, so this can potentially allow normal users to kill networking (or worse..) if mac80211 is built without CONFIG_MAC80211_MESH=y. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 15:57:36 +00:00
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
wdev_lock(wdev);
/* If not connected, get default parameters */
if (!wdev->mesh_id_len)
memcpy(&cur_params, &default_mesh_config, sizeof(cur_params));
else
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_get_mesh_config(rdev, dev, &cur_params);
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
wdev_unlock(wdev);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
if (err)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return err;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
/* Draw up a netlink message to send back */
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!msg)
return -ENOMEM;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
NL80211_CMD_GET_MESH_CONFIG);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
if (!hdr)
net/wireless/nl80211.c: Avoid call to genlmsg_cancel genlmsg_cancel subtracts some constants from its second argument before calling nlmsg_cancel. nlmsg_cancel then calls nlmsg_trim on the same arguments. nlmsg_trim tests for NULL before doing any computation, but a NULL second argument to genlmsg_cancel is no longer NULL due to the initial subtraction. Nothing else happens in this execution, so the call to genlmsg_cancel is simply unnecessary in this case. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression data; @@ if (data == NULL) { ... * genlmsg_cancel(..., data); ... return ...; } // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-01-28 14:17:11 +00:00
goto out;
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
pinfoattr = nla_nest_start(msg, NL80211_ATTR_MESH_CONFIG);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
if (!pinfoattr)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put_u16(msg, NL80211_MESHCONF_RETRY_TIMEOUT,
cur_params.dot11MeshRetryTimeout) ||
nla_put_u16(msg, NL80211_MESHCONF_CONFIRM_TIMEOUT,
cur_params.dot11MeshConfirmTimeout) ||
nla_put_u16(msg, NL80211_MESHCONF_HOLDING_TIMEOUT,
cur_params.dot11MeshHoldingTimeout) ||
nla_put_u16(msg, NL80211_MESHCONF_MAX_PEER_LINKS,
cur_params.dot11MeshMaxPeerLinks) ||
nla_put_u8(msg, NL80211_MESHCONF_MAX_RETRIES,
cur_params.dot11MeshMaxRetries) ||
nla_put_u8(msg, NL80211_MESHCONF_TTL,
cur_params.dot11MeshTTL) ||
nla_put_u8(msg, NL80211_MESHCONF_ELEMENT_TTL,
cur_params.element_ttl) ||
nla_put_u8(msg, NL80211_MESHCONF_AUTO_OPEN_PLINKS,
cur_params.auto_open_plinks) ||
Merge branch 'master' into for-davem Conflicts: drivers/net/wireless/iwlwifi/iwl-testmode.c net/wireless/nl80211.c
2012-04-12 18:25:14 +00:00
nla_put_u32(msg, NL80211_MESHCONF_SYNC_OFFSET_MAX_NEIGHBOR,
cur_params.dot11MeshNbrOffsetMaxNeighbor) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u8(msg, NL80211_MESHCONF_HWMP_MAX_PREQ_RETRIES,
cur_params.dot11MeshHWMPmaxPREQretries) ||
nla_put_u32(msg, NL80211_MESHCONF_PATH_REFRESH_TIME,
cur_params.path_refresh_time) ||
nla_put_u16(msg, NL80211_MESHCONF_MIN_DISCOVERY_TIMEOUT,
cur_params.min_discovery_timeout) ||
nla_put_u32(msg, NL80211_MESHCONF_HWMP_ACTIVE_PATH_TIMEOUT,
cur_params.dot11MeshHWMPactivePathTimeout) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_PREQ_MIN_INTERVAL,
cur_params.dot11MeshHWMPpreqMinInterval) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_PERR_MIN_INTERVAL,
cur_params.dot11MeshHWMPperrMinInterval) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME,
cur_params.dot11MeshHWMPnetDiameterTraversalTime) ||
nla_put_u8(msg, NL80211_MESHCONF_HWMP_ROOTMODE,
cur_params.dot11MeshHWMPRootMode) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_RANN_INTERVAL,
cur_params.dot11MeshHWMPRannInterval) ||
nla_put_u8(msg, NL80211_MESHCONF_GATE_ANNOUNCEMENTS,
cur_params.dot11MeshGateAnnouncementProtocol) ||
nla_put_u8(msg, NL80211_MESHCONF_FORWARDING,
cur_params.dot11MeshForwarding) ||
nl80211: Use signed function for a signed variable The rssi_threshold is defined as s32. Signed-off-by: Masashi Honma <masashi.honma@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-16 01:57:17 +00:00
nla_put_s32(msg, NL80211_MESHCONF_RSSI_THRESHOLD,
{nl,cfg,mac}80211: Allow user to see/configure HT protection mode This patch introduces a new mesh configuration parameter "ht_opmode" and will allow user to check the current HT protection mode selected. Users could configure the protection mode by the command "iw mesh_iface set mesh_param mesh_ht_protection_mode=2". The default protection mode of mesh is set to non-HT mixed mode. Signed-off-by: Ashok Nagarajan <ashok@cozybit.com> Reviewed-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-30 21:20:32 +00:00
cur_params.rssi_threshold) ||
nla_put_u32(msg, NL80211_MESHCONF_HT_OPMODE,
{nl,cfg,mac}80211: implement dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout Add the mesh configuration parameters dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout to be used by proactive PREQ mechanism. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:06 +00:00
cur_params.ht_opmode) ||
nla_put_u32(msg, NL80211_MESHCONF_HWMP_PATH_TO_ROOT_TIMEOUT,
cur_params.dot11MeshHWMPactivePathToRootTimeout) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_ROOT_INTERVAL,
{nl,cfg,mac}80211: implement dot11MeshHWMPconfirmationInterval As defined in section 13.10.9.3 Case D (802.11-2012), this control variable is used to limit the mesh STA to send only one PREQ to a root mesh STA within this interval of time (in TUs). The default value for this variable is set to 2000 TUs. However, for current implementation, the maximum configurable of dot11MeshHWMPconfirmationInterval is restricted by dot11MeshHWMPactivePathTimeout. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:10 +00:00
cur_params.dot11MeshHWMProotInterval) ||
nla_put_u16(msg, NL80211_MESHCONF_HWMP_CONFIRMATION_INTERVAL,
{cfg,nl}80211: mesh power mode primitives and userspace access Add the nl80211_mesh_power_mode enumeration which holds possible values for the mesh power mode. These modes are unknown, active, light sleep and deep sleep. Add power_mode entry to the mesh config structure to hold the user-configured default mesh power mode. This value will be used for new peer links. Add the dot11MeshAwakeWindowDuration value to the mesh config. The awake window is a duration in TU describing how long the STA will stay awake after transmitting its beacon in PS mode. Add access routines to: - get/set local link-specific power mode (STA) - get remote STA's link-specific power mode (STA) - get remote STA's non-peer power mode (STA) - get/set default mesh power mode (mesh config) - get/set mesh awake window duration (mesh config) All config changes may be done at mesh runtime and take effect immediately. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Ivan Bezyazychnyy <ivan.bezyazychnyy@gmail.com> Signed-off-by: Mike Krinkin <krinkin.m.u@gmail.com> [fix commit message line length, error handling in set station] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:52 +00:00
cur_params.dot11MeshHWMPconfirmationInterval) ||
nla_put_u32(msg, NL80211_MESHCONF_POWER_MODE,
cur_params.power_mode) ||
nla_put_u16(msg, NL80211_MESHCONF_AWAKE_WINDOW,
{nl,cfg}80211: make peer link expiration time configurable If a STA has a peer that it hasn't seen any tx activity from for a certain length of time, the peer link is expired. This means the inactive STA is removed from the list of peers and that STA is not considered a peer again unless it re-peers. Previously, this inactivity time was always 30 minutes. Now, add it to the mesh configuration and allow it to be configured. Retain 30 minutes as a default value. Signed-off-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 16:53:39 +00:00
cur_params.dot11MeshAwakeWindowDuration) ||
nla_put_u32(msg, NL80211_MESHCONF_PLINK_TIMEOUT,
cur_params.plink_timeout))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
nla_nest_end(msg, pinfoattr);
genlmsg_end(msg, hdr);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
nla_put_failure:
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
genlmsg_cancel(msg, hdr);
net/wireless/nl80211.c: Avoid call to genlmsg_cancel genlmsg_cancel subtracts some constants from its second argument before calling nlmsg_cancel. nlmsg_cancel then calls nlmsg_trim on the same arguments. nlmsg_trim tests for NULL before doing any computation, but a NULL second argument to genlmsg_cancel is no longer NULL due to the initial subtraction. Nothing else happens in this execution, so the call to genlmsg_cancel is simply unnecessary in this case. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression data; @@ if (data == NULL) { ... * genlmsg_cancel(..., data); ... return ...; } // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-01-28 14:17:11 +00:00
out:
nl80211: Fix memory leaks In case of errors during message composing msg should be freed after canceling. Signed-off-by: Yuri Kululin <ext-yuri.kululin@nokia.com> Signed-off-by: Yuri Ershov <ext-yuri.ershov@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-29 11:08:07 +00:00
nlmsg_free(msg);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -ENOBUFS;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
}
const: struct nla_policy Make remaining netlink policies as const. Fixup coding style where needed. Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2010-02-18 08:14:31 +00:00
static const struct nla_policy nl80211_meshconf_params_policy[NL80211_MESHCONF_ATTR_MAX+1] = {
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
[NL80211_MESHCONF_RETRY_TIMEOUT] = { .type = NLA_U16 },
[NL80211_MESHCONF_CONFIRM_TIMEOUT] = { .type = NLA_U16 },
[NL80211_MESHCONF_HOLDING_TIMEOUT] = { .type = NLA_U16 },
[NL80211_MESHCONF_MAX_PEER_LINKS] = { .type = NLA_U16 },
[NL80211_MESHCONF_MAX_RETRIES] = { .type = NLA_U8 },
[NL80211_MESHCONF_TTL] = { .type = NLA_U8 },
nl80211/mac80211: define and allow configuring mesh element TTL The TTL in path selection information elements is different from the mesh ttl used in mesh data frames. Version 7.03 of the 11s draft calls this ttl 'Element TTL'. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:40 +00:00
[NL80211_MESHCONF_ELEMENT_TTL] = { .type = NLA_U8 },
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
[NL80211_MESHCONF_AUTO_OPEN_PLINKS] = { .type = NLA_U8 },
{nl,cfg}80211: Support for mesh synchronization Report Toffset to userspace. Let userspace select the mesh synchronization method. Signed-off-by: Marco Porsch <marco.porsch@s2005.tu-chemnitz.de> Signed-off-by: Pavel Zubarev <pavel.zubarev@gmail.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-31 18:31:33 +00:00
[NL80211_MESHCONF_SYNC_OFFSET_MAX_NEIGHBOR] = { .type = NLA_U32 },
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
[NL80211_MESHCONF_HWMP_MAX_PREQ_RETRIES] = { .type = NLA_U8 },
[NL80211_MESHCONF_PATH_REFRESH_TIME] = { .type = NLA_U32 },
[NL80211_MESHCONF_MIN_DISCOVERY_TIMEOUT] = { .type = NLA_U16 },
[NL80211_MESHCONF_HWMP_ACTIVE_PATH_TIMEOUT] = { .type = NLA_U32 },
[NL80211_MESHCONF_HWMP_PREQ_MIN_INTERVAL] = { .type = NLA_U16 },
{nl,cfg,mac}80211: implement dot11MeshHWMPperrMinInterval As per 802.11mb 13.9.11.3 Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-25 01:15:24 +00:00
[NL80211_MESHCONF_HWMP_PERR_MIN_INTERVAL] = { .type = NLA_U16 },
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
[NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME] = { .type = NLA_U16 },
{nl,mac}80211: add missing root mode meshconf entries This fix allows userspace to mark a meshif as a root node. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-09 23:45:09 +00:00
[NL80211_MESHCONF_HWMP_ROOTMODE] = { .type = NLA_U8 },
{nl,cfg,mac}80211: let userspace set RANN interval Allow userspace to set Root Announcement Interval for our mesh interface. Also, RANN interval is now in proper units of TUs. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-09 23:45:10 +00:00
[NL80211_MESHCONF_HWMP_RANN_INTERVAL] = { .type = NLA_U16 },
{nl,cfg,mac}80211: let userspace make meshif mesh gate Allow userspace to set NL80211_MESHCONF_GATE_ANNOUNCEMENTS attribute, which will advertise this mesh node as being a mesh gate. NL80211_HWMP_ROOTMODE must be set or this will do nothing. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-09 23:45:11 +00:00
[NL80211_MESHCONF_GATE_ANNOUNCEMENTS] = { .type = NLA_U8 },
{nl,cfg,mac}80211: Add support of setting non-forwarding entity in Mesh A mesh node that joins the mesh network is by default a forwarding entity. This patch allows the mesh node to set as non-forwarding entity. Whenever dot11MeshForwarding is set to 0, the mesh node can prevent itself from forwarding the traffic which is not destined to him. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-20 17:02:16 +00:00
[NL80211_MESHCONF_FORWARDING] = { .type = NLA_U8 },
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
[NL80211_MESHCONF_RSSI_THRESHOLD] = { .type = NLA_U32 },
[NL80211_MESHCONF_HT_OPMODE] = { .type = NLA_U16 },
{nl,cfg,mac}80211: implement dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout Add the mesh configuration parameters dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout to be used by proactive PREQ mechanism. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:06 +00:00
[NL80211_MESHCONF_HWMP_PATH_TO_ROOT_TIMEOUT] = { .type = NLA_U32 },
[NL80211_MESHCONF_HWMP_ROOT_INTERVAL] = { .type = NLA_U16 },
{nl,cfg,mac}80211: implement dot11MeshHWMPconfirmationInterval As defined in section 13.10.9.3 Case D (802.11-2012), this control variable is used to limit the mesh STA to send only one PREQ to a root mesh STA within this interval of time (in TUs). The default value for this variable is set to 2000 TUs. However, for current implementation, the maximum configurable of dot11MeshHWMPconfirmationInterval is restricted by dot11MeshHWMPactivePathTimeout. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:10 +00:00
[NL80211_MESHCONF_HWMP_CONFIRMATION_INTERVAL] = { .type = NLA_U16 },
{cfg,nl}80211: mesh power mode primitives and userspace access Add the nl80211_mesh_power_mode enumeration which holds possible values for the mesh power mode. These modes are unknown, active, light sleep and deep sleep. Add power_mode entry to the mesh config structure to hold the user-configured default mesh power mode. This value will be used for new peer links. Add the dot11MeshAwakeWindowDuration value to the mesh config. The awake window is a duration in TU describing how long the STA will stay awake after transmitting its beacon in PS mode. Add access routines to: - get/set local link-specific power mode (STA) - get remote STA's link-specific power mode (STA) - get remote STA's non-peer power mode (STA) - get/set default mesh power mode (mesh config) - get/set mesh awake window duration (mesh config) All config changes may be done at mesh runtime and take effect immediately. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Ivan Bezyazychnyy <ivan.bezyazychnyy@gmail.com> Signed-off-by: Mike Krinkin <krinkin.m.u@gmail.com> [fix commit message line length, error handling in set station] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:52 +00:00
[NL80211_MESHCONF_POWER_MODE] = { .type = NLA_U32 },
[NL80211_MESHCONF_AWAKE_WINDOW] = { .type = NLA_U16 },
{nl,cfg}80211: make peer link expiration time configurable If a STA has a peer that it hasn't seen any tx activity from for a certain length of time, the peer link is expired. This means the inactive STA is removed from the list of peers and that STA is not considered a peer again unless it re-peers. Previously, this inactivity time was always 30 minutes. Now, add it to the mesh configuration and allow it to be configured. Retain 30 minutes as a default value. Signed-off-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 16:53:39 +00:00
[NL80211_MESHCONF_PLINK_TIMEOUT] = { .type = NLA_U32 },
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
};
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
static const struct nla_policy
nl80211_mesh_setup_params_policy[NL80211_MESH_SETUP_ATTR_MAX+1] = {
{nl,cfg}80211: Support for mesh synchronization Report Toffset to userspace. Let userspace select the mesh synchronization method. Signed-off-by: Marco Porsch <marco.porsch@s2005.tu-chemnitz.de> Signed-off-by: Pavel Zubarev <pavel.zubarev@gmail.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-31 18:31:33 +00:00
[NL80211_MESH_SETUP_ENABLE_VENDOR_SYNC] = { .type = NLA_U8 },
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
[NL80211_MESH_SETUP_ENABLE_VENDOR_PATH_SEL] = { .type = NLA_U8 },
[NL80211_MESH_SETUP_ENABLE_VENDOR_METRIC] = { .type = NLA_U8 },
cfg80211/nl80211: Add userspace authentication flag to mesh setup During mesh setup, use NL80211_MESH_SETUP_USERSPACE_AUTH flag to create a secure mesh and route management frames to userspace. Also, NL80211_CMD_GET_WIPHY now returns a flag NL80211_SUPPORT_MESH_AUTH if the wiphy's mesh implementation supports routing of mesh auth frames to userspace. This is useful for forward compatibility between old kernels and new userspace tools. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:28 +00:00
[NL80211_MESH_SETUP_USERSPACE_AUTH] = { .type = NLA_FLAG },
cfg80211: Userspace may inform kernel of mesh auth method. Authentication takes place in userspace, but the beacon is generated in the kernel. Allow userspace to inform the kernel of the authentication method so the appropriate mesh config IE can be set prior to beacon generation when joining the MBSS. Signed-off-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 18:45:59 +00:00
[NL80211_MESH_SETUP_AUTH_PROTOCOL] = { .type = NLA_U8 },
nl80211: explicit userspace MPM Secure mesh had the implicit requirement that the Mesh Peering Management entity be in userspace. However userspace might want to implement an open MPM as well, so specify a mesh setup parameter to indicate this. Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:10 +00:00
[NL80211_MESH_SETUP_USERSPACE_MPM] = { .type = NLA_FLAG },
nl80211: rename NL80211_MESH_SETUP_VENDOR_PATH_SEL_IE To NL80211_MESH_SETUP_IE. This reflects our ability to insert any ie into a mesh beacon, not simply path selection ies. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:27 +00:00
[NL80211_MESH_SETUP_IE] = { .type = NLA_BINARY,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
.len = IEEE80211_MAX_DATA_LEN },
nl80211: Introduce NL80211_MESH_SETUP_USERSPACE_AMPE Introduce a new configuration option to support AMPE from userspace. Prior to this series we only supported authentication in userspace: an authentication daemon would authenticate peer candidates in userspace and hand them over to the kernel. From that point the mesh stack would take over and establish a peer link (Mesh Peering Management). These patches introduce support for Authenticated Mesh Peering Exchange in userspace. The userspace daemon implements the AMPE protocol and on successfull completion create mesh peers and install encryption keys. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:07 +00:00
[NL80211_MESH_SETUP_USERSPACE_AMPE] = { .type = NLA_FLAG },
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
};
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
static int nl80211_check_bool(const struct nlattr *nla, u8 min, u8 max, bool *out)
{
u8 val = nla_get_u8(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
static int nl80211_check_u8(const struct nlattr *nla, u8 min, u8 max, u8 *out)
{
u8 val = nla_get_u8(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
static int nl80211_check_u16(const struct nlattr *nla, u16 min, u16 max, u16 *out)
{
u16 val = nla_get_u16(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
static int nl80211_check_u32(const struct nlattr *nla, u32 min, u32 max, u32 *out)
{
u32 val = nla_get_u32(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
static int nl80211_check_s32(const struct nlattr *nla, s32 min, s32 max, s32 *out)
{
s32 val = nla_get_s32(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
nl80211: explicitly check enum nl80211_mesh_power_mode Different gcc versions appear to be treating enum with different signedness, causing warnings with the out parameter one way or the other. Just use the correct type to avoid all that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 12:59:53 +00:00
static int nl80211_check_power_mode(const struct nlattr *nla,
enum nl80211_mesh_power_mode min,
enum nl80211_mesh_power_mode max,
enum nl80211_mesh_power_mode *out)
{
u32 val = nla_get_u32(nla);
if (val < min || val > max)
return -EINVAL;
*out = val;
return 0;
}
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
static int nl80211_parse_mesh_config(struct genl_info *info,
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
struct mesh_config *cfg,
u32 *mask_out)
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
{
struct nlattr *tb[NL80211_MESHCONF_ATTR_MAX + 1];
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
u32 mask = 0;
nl80211: correct checks for NL80211_MESHCONF_HT_OPMODE value Previously, NL80211_MESHCONF_HT_OPMODE validation rejected correct flag combinations, e.g. IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED | IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT. Doing just a range-check allows setting flags that don't exist (0x8) and invalid flag combinations. Implements some checks based on IEEE 802.11 2012 8.4.2.59 "HT Operation element". Signed-off-by: Masashi Honma <masashi.honma@gmail.com> [reword commit message, simplify a bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 01:07:44 +00:00
u16 ht_opmode;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
#define FILL_IN_MESH_PARAM_IF_SET(tb, cfg, param, min, max, mask, attr, fn) \
do { \
if (tb[attr]) { \
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
if (fn(tb[attr], min, max, &cfg->param)) \
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
return -EINVAL; \
mask |= (1 << (attr - 1)); \
} \
} while (0)
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
if (!info->attrs[NL80211_ATTR_MESH_CONFIG])
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
return -EINVAL;
if (nla_parse_nested(tb, NL80211_MESHCONF_ATTR_MAX,
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
info->attrs[NL80211_ATTR_MESH_CONFIG],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_meshconf_params_policy, info->extack))
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
return -EINVAL;
/* This makes sure that there aren't more than 32 mesh config
* parameters (otherwise our bitfield scheme would not work.) */
BUILD_BUG_ON(NL80211_MESHCONF_ATTR_MAX > 32);
/* Fill in the params struct */
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshRetryTimeout, 1, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_RETRY_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshConfirmTimeout, 1, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_CONFIRM_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHoldingTimeout, 1, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_HOLDING_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshMaxPeerLinks, 0, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_MAX_PEER_LINKS,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshMaxRetries, 0, 16,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_MAX_RETRIES,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u8);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshTTL, 1, 255,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
mask, NL80211_MESHCONF_TTL, nl80211_check_u8);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, element_ttl, 1, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_ELEMENT_TTL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u8);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, auto_open_plinks, 0, 1,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_AUTO_OPEN_PLINKS,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_bool);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshNbrOffsetMaxNeighbor,
1, 255, mask,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
NL80211_MESHCONF_SYNC_OFFSET_MAX_NEIGHBOR,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u32);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPmaxPREQretries, 0, 255,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_HWMP_MAX_PREQ_RETRIES,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u8);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, path_refresh_time, 1, 65535,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_PATH_REFRESH_TIME,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u32);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, min_discovery_timeout, 1, 65535,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_MIN_DISCOVERY_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPactivePathTimeout,
1, 65535, mask,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
NL80211_MESHCONF_HWMP_ACTIVE_PATH_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u32);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPpreqMinInterval,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
1, 65535, mask,
NL80211_MESHCONF_HWMP_PREQ_MIN_INTERVAL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
{nl,cfg,mac}80211: implement dot11MeshHWMPperrMinInterval As per 802.11mb 13.9.11.3 Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-25 01:15:24 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPperrMinInterval,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
1, 65535, mask,
NL80211_MESHCONF_HWMP_PERR_MIN_INTERVAL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
dot11MeshHWMPnetDiameterTraversalTime,
1, 65535, mask,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
NL80211_MESHCONF_HWMP_NET_DIAM_TRVS_TIME,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPRootMode, 0, 4,
mask, NL80211_MESHCONF_HWMP_ROOTMODE,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u8);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPRannInterval, 1, 65535,
mask, NL80211_MESHCONF_HWMP_RANN_INTERVAL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
mac80211: add nl80211/cfg80211 handling of the new mesh root mode option. Signed-off-by: Rui Paulo <rpaulo@gmail.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Reviewed-by: Andrey Yurovsky <andrey@cozybit.com> Tested-by: Brian Cavagnolo <brian@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-09 23:46:57 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
dot11MeshGateAnnouncementProtocol, 0, 1,
mask, NL80211_MESHCONF_GATE_ANNOUNCEMENTS,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_bool);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshForwarding, 0, 1,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_FORWARDING,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_bool);
nl80211: fix the setting of RSSI threshold value for mesh RSSI threshold value used for mesh peering should be in negative value. After range checks to mesh parameters is introduced, this is not allowed. Fix this. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-07-11 10:24:03 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, rssi_threshold, -255, 0,
{nl,cfg,mac}80211: fix the coding style related to mesh parameters fix the coding style related to mesh parameters, especially the indentation, as pointed out by Johannes Berg. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-11 03:59:36 +00:00
mask, NL80211_MESHCONF_RSSI_THRESHOLD,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_s32);
nl80211: correct checks for NL80211_MESHCONF_HT_OPMODE value Previously, NL80211_MESHCONF_HT_OPMODE validation rejected correct flag combinations, e.g. IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED | IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT. Doing just a range-check allows setting flags that don't exist (0x8) and invalid flag combinations. Implements some checks based on IEEE 802.11 2012 8.4.2.59 "HT Operation element". Signed-off-by: Masashi Honma <masashi.honma@gmail.com> [reword commit message, simplify a bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 01:07:44 +00:00
/*
* Check HT operation mode based on
* IEEE 802.11 2012 8.4.2.59 HT Operation element.
*/
if (tb[NL80211_MESHCONF_HT_OPMODE]) {
ht_opmode = nla_get_u16(tb[NL80211_MESHCONF_HT_OPMODE]);
if (ht_opmode & ~(IEEE80211_HT_OP_MODE_PROTECTION |
IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT |
IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT))
return -EINVAL;
if ((ht_opmode & IEEE80211_HT_OP_MODE_NON_GF_STA_PRSNT) &&
(ht_opmode & IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT))
return -EINVAL;
switch (ht_opmode & IEEE80211_HT_OP_MODE_PROTECTION) {
case IEEE80211_HT_OP_MODE_PROTECTION_NONE:
case IEEE80211_HT_OP_MODE_PROTECTION_20MHZ:
if (ht_opmode & IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT)
return -EINVAL;
break;
case IEEE80211_HT_OP_MODE_PROTECTION_NONMEMBER:
case IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED:
if (!(ht_opmode & IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT))
return -EINVAL;
break;
}
cfg->ht_opmode = ht_opmode;
nl80211: Fix mesh HT operation check A previous change to fix checks for NL80211_MESHCONF_HT_OPMODE missed setting the flag when replacing FILL_IN_MESH_PARAM_IF_SET with checking codes. This results in dropping the received HT operation value when called by nl80211_update_mesh_config(). Fix this by setting the flag properly. Fixes: 9757235f451c ("nl80211: correct checks for NL80211_MESHCONF_HT_OPMODE value") Signed-off-by: Masashi Honma <masashi.honma@gmail.com> [rewrite commit message to use Fixes: line] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-25 23:56:13 +00:00
mask |= (1 << (NL80211_MESHCONF_HT_OPMODE - 1));
nl80211: correct checks for NL80211_MESHCONF_HT_OPMODE value Previously, NL80211_MESHCONF_HT_OPMODE validation rejected correct flag combinations, e.g. IEEE80211_HT_OP_MODE_PROTECTION_NONHT_MIXED | IEEE80211_HT_OP_MODE_NON_HT_STA_PRSNT. Doing just a range-check allows setting flags that don't exist (0x8) and invalid flag combinations. Implements some checks based on IEEE 802.11 2012 8.4.2.59 "HT Operation element". Signed-off-by: Masashi Honma <masashi.honma@gmail.com> [reword commit message, simplify a bit] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 01:07:44 +00:00
}
{nl,cfg,mac}80211: implement dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout Add the mesh configuration parameters dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout to be used by proactive PREQ mechanism. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:06 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMPactivePathToRootTimeout,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
1, 65535, mask,
{nl,cfg,mac}80211: implement dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout Add the mesh configuration parameters dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout to be used by proactive PREQ mechanism. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:06 +00:00
NL80211_MESHCONF_HWMP_PATH_TO_ROOT_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u32);
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshHWMProotInterval, 1, 65535,
{nl,cfg,mac}80211: implement dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout Add the mesh configuration parameters dot11MeshHWMProotInterval and dot11MeshHWMPactivePathToRootTimeout to be used by proactive PREQ mechanism. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:06 +00:00
mask, NL80211_MESHCONF_HWMP_ROOT_INTERVAL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
{nl,cfg,mac}80211: implement dot11MeshHWMPconfirmationInterval As defined in section 13.10.9.3 Case D (802.11-2012), this control variable is used to limit the mesh STA to send only one PREQ to a root mesh STA within this interval of time (in TUs). The default value for this variable is set to 2000 TUs. However, for current implementation, the maximum configurable of dot11MeshHWMPconfirmationInterval is restricted by dot11MeshHWMPactivePathTimeout. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:10 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg,
nl80211: add range checks to mesh parameters Ranges are taken from IEEE 802.11-2012, common sense or current implementation requirements. Signed-off-by: Marco Porsch <marco@cozybit.com> Acked-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:48 +00:00
dot11MeshHWMPconfirmationInterval,
1, 65535, mask,
{nl,cfg,mac}80211: implement dot11MeshHWMPconfirmationInterval As defined in section 13.10.9.3 Case D (802.11-2012), this control variable is used to limit the mesh STA to send only one PREQ to a root mesh STA within this interval of time (in TUs). The default value for this variable is set to 2000 TUs. However, for current implementation, the maximum configurable of dot11MeshHWMPconfirmationInterval is restricted by dot11MeshHWMPactivePathTimeout. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> [line-break commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 18:06:10 +00:00
NL80211_MESHCONF_HWMP_CONFIRMATION_INTERVAL,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u16);
{cfg,nl}80211: mesh power mode primitives and userspace access Add the nl80211_mesh_power_mode enumeration which holds possible values for the mesh power mode. These modes are unknown, active, light sleep and deep sleep. Add power_mode entry to the mesh config structure to hold the user-configured default mesh power mode. This value will be used for new peer links. Add the dot11MeshAwakeWindowDuration value to the mesh config. The awake window is a duration in TU describing how long the STA will stay awake after transmitting its beacon in PS mode. Add access routines to: - get/set local link-specific power mode (STA) - get remote STA's link-specific power mode (STA) - get remote STA's non-peer power mode (STA) - get/set default mesh power mode (mesh config) - get/set mesh awake window duration (mesh config) All config changes may be done at mesh runtime and take effect immediately. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Ivan Bezyazychnyy <ivan.bezyazychnyy@gmail.com> Signed-off-by: Mike Krinkin <krinkin.m.u@gmail.com> [fix commit message line length, error handling in set station] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:52 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, power_mode,
NL80211_MESH_POWER_ACTIVE,
NL80211_MESH_POWER_MAX,
mask, NL80211_MESHCONF_POWER_MODE,
nl80211: explicitly check enum nl80211_mesh_power_mode Different gcc versions appear to be treating enum with different signedness, causing warnings with the out parameter one way or the other. Just use the correct type to avoid all that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 12:59:53 +00:00
nl80211_check_power_mode);
{cfg,nl}80211: mesh power mode primitives and userspace access Add the nl80211_mesh_power_mode enumeration which holds possible values for the mesh power mode. These modes are unknown, active, light sleep and deep sleep. Add power_mode entry to the mesh config structure to hold the user-configured default mesh power mode. This value will be used for new peer links. Add the dot11MeshAwakeWindowDuration value to the mesh config. The awake window is a duration in TU describing how long the STA will stay awake after transmitting its beacon in PS mode. Add access routines to: - get/set local link-specific power mode (STA) - get remote STA's link-specific power mode (STA) - get remote STA's non-peer power mode (STA) - get/set default mesh power mode (mesh config) - get/set mesh awake window duration (mesh config) All config changes may be done at mesh runtime and take effect immediately. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Ivan Bezyazychnyy <ivan.bezyazychnyy@gmail.com> Signed-off-by: Mike Krinkin <krinkin.m.u@gmail.com> [fix commit message line length, error handling in set station] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:52 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, dot11MeshAwakeWindowDuration,
0, 65535, mask,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
NL80211_MESHCONF_AWAKE_WINDOW, nl80211_check_u16);
nl/mac80211: allow zero plink timeout to disable STA expiration Both wpa_supplicant and mac80211 have and inactivity timer. By default wpa_supplicant will be timed out in 5 minutes and mac80211's it is 30 minutes. If wpa_supplicant uses a longer timer than mac80211, it will get unexpected disconnection by mac80211. Using 0xffffffff instead as the configured value could solve this w/o changing the code, but due to integer overflow in the expression used this doesn't work. The expression is: (current jiffies) > (frame Rx jiffies + NL80211_MESHCONF_PLINK_TIMEOUT * 250) On 32bit system, the right side would overflow and be a very small value if NL80211_MESHCONF_PLINK_TIMEOUT is sufficiently large, causing unexpectedly early disconnections. Instead allow disabling the inactivity timer to avoid this situation, by passing the (previously invalid and useless) value 0. Signed-off-by: Masashi Honma <masashi.honma@gmail.com> [reword/rewrap commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-24 13:42:16 +00:00
FILL_IN_MESH_PARAM_IF_SET(tb, cfg, plink_timeout, 0, 0xffffffff,
{nl,cfg}80211: make peer link expiration time configurable If a STA has a peer that it hasn't seen any tx activity from for a certain length of time, the peer link is expired. This means the inactive STA is removed from the list of peers and that STA is not considered a peer again unless it re-peers. Previously, this inactivity time was always 30 minutes. Now, add it to the mesh configuration and allow it to be configured. Retain 30 minutes as a default value. Signed-off-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 16:53:39 +00:00
mask, NL80211_MESHCONF_PLINK_TIMEOUT,
nl80211: improve nl80211_parse_mesh_config type checking When building a kernel with W=1, the nl80211.c file causes a number of warnings, all about the same problem: net/wireless/nl80211.c: In function 'nl80211_parse_mesh_config': net/wireless/nl80211.c:5287:103: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5290:96: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5293:124: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5295:148: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5298:106: error: comparison is always false due to limited range of data type [-Werror=type-limits] net/wireless/nl80211.c:5305:116: error: comparison is always false due to limited range of data type [-Werror=type-limits] The problem is that gcc does not notice that the check is generate by a macro, so it complains about comparing an unsigned type against 0. I've tried to come up with a way to rephrase that code in a way that avoids the warnings and otherwise improves the code as well. This uses a set of new helper functions that perform the range checking, and should provide slightly better type safety than the older patch, at the expense of adding 44 lines to the code. Binary code size is basically unchanged though (20 bytes added to 126561 bytes .text). Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2016-06-15 20:29:41 +00:00
nl80211_check_u32);
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
if (mask_out)
*mask_out = mask;
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
return 0;
#undef FILL_IN_MESH_PARAM_IF_SET
}
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
static int nl80211_parse_mesh_setup(struct genl_info *info,
struct mesh_setup *setup)
{
nl80211: explicit userspace MPM Secure mesh had the implicit requirement that the Mesh Peering Management entity be in userspace. However userspace might want to implement an open MPM as well, so specify a mesh setup parameter to indicate this. Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:10 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
struct nlattr *tb[NL80211_MESH_SETUP_ATTR_MAX + 1];
if (!info->attrs[NL80211_ATTR_MESH_SETUP])
return -EINVAL;
if (nla_parse_nested(tb, NL80211_MESH_SETUP_ATTR_MAX,
info->attrs[NL80211_ATTR_MESH_SETUP],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_mesh_setup_params_policy, info->extack))
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
return -EINVAL;
{nl,cfg}80211: Support for mesh synchronization Report Toffset to userspace. Let userspace select the mesh synchronization method. Signed-off-by: Marco Porsch <marco.porsch@s2005.tu-chemnitz.de> Signed-off-by: Pavel Zubarev <pavel.zubarev@gmail.com> Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-31 18:31:33 +00:00
if (tb[NL80211_MESH_SETUP_ENABLE_VENDOR_SYNC])
setup->sync_method =
(nla_get_u8(tb[NL80211_MESH_SETUP_ENABLE_VENDOR_SYNC])) ?
IEEE80211_SYNC_METHOD_VENDOR :
IEEE80211_SYNC_METHOD_NEIGHBOR_OFFSET;
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
if (tb[NL80211_MESH_SETUP_ENABLE_VENDOR_PATH_SEL])
setup->path_sel_proto =
(nla_get_u8(tb[NL80211_MESH_SETUP_ENABLE_VENDOR_PATH_SEL])) ?
IEEE80211_PATH_PROTOCOL_VENDOR :
IEEE80211_PATH_PROTOCOL_HWMP;
if (tb[NL80211_MESH_SETUP_ENABLE_VENDOR_METRIC])
setup->path_metric =
(nla_get_u8(tb[NL80211_MESH_SETUP_ENABLE_VENDOR_METRIC])) ?
IEEE80211_PATH_METRIC_VENDOR :
IEEE80211_PATH_METRIC_AIRTIME;
nl80211: rename NL80211_MESH_SETUP_VENDOR_PATH_SEL_IE To NL80211_MESH_SETUP_IE. This reflects our ability to insert any ie into a mesh beacon, not simply path selection ies. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:27 +00:00
if (tb[NL80211_MESH_SETUP_IE]) {
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
struct nlattr *ieattr =
nl80211: rename NL80211_MESH_SETUP_VENDOR_PATH_SEL_IE To NL80211_MESH_SETUP_IE. This reflects our ability to insert any ie into a mesh beacon, not simply path selection ies. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:27 +00:00
tb[NL80211_MESH_SETUP_IE];
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
if (!is_valid_ie_attr(ieattr))
return -EINVAL;
nl80211: rename NL80211_MESH_SETUP_VENDOR_PATH_SEL_IE To NL80211_MESH_SETUP_IE. This reflects our ability to insert any ie into a mesh beacon, not simply path selection ies. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:27 +00:00
setup->ie = nla_data(ieattr);
setup->ie_len = nla_len(ieattr);
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
}
nl80211: explicit userspace MPM Secure mesh had the implicit requirement that the Mesh Peering Management entity be in userspace. However userspace might want to implement an open MPM as well, so specify a mesh setup parameter to indicate this. Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:10 +00:00
if (tb[NL80211_MESH_SETUP_USERSPACE_MPM] &&
!(rdev->wiphy.features & NL80211_FEATURE_USERSPACE_MPM))
return -EINVAL;
setup->user_mpm = nla_get_flag(tb[NL80211_MESH_SETUP_USERSPACE_MPM]);
nl80211: Introduce NL80211_MESH_SETUP_USERSPACE_AMPE Introduce a new configuration option to support AMPE from userspace. Prior to this series we only supported authentication in userspace: an authentication daemon would authenticate peer candidates in userspace and hand them over to the kernel. From that point the mesh stack would take over and establish a peer link (Mesh Peering Management). These patches introduce support for Authenticated Mesh Peering Exchange in userspace. The userspace daemon implements the AMPE protocol and on successfull completion create mesh peers and install encryption keys. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-03 23:57:07 +00:00
setup->is_authenticated = nla_get_flag(tb[NL80211_MESH_SETUP_USERSPACE_AUTH]);
setup->is_secure = nla_get_flag(tb[NL80211_MESH_SETUP_USERSPACE_AMPE]);
nl80211: explicit userspace MPM Secure mesh had the implicit requirement that the Mesh Peering Management entity be in userspace. However userspace might want to implement an open MPM as well, so specify a mesh setup parameter to indicate this. Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:10 +00:00
if (setup->is_secure)
setup->user_mpm = true;
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
cfg80211: Userspace may inform kernel of mesh auth method. Authentication takes place in userspace, but the beacon is generated in the kernel. Allow userspace to inform the kernel of the authentication method so the appropriate mesh config IE can be set prior to beacon generation when joining the MBSS. Signed-off-by: Colleen Twitty <colleen@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 18:45:59 +00:00
if (tb[NL80211_MESH_SETUP_AUTH_PROTOCOL]) {
if (!setup->user_mpm)
return -EINVAL;
setup->auth_id =
nla_get_u8(tb[NL80211_MESH_SETUP_AUTH_PROTOCOL]);
}
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
return 0;
}
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
static int nl80211_update_mesh_config(struct sk_buff *skb,
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
struct genl_info *info)
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
struct mesh_config cfg;
u32 mask;
int err;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
if (wdev->iftype != NL80211_IFTYPE_MESH_POINT)
return -EOPNOTSUPP;
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
if (!rdev->ops->update_mesh_config)
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
return -EOPNOTSUPP;
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
err = nl80211_parse_mesh_config(info, &cfg, &mask);
nl80211: refactor mesh parameter parsing I'm going to need this in a new place later. Tested-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:43 +00:00
if (err)
return err;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
wdev_lock(wdev);
if (!wdev->mesh_id_len)
err = -ENOLINK;
if (!err)
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_update_mesh_config(rdev, dev, mask, &cfg);
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
wdev_unlock(wdev);
return err;
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
}
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
static int nl80211_put_regdom(const struct ieee80211_regdomain *regdom,
struct sk_buff *msg)
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
{
struct nlattr *nl_reg_rules;
unsigned int i;
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
if (nla_put_string(msg, NL80211_ATTR_REG_ALPHA2, regdom->alpha2) ||
(regdom->dfs_region &&
nla_put_u8(msg, NL80211_ATTR_DFS_REGION, regdom->dfs_region)))
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
goto nla_put_failure;
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
nl_reg_rules = nla_nest_start(msg, NL80211_ATTR_REG_RULES);
if (!nl_reg_rules)
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
goto nla_put_failure;
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
for (i = 0; i < regdom->n_reg_rules; i++) {
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
struct nlattr *nl_reg_rule;
const struct ieee80211_reg_rule *reg_rule;
const struct ieee80211_freq_range *freq_range;
const struct ieee80211_power_rule *power_rule;
cfg80211: regulatory introduce maximum bandwidth calculation In case we will get regulatory request with rule where max_bandwidth_khz is set to 0 handle this case as a special one. If max_bandwidth_khz == 0 we should calculate maximum available bandwidth base on all frequency contiguous rules. In case we need auto calculation we just have to set: country PL: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ AUTO), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS This mean we will calculate maximum bw for rules where AUTO (N/A) were set, 160MHz (5330 - 5170) in example above. So we will get: (5170 - 5250 @ 160), (N/A, 20) (5250 - 5330 @ 160), (N/A, 20), DFS In other case: country FR: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS We will get 80MHz (5250 - 5170): (5170 - 5250 @ 80), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS Base on this calculations we will set correct channel bandwidth flags (eg. IEEE80211_CHAN_NO_80MHZ). We don't need any changes in CRDA or internal regulatory. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> [extend nl80211 description a bit, fix typo] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-30 08:52:20 +00:00
unsigned int max_bandwidth_khz;
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
reg_rule = &regdom->reg_rules[i];
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
freq_range = &reg_rule->freq_range;
power_rule = &reg_rule->power_rule;
nl_reg_rule = nla_nest_start(msg, i);
if (!nl_reg_rule)
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
goto nla_put_failure;
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
cfg80211: regulatory introduce maximum bandwidth calculation In case we will get regulatory request with rule where max_bandwidth_khz is set to 0 handle this case as a special one. If max_bandwidth_khz == 0 we should calculate maximum available bandwidth base on all frequency contiguous rules. In case we need auto calculation we just have to set: country PL: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ AUTO), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS This mean we will calculate maximum bw for rules where AUTO (N/A) were set, 160MHz (5330 - 5170) in example above. So we will get: (5170 - 5250 @ 160), (N/A, 20) (5250 - 5330 @ 160), (N/A, 20), DFS In other case: country FR: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS We will get 80MHz (5250 - 5170): (5170 - 5250 @ 80), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS Base on this calculations we will set correct channel bandwidth flags (eg. IEEE80211_CHAN_NO_80MHZ). We don't need any changes in CRDA or internal regulatory. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> [extend nl80211 description a bit, fix typo] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-30 08:52:20 +00:00
max_bandwidth_khz = freq_range->max_bandwidth_khz;
if (!max_bandwidth_khz)
max_bandwidth_khz = reg_get_max_bandwidth(regdom,
reg_rule);
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_REG_RULE_FLAGS,
reg_rule->flags) ||
nla_put_u32(msg, NL80211_ATTR_FREQ_RANGE_START,
freq_range->start_freq_khz) ||
nla_put_u32(msg, NL80211_ATTR_FREQ_RANGE_END,
freq_range->end_freq_khz) ||
nla_put_u32(msg, NL80211_ATTR_FREQ_RANGE_MAX_BW,
cfg80211: regulatory introduce maximum bandwidth calculation In case we will get regulatory request with rule where max_bandwidth_khz is set to 0 handle this case as a special one. If max_bandwidth_khz == 0 we should calculate maximum available bandwidth base on all frequency contiguous rules. In case we need auto calculation we just have to set: country PL: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ AUTO), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS This mean we will calculate maximum bw for rules where AUTO (N/A) were set, 160MHz (5330 - 5170) in example above. So we will get: (5170 - 5250 @ 160), (N/A, 20) (5250 - 5330 @ 160), (N/A, 20), DFS In other case: country FR: DFS-ETSI (2402 - 2482 @ 40), (N/A, 20) (5170 - 5250 @ AUTO), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS (5490 - 5710 @ 80), (N/A, 27), DFS We will get 80MHz (5250 - 5170): (5170 - 5250 @ 80), (N/A, 20) (5250 - 5330 @ 80), (N/A, 20), DFS Base on this calculations we will set correct channel bandwidth flags (eg. IEEE80211_CHAN_NO_80MHZ). We don't need any changes in CRDA or internal regulatory. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> [extend nl80211 description a bit, fix typo] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-30 08:52:20 +00:00
max_bandwidth_khz) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_ATTR_POWER_RULE_MAX_ANT_GAIN,
power_rule->max_antenna_gain) ||
nla_put_u32(msg, NL80211_ATTR_POWER_RULE_MAX_EIRP,
cfg80211: regulatory: allow getting DFS CAC time from userspace Introduce DFS CAC time as a regd param, configured per REG_RULE and set per channel in cfg80211. DFS CAC time is close connected with regulatory database configuration. Instead of using hardcoded values, get DFS CAC time form regulatory database. Pass DFS CAC time to user mode (mainly for iw reg get, iw list, iw info). Allow setting DFS CAC time via CRDA. Add support for internal regulatory database. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> [rewrap commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-21 18:46:12 +00:00
power_rule->max_eirp) ||
nla_put_u32(msg, NL80211_ATTR_DFS_CAC_TIME,
reg_rule->dfs_cac_ms))
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
goto nla_put_failure;
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
nla_nest_end(msg, nl_reg_rule);
}
nla_nest_end(msg, nl_reg_rules);
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
return 0;
nla_put_failure:
return -EMSGSIZE;
}
static int nl80211_get_reg_do(struct sk_buff *skb, struct genl_info *info)
{
const struct ieee80211_regdomain *regdom = NULL;
struct cfg80211_registered_device *rdev;
struct wiphy *wiphy = NULL;
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOBUFS;
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
NL80211_CMD_GET_REG);
if (!hdr)
goto put_failure;
if (info->attrs[NL80211_ATTR_WIPHY]) {
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
bool self_managed;
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
rdev = cfg80211_get_dev_from_info(genl_info_net(info), info);
if (IS_ERR(rdev)) {
nlmsg_free(msg);
return PTR_ERR(rdev);
}
wiphy = &rdev->wiphy;
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
self_managed = wiphy->regulatory_flags &
REGULATORY_WIPHY_SELF_MANAGED;
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
regdom = get_wiphy_regdom(wiphy);
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
/* a self-managed-reg device must have a private regdom */
if (WARN_ON(!regdom && self_managed)) {
nlmsg_free(msg);
return -EINVAL;
}
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
if (regdom &&
nla_put_u32(msg, NL80211_ATTR_WIPHY, get_wiphy_idx(wiphy)))
goto nla_put_failure;
}
if (!wiphy && reg_last_request_cell_base() &&
nla_put_u32(msg, NL80211_ATTR_USER_REG_HINT_TYPE,
NL80211_USER_REG_HINT_CELL_BASE))
goto nla_put_failure;
rcu_read_lock();
if (!regdom)
regdom = rcu_dereference(cfg80211_regdomain);
if (nl80211_put_regdom(regdom, msg))
goto nla_put_failure_rcu;
rcu_read_unlock();
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
genlmsg_end(msg, hdr);
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
return genlmsg_reply(msg, info);
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
regulatory: use RCU to protect global and wiphy regdomains To simplify the locking and not require cfg80211_mutex (which nl80211 uses to access the global regdomain) and also to make it possible for drivers to access their wiphy->regd safely, use RCU to protect these pointers. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-06 14:47:38 +00:00
nla_put_failure_rcu:
rcu_read_unlock();
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
net/wireless/nl80211.c: Avoid call to genlmsg_cancel genlmsg_cancel subtracts some constants from its second argument before calling nlmsg_cancel. nlmsg_cancel then calls nlmsg_trim on the same arguments. nlmsg_trim tests for NULL before doing any computation, but a NULL second argument to genlmsg_cancel is no longer NULL due to the initial subtraction. Nothing else happens in this execution, so the call to genlmsg_cancel is simply unnecessary in this case. The semantic match that finds this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ expression data; @@ if (data == NULL) { ... * genlmsg_cancel(..., data); ... return ...; } // </smpl> Signed-off-by: Julia Lawall <julia@diku.dk> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-01-28 14:17:11 +00:00
put_failure:
nl80211: Fix memory leaks In case of errors during message composing msg should be freed after canceling. Signed-off-by: Yuri Kululin <ext-yuri.kululin@nokia.com> Signed-off-by: Yuri Ershov <ext-yuri.ershov@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-29 11:08:07 +00:00
nlmsg_free(msg);
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
return -EMSGSIZE;
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
}
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
static int nl80211_send_regdom(struct sk_buff *msg, struct netlink_callback *cb,
u32 seq, int flags, struct wiphy *wiphy,
const struct ieee80211_regdomain *regdom)
{
void *hdr = nl80211hdr_put(msg, NETLINK_CB(cb->skb).portid, seq, flags,
NL80211_CMD_GET_REG);
if (!hdr)
return -1;
genl_dump_check_consistent(cb, hdr, &nl80211_fam);
if (nl80211_put_regdom(regdom, msg))
goto nla_put_failure;
if (!wiphy && reg_last_request_cell_base() &&
nla_put_u32(msg, NL80211_ATTR_USER_REG_HINT_TYPE,
NL80211_USER_REG_HINT_CELL_BASE))
goto nla_put_failure;
if (wiphy &&
nla_put_u32(msg, NL80211_ATTR_WIPHY, get_wiphy_idx(wiphy)))
goto nla_put_failure;
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
if (wiphy && wiphy->regulatory_flags & REGULATORY_WIPHY_SELF_MANAGED &&
nla_put_flag(msg, NL80211_ATTR_WIPHY_SELF_MANAGED_REG))
goto nla_put_failure;
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
static int nl80211_get_reg_dump(struct sk_buff *skb,
struct netlink_callback *cb)
{
const struct ieee80211_regdomain *regdom = NULL;
struct cfg80211_registered_device *rdev;
int err, reg_idx, start = cb->args[2];
rtnl_lock();
if (cfg80211_regdomain && start == 0) {
err = nl80211_send_regdom(skb, cb, cb->nlh->nlmsg_seq,
NLM_F_MULTI, NULL,
rtnl_dereference(cfg80211_regdomain));
if (err < 0)
goto out_err;
}
/* the global regdom is idx 0 */
reg_idx = 1;
list_for_each_entry(rdev, &cfg80211_rdev_list, list) {
regdom = get_wiphy_regdom(&rdev->wiphy);
if (!regdom)
continue;
if (++reg_idx <= start)
continue;
err = nl80211_send_regdom(skb, cb, cb->nlh->nlmsg_seq,
NLM_F_MULTI, &rdev->wiphy, regdom);
if (err < 0) {
reg_idx--;
break;
}
}
cb->args[2] = reg_idx;
err = skb->len;
out_err:
rtnl_unlock();
return err;
}
cfg80211: reg: make CRDA support optional If there's a built-in regulatory database, there may be little point in also calling out to CRDA and failing if the system is configured that way. Allow removing CRDA support to save ~1K kernel size. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-15 07:25:18 +00:00
#ifdef CONFIG_CFG80211_CRDA_SUPPORT
static const struct nla_policy reg_rule_policy[NL80211_REG_RULE_ATTR_MAX + 1] = {
[NL80211_ATTR_REG_RULE_FLAGS] = { .type = NLA_U32 },
[NL80211_ATTR_FREQ_RANGE_START] = { .type = NLA_U32 },
[NL80211_ATTR_FREQ_RANGE_END] = { .type = NLA_U32 },
[NL80211_ATTR_FREQ_RANGE_MAX_BW] = { .type = NLA_U32 },
[NL80211_ATTR_POWER_RULE_MAX_ANT_GAIN] = { .type = NLA_U32 },
[NL80211_ATTR_POWER_RULE_MAX_EIRP] = { .type = NLA_U32 },
[NL80211_ATTR_DFS_CAC_TIME] = { .type = NLA_U32 },
};
static int parse_reg_rule(struct nlattr *tb[],
struct ieee80211_reg_rule *reg_rule)
{
struct ieee80211_freq_range *freq_range = &reg_rule->freq_range;
struct ieee80211_power_rule *power_rule = &reg_rule->power_rule;
if (!tb[NL80211_ATTR_REG_RULE_FLAGS])
return -EINVAL;
if (!tb[NL80211_ATTR_FREQ_RANGE_START])
return -EINVAL;
if (!tb[NL80211_ATTR_FREQ_RANGE_END])
return -EINVAL;
if (!tb[NL80211_ATTR_FREQ_RANGE_MAX_BW])
return -EINVAL;
if (!tb[NL80211_ATTR_POWER_RULE_MAX_EIRP])
return -EINVAL;
reg_rule->flags = nla_get_u32(tb[NL80211_ATTR_REG_RULE_FLAGS]);
freq_range->start_freq_khz =
nla_get_u32(tb[NL80211_ATTR_FREQ_RANGE_START]);
freq_range->end_freq_khz =
nla_get_u32(tb[NL80211_ATTR_FREQ_RANGE_END]);
freq_range->max_bandwidth_khz =
nla_get_u32(tb[NL80211_ATTR_FREQ_RANGE_MAX_BW]);
power_rule->max_eirp =
nla_get_u32(tb[NL80211_ATTR_POWER_RULE_MAX_EIRP]);
if (tb[NL80211_ATTR_POWER_RULE_MAX_ANT_GAIN])
power_rule->max_antenna_gain =
nla_get_u32(tb[NL80211_ATTR_POWER_RULE_MAX_ANT_GAIN]);
if (tb[NL80211_ATTR_DFS_CAC_TIME])
reg_rule->dfs_cac_ms =
nla_get_u32(tb[NL80211_ATTR_DFS_CAC_TIME]);
return 0;
}
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
static int nl80211_set_reg(struct sk_buff *skb, struct genl_info *info)
{
struct nlattr *tb[NL80211_REG_RULE_ATTR_MAX + 1];
struct nlattr *nl_reg_rule;
cfg80211: remove unneeded initialisations in nl80211_set_reg Some variables are assigned unconditionally, remove their initialisations to help avoid introducing errors later. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-28 13:54:31 +00:00
char *alpha2;
int rem_reg_rules, r;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
u32 num_rules = 0, rule_idx = 0, size_of_regd;
cfg80211: use enum nl80211_dfs_regions for dfs_region everywhere u8 was used in some other places, just stick to the enum, this forces us to express the values that are expected. Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 17:54:02 +00:00
enum nl80211_dfs_regions dfs_region = NL80211_DFS_UNSET;
cfg80211: remove unneeded initialisations in nl80211_set_reg Some variables are assigned unconditionally, remove their initialisations to help avoid introducing errors later. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-28 13:54:31 +00:00
struct ieee80211_regdomain *rd;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
if (!info->attrs[NL80211_ATTR_REG_ALPHA2])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_REG_RULES])
return -EINVAL;
alpha2 = nla_data(info->attrs[NL80211_ATTR_REG_ALPHA2]);
cfg80211: process regulatory DFS region for countries The wireless-regdb now has support for mapping a country to one DFS region. CRDA sends this to us now so process it so we can provide that hint to drivers. This will later be used by code for processing DFS in a way that meets the criteria for the DFS region the country belongs to. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-11 17:59:02 +00:00
if (info->attrs[NL80211_ATTR_DFS_REGION])
dfs_region = nla_get_u8(info->attrs[NL80211_ATTR_DFS_REGION]);
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
nla_for_each_nested(nl_reg_rule, info->attrs[NL80211_ATTR_REG_RULES],
regulatory: code cleanup Clean up various things like indentation, extra parentheses, too many/few line breaks, etc. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-03 16:21:11 +00:00
rem_reg_rules) {
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
num_rules++;
if (num_rules > NL80211_MAX_SUPP_REG_RULES)
cfg80211: return immediately if num reg rules > NL80211_MAX_SUPP_REG_RULES This has no functional change except we save a kfree(rd) and allows us to clean this code up a bit after this. We do avoid an unnecessary kfree(NULL) but calling that was OK too. Cc: stable@kernel.org Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-13 21:04:39 +00:00
return -EINVAL;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
}
cfg80211: check regulatory request alpha2 early Currently nl80211 allows userspace to send the kernel a bogus regulatory domain with at most 32 rules set and it won't reject it until after its allocated memory. Let's be smart about it and take advantage that the last_request is now available under RTNL and check if the alpha2 matches an expected request and reject any bogus userspace requests prior to hitting the memory allocator. Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-05 17:18:01 +00:00
if (!reg_is_valid_request(alpha2))
return -EINVAL;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
size_of_regd = sizeof(struct ieee80211_regdomain) +
regulatory: code cleanup Clean up various things like indentation, extra parentheses, too many/few line breaks, etc. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-03 16:21:11 +00:00
num_rules * sizeof(struct ieee80211_reg_rule);
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
rd = kzalloc(size_of_regd, GFP_KERNEL);
regulatory: fix reg_is_valid_request handling There's a bug with the world regulatory domain, it can be updated any time which is different from all other regdomains that can only be updated once after a request for them. Fix this by adding a check for "processed" to the reg_is_valid_request() function and clear that when doing a request. While looking at this I also found another locking bug, last_request is protected by the reg_mutex not the cfg80211_mutex so the code in nl80211 is racy. Remove that code as it only tries to prevent an allocation in an error case, which isn't necessary. Then the function can also become static and locking in nl80211 can have a smaller scope. Also change __set_regdom() to do the checks earlier and not different for world/other regdomains. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-03 23:48:59 +00:00
if (!rd)
return -ENOMEM;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
rd->n_reg_rules = num_rules;
rd->alpha2[0] = alpha2[0];
rd->alpha2[1] = alpha2[1];
cfg80211: process regulatory DFS region for countries The wireless-regdb now has support for mapping a country to one DFS region. CRDA sends this to us now so process it so we can provide that hint to drivers. This will later be used by code for processing DFS in a way that meets the criteria for the DFS region the country belongs to. Signed-off-by: Luis R. Rodriguez <mcgrof@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-11 17:59:02 +00:00
/*
* Disable DFS master mode if the DFS region was
* not supported or known on this kernel.
*/
if (reg_supported_dfs_region(dfs_region))
rd->dfs_region = dfs_region;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
nla_for_each_nested(nl_reg_rule, info->attrs[NL80211_ATTR_REG_RULES],
regulatory: code cleanup Clean up various things like indentation, extra parentheses, too many/few line breaks, etc. Acked-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-12-03 16:21:11 +00:00
rem_reg_rules) {
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
r = nla_parse_nested(tb, NL80211_REG_RULE_ATTR_MAX,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl_reg_rule, reg_rule_policy,
info->extack);
nl80211: check nla_parse() return values If there's a policy, then nla_parse() return values must be checked, otherwise the policy is useless and there's nothing that ensures the attributes are actually what we expect them to be. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:17:47 +00:00
if (r)
goto bad_reg;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
r = parse_reg_rule(tb, &rd->reg_rules[rule_idx]);
if (r)
goto bad_reg;
rule_idx++;
cfg80211: cleanup return calls on nl80211_set_reg() This has no functional change, but it will make the race fix easier to spot in my next patch. Cc: stable@kernel.org Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-13 21:04:40 +00:00
if (rule_idx > NL80211_MAX_SUPP_REG_RULES) {
r = -EINVAL;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
goto bad_reg;
cfg80211: cleanup return calls on nl80211_set_reg() This has no functional change, but it will make the race fix easier to spot in my next patch. Cc: stable@kernel.org Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-13 21:04:40 +00:00
}
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
}
nl80211: clarify nl80211_set_reg() success path Setting rd to NULL to avoid freeing it, just to be able to return from the function in a single place, doesn't make much sense. Return the set_regdom() return value directly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-06-09 08:40:09 +00:00
/* set_regdom takes ownership of rd */
return set_regdom(rd, REGD_SOURCE_CRDA);
wireless: make regdom passing semantics simpler The regdom struct is given to the core, so it might as well free it in error conditions. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-24 18:32:20 +00:00
bad_reg:
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
kfree(rd);
cfg80211: cleanup return calls on nl80211_set_reg() This has no functional change, but it will make the race fix easier to spot in my next patch. Cc: stable@kernel.org Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-13 21:04:40 +00:00
return r;
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
}
cfg80211: reg: make CRDA support optional If there's a built-in regulatory database, there may be little point in also calling out to CRDA and failing if the system is configured that way. Allow removing CRDA support to save ~1K kernel size. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-15 07:25:18 +00:00
#endif /* CONFIG_CFG80211_CRDA_SUPPORT */
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
cfg80211: prohibit scanning the same channel more than once It isn't very useful to scan the same channel more than once during a given scan, and some hardware (notably iwlwifi) can only scan a limited number of channels at a time. To prevent any overflows, simply disallow scanning any channel multiple times in a given scan command. This is a small change in the userspace ABI, but the only user, wpa_supplicant, never asks for a scan with the same frequency listed twice. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-17 15:41:49 +00:00
static int validate_scan_freqs(struct nlattr *freqs)
{
struct nlattr *attr1, *attr2;
int n_channels = 0, tmp1, tmp2;
cfg80211: Validate frequencies nested in NL80211_ATTR_SCAN_FREQUENCIES validate_scan_freqs() retrieves frequencies from attributes nested in the attribute NL80211_ATTR_SCAN_FREQUENCIES with nla_get_u32(), which reads 4 bytes from each attribute without validating the size of data received. Attributes nested in NL80211_ATTR_SCAN_FREQUENCIES don't have an nla policy. Validate size of each attribute before parsing to avoid potential buffer overread. Fixes: 2a519311926 ("cfg80211/nl80211: scanning (and mac80211 update to use it)") Cc: stable@vger.kernel.org Signed-off-by: Srinivas Dasari <dasaris@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-07-06 22:43:42 +00:00
nla_for_each_nested(attr1, freqs, tmp1)
if (nla_len(attr1) != sizeof(u32))
return 0;
cfg80211: prohibit scanning the same channel more than once It isn't very useful to scan the same channel more than once during a given scan, and some hardware (notably iwlwifi) can only scan a limited number of channels at a time. To prevent any overflows, simply disallow scanning any channel multiple times in a given scan command. This is a small change in the userspace ABI, but the only user, wpa_supplicant, never asks for a scan with the same frequency listed twice. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-17 15:41:49 +00:00
nla_for_each_nested(attr1, freqs, tmp1) {
n_channels++;
/*
* Some hardware has a limited channel list for
* scanning, and it is pretty much nonsensical
* to scan for a channel twice, so disallow that
* and don't require drivers to check that the
* channel list they get isn't longer than what
* they can scan, as long as they can scan all
* the channels they registered at once.
*/
nla_for_each_nested(attr2, freqs, tmp2)
if (attr1 != attr2 &&
nla_get_u32(attr1) == nla_get_u32(attr2))
return 0;
}
return n_channels;
}
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
static bool is_band_valid(struct wiphy *wiphy, enum nl80211_band b)
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
{
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
return b < NUM_NL80211_BANDS && wiphy->bands[b];
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
}
static int parse_bss_select(struct nlattr *nla, struct wiphy *wiphy,
struct cfg80211_bss_selection *bss_select)
{
struct nlattr *attr[NL80211_BSS_SELECT_ATTR_MAX + 1];
struct nlattr *nest;
int err;
bool found = false;
int i;
/* only process one nested attribute */
nest = nla_data(nla);
if (!nla_ok(nest, nla_len(nest)))
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(attr, NL80211_BSS_SELECT_ATTR_MAX, nest,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_bss_select_policy, NULL);
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
if (err)
return err;
/* only one attribute may be given */
for (i = 0; i <= NL80211_BSS_SELECT_ATTR_MAX; i++) {
if (attr[i]) {
if (found)
return -EINVAL;
found = true;
}
}
bss_select->behaviour = __NL80211_BSS_SELECT_ATTR_INVALID;
if (attr[NL80211_BSS_SELECT_ATTR_RSSI])
bss_select->behaviour = NL80211_BSS_SELECT_ATTR_RSSI;
if (attr[NL80211_BSS_SELECT_ATTR_BAND_PREF]) {
bss_select->behaviour = NL80211_BSS_SELECT_ATTR_BAND_PREF;
bss_select->param.band_pref =
nla_get_u32(attr[NL80211_BSS_SELECT_ATTR_BAND_PREF]);
if (!is_band_valid(wiphy, bss_select->param.band_pref))
return -EINVAL;
}
if (attr[NL80211_BSS_SELECT_ATTR_RSSI_ADJUST]) {
struct nl80211_bss_select_rssi_adjust *adj_param;
adj_param = nla_data(attr[NL80211_BSS_SELECT_ATTR_RSSI_ADJUST]);
bss_select->behaviour = NL80211_BSS_SELECT_ATTR_RSSI_ADJUST;
bss_select->param.adjust.band = adj_param->band;
bss_select->param.adjust.delta = adj_param->delta;
if (!is_band_valid(wiphy, bss_select->param.adjust.band))
return -EINVAL;
}
/* user-space did not provide behaviour attribute */
if (bss_select->behaviour == __NL80211_BSS_SELECT_ATTR_INVALID)
return -EINVAL;
if (!(wiphy->bss_select_support & BIT(bss_select->behaviour)))
return -EINVAL;
return 0;
}
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
static int nl80211_parse_random_mac(struct nlattr **attrs,
u8 *mac_addr, u8 *mac_addr_mask)
{
int i;
if (!attrs[NL80211_ATTR_MAC] && !attrs[NL80211_ATTR_MAC_MASK]) {
wireless: Use eth_<foo>_addr instead of memset Use the built-in function instead of memset. Signed-off-by: Joe Perches <joe@perches.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-03 03:54:58 +00:00
eth_zero_addr(mac_addr);
eth_zero_addr(mac_addr_mask);
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
mac_addr[0] = 0x2;
mac_addr_mask[0] = 0x3;
return 0;
}
/* need both or none */
if (!attrs[NL80211_ATTR_MAC] || !attrs[NL80211_ATTR_MAC_MASK])
return -EINVAL;
memcpy(mac_addr, nla_data(attrs[NL80211_ATTR_MAC]), ETH_ALEN);
memcpy(mac_addr_mask, nla_data(attrs[NL80211_ATTR_MAC_MASK]), ETH_ALEN);
/* don't allow or configure an mcast address */
if (!is_multicast_ether_addr(mac_addr_mask) ||
is_multicast_ether_addr(mac_addr))
return -EINVAL;
/*
* allow users to pass a MAC address that has bits set outside
* of the mask, but don't bother drivers with having to deal
* with such bits
*/
for (i = 0; i < ETH_ALEN; i++)
mac_addr[i] &= mac_addr_mask[i];
return 0;
}
cfg80211: Disallow moving out of operating DFS channel in non-ETSI For non-ETSI regulatory domain, CAC result on DFS channel may not be valid once moving out of that channel (as done during remain-on-channel, scannning and off-channel tx). Running CAC on an operating DFS channel after every off-channel operation will only add complexity and disturb the current link. Better do not allow any off-channel switch from a DFS operating channel in non-ETSI domain. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-27 11:34:34 +00:00
static bool cfg80211_off_channel_oper_allowed(struct wireless_dev *wdev)
{
ASSERT_WDEV_LOCK(wdev);
if (!cfg80211_beaconing_iface_active(wdev))
return true;
if (!(wdev->chandef.chan->flags & IEEE80211_CHAN_RADAR))
return true;
return regulatory_pre_cac_allowed(wdev->wiphy);
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
static int nl80211_trigger_scan(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
struct cfg80211_scan_request *request;
struct nlattr *attr;
struct wiphy *wiphy;
cfg80211: prohibit scanning the same channel more than once It isn't very useful to scan the same channel more than once during a given scan, and some hardware (notably iwlwifi) can only scan a limited number of channels at a time. To prevent any overflows, simply disallow scanning any channel multiple times in a given scan command. This is a small change in the userspace ABI, but the only user, wpa_supplicant, never asks for a scan with the same frequency listed twice. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-17 15:41:49 +00:00
int err, tmp, n_ssids = 0, n_channels, i;
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
size_t ie_len;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
wiphy = &rdev->wiphy;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
if (wdev->iftype == NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->scan)
return -EOPNOTSUPP;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
if (rdev->scan_req || rdev->scan_msg) {
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
err = -EBUSY;
goto unlock;
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (info->attrs[NL80211_ATTR_SCAN_FREQUENCIES]) {
cfg80211: prohibit scanning the same channel more than once It isn't very useful to scan the same channel more than once during a given scan, and some hardware (notably iwlwifi) can only scan a limited number of channels at a time. To prevent any overflows, simply disallow scanning any channel multiple times in a given scan command. This is a small change in the userspace ABI, but the only user, wpa_supplicant, never asks for a scan with the same frequency listed twice. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-17 15:41:49 +00:00
n_channels = validate_scan_freqs(
info->attrs[NL80211_ATTR_SCAN_FREQUENCIES]);
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
if (!n_channels) {
err = -EINVAL;
goto unlock;
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
} else {
cfg80211: Add a function to get the number of supported channels Add a utility function to get the number of channels supported by the device, and update the places in the code that need this data. Signed-off-by: Ilan Peer <ilan.peer@intel.com> [replace another occurrence in libertas, fix kernel-doc, fix bugs] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-09 09:37:23 +00:00
n_channels = ieee80211_get_num_supported_channels(wiphy);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
}
if (info->attrs[NL80211_ATTR_SCAN_SSIDS])
nla_for_each_nested(attr, info->attrs[NL80211_ATTR_SCAN_SSIDS], tmp)
n_ssids++;
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
if (n_ssids > wiphy->max_scan_ssids) {
err = -EINVAL;
goto unlock;
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
if (info->attrs[NL80211_ATTR_IE])
ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
else
ie_len = 0;
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
if (ie_len > wiphy->max_scan_ie_len) {
err = -EINVAL;
goto unlock;
}
cfg80211: introduce scan IE limit attribute This patch introduces a new attribute for a wiphy that tells userspace how long the information elements added to a probe request frame can be at most. It also updates the at76 to advertise that it cannot support that, and, for now until I can fix that, iwlwifi too. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-31 10:12:05 +00:00
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
request = kzalloc(sizeof(*request)
nl80211: remove some stack variables in trigger_scan and start_sched_scan Some stack variables (name *ssid and *channel) are only used to define the size of the memory block that needs to be allocated for the request structure in the nl80211_trigger_scan() and nl80211_start_sched_scan() functions. This is unnecessary because the sizes of the actual elements in the structure can be used instead. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-18 08:42:03 +00:00
+ sizeof(*request->ssids) * n_ssids
+ sizeof(*request->channels) * n_channels
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
+ ie_len, GFP_KERNEL);
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
if (!request) {
err = -ENOMEM;
goto unlock;
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (n_ssids)
cfg80211: fix alignment problem in scan request The memory layout for scan requests was rather wrong, we put the scan SSIDs before the channels which could lead to the channel pointers being unaligned in memory. It turns out that using a pointer to the channel array isn't necessary anyway since we can embed a zero-length array into the struct. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-08-07 15:54:07 +00:00
request->ssids = (void *)&request->channels[n_channels];
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
request->n_ssids = n_ssids;
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
if (ie_len) {
nl80211: suppress smatch warnings smatch warns that we once checked request->ssids in two functions and then unconditionally used it later again. This is actually fine, because the code has a relationship between attrs[NL80211_ATTR_SCAN_SSIDS], n_ssids and request->ssids, but smatch isn't smart enough to realize that. Suppress the warnings by always checking just n_ssids - that way smatch won't know that request->ssids could be NULL, and since it is only NULL when n_ssids is 0 we still check everything correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:25:20 +00:00
if (n_ssids)
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
request->ie = (void *)(request->ssids + n_ssids);
else
request->ie = (void *)(request->channels + n_channels);
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: validate scan channels Currently it is possible to request a scan on only disabled channels, which could be problematic for some drivers. Reject such scans, and also ignore disabled channels that are given. This resuls in the scan begin/end event only including channels that are actually used. This makes the mac80211 check for disabled channels superfluous. At the same time, remove the no-IBSS check from mac80211 -- nothing says that we should not find any networks on channels that cannot be used for an IBSS, even when operating in IBSS mode. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-02 12:32:03 +00:00
i = 0;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (info->attrs[NL80211_ATTR_SCAN_FREQUENCIES]) {
/* user specified, bail out if channel not found */
nla_for_each_nested(attr, info->attrs[NL80211_ATTR_SCAN_FREQUENCIES], tmp) {
cfg80211: validate scan channels Currently it is possible to request a scan on only disabled channels, which could be problematic for some drivers. Reject such scans, and also ignore disabled channels that are given. This resuls in the scan begin/end event only including channels that are actually used. This makes the mac80211 check for disabled channels superfluous. At the same time, remove the no-IBSS check from mac80211 -- nothing says that we should not find any networks on channels that cannot be used for an IBSS, even when operating in IBSS mode. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-02 12:32:03 +00:00
struct ieee80211_channel *chan;
chan = ieee80211_get_channel(wiphy, nla_get_u32(attr));
if (!chan) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
err = -EINVAL;
goto out_free;
}
cfg80211: validate scan channels Currently it is possible to request a scan on only disabled channels, which could be problematic for some drivers. Reject such scans, and also ignore disabled channels that are given. This resuls in the scan begin/end event only including channels that are actually used. This makes the mac80211 check for disabled channels superfluous. At the same time, remove the no-IBSS check from mac80211 -- nothing says that we should not find any networks on channels that cannot be used for an IBSS, even when operating in IBSS mode. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-02 12:32:03 +00:00
/* ignore disabled channels */
if (chan->flags & IEEE80211_CHAN_DISABLED)
continue;
request->channels[i] = chan;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
i++;
}
} else {
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
enum nl80211_band band;
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
/* all channels */
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
for (band = 0; band < NUM_NL80211_BANDS; band++) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
int j;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (!wiphy->bands[band])
continue;
for (j = 0; j < wiphy->bands[band]->n_channels; j++) {
cfg80211: validate scan channels Currently it is possible to request a scan on only disabled channels, which could be problematic for some drivers. Reject such scans, and also ignore disabled channels that are given. This resuls in the scan begin/end event only including channels that are actually used. This makes the mac80211 check for disabled channels superfluous. At the same time, remove the no-IBSS check from mac80211 -- nothing says that we should not find any networks on channels that cannot be used for an IBSS, even when operating in IBSS mode. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-02 12:32:03 +00:00
struct ieee80211_channel *chan;
chan = &wiphy->bands[band]->channels[j];
if (chan->flags & IEEE80211_CHAN_DISABLED)
continue;
request->channels[i] = chan;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
i++;
}
}
}
cfg80211: validate scan channels Currently it is possible to request a scan on only disabled channels, which could be problematic for some drivers. Reject such scans, and also ignore disabled channels that are given. This resuls in the scan begin/end event only including channels that are actually used. This makes the mac80211 check for disabled channels superfluous. At the same time, remove the no-IBSS check from mac80211 -- nothing says that we should not find any networks on channels that cannot be used for an IBSS, even when operating in IBSS mode. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-02 12:32:03 +00:00
if (!i) {
err = -EINVAL;
goto out_free;
}
request->n_channels = i;
cfg80211: Disallow moving out of operating DFS channel in non-ETSI For non-ETSI regulatory domain, CAC result on DFS channel may not be valid once moving out of that channel (as done during remain-on-channel, scannning and off-channel tx). Running CAC on an operating DFS channel after every off-channel operation will only add complexity and disturb the current link. Better do not allow any off-channel switch from a DFS operating channel in non-ETSI domain. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-27 11:34:34 +00:00
wdev_lock(wdev);
if (!cfg80211_off_channel_oper_allowed(wdev)) {
struct ieee80211_channel *chan;
if (request->n_channels != 1) {
wdev_unlock(wdev);
err = -EBUSY;
goto out_free;
}
chan = request->channels[0];
if (chan->center_freq != wdev->chandef.chan->center_freq) {
wdev_unlock(wdev);
err = -EBUSY;
goto out_free;
}
}
wdev_unlock(wdev);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
i = 0;
nl80211: suppress smatch warnings smatch warns that we once checked request->ssids in two functions and then unconditionally used it later again. This is actually fine, because the code has a relationship between attrs[NL80211_ATTR_SCAN_SSIDS], n_ssids and request->ssids, but smatch isn't smart enough to realize that. Suppress the warnings by always checking just n_ssids - that way smatch won't know that request->ssids could be NULL, and since it is only NULL when n_ssids is 0 we still check everything correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:25:20 +00:00
if (n_ssids) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nla_for_each_nested(attr, info->attrs[NL80211_ATTR_SCAN_SSIDS], tmp) {
nl80211: fix overflow in ssid_len When one of the SSID's length passed in a scan or sched_scan request is larger than 255, there will be an overflow in the u8 that is used to store the length before checking. This causes the check to fail and we overrun the buffer when copying the SSID. Fix this by checking the nl80211 attribute length before copying it to the struct. This is a follow up for the previous commit 208c72f4fe44fe09577e7975ba0e7fa0278f3d03, which didn't fix the problem entirely. Reported-by: Ido Yariv <ido@wizery.com> Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-07 17:42:26 +00:00
if (nla_len(attr) > IEEE80211_MAX_SSID_LEN) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
err = -EINVAL;
goto out_free;
}
nl80211: fix overflow in ssid_len When one of the SSID's length passed in a scan or sched_scan request is larger than 255, there will be an overflow in the u8 that is used to store the length before checking. This causes the check to fail and we overrun the buffer when copying the SSID. Fix this by checking the nl80211 attribute length before copying it to the struct. This is a follow up for the previous commit 208c72f4fe44fe09577e7975ba0e7fa0278f3d03, which didn't fix the problem entirely. Reported-by: Ido Yariv <ido@wizery.com> Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-07 17:42:26 +00:00
request->ssids[i].ssid_len = nla_len(attr);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
memcpy(request->ssids[i].ssid, nla_data(attr), nla_len(attr));
i++;
}
}
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
if (info->attrs[NL80211_ATTR_IE]) {
request->ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
mac80211: pass all probe request IEs to driver Instead of just passing the cfg80211-requested IEs, pass the locally generated ones as well. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-01 09:58:36 +00:00
memcpy((void *)request->ie,
nla_data(info->attrs[NL80211_ATTR_IE]),
nl80211: Optional IEs into scan request This extends the NL80211_CMD_TRIGGER_SCAN command to allow applications to specify a set of information element(s) to be added into Probe Request frames with NL80211_ATTR_IE. This provides support for the MLME-SCAN.request primitive parameter VendorSpecificInfo and can be used, e.g., to implement WPS scanning. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-16 17:39:13 +00:00
request->ie_len);
}
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
for (i = 0; i < NUM_NL80211_BANDS; i++)
cfg80211: fix scan crash on single-band cards commit 58389c69150e6032504dfcd3edca6b1975c8b5bc Author: Johannes Berg <johannes.berg@intel.com> Date: Mon Jul 18 18:08:35 2011 +0200 cfg80211: allow userspace to control supported rates in scan made single-band cards crash since it would always access all wiphy->bands[]. Fix this and reject any attempts in the new helper ieee80211_get_ratemask() to do the same, rejecting rates configuration for unsupported bands. Reported-by: Pavel Roskin <proski@gnu.org> Tested-by: Pavel Roskin <proski@gnu.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-19 22:52:16 +00:00
if (wiphy->bands[i])
request->rates[i] =
(1 << wiphy->bands[i]->n_bitrates) - 1;
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
if (info->attrs[NL80211_ATTR_SCAN_SUPP_RATES]) {
nla_for_each_nested(attr,
info->attrs[NL80211_ATTR_SCAN_SUPP_RATES],
tmp) {
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
enum nl80211_band band = nla_type(attr);
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
if (band < 0 || band >= NUM_NL80211_BANDS) {
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
err = -EINVAL;
goto out_free;
}
cfg80211: ignore supported rates for nonexistant bands on scan Fixes wpa_supplicant p2p_find on 5GHz-only devices Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-20 18:40:41 +00:00
if (!wiphy->bands[band])
continue;
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
err = ieee80211_get_ratemask(wiphy->bands[band],
nla_data(attr),
nla_len(attr),
&request->rates[band]);
if (err)
goto out_free;
}
}
nl80211: support beacon report scanning Beacon report radio measurement requires reporting observed BSSs on the channels specified in the beacon request. If the measurement mode is set to passive or active, it requires actually performing a scan (passive or active, accordingly), and reporting the time that the scan was started and the time each beacon/probe was received (both in terms of TSF of the BSS of the requesting AP). If the request mode is table, this information is optional. In addition, the radio measurement request specifies the channel dwell time for the measurement. In order to use scan for beacon report when the mode is active or passive, add a parameter to scan request that specifies the channel dwell time, and add scan start time and beacon received time to scan results information. Supporting beacon report is required for Multi Band Operation (MBO). Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-05 14:10:13 +00:00
if (info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]) {
if (!wiphy_ext_feature_isset(wiphy,
NL80211_EXT_FEATURE_SET_SCAN_DWELL)) {
err = -EOPNOTSUPP;
goto out_free;
}
request->duration =
nla_get_u16(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION]);
request->duration_mandatory =
nla_get_flag(info->attrs[NL80211_ATTR_MEASUREMENT_DURATION_MANDATORY]);
}
cfg80211: add scan flag to indicate its priority Add NL80211_SCAN_FLAG_LOW_PRIORITY flag support. It tells drivers that this is a low priority scan request, so that they can take necessary action. Drivers need to advertise low priority scan capability during registration. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:32 +00:00
if (info->attrs[NL80211_ATTR_SCAN_FLAGS]) {
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
request->flags = nla_get_u32(
info->attrs[NL80211_ATTR_SCAN_FLAGS]);
cfg80211: don't allow drivers to unset NL80211_FEATURE_SCAN_FLUSH As the flag is entirely implemented in cfg80211, it should have been a global feature flag (which I believe didn't exist at the time). However, there's no reason to allow drivers to unset the flag, so don't allow it and remove the validation of NL80211_SCAN_FLAG_FLUSH. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-26 15:14:38 +00:00
if ((request->flags & NL80211_SCAN_FLAG_LOW_PRIORITY) &&
!(wiphy->features & NL80211_FEATURE_LOW_PRIORITY_SCAN)) {
cfg80211: add scan flag to indicate its priority Add NL80211_SCAN_FLAG_LOW_PRIORITY flag support. It tells drivers that this is a low priority scan request, so that they can take necessary action. Drivers need to advertise low priority scan capability during registration. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:32 +00:00
err = -EOPNOTSUPP;
goto out_free;
}
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
if (request->flags & NL80211_SCAN_FLAG_RANDOM_ADDR) {
if (!(wiphy->features &
NL80211_FEATURE_SCAN_RANDOM_MAC_ADDR)) {
err = -EOPNOTSUPP;
goto out_free;
}
if (wdev->current_bss) {
err = -EOPNOTSUPP;
goto out_free;
}
err = nl80211_parse_random_mac(info->attrs,
request->mac_addr,
request->mac_addr_mask);
if (err)
goto out_free;
}
cfg80211: add scan flag to indicate its priority Add NL80211_SCAN_FLAG_LOW_PRIORITY flag support. It tells drivers that this is a low priority scan request, so that they can take necessary action. Drivers need to advertise low priority scan capability during registration. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:32 +00:00
}
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
nl80211/cfg80211: Add support to disable CCK rate for management frame Add a new nl80211 attribute to specify whether to send the management frames in CCK rate or not. As of now the wpa_supplicant is disabling CCK rate at P2P init itself. So this patch helps to send P2P probe request/probe response/action frames being sent at non CCK rate in 2GHz without disabling 11b rates. This attribute is used with NL80211_CMD_TRIGGER_SCAN and NL80211_CMD_FRAME commands to disable CCK rate for management frame transmission. Cc: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-25 09:23:30 +00:00
request->no_cck =
nla_get_flag(info->attrs[NL80211_ATTR_TX_NO_CCK_RATE]);
nl80211: Use different attrs for BSSID and random MAC addr in scan req NL80211_ATTR_MAC was used to set both the specific BSSID to be scanned and the random MAC address to be used when privacy is enabled. When both the features are enabled, both the BSSID and the local MAC address were getting same value causing Probe Request frames to go with unintended DA. Hence, this has been fixed by using a different NL80211_ATTR_BSSID attribute to set the specific BSSID (which was the more recent addition in cfg80211) for a scan. Backwards compatibility with old userspace software is maintained to some extent by allowing NL80211_ATTR_MAC to be used to set the specific BSSID when scanning without enabling random MAC address use. Scanning with random source MAC address was introduced by commit ad2b26abc157 ("cfg80211: allow drivers to support random MAC addresses for scan") and the issue was introduced with the addition of the second user for the same attribute in commit 818965d39177 ("cfg80211: Allow a scan request for a specific BSSID"). Fixes: 818965d39177 ("cfg80211: Allow a scan request for a specific BSSID") Signed-off-by: Vamsi Krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-02 21:59:08 +00:00
/* Initial implementation used NL80211_ATTR_MAC to set the specific
* BSSID to scan for. This was problematic because that same attribute
* was already used for another purpose (local random MAC address). The
* NL80211_ATTR_BSSID attribute was added to fix this. For backwards
* compatibility with older userspace components, also use the
* NL80211_ATTR_MAC value here if it can be determined to be used for
* the specific BSSID use case instead of the random MAC address
* (NL80211_ATTR_SCAN_FLAGS is used to enable random MAC address use).
*/
if (info->attrs[NL80211_ATTR_BSSID])
memcpy(request->bssid,
nla_data(info->attrs[NL80211_ATTR_BSSID]), ETH_ALEN);
else if (!(request->flags & NL80211_SCAN_FLAG_RANDOM_ADDR) &&
info->attrs[NL80211_ATTR_MAC])
cfg80211: Allow a scan request for a specific BSSID This allows scans for a specific BSSID to be optimized by the user space application by requesting the driver to set the Probe Request frame BSSID field (Address 3) to the specified BSSID instead of the wildcard BSSID. This prevents other APs from replying which reduces airtime need and latency in getting the response from the target AP through. This is an optimization and as such, it is acceptable for some of the drivers not to support the mechanism. If not supported, the wildcard BSSID will be used and more responses may be received. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-02-26 20:12:47 +00:00
memcpy(request->bssid, nla_data(info->attrs[NL80211_ATTR_MAC]),
ETH_ALEN);
else
eth_broadcast_addr(request->bssid);
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
request->wdev = wdev;
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
request->wiphy = &rdev->wiphy;
cfg80211: add support for flushing old scan results Add an NL80211_SCAN_FLAG_FLUSH flag that causes old bss cache entries to be flushed on scan completion. This is useful for collecting guaranteed fresh scan/survey result (e.g. on resume). For normal scan, flushing only happens on successful completion of a scan; i.e. it does not happen if the scan is aborted. For scheduled scan, previous scan results are flushed everytime when we get new scan results. This feature is enabled by default. Drivers can disable it by unsetting the NL80211_FEATURE_SCAN_FLUSH flag. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [invert polarity of feature flag to account for old kernels] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:34 +00:00
request->scan_start = jiffies;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
rdev->scan_req = request;
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_scan(rdev, request);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
if (!err) {
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
nl80211_send_scan_start(rdev, wdev);
if (wdev->netdev)
dev_hold(wdev->netdev);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
} else {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
out_free:
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
rdev->scan_req = NULL;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
kfree(request);
}
nl80211: rework locking When I added scanning to cfg80211, we got a lock dependency like this: rtnl --> cfg80211_mtx nl80211, on the other hand, has the reverse lock dependency: cfg80211_mtx --> rtnl which clearly is a bad idea. This patch reworks nl80211 to take these two locks in the other order to fix the possible, and easily triggerable, deadlock. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-12 08:55:09 +00:00
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
unlock:
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
return err;
}
cfg80211: Add support for aborting an ongoing scan Implement new functionality for aborting an ongoing scan. Add NL80211_CMD_ABORT_SCAN to the nl80211 interface. After aborting the scan, driver shall provide the scan status by calling cfg80211_scan_done(). Reviewed-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Sunil Dutt <usdutt@qti.qualcomm.com> [change command to take wdev instead of netdev so that it can be used on p2p-device scans] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-30 13:44:49 +00:00
static int nl80211_abort_scan(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
if (!rdev->ops->abort_scan)
return -EOPNOTSUPP;
if (rdev->scan_msg)
return 0;
if (!rdev->scan_req)
return -ENOENT;
rdev_abort_scan(rdev, wdev);
return 0;
}
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
static int
nl80211_parse_sched_scan_plans(struct wiphy *wiphy, int n_plans,
struct cfg80211_sched_scan_request *request,
struct nlattr **attrs)
{
int tmp, err, i = 0;
struct nlattr *attr;
if (!attrs[NL80211_ATTR_SCHED_SCAN_PLANS]) {
u32 interval;
/*
* If scan plans are not specified,
nl80211: check NL80211_ATTR_SCHED_SCAN_INTERVAL only once The presence of the NL80211_ATTR_SCHED_SCAN_INTERVAL attribute was checked in nl80211_parse_sched_scan() and nl80211_parse_sched_scan_plans() which might be a bit redundant so removing one. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-11-17 09:02:40 +00:00
* %NL80211_ATTR_SCHED_SCAN_INTERVAL will be specified. In this
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
* case one scan plan will be set with the specified scan
* interval and infinite number of iterations.
*/
interval = nla_get_u32(attrs[NL80211_ATTR_SCHED_SCAN_INTERVAL]);
if (!interval)
return -EINVAL;
request->scan_plans[0].interval =
DIV_ROUND_UP(interval, MSEC_PER_SEC);
if (!request->scan_plans[0].interval)
return -EINVAL;
if (request->scan_plans[0].interval >
wiphy->max_sched_scan_plan_interval)
request->scan_plans[0].interval =
wiphy->max_sched_scan_plan_interval;
return 0;
}
nla_for_each_nested(attr, attrs[NL80211_ATTR_SCHED_SCAN_PLANS], tmp) {
struct nlattr *plan[NL80211_SCHED_SCAN_PLAN_MAX + 1];
if (WARN_ON(i >= n_plans))
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(plan, NL80211_SCHED_SCAN_PLAN_MAX,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
attr, nl80211_plan_policy, NULL);
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
if (err)
return err;
if (!plan[NL80211_SCHED_SCAN_PLAN_INTERVAL])
return -EINVAL;
request->scan_plans[i].interval =
nla_get_u32(plan[NL80211_SCHED_SCAN_PLAN_INTERVAL]);
if (!request->scan_plans[i].interval ||
request->scan_plans[i].interval >
wiphy->max_sched_scan_plan_interval)
return -EINVAL;
if (plan[NL80211_SCHED_SCAN_PLAN_ITERATIONS]) {
request->scan_plans[i].iterations =
nla_get_u32(plan[NL80211_SCHED_SCAN_PLAN_ITERATIONS]);
if (!request->scan_plans[i].iterations ||
(request->scan_plans[i].iterations >
wiphy->max_sched_scan_plan_iterations))
return -EINVAL;
} else if (i < n_plans - 1) {
/*
* All scan plans but the last one must specify
* a finite number of iterations
*/
return -EINVAL;
}
i++;
}
/*
* The last scan plan must not specify the number of
* iterations, it is supposed to run infinitely
*/
if (request->scan_plans[n_plans - 1].iterations)
return -EINVAL;
return 0;
}
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
static struct cfg80211_sched_scan_request *
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
nl80211_parse_sched_scan(struct wiphy *wiphy, struct wireless_dev *wdev,
nl80211: fix validation of scheduled scan info for wowlan netdetect For wowlan netdetect a separate limit is defined for the number of matchsets. Currently, this limit is ignored and the regular limit for scheduled scan matchsets, ie. struct wiphy::max_match_sets, is used for the net-detect case as well. Cc: Johannes Berg <johannes@sipsolutions.net> Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-27 12:27:44 +00:00
struct nlattr **attrs, int max_match_sets)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
{
struct cfg80211_sched_scan_request *request;
struct nlattr *attr;
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
int err, tmp, n_ssids = 0, n_match_sets = 0, n_channels, i, n_plans = 0;
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
enum nl80211_band band;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
size_t ie_len;
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
struct nlattr *tb[NL80211_SCHED_SCAN_MATCH_ATTR_MAX + 1];
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
s32 default_match_rssi = NL80211_SCAN_RSSI_THOLD_OFF;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (!is_valid_ie_attr(attrs[NL80211_ATTR_IE]))
return ERR_PTR(-EINVAL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCAN_FREQUENCIES]) {
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
n_channels = validate_scan_freqs(
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
attrs[NL80211_ATTR_SCAN_FREQUENCIES]);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
if (!n_channels)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(-EINVAL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
} else {
cfg80211: Add a function to get the number of supported channels Add a utility function to get the number of channels supported by the device, and update the places in the code that need this data. Signed-off-by: Ilan Peer <ilan.peer@intel.com> [replace another occurrence in libertas, fix kernel-doc, fix bugs] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-09 09:37:23 +00:00
n_channels = ieee80211_get_num_supported_channels(wiphy);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
}
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCAN_SSIDS])
nla_for_each_nested(attr, attrs[NL80211_ATTR_SCAN_SSIDS],
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
tmp)
n_ssids++;
nl80211/cfg80211: add max_sched_scan_ssids in the hw description Some chips can scan more SSIDs with a single scheduled scan command than with a single normal scan command (eg. wl12xx chips). To support this, this patch creates a separate hardware description element that describes the amount of SSIDs supported in scheduled scans. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 11:57:28 +00:00
if (n_ssids > wiphy->max_sched_scan_ssids)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(-EINVAL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
/*
* First, count the number of 'real' matchsets. Due to an issue with
* the old implementation, matchsets containing only the RSSI attribute
* (NL80211_SCHED_SCAN_MATCH_ATTR_RSSI) are considered as the 'default'
* RSSI for all matchsets, rather than their own matchset for reporting
* all APs with a strong RSSI. This is needed to be compatible with
* older userspace that treated a matchset with only the RSSI as the
* global RSSI for all other matchsets - if there are other matchsets.
*/
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCHED_SCAN_MATCH]) {
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
nla_for_each_nested(attr,
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
attrs[NL80211_ATTR_SCHED_SCAN_MATCH],
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
tmp) {
struct nlattr *rssi;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb,
NL80211_SCHED_SCAN_MATCH_ATTR_MAX,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
attr, nl80211_match_policy,
NULL);
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
if (err)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(err);
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
/* SSID and BSSID are mutually exclusive */
if (tb[NL80211_SCHED_SCAN_MATCH_ATTR_SSID] &&
tb[NL80211_SCHED_SCAN_MATCH_ATTR_BSSID])
return ERR_PTR(-EINVAL);
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
/* add other standalone attributes here */
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
if (tb[NL80211_SCHED_SCAN_MATCH_ATTR_SSID] ||
tb[NL80211_SCHED_SCAN_MATCH_ATTR_BSSID]) {
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
n_match_sets++;
continue;
}
rssi = tb[NL80211_SCHED_SCAN_MATCH_ATTR_RSSI];
if (rssi)
default_match_rssi = nla_get_s32(rssi);
}
}
/* However, if there's no other matchset, add the RSSI one */
if (!n_match_sets && default_match_rssi != NL80211_SCAN_RSSI_THOLD_OFF)
n_match_sets = 1;
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
nl80211: fix validation of scheduled scan info for wowlan netdetect For wowlan netdetect a separate limit is defined for the number of matchsets. Currently, this limit is ignored and the regular limit for scheduled scan matchsets, ie. struct wiphy::max_match_sets, is used for the net-detect case as well. Cc: Johannes Berg <johannes@sipsolutions.net> Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-27 12:27:44 +00:00
if (n_match_sets > max_match_sets)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(-EINVAL);
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_IE])
ie_len = nla_len(attrs[NL80211_ATTR_IE]);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
else
ie_len = 0;
nl80211/cfg80211: add max_sched_scan_ie_len in the hw description Some chips may support different lengths of user-supplied IEs with a single scheduled scan command than with a single normal scan command. To support this, this patch creates a separate hardware description element that describes the maximum size of user-supplied information element data supported in scheduled scans. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 11:57:29 +00:00
if (ie_len > wiphy->max_sched_scan_ie_len)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(-EINVAL);
cfg80211: fix deadlock with rfkill/sched_scan by adding new mutex There was a deadlock when rfkill-blocking a wireless interface, because we were locking the rdev mutex on NETDEV_GOING_DOWN to stop sched_scans that were eventually running. The rfkill block code was already holding a mutex under rdev: kernel: ======================================================= kernel: [ INFO: possible circular locking dependency detected ] kernel: 3.0.0-rc1-00049-g1fa7b6a #57 kernel: ------------------------------------------------------- kernel: kworker/0:1/4525 is trying to acquire lock: kernel: (&rdev->mtx){+.+.+.}, at: [<ffffffff8164c831>] cfg80211_netdev_notifier_call+0x131/0x5b0 kernel: kernel: but task is already holding lock: kernel: (&rdev->devlist_mtx){+.+.+.}, at: [<ffffffff8164dcef>] cfg80211_rfkill_set_block+0x4f/0xa0 kernel: kernel: which lock already depends on the new lock. To fix this, add a new mutex specifically for sched_scan, to protect the sched_scan_req element in the rdev struct, instead of using the global rdev mutex. Reported-by: Duane Griffin <duaneg@dghda.com> Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-30 05:32:41 +00:00
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
if (attrs[NL80211_ATTR_SCHED_SCAN_PLANS]) {
/*
* NL80211_ATTR_SCHED_SCAN_INTERVAL must not be specified since
* each scan plan already specifies its own interval
*/
if (attrs[NL80211_ATTR_SCHED_SCAN_INTERVAL])
return ERR_PTR(-EINVAL);
nla_for_each_nested(attr,
attrs[NL80211_ATTR_SCHED_SCAN_PLANS], tmp)
n_plans++;
} else {
/*
* The scan interval attribute is kept for backward
* compatibility. If no scan plans are specified and sched scan
* interval is specified, one scan plan will be set with this
* scan interval and infinite number of iterations.
*/
if (!attrs[NL80211_ATTR_SCHED_SCAN_INTERVAL])
return ERR_PTR(-EINVAL);
n_plans = 1;
}
if (!n_plans || n_plans > wiphy->max_sched_scan_plans)
return ERR_PTR(-EINVAL);
cfg80211: Add support to sched scan to report better BSSs Enhance sched scan to support option of finding a better BSS while in connected state. Firmware scans the medium and reports when it finds a known BSS which has better RSSI than the current connected BSS. New attributes to specify the relative RSSI (compared to the current BSS) are added to the sched scan to implement this. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:20 +00:00
if (!wiphy_ext_feature_isset(
wiphy, NL80211_EXT_FEATURE_SCHED_SCAN_RELATIVE_RSSI) &&
(attrs[NL80211_ATTR_SCHED_SCAN_RELATIVE_RSSI] ||
attrs[NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST]))
return ERR_PTR(-EINVAL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
request = kzalloc(sizeof(*request)
nl80211: remove some stack variables in trigger_scan and start_sched_scan Some stack variables (name *ssid and *channel) are only used to define the size of the memory block that needs to be allocated for the request structure in the nl80211_trigger_scan() and nl80211_start_sched_scan() functions. This is unnecessary because the sizes of the actual elements in the structure can be used instead. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-18 08:42:03 +00:00
+ sizeof(*request->ssids) * n_ssids
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
+ sizeof(*request->match_sets) * n_match_sets
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
+ sizeof(*request->scan_plans) * n_plans
nl80211: remove some stack variables in trigger_scan and start_sched_scan Some stack variables (name *ssid and *channel) are only used to define the size of the memory block that needs to be allocated for the request structure in the nl80211_trigger_scan() and nl80211_start_sched_scan() functions. This is unnecessary because the sizes of the actual elements in the structure can be used instead. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-18 08:42:03 +00:00
+ sizeof(*request->channels) * n_channels
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
+ ie_len, GFP_KERNEL);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (!request)
return ERR_PTR(-ENOMEM);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
if (n_ssids)
request->ssids = (void *)&request->channels[n_channels];
request->n_ssids = n_ssids;
if (ie_len) {
nl80211: suppress smatch warnings smatch warns that we once checked request->ssids in two functions and then unconditionally used it later again. This is actually fine, because the code has a relationship between attrs[NL80211_ATTR_SCAN_SSIDS], n_ssids and request->ssids, but smatch isn't smart enough to realize that. Suppress the warnings by always checking just n_ssids - that way smatch won't know that request->ssids could be NULL, and since it is only NULL when n_ssids is 0 we still check everything correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:25:20 +00:00
if (n_ssids)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
request->ie = (void *)(request->ssids + n_ssids);
else
request->ie = (void *)(request->channels + n_channels);
}
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
if (n_match_sets) {
if (request->ie)
request->match_sets = (void *)(request->ie + ie_len);
nl80211: suppress smatch warnings smatch warns that we once checked request->ssids in two functions and then unconditionally used it later again. This is actually fine, because the code has a relationship between attrs[NL80211_ATTR_SCAN_SSIDS], n_ssids and request->ssids, but smatch isn't smart enough to realize that. Suppress the warnings by always checking just n_ssids - that way smatch won't know that request->ssids could be NULL, and since it is only NULL when n_ssids is 0 we still check everything correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:25:20 +00:00
else if (n_ssids)
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
request->match_sets =
(void *)(request->ssids + n_ssids);
else
request->match_sets =
(void *)(request->channels + n_channels);
}
request->n_match_sets = n_match_sets;
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
if (n_match_sets)
request->scan_plans = (void *)(request->match_sets +
n_match_sets);
else if (request->ie)
request->scan_plans = (void *)(request->ie + ie_len);
else if (n_ssids)
request->scan_plans = (void *)(request->ssids + n_ssids);
else
request->scan_plans = (void *)(request->channels + n_channels);
request->n_scan_plans = n_plans;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
i = 0;
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCAN_FREQUENCIES]) {
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
/* user specified, bail out if channel not found */
nla_for_each_nested(attr,
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
attrs[NL80211_ATTR_SCAN_FREQUENCIES],
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
tmp) {
struct ieee80211_channel *chan;
chan = ieee80211_get_channel(wiphy, nla_get_u32(attr));
if (!chan) {
err = -EINVAL;
goto out_free;
}
/* ignore disabled channels */
if (chan->flags & IEEE80211_CHAN_DISABLED)
continue;
request->channels[i] = chan;
i++;
}
} else {
/* all channels */
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
for (band = 0; band < NUM_NL80211_BANDS; band++) {
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
int j;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
if (!wiphy->bands[band])
continue;
for (j = 0; j < wiphy->bands[band]->n_channels; j++) {
struct ieee80211_channel *chan;
chan = &wiphy->bands[band]->channels[j];
if (chan->flags & IEEE80211_CHAN_DISABLED)
continue;
request->channels[i] = chan;
i++;
}
}
}
if (!i) {
err = -EINVAL;
goto out_free;
}
request->n_channels = i;
i = 0;
nl80211: suppress smatch warnings smatch warns that we once checked request->ssids in two functions and then unconditionally used it later again. This is actually fine, because the code has a relationship between attrs[NL80211_ATTR_SCAN_SSIDS], n_ssids and request->ssids, but smatch isn't smart enough to realize that. Suppress the warnings by always checking just n_ssids - that way smatch won't know that request->ssids could be NULL, and since it is only NULL when n_ssids is 0 we still check everything correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-23 10:25:20 +00:00
if (n_ssids) {
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
nla_for_each_nested(attr, attrs[NL80211_ATTR_SCAN_SSIDS],
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
tmp) {
nl80211: fix overflow in ssid_len When one of the SSID's length passed in a scan or sched_scan request is larger than 255, there will be an overflow in the u8 that is used to store the length before checking. This causes the check to fail and we overrun the buffer when copying the SSID. Fix this by checking the nl80211 attribute length before copying it to the struct. This is a follow up for the previous commit 208c72f4fe44fe09577e7975ba0e7fa0278f3d03, which didn't fix the problem entirely. Reported-by: Ido Yariv <ido@wizery.com> Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-07 17:42:26 +00:00
if (nla_len(attr) > IEEE80211_MAX_SSID_LEN) {
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
err = -EINVAL;
goto out_free;
}
nl80211: fix overflow in ssid_len When one of the SSID's length passed in a scan or sched_scan request is larger than 255, there will be an overflow in the u8 that is used to store the length before checking. This causes the check to fail and we overrun the buffer when copying the SSID. Fix this by checking the nl80211 attribute length before copying it to the struct. This is a follow up for the previous commit 208c72f4fe44fe09577e7975ba0e7fa0278f3d03, which didn't fix the problem entirely. Reported-by: Ido Yariv <ido@wizery.com> Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-07 17:42:26 +00:00
request->ssids[i].ssid_len = nla_len(attr);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
memcpy(request->ssids[i].ssid, nla_data(attr),
nla_len(attr));
i++;
}
}
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
i = 0;
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCHED_SCAN_MATCH]) {
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
nla_for_each_nested(attr,
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
attrs[NL80211_ATTR_SCHED_SCAN_MATCH],
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
tmp) {
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
struct nlattr *ssid, *bssid, *rssi;
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb,
NL80211_SCHED_SCAN_MATCH_ATTR_MAX,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
attr, nl80211_match_policy,
NULL);
nl80211: check nla_parse() return values If there's a policy, then nla_parse() return values must be checked, otherwise the policy is useless and there's nothing that ensures the attributes are actually what we expect them to be. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:17:47 +00:00
if (err)
goto out_free;
nl80211: fix sched scan match attribute name It should be NL80211_SCHED_SCAN_MATCH_ATTR_SSID as documented, not NL80211_ATTR_SCHED_SCAN_MATCH_SSID. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-13 09:17:11 +00:00
ssid = tb[NL80211_SCHED_SCAN_MATCH_ATTR_SSID];
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
bssid = tb[NL80211_SCHED_SCAN_MATCH_ATTR_BSSID];
if (ssid || bssid) {
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
if (WARN_ON(i >= n_match_sets)) {
/* this indicates a programming error,
* the loop above should have verified
* things properly
*/
err = -EINVAL;
goto out_free;
}
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
if (ssid) {
if (nla_len(ssid) > IEEE80211_MAX_SSID_LEN) {
err = -EINVAL;
goto out_free;
}
memcpy(request->match_sets[i].ssid.ssid,
nla_data(ssid), nla_len(ssid));
request->match_sets[i].ssid.ssid_len =
nla_len(ssid);
}
if (bssid) {
if (nla_len(bssid) != ETH_ALEN) {
err = -EINVAL;
goto out_free;
}
memcpy(request->match_sets[i].bssid,
nla_data(bssid), ETH_ALEN);
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
}
nl80211: add support for BSSIDs in scheduled scan matchsets This patch allows for the scheduled scan request to specify matchsets for specific BSSIDs. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [docs, netlink policy fix] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:01 +00:00
nl80211: Fix spelling Fix 'implementation' spelling, reported by checkpatch.pl Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:54:10 +00:00
/* special attribute - old implementation w/a */
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
request->match_sets[i].rssi_thold =
default_match_rssi;
rssi = tb[NL80211_SCHED_SCAN_MATCH_ATTR_RSSI];
if (rssi)
request->match_sets[i].rssi_thold =
nla_get_s32(rssi);
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
}
i++;
}
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
/* there was no other matchset, so the RSSI one is alone */
nl80211: check matches array length before acessing it If the userspace passes a malformed sched scan request (or a net detect wowlan configuration) by adding a NL80211_ATTR_SCHED_SCAN_MATCH attribute without any nested matchsets, a NULL pointer dereference will occur. Fix this by checking that we do have matchsets in our array before trying to access it. BUG: unable to handle kernel NULL pointer dereference at 0000000000000024 IP: [<ffffffffa002fd69>] nl80211_parse_sched_scan.part.67+0x6e9/0x900 [cfg80211] PGD 865c067 PUD 865b067 PMD 0 Oops: 0002 [#1] SMP Modules linked in: iwlmvm(O) iwlwifi(O) mac80211(O) cfg80211(O) compat(O) [last unloaded: compat] CPU: 2 PID: 2442 Comm: iw Tainted: G O 3.17.2 #31 Hardware name: Bochs Bochs, BIOS Bochs 01/01/2011 task: ffff880013800790 ti: ffff880008d80000 task.ti: ffff880008d80000 RIP: 0010:[<ffffffffa002fd69>] [<ffffffffa002fd69>] nl80211_parse_sched_scan.part.67+0x6e9/0x900 [cfg80211] RSP: 0018:ffff880008d838d0 EFLAGS: 00010293 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000 RDX: 000000000000143c RSI: 0000000000000000 RDI: ffff880008ee8dd0 RBP: ffff880008d83948 R08: 0000000000000002 R09: 0000000000000019 R10: ffff88001d1b3c40 R11: 0000000000000002 R12: ffff880019e85e00 R13: 00000000fffffed4 R14: ffff880009757800 R15: 0000000000001388 FS: 00007fa3b6d13700(0000) GS:ffff88003e200000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000024 CR3: 0000000008670000 CR4: 00000000000006e0 Stack: ffff880009757800 ffff880000000001 0000000000000000 ffff880008ee84e0 0000000000000000 ffff880009757800 00000000fffffed4 ffff880008d83948 ffffffff814689c9 ffff880009757800 ffff880008ee8000 0000000000000000 Call Trace: [<ffffffff814689c9>] ? nla_parse+0xb9/0x120 [<ffffffffa00306de>] nl80211_set_wowlan+0x75e/0x960 [cfg80211] [<ffffffff810bf3d5>] ? mark_held_locks+0x75/0xa0 [<ffffffff8161a77b>] genl_family_rcv_msg+0x18b/0x360 [<ffffffff810bf66d>] ? trace_hardirqs_on+0xd/0x10 [<ffffffff8161a9d4>] genl_rcv_msg+0x84/0xc0 [<ffffffff8161a950>] ? genl_family_rcv_msg+0x360/0x360 [<ffffffff81618e79>] netlink_rcv_skb+0xa9/0xd0 [<ffffffff81619458>] genl_rcv+0x28/0x40 [<ffffffff816184a5>] netlink_unicast+0x105/0x180 [<ffffffff8161886f>] netlink_sendmsg+0x34f/0x7a0 [<ffffffff8105a097>] ? kvm_clock_read+0x27/0x40 [<ffffffff815c644d>] sock_sendmsg+0x8d/0xc0 [<ffffffff811a75c9>] ? might_fault+0xb9/0xc0 [<ffffffff811a756e>] ? might_fault+0x5e/0xc0 [<ffffffff815d5d26>] ? verify_iovec+0x56/0xe0 [<ffffffff815c73e0>] ___sys_sendmsg+0x3d0/0x3e0 [<ffffffff810a7be8>] ? sched_clock_cpu+0x98/0xd0 [<ffffffff810611b4>] ? __do_page_fault+0x254/0x580 [<ffffffff810bb39f>] ? up_read+0x1f/0x40 [<ffffffff810611b4>] ? __do_page_fault+0x254/0x580 [<ffffffff812146ed>] ? __fget_light+0x13d/0x160 [<ffffffff815c7b02>] __sys_sendmsg+0x42/0x80 [<ffffffff815c7b52>] SyS_sendmsg+0x12/0x20 [<ffffffff81751f69>] system_call_fastpath+0x16/0x1b Fixes: ea73cbce4e1f ("nl80211: fix scheduled scan RSSI matchset attribute confusion") Cc: stable@vger.kernel.org [3.15+] Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-01 09:32:09 +00:00
if (i == 0 && n_match_sets)
nl80211: fix scheduled scan RSSI matchset attribute confusion The scheduled scan matchsets were intended to be a list of filters, with the found BSS having to pass at least one of them to be passed to the host. When the RSSI attribute was added, however, this was broken and currently wpa_supplicant adds that attribute in its own matchset; however, it doesn't intend that to mean that anything that passes the RSSI filter should be passed to the host, instead it wants it to mean that everything needs to also have higher RSSI. This is semantically problematic because we have a list of filters like [ SSID1, SSID2, SSID3, RSSI ] with no real indication which one should be OR'ed and which one AND'ed. To fix this, move the RSSI filter attribute into each matchset. As we need to stay backward compatible, treat a matchset with only the RSSI attribute as a "default RSSI filter" for all other matchsets, but only if there are other matchsets (an RSSI-only matchset by itself is still desirable.) To make driver implementation easier, keep a global min_rssi_thold for the entire request as well. The only affected driver is ath6kl. I found this when I looked into the code after Raja Mani submitted a patch fixing the n_match_sets calculation to disregard the RSSI, but that patch didn't address the semantic issue. Reported-by: Raja Mani <rmani@qti.qualcomm.com> Acked-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 09:53:53 +00:00
request->match_sets[0].rssi_thold = default_match_rssi;
request->min_rssi_thold = INT_MAX;
for (i = 0; i < n_match_sets; i++)
request->min_rssi_thold =
min(request->match_sets[i].rssi_thold,
request->min_rssi_thold);
} else {
request->min_rssi_thold = NL80211_SCAN_RSSI_THOLD_OFF;
nl80211/cfg80211: add match filtering for sched_scan Introduce filtering for scheduled scans to reduce the number of unnecessary results (which cause useless wake-ups). Add a new nested attribute where sets of parameters to be matched can be passed when starting a scheduled scan. Only scan results that match any of the sets will be returned. At this point, the set consists of a single parameter, an SSID. This can be easily extended in the future to support more complex matches. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-08-31 13:01:48 +00:00
}
nl80211: use ie_len in scheduled scan We've already checked the IE length and assigned request->ie based on that, so continue using it to make the code a bit clearer. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-04 20:01:25 +00:00
if (ie_len) {
request->ie_len = ie_len;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
memcpy((void *)request->ie,
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
nla_data(attrs[NL80211_ATTR_IE]),
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
request->ie_len);
}
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (attrs[NL80211_ATTR_SCAN_FLAGS]) {
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
request->flags = nla_get_u32(
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
attrs[NL80211_ATTR_SCAN_FLAGS]);
cfg80211: don't allow drivers to unset NL80211_FEATURE_SCAN_FLUSH As the flag is entirely implemented in cfg80211, it should have been a global feature flag (which I believe didn't exist at the time). However, there's no reason to allow drivers to unset the flag, so don't allow it and remove the validation of NL80211_SCAN_FLAG_FLUSH. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-26 15:14:38 +00:00
if ((request->flags & NL80211_SCAN_FLAG_LOW_PRIORITY) &&
!(wiphy->features & NL80211_FEATURE_LOW_PRIORITY_SCAN)) {
cfg80211: add scan flag to indicate its priority Add NL80211_SCAN_FLAG_LOW_PRIORITY flag support. It tells drivers that this is a low priority scan request, so that they can take necessary action. Drivers need to advertise low priority scan capability during registration. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:32 +00:00
err = -EOPNOTSUPP;
goto out_free;
}
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
if (request->flags & NL80211_SCAN_FLAG_RANDOM_ADDR) {
u32 flg = NL80211_FEATURE_SCHED_SCAN_RANDOM_MAC_ADDR;
if (!wdev) /* must be net-detect */
flg = NL80211_FEATURE_ND_RANDOM_MAC_ADDR;
if (!(wiphy->features & flg)) {
err = -EOPNOTSUPP;
goto out_free;
}
if (wdev && wdev->current_bss) {
err = -EOPNOTSUPP;
goto out_free;
}
err = nl80211_parse_random_mac(attrs, request->mac_addr,
request->mac_addr_mask);
if (err)
goto out_free;
}
cfg80211: add scan flag to indicate its priority Add NL80211_SCAN_FLAG_LOW_PRIORITY flag support. It tells drivers that this is a low priority scan request, so that they can take necessary action. Drivers need to advertise low priority scan capability during registration. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:32 +00:00
}
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
nl80211: add an attribute to allow delaying the first scheduled scan cycle The userspace may want to delay the the first scheduled scan or net-detect cycle. Add an optional attribute to the scheduled scan configuration to pass the delay to be (optionally) used by the driver. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> [add the attribute to the policy to validate it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-16 14:04:09 +00:00
if (attrs[NL80211_ATTR_SCHED_SCAN_DELAY])
request->delay =
nla_get_u32(attrs[NL80211_ATTR_SCHED_SCAN_DELAY]);
cfg80211: Add support to sched scan to report better BSSs Enhance sched scan to support option of finding a better BSS while in connected state. Firmware scans the medium and reports when it finds a known BSS which has better RSSI than the current connected BSS. New attributes to specify the relative RSSI (compared to the current BSS) are added to the sched scan to implement this. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:20 +00:00
if (attrs[NL80211_ATTR_SCHED_SCAN_RELATIVE_RSSI]) {
request->relative_rssi = nla_get_s8(
attrs[NL80211_ATTR_SCHED_SCAN_RELATIVE_RSSI]);
request->relative_rssi_set = true;
}
if (request->relative_rssi_set &&
attrs[NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST]) {
struct nl80211_bss_select_rssi_adjust *rssi_adjust;
rssi_adjust = nla_data(
attrs[NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST]);
request->rssi_adjust.band = rssi_adjust->band;
request->rssi_adjust.delta = rssi_adjust->delta;
if (!is_band_valid(wiphy, request->rssi_adjust.band)) {
err = -EINVAL;
goto out_free;
}
}
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
err = nl80211_parse_sched_scan_plans(wiphy, n_plans, request, attrs);
if (err)
goto out_free;
cfg80211: add support for flushing old scan results Add an NL80211_SCAN_FLAG_FLUSH flag that causes old bss cache entries to be flushed on scan completion. This is useful for collecting guaranteed fresh scan/survey result (e.g. on resume). For normal scan, flushing only happens on successful completion of a scan; i.e. it does not happen if the scan is aborted. For scheduled scan, previous scan results are flushed everytime when we get new scan results. This feature is enabled by default. Drivers can disable it by unsetting the NL80211_FEATURE_SCAN_FLUSH flag. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [invert polarity of feature flag to account for old kernels] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:34 +00:00
request->scan_start = jiffies;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return request;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
out_free:
kfree(request);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return ERR_PTR(err);
}
static int nl80211_start_sched_scan(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: allow drivers to support random MAC addresses for scan Add the necessary feature flags and a scan flag to support using random MAC addresses for scan while unassociated. The configuration for this supports an arbitrary MAC address value and mask, so that any kind of configuration (e.g. fixed OUI or full 46-bit random) can be requested. Full 46-bit random is the default when no other configuration is passed. Also add a small helper function to use the addr/mask correctly. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-12 19:39:05 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
struct cfg80211_sched_scan_request *sched_scan_req;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
bool want_multi;
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
int err;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
if (!rdev->wiphy.max_sched_scan_reqs || !rdev->ops->sched_scan_start)
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return -EOPNOTSUPP;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
want_multi = info->attrs[NL80211_ATTR_SCHED_SCAN_MULTI];
err = cfg80211_sched_scan_req_possible(rdev, want_multi);
if (err)
return err;
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
sched_scan_req = nl80211_parse_sched_scan(&rdev->wiphy, wdev,
nl80211: fix validation of scheduled scan info for wowlan netdetect For wowlan netdetect a separate limit is defined for the number of matchsets. Currently, this limit is ignored and the regular limit for scheduled scan matchsets, ie. struct wiphy::max_match_sets, is used for the net-detect case as well. Cc: Johannes Berg <johannes@sipsolutions.net> Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-27 12:27:44 +00:00
info->attrs,
rdev->wiphy.max_match_sets);
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
err = PTR_ERR_OR_ZERO(sched_scan_req);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (err)
goto out_err;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
/* leave request id zero for legacy request
* or if driver does not support multi-scheduled scan
*/
if (want_multi && rdev->wiphy.max_sched_scan_reqs > 1) {
while (!sched_scan_req->reqid)
sched_scan_req->reqid = rdev->wiphy.cookie_counter++;
}
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
err = rdev_sched_scan_start(rdev, dev, sched_scan_req);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
if (err)
goto out_free;
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
sched_scan_req->dev = dev;
sched_scan_req->wiphy = &rdev->wiphy;
nl80211: Stop scheduled scan if netlink client disappears An attribute NL80211_ATTR_SOCKET_OWNER can be set by the scan initiator. If present, the attribute will cause the scan to be stopped if the client dies. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:39 +00:00
if (info->attrs[NL80211_ATTR_SOCKET_OWNER])
sched_scan_req->owner_nlportid = info->snd_portid;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
cfg80211_add_sched_scan_req(rdev, sched_scan_req);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
nl80211_send_sched_scan(sched_scan_req, NL80211_CMD_START_SCHED_SCAN);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
return 0;
out_free:
nl80211: Convert sched_scan_req pointer to RCU pointer Because of possible races when accessing sched_scan_req pointer in rdev, the sched_scan_req is converted to RCU pointer. Signed-off-by: Jukka Rissanen <jukka.rissanen@linux.intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 11:25:38 +00:00
kfree(sched_scan_req);
cfg80211: refactor nl80211_start_sched_scan so it can be reused For net detect, we will need to reuse most of the scheduled scan parsing function, but not all, so split out the attributes parsing part out of the main start sched_scan function. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-10 14:13:46 +00:00
out_err:
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
return err;
}
static int nl80211_stop_sched_scan(struct sk_buff *skb,
struct genl_info *info)
{
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
struct cfg80211_sched_scan_request *req;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
u64 cookie;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
if (!rdev->wiphy.max_sched_scan_reqs || !rdev->ops->sched_scan_stop)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
return -EOPNOTSUPP;
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
if (info->attrs[NL80211_ATTR_COOKIE]) {
cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]);
return __cfg80211_stop_sched_scan(rdev, cookie, false);
}
req = list_first_or_null_rcu(&rdev->sched_scan_req_list,
struct cfg80211_sched_scan_request,
list);
if (!req || req->reqid ||
(req->owner_nlportid &&
req->owner_nlportid != info->snd_portid))
return -ENOENT;
return cfg80211_stop_sched_scan_req(rdev, req, false);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
}
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
static int nl80211_start_radar_detection(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_chan_def chandef;
cfg80211: DFS check dfs_region before usage Check the DFS region before channel availability check. Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-25 19:56:10 +00:00
enum nl80211_dfs_regions dfs_region;
cfg80211: DFS get CAC time from regulatory database Send Channel Availability Check time as a parameter of start_radar_detection() callback. Get CAC time from regulatory database. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-21 18:46:13 +00:00
unsigned int cac_time_ms;
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
int err;
cfg80211: DFS check dfs_region before usage Check the DFS region before channel availability check. Signed-off-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-25 19:56:10 +00:00
dfs_region = reg_get_dfs_region(wdev->wiphy);
if (dfs_region == NL80211_DFS_UNSET)
return -EINVAL;
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
err = nl80211_parse_chandef(rdev, info, &chandef);
if (err)
return err;
nl80211: allow CAC only if no operation is going on A CAC should fail if it is triggered while the interface is already running. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-09-03 17:43:18 +00:00
if (netif_carrier_ok(dev))
return -EBUSY;
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
if (wdev->cac_started)
return -EBUSY;
cfg80211/mac80211: refactor cfg80211_chandef_dfs_required() Some interface types don't require DFS (such as STATION, P2P_CLIENT etc). In order to centralize these decisions, make cfg80211_chandef_dfs_required() take the iftype into consideration. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-18 09:40:36 +00:00
err = cfg80211_chandef_dfs_required(wdev->wiphy, &chandef,
cfg80211: pass the actual iftype when calling cfg80211_chandef_dfs_required() There is no need to pass NL80211_IFTYPE_UNSPECIFIED when calling cfg80211_chandef_dfs_required() since we always already have the interface type. So, pass the actual interface type instead. Additionally, have cfg80211_chandef_dfs_required() WARN if the passed interface type is NL80211_IFTYPE_UNSPECIFIED, so we can detect problems more easily. Tested-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Reported-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-15 10:05:39 +00:00
wdev->iftype);
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
if (err < 0)
return err;
if (err == 0)
return -EINVAL;
cfg80211: DFS check chandef usable before CAC Check chandef we get in CAC request is usable for CAC. All channels have to be DFS channels. Allow DFS_USABLE and DFS_AVAILABLE channels mix. At least one channel has to be DFS_USABLE (require CAC). Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Reviewed-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-05 13:48:48 +00:00
if (!cfg80211_chandef_dfs_usable(wdev->wiphy, &chandef))
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
return -EINVAL;
if (!rdev->ops->start_radar_detection)
return -EOPNOTSUPP;
cfg80211: DFS get CAC time from regulatory database Send Channel Availability Check time as a parameter of start_radar_detection() callback. Get CAC time from regulatory database. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-21 18:46:13 +00:00
cac_time_ms = cfg80211_chandef_dfs_cac_time(&rdev->wiphy, &chandef);
if (WARN_ON(!cac_time_ms))
cac_time_ms = IEEE80211_DFS_MIN_CAC_TIME_MS;
cfg80211: Add missing tracing to cfg80211 Add missing tracing for: 1. start_radar_detection() 2. set_mcast_rates() 3. set_coalesce() Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-22 19:27:46 +00:00
err = rdev_start_radar_detection(rdev, dev, &chandef, cac_time_ms);
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
if (!err) {
cfg80211: consider existing DFS interfaces It was possible to break interface combinations in the following way: combo 1: iftype = AP, num_ifaces = 2, num_chans = 2, combo 2: iftype = AP, num_ifaces = 1, num_chans = 1, radar = HT20 With the above interface combinations it was possible to: step 1. start AP on DFS channel by matching combo 2 step 2. start AP on non-DFS channel by matching combo 1 This was possible beacuse (step 2) did not consider if other interfaces require radar detection. The patch changes how cfg80211 tracks channels - instead of channel itself now a complete chandef is stored. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-29 13:22:27 +00:00
wdev->chandef = chandef;
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
wdev->cac_started = true;
wdev->cac_start_time = jiffies;
cfg80211: DFS get CAC time from regulatory database Send Channel Availability Check time as a parameter of start_radar_detection() callback. Get CAC time from regulatory database. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-21 18:46:13 +00:00
wdev->cac_time_ms = cac_time_ms;
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
}
return err;
}
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
static int nl80211_channel_switch(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_csa_settings params;
/* csa_attrs is defined static to avoid waste of stack size - this
* function is called under RTNL lock, so this should not be a problem.
*/
static struct nlattr *csa_attrs[NL80211_ATTR_MAX+1];
int err;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
bool need_new_beacon = false;
wireless: Require HANDLE_DFS flag to switch channel for non-AP mode In the case the channel should be switched to one requiring DFS we need to make sure that userspace will handle radar events when they happen. For AP mode this is assumed to be the case, as a manager like hostapd is required. However IBSS and MESH modes can work without further userspace assistance, so refuse to use DFS channels unless userspace vouches that it handles DFS. NOTE: Userspace should have already flagged support earlier during mesh or IBSS setup. However, this information is not readily accessible currently. Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net> [sw: style cleanups] Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-05-16 09:23:12 +00:00
bool need_handle_dfs_flag = true;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
int len, i;
nl80211: sanity check the channel switch counter value The nl80211 channel switch count attribute (NL80211_ATTR_CH_SWITCH_COUNT) is specified as u32, but the specification uses u8 for the counter. To make sure strange things don't happen without informing the user, sanity check the value and return -EINVAL if it doesn't fit in u8. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-08 06:48:34 +00:00
u32 cs_count;
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
if (!rdev->ops->channel_switch ||
!(rdev->wiphy.flags & WIPHY_FLAG_HAS_CHANNEL_SWITCH))
return -EOPNOTSUPP;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
switch (dev->ieee80211_ptr->iftype) {
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_P2P_GO:
need_new_beacon = true;
wireless: Require HANDLE_DFS flag to switch channel for non-AP mode In the case the channel should be switched to one requiring DFS we need to make sure that userspace will handle radar events when they happen. For AP mode this is assumed to be the case, as a manager like hostapd is required. However IBSS and MESH modes can work without further userspace assistance, so refuse to use DFS channels unless userspace vouches that it handles DFS. NOTE: Userspace should have already flagged support earlier during mesh or IBSS setup. However, this information is not readily accessible currently. Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net> [sw: style cleanups] Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-05-16 09:23:12 +00:00
/* For all modes except AP the handle_dfs flag needs to be
* supplied to tell the kernel that userspace will handle radar
* events when they happen. Otherwise a switch to a channel
* requiring DFS will be rejected.
*/
need_handle_dfs_flag = false;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
/* useless if AP is not running */
if (!wdev->beacon_interval)
nl80211: check channel switch validity better Before allowing userspace to initiate a channel switch, check that it's actually connected in some sense. Also use a more appropriate error code for the not connected case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:05:27 +00:00
return -ENOTCONN;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
break;
case NL80211_IFTYPE_ADHOC:
nl80211: check channel switch validity better Before allowing userspace to initiate a channel switch, check that it's actually connected in some sense. Also use a more appropriate error code for the not connected case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:05:27 +00:00
if (!wdev->ssid_len)
return -ENOTCONN;
break;
{nl,cfg,mac}80211: enable the triggering of CSA frame in mesh Allow the triggering of CSA frame using mesh interface. The rules are more or less same with IBSS, such as not allowed to change between the band and channel width has to be same from the previous mode. Also, move the ieee80211_send_action_csa to a common space so that it can be re-used by mesh interface. Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-15 02:08:28 +00:00
case NL80211_IFTYPE_MESH_POINT:
nl80211: check channel switch validity better Before allowing userspace to initiate a channel switch, check that it's actually connected in some sense. Also use a more appropriate error code for the not connected case. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:05:27 +00:00
if (!wdev->mesh_id_len)
return -ENOTCONN;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
break;
default:
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EOPNOTSUPP;
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
}
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
memset(&params, 0, sizeof(params));
if (!info->attrs[NL80211_ATTR_WIPHY_FREQ] ||
!info->attrs[NL80211_ATTR_CH_SWITCH_COUNT])
return -EINVAL;
/* only important for AP, IBSS and mesh create IEs internally */
nl80211: fix channel switch parsing The nl80211 attribute NL80211_ATTR_CSA_C_OFF_BEACON should be nested inside NL80211_ATTR_CSA_IES, but commit ee4bc9e75811d2c0cb5f2a2fc5b5 ("nl80211: enable IBSS support for channel switch announcements") added a check in the outer message attributes. Fix channel switch calls by removing the erroneus condition. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [reword commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-17 08:52:17 +00:00
if (need_new_beacon && !info->attrs[NL80211_ATTR_CSA_IES])
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
nl80211: sanity check the channel switch counter value The nl80211 channel switch count attribute (NL80211_ATTR_CH_SWITCH_COUNT) is specified as u32, but the specification uses u8 for the counter. To make sure strange things don't happen without informing the user, sanity check the value and return -EINVAL if it doesn't fit in u8. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-08 06:48:34 +00:00
/* Even though the attribute is u32, the specification says
* u8, so let's make sure we don't overflow.
*/
cs_count = nla_get_u32(info->attrs[NL80211_ATTR_CH_SWITCH_COUNT]);
if (cs_count > 255)
return -EINVAL;
params.count = cs_count;
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
if (!need_new_beacon)
goto skip_beacons;
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
err = nl80211_parse_beacon(info->attrs, &params.beacon_after);
if (err)
return err;
err = nla_parse_nested(csa_attrs, NL80211_ATTR_MAX,
info->attrs[NL80211_ATTR_CSA_IES],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_policy, info->extack);
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
if (err)
return err;
err = nl80211_parse_beacon(csa_attrs, &params.beacon_csa);
if (err)
return err;
if (!csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON])
return -EINVAL;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
len = nla_len(csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]);
if (!len || (len % sizeof(u16)))
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
params.n_counter_offsets_beacon = len / sizeof(u16);
if (rdev->wiphy.max_num_csa_counters &&
(params.n_counter_offsets_beacon >
rdev->wiphy.max_num_csa_counters))
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
params.counter_offsets_beacon =
nla_data(csa_attrs[NL80211_ATTR_CSA_C_OFF_BEACON]);
/* sanity checks - counters should fit and be the same */
for (i = 0; i < params.n_counter_offsets_beacon; i++) {
u16 offset = params.counter_offsets_beacon[i];
if (offset >= params.beacon_csa.tail_len)
return -EINVAL;
if (params.beacon_csa.tail[offset] != params.count)
return -EINVAL;
}
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
if (csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]) {
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
len = nla_len(csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]);
if (!len || (len % sizeof(u16)))
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
params.n_counter_offsets_presp = len / sizeof(u16);
if (rdev->wiphy.max_num_csa_counters &&
nl80211: validate number of probe response CSA counters Due to an apparent copy/paste bug, the number of counters for the beacon configuration were checked twice, instead of checking the number of probe response counters. Fix this to check the number of probe response counters before parsing those. Cc: stable@vger.kernel.org Fixes: 9a774c78e211 ("cfg80211: Support multiple CSA counters") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 13:53:55 +00:00
(params.n_counter_offsets_presp >
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
rdev->wiphy.max_num_csa_counters))
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
cfg80211: Support multiple CSA counters Change the type of NL80211_ATTR_CSA_C_OFF_BEACON and NL80211_ATTR_CSA_C_OFF_PRESP to be NLA_BINARY which allows userspace to use beacons and probe responses with multiple CSA counters. This isn't breaking the API since userspace can continue to use nla_put_u16 for this attributes, which is equivalent to a single element u16 array. In addition advertise max number of supported CSA counters. This is needed when using CSA and eCSA IEs together. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:46 +00:00
params.counter_offsets_presp =
nla_data(csa_attrs[NL80211_ATTR_CSA_C_OFF_PRESP]);
/* sanity checks - counters should fit and be the same */
for (i = 0; i < params.n_counter_offsets_presp; i++) {
u16 offset = params.counter_offsets_presp[i];
if (offset >= params.beacon_csa.probe_resp_len)
return -EINVAL;
if (params.beacon_csa.probe_resp[offset] !=
params.count)
return -EINVAL;
}
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
}
nl80211: enable IBSS support for channel switch announcements Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-28 11:41:33 +00:00
skip_beacons:
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
err = nl80211_parse_chandef(rdev, info, &params.chandef);
if (err)
return err;
cfg80211: use RTNL locked reg_can_beacon for IR-relaxation The RTNL is required to check for IR-relaxation conditions that allow more channels to beacon. Export an RTNL locked version of reg_can_beacon and use it where possible in AP/STA interface type flows, where IR-relaxation may be applicable. Fixes: 06f207fc5418 ("cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA") Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-07-08 12:41:44 +00:00
if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &params.chandef,
wdev->iftype))
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
return -EINVAL;
cfg80211/mac80211: refactor cfg80211_chandef_dfs_required() Some interface types don't require DFS (such as STATION, P2P_CLIENT etc). In order to centralize these decisions, make cfg80211_chandef_dfs_required() take the iftype into consideration. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-18 09:40:36 +00:00
err = cfg80211_chandef_dfs_required(wdev->wiphy,
&params.chandef,
wdev->iftype);
if (err < 0)
return err;
wireless: Require HANDLE_DFS flag to switch channel for non-AP mode In the case the channel should be switched to one requiring DFS we need to make sure that userspace will handle radar events when they happen. For AP mode this is assumed to be the case, as a manager like hostapd is required. However IBSS and MESH modes can work without further userspace assistance, so refuse to use DFS channels unless userspace vouches that it handles DFS. NOTE: Userspace should have already flagged support earlier during mesh or IBSS setup. However, this information is not readily accessible currently. Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net> [sw: style cleanups] Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-05-16 09:23:12 +00:00
if (err > 0) {
cfg80211/mac80211: refactor cfg80211_chandef_dfs_required() Some interface types don't require DFS (such as STATION, P2P_CLIENT etc). In order to centralize these decisions, make cfg80211_chandef_dfs_required() take the iftype into consideration. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-18 09:40:36 +00:00
params.radar_required = true;
wireless: Require HANDLE_DFS flag to switch channel for non-AP mode In the case the channel should be switched to one requiring DFS we need to make sure that userspace will handle radar events when they happen. For AP mode this is assumed to be the case, as a manager like hostapd is required. However IBSS and MESH modes can work without further userspace assistance, so refuse to use DFS channels unless userspace vouches that it handles DFS. NOTE: Userspace should have already flagged support earlier during mesh or IBSS setup. However, this information is not readily accessible currently. Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net> [sw: style cleanups] Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-05-16 09:23:12 +00:00
if (need_handle_dfs_flag &&
!nla_get_flag(info->attrs[NL80211_ATTR_HANDLE_DFS])) {
return -EINVAL;
}
}
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
if (info->attrs[NL80211_ATTR_CH_SWITCH_BLOCK_TX])
params.block_tx = true;
cfg80211: protect beacon changing functions with wdev-lock To avoid race conditions in functions which modify the beacon information, lock these using the wdev lock. This is especially required to avoid problems for csa handling functions which modify beacons but can not be called under rtnl lock. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:49 +00:00
wdev_lock(wdev);
err = rdev_channel_switch(rdev, dev, &params);
wdev_unlock(wdev);
return err;
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
}
nl80211: use netlink consistent dump feature for BSS dumps Use the new consistent dump feature from (generic) netlink to advertise when dumps are incomplete. Readers may note that this does not initialize the rdev->bss_generation counter to a non-zero value. This is still OK since the value is modified only under spinlock when the list is modified. Since the dump code holds the spinlock, the value will either be > 0 already, or the list will still be empty in which case a consistent dump will actually be made (and be empty). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-21 07:45:33 +00:00
static int nl80211_send_bss(struct sk_buff *msg, struct netlink_callback *cb,
u32 seq, int flags,
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
struct cfg80211_registered_device *rdev,
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
struct wireless_dev *wdev,
struct cfg80211_internal_bss *intbss)
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
struct cfg80211_bss *res = &intbss->pub;
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
const struct cfg80211_bss_ies *ies;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
void *hdr;
struct nlattr *bss;
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
ASSERT_WDEV_LOCK(wdev);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, NETLINK_CB(cb->skb).portid, seq, flags,
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
NL80211_CMD_NEW_SCAN_RESULTS);
if (!hdr)
return -1;
nl80211: use netlink consistent dump feature for BSS dumps Use the new consistent dump feature from (generic) netlink to advertise when dumps are incomplete. Readers may note that this does not initialize the rdev->bss_generation counter to a non-zero value. This is still OK since the value is modified only under spinlock when the list is modified. Since the dump code holds the spinlock, the value will either be > 0 already, or the list will still be empty in which case a consistent dump will actually be made (and be empty). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-21 07:45:33 +00:00
genl_dump_check_consistent(cb, hdr, &nl80211_fam);
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (nla_put_u32(msg, NL80211_ATTR_GENERATION, rdev->bss_generation))
goto nla_put_failure;
if (wdev->netdev &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_ATTR_IFINDEX, wdev->netdev->ifindex))
goto nla_put_failure;
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
goto nla_put_failure;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
bss = nla_nest_start(msg, NL80211_ATTR_BSS);
if (!bss)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if ((!is_zero_ether_addr(res->bssid) &&
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
nla_put(msg, NL80211_BSS_BSSID, ETH_ALEN, res->bssid)))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
rcu_read_lock();
cfg80211: clarify BSS probe response vs. beacon data There are a few possible cases of where BSS data came from: 1) only a beacon has been received 2) only a probe response has been received 3) the driver didn't report what it received (this happens when using cfg80211_inform_bss[_width]()) 4) both probe response and beacon data has been received Unfortunately, in the userspace API, a few things weren't there: a) there was no way to differentiate cases 1) and 4) above without comparing the data of the IEs b) the TSF was always from the last frame, instead of being exposed for beacon/probe response separately like IEs Fix this by i) exporting a new flag attribute that indicates whether or not probe response data has been received - this addresses (a) ii) exporting a BEACON_TSF attribute that holds the beacon's TSF if a beacon has been received iii) not exporting the beacon attributes in case (3) above as that would just lead userspace into thinking the data actually came from a beacon when that isn't clear To implement this, track inside the IEs struct whether or not it (definitely) came from a beacon. Reported-by: William Seto Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-08-12 18:34:30 +00:00
/* indicate whether we have probe response data or not */
if (rcu_access_pointer(res->proberesp_ies) &&
nla_put_flag(msg, NL80211_BSS_PRESP_DATA))
goto fail_unlock_rcu;
/* this pointer prefers to be pointed to probe response data
* but is always valid
*/
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
ies = rcu_dereference(res->ies);
cfg80211: move TSF into IEs While technically the TSF isn't an IE, it can be necessary to distinguish between the TSF from a beacon and a probe response, in particular in order to know the next DTIM TBTT, as not all APs are spec compliant wrt. TSF==0 being a DTIM TBTT and thus the DTIM count needs to be taken into account as well. To allow this, move the TSF into the IE struct so it can be known whence it came. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-05 15:54:31 +00:00
if (ies) {
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_BSS_TSF, ies->tsf,
NL80211_BSS_PAD))
cfg80211: move TSF into IEs While technically the TSF isn't an IE, it can be necessary to distinguish between the TSF from a beacon and a probe response, in particular in order to know the next DTIM TBTT, as not all APs are spec compliant wrt. TSF==0 being a DTIM TBTT and thus the DTIM count needs to be taken into account as well. To allow this, move the TSF into the IE struct so it can be known whence it came. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-05 15:54:31 +00:00
goto fail_unlock_rcu;
if (ies->len && nla_put(msg, NL80211_BSS_INFORMATION_ELEMENTS,
ies->len, ies->data))
goto fail_unlock_rcu;
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
}
cfg80211: clarify BSS probe response vs. beacon data There are a few possible cases of where BSS data came from: 1) only a beacon has been received 2) only a probe response has been received 3) the driver didn't report what it received (this happens when using cfg80211_inform_bss[_width]()) 4) both probe response and beacon data has been received Unfortunately, in the userspace API, a few things weren't there: a) there was no way to differentiate cases 1) and 4) above without comparing the data of the IEs b) the TSF was always from the last frame, instead of being exposed for beacon/probe response separately like IEs Fix this by i) exporting a new flag attribute that indicates whether or not probe response data has been received - this addresses (a) ii) exporting a BEACON_TSF attribute that holds the beacon's TSF if a beacon has been received iii) not exporting the beacon attributes in case (3) above as that would just lead userspace into thinking the data actually came from a beacon when that isn't clear To implement this, track inside the IEs struct whether or not it (definitely) came from a beacon. Reported-by: William Seto Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-08-12 18:34:30 +00:00
/* and this pointer is always (unless driver didn't know) beacon data */
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
ies = rcu_dereference(res->beacon_ies);
cfg80211: clarify BSS probe response vs. beacon data There are a few possible cases of where BSS data came from: 1) only a beacon has been received 2) only a probe response has been received 3) the driver didn't report what it received (this happens when using cfg80211_inform_bss[_width]()) 4) both probe response and beacon data has been received Unfortunately, in the userspace API, a few things weren't there: a) there was no way to differentiate cases 1) and 4) above without comparing the data of the IEs b) the TSF was always from the last frame, instead of being exposed for beacon/probe response separately like IEs Fix this by i) exporting a new flag attribute that indicates whether or not probe response data has been received - this addresses (a) ii) exporting a BEACON_TSF attribute that holds the beacon's TSF if a beacon has been received iii) not exporting the beacon attributes in case (3) above as that would just lead userspace into thinking the data actually came from a beacon when that isn't clear To implement this, track inside the IEs struct whether or not it (definitely) came from a beacon. Reported-by: William Seto Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-08-12 18:34:30 +00:00
if (ies && ies->from_beacon) {
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_BSS_BEACON_TSF, ies->tsf,
NL80211_BSS_PAD))
cfg80211: move TSF into IEs While technically the TSF isn't an IE, it can be necessary to distinguish between the TSF from a beacon and a probe response, in particular in order to know the next DTIM TBTT, as not all APs are spec compliant wrt. TSF==0 being a DTIM TBTT and thus the DTIM count needs to be taken into account as well. To allow this, move the TSF into the IE struct so it can be known whence it came. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-05 15:54:31 +00:00
goto fail_unlock_rcu;
if (ies->len && nla_put(msg, NL80211_BSS_BEACON_IES,
ies->len, ies->data))
goto fail_unlock_rcu;
cfg80211: fix BSS struct IE access races When a BSS struct is updated, the IEs are currently overwritten or freed. This can lead to races if some other CPU is accessing the BSS struct and using the IEs concurrently. Fix this by always allocating the IEs in a new struct that holds the data and length and protecting access to this new struct with RCU. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 00:25:20 +00:00
}
rcu_read_unlock();
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (res->beacon_interval &&
nla_put_u16(msg, NL80211_BSS_BEACON_INTERVAL, res->beacon_interval))
goto nla_put_failure;
if (nla_put_u16(msg, NL80211_BSS_CAPABILITY, res->capability) ||
nla_put_u32(msg, NL80211_BSS_FREQUENCY, res->channel->center_freq) ||
nl80211: add scan width to bss and scan request structs To allow scanning and working with 5 MHz and 10 MHz BSS, extend the inform bss commands and add wrappers to take 5 and 10 MHz bss into account. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-07-08 14:55:49 +00:00
nla_put_u32(msg, NL80211_BSS_CHAN_WIDTH, res->scan_width) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_BSS_SEEN_MS_AGO,
jiffies_to_msecs(jiffies - intbss->ts)))
goto nla_put_failure;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: support beacon report scanning Beacon report radio measurement requires reporting observed BSSs on the channels specified in the beacon request. If the measurement mode is set to passive or active, it requires actually performing a scan (passive or active, accordingly), and reporting the time that the scan was started and the time each beacon/probe was received (both in terms of TSF of the BSS of the requesting AP). If the request mode is table, this information is optional. In addition, the radio measurement request specifies the channel dwell time for the measurement. In order to use scan for beacon report when the mode is active or passive, add a parameter to scan request that specifies the channel dwell time, and add scan start time and beacon received time to scan results information. Supporting beacon report is required for Multi Band Operation (MBO). Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-05 14:10:13 +00:00
if (intbss->parent_tsf &&
(nla_put_u64_64bit(msg, NL80211_BSS_PARENT_TSF,
intbss->parent_tsf, NL80211_BSS_PAD) ||
nla_put(msg, NL80211_BSS_PARENT_BSSID, ETH_ALEN,
intbss->parent_bssid)))
goto nla_put_failure;
nl80211: allow BSS data to include CLOCK_BOOTTIME timestamp For location and connectivity services, userspace would often like to know the time when the BSS was last seen. The current "last seen" value is calculated in a way that makes it less useful, especially if the system suspended in the meantime. Add the ability for the driver to report a real CLOCK_BOOTTIME stamp that can then be reported to userspace (if present). Drivers wishing to use this must be converted to the new API to call cfg80211_inform_bss_data() or cfg80211_inform_bss_frame_data(). They need to ensure the reported value is accurate enough even when the frame might have been buffered in the device (e.g. firmware.) Signed-off-by: Dmitry Shmidt <dimitrysh@google.com> [modified to use struct, inlines] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-07 09:32:53 +00:00
if (intbss->ts_boottime &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_BSS_LAST_SEEN_BOOTTIME,
intbss->ts_boottime, NL80211_BSS_PAD))
nl80211: allow BSS data to include CLOCK_BOOTTIME timestamp For location and connectivity services, userspace would often like to know the time when the BSS was last seen. The current "last seen" value is calculated in a way that makes it less useful, especially if the system suspended in the meantime. Add the ability for the driver to report a real CLOCK_BOOTTIME stamp that can then be reported to userspace (if present). Drivers wishing to use this must be converted to the new API to call cfg80211_inform_bss_data() or cfg80211_inform_bss_frame_data(). They need to ensure the reported value is accurate enough even when the frame might have been buffered in the device (e.g. firmware.) Signed-off-by: Dmitry Shmidt <dimitrysh@google.com> [modified to use struct, inlines] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-07 09:32:53 +00:00
goto nla_put_failure;
cfg80211: clean up signal type It wasn't a good idea to make the signal type a per-BSS option, although then it is closer to the actual value. Move it to be a per-wiphy setting, update mac80211 to match. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-18 17:45:06 +00:00
switch (rdev->wiphy.signal_type) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
case CFG80211_SIGNAL_TYPE_MBM:
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_BSS_SIGNAL_MBM, res->signal))
goto nla_put_failure;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
break;
case CFG80211_SIGNAL_TYPE_UNSPEC:
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u8(msg, NL80211_BSS_SIGNAL_UNSPEC, res->signal))
goto nla_put_failure;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
break;
default:
break;
}
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
switch (wdev->iftype) {
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
case NL80211_IFTYPE_P2P_CLIENT:
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
case NL80211_IFTYPE_STATION:
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (intbss == wdev->current_bss &&
nla_put_u32(msg, NL80211_BSS_STATUS,
NL80211_BSS_STATUS_ASSOCIATED))
goto nla_put_failure;
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
break;
case NL80211_IFTYPE_ADHOC:
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (intbss == wdev->current_bss &&
nla_put_u32(msg, NL80211_BSS_STATUS,
NL80211_BSS_STATUS_IBSS_JOINED))
goto nla_put_failure;
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
break;
default:
break;
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nla_nest_end(msg, bss);
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cfg80211: move TSF into IEs While technically the TSF isn't an IE, it can be necessary to distinguish between the TSF from a beacon and a probe response, in particular in order to know the next DTIM TBTT, as not all APs are spec compliant wrt. TSF==0 being a DTIM TBTT and thus the DTIM count needs to be taken into account as well. To allow this, move the TSF into the IE struct so it can be known whence it came. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-05 15:54:31 +00:00
fail_unlock_rcu:
rcu_read_unlock();
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
static int nl80211_dump_scan(struct sk_buff *skb, struct netlink_callback *cb)
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
struct cfg80211_registered_device *rdev;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
struct cfg80211_internal_bss *scan;
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
struct wireless_dev *wdev;
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
int start = cb->args[2], idx = 0;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
int err;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
err = nl80211_prepare_wdev_dump(skb, cb, &rdev, &wdev);
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (err) {
rtnl_unlock();
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
return err;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
}
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
wdev_lock(wdev);
spin_lock_bh(&rdev->bss_lock);
cfg80211_bss_expire(rdev);
nl80211: use netlink consistent dump feature for BSS dumps Use the new consistent dump feature from (generic) netlink to advertise when dumps are incomplete. Readers may note that this does not initialize the rdev->bss_generation counter to a non-zero value. This is still OK since the value is modified only under spinlock when the list is modified. Since the dump code holds the spinlock, the value will either be > 0 already, or the list will still be empty in which case a consistent dump will actually be made (and be empty). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-21 07:45:33 +00:00
cb->seq = rdev->bss_generation;
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
list_for_each_entry(scan, &rdev->bss_list, list) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (++idx <= start)
continue;
nl80211: use netlink consistent dump feature for BSS dumps Use the new consistent dump feature from (generic) netlink to advertise when dumps are incomplete. Readers may note that this does not initialize the rdev->bss_generation counter to a non-zero value. This is still OK since the value is modified only under spinlock when the list is modified. Since the dump code holds the spinlock, the value will either be > 0 already, or the list will still be empty in which case a consistent dump will actually be made (and be empty). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-21 07:45:33 +00:00
if (nl80211_send_bss(skb, cb,
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
rdev, wdev, scan) < 0) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
idx--;
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
break;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
}
}
nl80211: report BSS status When connected to a BSS, or joined to an IBSS, we'll want to know in userspace without using wireless extensions, so report the BSS status in the BSS list. Userspace can query the BSS list, display all the information and retrieve the station information as well. For example (from hwsim): $ iw dev wlan1 scan dump BSS 02:00:00:00:00:00 (on wlan1) -- associated freq: 2462 beacon interval: 100 capability: ESS ShortSlotTime (0x0401) signal: -50.00 dBm SSID: j Supported rates: 1.0* 2.0* 5.5* 11.0* 6.0 9.0 12.0 18.0 DS Paramater set: channel 11 ERP: <no flags> Extended supported rates: 24.0 36.0 48.0 54.0 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-10 16:42:31 +00:00
spin_unlock_bh(&rdev->bss_lock);
wdev_unlock(wdev);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
cb->args[2] = idx;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_unlock();
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
return skb->len;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
}
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
static int nl80211_send_survey(struct sk_buff *msg, u32 portid, u32 seq,
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
int flags, struct net_device *dev,
bool allow_radio_stats,
struct survey_info *survey)
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
{
void *hdr;
struct nlattr *infoattr;
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
/* skip radio stats if userspace didn't request them */
if (!survey->channel && !allow_radio_stats)
return 0;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, portid, seq, flags,
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
NL80211_CMD_NEW_SURVEY_RESULTS);
if (!hdr)
return -ENOMEM;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex))
goto nla_put_failure;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
infoattr = nla_nest_start(msg, NL80211_ATTR_SURVEY_INFO);
if (!infoattr)
goto nla_put_failure;
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
if (survey->channel &&
nla_put_u32(msg, NL80211_SURVEY_INFO_FREQUENCY,
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
survey->channel->center_freq))
goto nla_put_failure;
if ((survey->filled & SURVEY_INFO_NOISE_DBM) &&
nla_put_u8(msg, NL80211_SURVEY_INFO_NOISE, survey->noise))
goto nla_put_failure;
if ((survey->filled & SURVEY_INFO_IN_USE) &&
nla_put_flag(msg, NL80211_SURVEY_INFO_IN_USE))
goto nla_put_failure;
cfg80211: remove "channel" from survey names All of the survey data is (currently) per channel anyway, so having the word "channel" in the name does nothing. In the next patch I'll introduce global data to the survey, where the word "channel" is actually confusing. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:35:34 +00:00
if ((survey->filled & SURVEY_INFO_TIME) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME,
survey->time, NL80211_SURVEY_INFO_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: remove "channel" from survey names All of the survey data is (currently) per channel anyway, so having the word "channel" in the name does nothing. In the next patch I'll introduce global data to the survey, where the word "channel" is actually confusing. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:35:34 +00:00
if ((survey->filled & SURVEY_INFO_TIME_BUSY) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME_BUSY,
survey->time_busy, NL80211_SURVEY_INFO_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: remove "channel" from survey names All of the survey data is (currently) per channel anyway, so having the word "channel" in the name does nothing. In the next patch I'll introduce global data to the survey, where the word "channel" is actually confusing. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:35:34 +00:00
if ((survey->filled & SURVEY_INFO_TIME_EXT_BUSY) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME_EXT_BUSY,
survey->time_ext_busy, NL80211_SURVEY_INFO_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: remove "channel" from survey names All of the survey data is (currently) per channel anyway, so having the word "channel" in the name does nothing. In the next patch I'll introduce global data to the survey, where the word "channel" is actually confusing. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:35:34 +00:00
if ((survey->filled & SURVEY_INFO_TIME_RX) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME_RX,
survey->time_rx, NL80211_SURVEY_INFO_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: remove "channel" from survey names All of the survey data is (currently) per channel anyway, so having the word "channel" in the name does nothing. In the next patch I'll introduce global data to the survey, where the word "channel" is actually confusing. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:35:34 +00:00
if ((survey->filled & SURVEY_INFO_TIME_TX) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME_TX,
survey->time_tx, NL80211_SURVEY_INFO_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: add scan time to survey data Add the time spent scanning to the survey data so it can be reported by drivers that collect such information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:44:11 +00:00
if ((survey->filled & SURVEY_INFO_TIME_SCAN) &&
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_SURVEY_INFO_TIME_SCAN,
survey->time_scan, NL80211_SURVEY_INFO_PAD))
cfg80211: add scan time to survey data Add the time spent scanning to the survey data so it can be reported by drivers that collect such information. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:44:11 +00:00
goto nla_put_failure;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
nla_nest_end(msg, infoattr);
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
static int nl80211_dump_survey(struct sk_buff *skb, struct netlink_callback *cb)
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
{
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
struct nlattr **attrbuf = genl_family_attrbuf(&nl80211_fam);
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
struct survey_info survey;
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
struct cfg80211_registered_device *rdev;
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
struct wireless_dev *wdev;
int survey_idx = cb->args[2];
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
int res;
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
bool radio_stats;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
res = nl80211_prepare_wdev_dump(skb, cb, &rdev, &wdev);
nl80211: reduce dumping boilerplate Consolidate boilerplate code needed for .dumpit calls operating on netdevs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:06 +00:00
if (res)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out_err;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
/* prepare_wdev_dump parsed the attributes */
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
radio_stats = attrbuf[NL80211_ATTR_SURVEY_RADIO_STATS];
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
if (!wdev->netdev) {
res = -EINVAL;
goto out_err;
}
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
if (!rdev->ops->dump_survey) {
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
res = -EOPNOTSUPP;
goto out_err;
}
while (1) {
cfg80211: change registered device pointer name Name "dev" is too common and ambiguous, let all the pointer name pointing to struct cfg80211_registered_device be "rdev". This can improve code readability and consistency(since other places have already called it rdev). Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:02 +00:00
res = rdev_dump_survey(rdev, wdev->netdev, survey_idx, &survey);
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
if (res == -ENOENT)
break;
if (res)
goto out_err;
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
/* don't send disabled channels, but do send non-channel data */
if (survey.channel &&
survey.channel->flags & IEEE80211_CHAN_DISABLED) {
cfg80211: skip disabled channels on channel survey The channel survey information will be empy for disabled channels so simply discard those entries. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-27 14:24:02 +00:00
survey_idx++;
continue;
}
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
if (nl80211_send_survey(skb,
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
NETLINK_CB(cb->skb).portid,
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
cfg80211: allow survey data to return global data Not all devices are able to report survey data (particularly time spent for various operations) per channel. As all these statistics already exist in survey data, allow such devices to report them (if userspace requested it) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 15:43:50 +00:00
wdev->netdev, radio_stats, &survey) < 0)
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
goto out;
survey_idx++;
}
out:
nl80211: allow using wdev identifiers to get scan results Most dump callbacks, including the scan results one, use the netdev to identify what to do, which is incorrect for the P2P_DEVICE support, it needs to be able to get the scan result from the wdev. Change all dumps to unify the code, but ones other than scan don't really support being executed on a wdev that has no netdev. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 23:02:55 +00:00
cb->args[2] = survey_idx;
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
res = skb->len;
out_err:
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_unlock();
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
return res;
}
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
static bool nl80211_valid_wpa_versions(u32 wpa_versions)
{
return !(wpa_versions & ~(NL80211_WPA_VERSION_1 |
NL80211_WPA_VERSION_2));
}
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
static int nl80211_authenticate(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
struct ieee80211_channel *chan;
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
const u8 *bssid, *ssid, *ie = NULL, *auth_data = NULL;
int err, ssid_len, ie_len = 0, auth_data_len = 0;
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
enum nl80211_auth_type auth_type;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
struct key_parse key;
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
bool local_state_change;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
nl80211: Require auth type for NL80211_CMD_AUTHENTICATE NL80211_ATTR_AUTH_TYPE is a required parameter for NL80211_CMD_AUTHENTICATE. We are currently (by chance) defaulting to open system authentication if the attribute is not specified. It is better to just reject the invalid command. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 18:52:47 +00:00
if (!info->attrs[NL80211_ATTR_AUTH_TYPE])
return -EINVAL;
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
if (!info->attrs[NL80211_ATTR_SSID])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_WIPHY_FREQ])
return -EINVAL;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
err = nl80211_parse_key(info, &key);
if (err)
return err;
if (key.idx >= 0) {
cfg80211/mac80211: allow per-station GTKs This adds API to allow adding per-station GTKs, updates mac80211 to support it, and also allows drivers to remove a key from hwaccel again when this may be necessary due to multiple GTKs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-05 17:39:30 +00:00
if (key.type != -1 && key.type != NL80211_KEYTYPE_GROUP)
return -EINVAL;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (!key.p.key || !key.p.key_len)
return -EINVAL;
if ((key.p.cipher != WLAN_CIPHER_SUITE_WEP40 ||
key.p.key_len != WLAN_KEY_LEN_WEP40) &&
(key.p.cipher != WLAN_CIPHER_SUITE_WEP104 ||
key.p.key_len != WLAN_KEY_LEN_WEP104))
return -EINVAL;
cfg80211: disallow shared key authentication with key index 4 Key index 4 can only be used for an IGTK, so the range checks for shared key authentication should treat 4 as an error, fix that in the code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-13 14:25:58 +00:00
if (key.idx > 3)
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
return -EINVAL;
} else {
key.p.key_len = 0;
key.p.key = NULL;
}
cfg80211: check if WEP is available for shared key auth When shared key auth is requested, cfg80211 should verify that the device is capable of WEP crypto which is required. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-10 07:46:42 +00:00
if (key.idx >= 0) {
int i;
bool ok = false;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211: check if WEP is available for shared key auth When shared key auth is requested, cfg80211 should verify that the device is capable of WEP crypto which is required. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-10 07:46:42 +00:00
for (i = 0; i < rdev->wiphy.n_cipher_suites; i++) {
if (key.p.cipher == rdev->wiphy.cipher_suites[i]) {
ok = true;
break;
}
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!ok)
return -EINVAL;
cfg80211: check if WEP is available for shared key auth When shared key auth is requested, cfg80211 should verify that the device is capable of WEP crypto which is required. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-10 07:46:42 +00:00
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->auth)
return -EOPNOTSUPP;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: Clean up connect params and channel fetching Addition of the frequency hints showed up couple of places in cfg80211 where pointers could be marked const and a shared function could be used to fetch a valid channel. Signed-off-by: Jouni Malinen <j@w1.fi> [fix mwifiex] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:44 +00:00
chan = nl80211_get_valid_chan(&rdev->wiphy,
info->attrs[NL80211_ATTR_WIPHY_FREQ]);
if (!chan)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
if (info->attrs[NL80211_ATTR_IE]) {
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
ie = nla_data(info->attrs[NL80211_ATTR_IE]);
ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
auth_type = nla_get_u32(info->attrs[NL80211_ATTR_AUTH_TYPE]);
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
if (!nl80211_valid_auth_type(rdev, auth_type, NL80211_CMD_AUTHENTICATE))
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: Add Fast Initial Link Setup (FILS) auth algs This defines authentication algorithms for FILS (IEEE 802.11ai). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:02 +00:00
if ((auth_type == NL80211_AUTHTYPE_SAE ||
auth_type == NL80211_AUTHTYPE_FILS_SK ||
auth_type == NL80211_AUTHTYPE_FILS_SK_PFS ||
auth_type == NL80211_AUTHTYPE_FILS_PK) &&
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
!info->attrs[NL80211_ATTR_AUTH_DATA])
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
return -EINVAL;
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
if (info->attrs[NL80211_ATTR_AUTH_DATA]) {
cfg80211: Add Fast Initial Link Setup (FILS) auth algs This defines authentication algorithms for FILS (IEEE 802.11ai). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:02 +00:00
if (auth_type != NL80211_AUTHTYPE_SAE &&
auth_type != NL80211_AUTHTYPE_FILS_SK &&
auth_type != NL80211_AUTHTYPE_FILS_SK_PFS &&
auth_type != NL80211_AUTHTYPE_FILS_PK)
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
return -EINVAL;
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
auth_data = nla_data(info->attrs[NL80211_ATTR_AUTH_DATA]);
auth_data_len = nla_len(info->attrs[NL80211_ATTR_AUTH_DATA]);
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
/* need to include at least Auth Transaction and Status Code */
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
if (auth_data_len < 4)
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
return -EINVAL;
}
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE];
cfg80211: stop tracking authenticated state To track authenticated state seems to have been a design mistake in cfg80211. It is possible to have out of band authentication (FT), tracking multiple authentications caused more problems than it ever helped, and the implementation in mac80211 is too complex. Remove all this complexity, and let userspace do whatever it wants to, mac80211 can deal with that just fine. Association is still tracked of course, but authentication no longer is. Local auth state changes are thus no longer of value, so ignore them completely. This will also help implement SAE -- asking the driver to do an authentication is now almost equivalent to sending an authentication frame, with the exception of shared key authentication which is still handled completely. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-20 12:55:25 +00:00
/*
* Since we no longer track auth state, ignore
* requests to only change local state.
*/
if (local_state_change)
return 0;
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_lock(dev->ieee80211_ptr);
err = cfg80211_mlme_auth(rdev, dev, chan, auth_type, bssid,
ssid, ssid_len, ie, ie_len,
key.p.key, key.p.key_len, key.idx,
cfg80211: Rename SAE_DATA to more generic AUTH_DATA This adds defines and nl80211 extensions to allow FILS Authentication to be implemented similarly to SAE. FILS does not need the special rules for the Authentication transaction number and Status code fields, but it does need to add non-IE fields. The previously used NL80211_ATTR_SAE_DATA can be reused for this to avoid having to duplicate that implementation. Rename that attribute to more generic NL80211_ATTR_AUTH_DATA (with backwards compatibility define for NL80211_SAE_DATA). Also document the special rules related to the Authentication transaction number and Status code fiels. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:41:58 +00:00
auth_data, auth_data_len);
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_unlock(dev->ieee80211_ptr);
return err;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
cfg80211: allow changing port control protocol Some vendor specified mechanisms for 802.1X-style functionality use a different protocol than EAP (even if EAP is vendor-extensible). Allow setting the ethertype for the protocol when a driver has support for this. The default if unspecified is EAP, of course. Note: This is suitable only for station mode, not for AP implementation. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-27 11:26:53 +00:00
static int nl80211_crypto_settings(struct cfg80211_registered_device *rdev,
struct genl_info *info,
nl80211: limit to one pairwise cipher for associate() In this case, only one cipher makes sense, unlike for connect() where it may be possible to have the card or driver select. No changes to mac80211 due to the way the structs are laid out -- but the loop in net/mac80211/cfg.c will degrade to just zero or one passes. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 19:36:37 +00:00
struct cfg80211_crypto_settings *settings,
int cipher_limit)
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
nl80211: add missing parameter clearing Jouni and Maxim reported an oops when using wpa_supplicant -Dnl80211, which seems to be due to random data being contained in the crypto settings for the assoc() command. This seems to be due to the missing memset here, so add it -- it's certainly missing but I'm not 100% certain that it will fix the problem. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-25 14:54:36 +00:00
memset(settings, 0, sizeof(*settings));
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
settings->control_port = info->attrs[NL80211_ATTR_CONTROL_PORT];
cfg80211: allow changing port control protocol Some vendor specified mechanisms for 802.1X-style functionality use a different protocol than EAP (even if EAP is vendor-extensible). Allow setting the ethertype for the protocol when a driver has support for this. The default if unspecified is EAP, of course. Note: This is suitable only for station mode, not for AP implementation. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-27 11:26:53 +00:00
if (info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]) {
u16 proto;
nl80211: Fix checkpatch warnings about blank lines This patch fixes the following checkpatch.pl issues - - Please don't use multiple blank lines - Blank lines aren't necessary before a close brace - Missing a blank line after declarations Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Kirtika Ruchandani <kirtika.ruchandani@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 02:51:23 +00:00
cfg80211: allow changing port control protocol Some vendor specified mechanisms for 802.1X-style functionality use a different protocol than EAP (even if EAP is vendor-extensible). Allow setting the ethertype for the protocol when a driver has support for this. The default if unspecified is EAP, of course. Note: This is suitable only for station mode, not for AP implementation. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-27 11:26:53 +00:00
proto = nla_get_u16(
info->attrs[NL80211_ATTR_CONTROL_PORT_ETHERTYPE]);
settings->control_port_ethertype = cpu_to_be16(proto);
if (!(rdev->wiphy.flags & WIPHY_FLAG_CONTROL_PORT_PROTOCOL) &&
proto != ETH_P_PAE)
return -EINVAL;
if (info->attrs[NL80211_ATTR_CONTROL_PORT_NO_ENCRYPT])
settings->control_port_no_encrypt = true;
} else
settings->control_port_ethertype = cpu_to_be16(ETH_P_PAE);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (info->attrs[NL80211_ATTR_CIPHER_SUITES_PAIRWISE]) {
void *data;
int len, i;
data = nla_data(info->attrs[NL80211_ATTR_CIPHER_SUITES_PAIRWISE]);
len = nla_len(info->attrs[NL80211_ATTR_CIPHER_SUITES_PAIRWISE]);
settings->n_ciphers_pairwise = len / sizeof(u32);
if (len % sizeof(u32))
return -EINVAL;
nl80211: limit to one pairwise cipher for associate() In this case, only one cipher makes sense, unlike for connect() where it may be possible to have the card or driver select. No changes to mac80211 due to the way the structs are laid out -- but the loop in net/mac80211/cfg.c will degrade to just zero or one passes. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 19:36:37 +00:00
if (settings->n_ciphers_pairwise > cipher_limit)
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return -EINVAL;
memcpy(settings->ciphers_pairwise, data, len);
for (i = 0; i < settings->n_ciphers_pairwise; i++)
cfg80211: Validate cipher suite against supported ciphers Instead of using a hardcoded list of cipher suites in nl80211.c, use a shared function in util.c to verify that the driver advertises support for the specified cipher. This provides more accurate validation of the values and allows vendor-specific cipher suites to be added in drivers. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-21 15:14:56 +00:00
if (!cfg80211_supported_cipher_suite(
&rdev->wiphy,
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
settings->ciphers_pairwise[i]))
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_CIPHER_SUITE_GROUP]) {
settings->cipher_group =
nla_get_u32(info->attrs[NL80211_ATTR_CIPHER_SUITE_GROUP]);
cfg80211: Validate cipher suite against supported ciphers Instead of using a hardcoded list of cipher suites in nl80211.c, use a shared function in util.c to verify that the driver advertises support for the specified cipher. This provides more accurate validation of the values and allows vendor-specific cipher suites to be added in drivers. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-21 15:14:56 +00:00
if (!cfg80211_supported_cipher_suite(&rdev->wiphy,
settings->cipher_group))
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_WPA_VERSIONS]) {
settings->wpa_versions =
nla_get_u32(info->attrs[NL80211_ATTR_WPA_VERSIONS]);
if (!nl80211_valid_wpa_versions(settings->wpa_versions))
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_AKM_SUITES]) {
void *data;
cfg80211: Remove strict validation of AKM suites NL80211_ATTR_AKM_SUITES can be used to configure new AKMs, like FT or the SHA-256 -based AKMs or FT from 802.11r/802.11w. In addition, vendor specific AKMs could be used. The current validation code for the connect command prevents cfg80211-based drivers from using these mechanisms even if the driver would not actually use this AKM value (i.e., it uses WPA/RSN IE from user space). mac80211-based drivers allow any AKM to be used since this value is not used there. Remove the unnecessary validation step in cfg80211 to allow drivers to decide what AKMs are supported. In theory, we could handle this by advertising supported AKMs, but that would not be very effective unless we enforce all drivers (including mac80211) to advertise the set of supported AKMs. This would require additional changes in many places whenever a new AKM is introduced even though no actually functionality changes may be required in most drivers. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-21 15:11:33 +00:00
int len;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
data = nla_data(info->attrs[NL80211_ATTR_AKM_SUITES]);
len = nla_len(info->attrs[NL80211_ATTR_AKM_SUITES]);
settings->n_akm_suites = len / sizeof(u32);
if (len % sizeof(u32))
return -EINVAL;
cfg80211: Fix validation of AKM suites Incorrect variable was used in validating the akm_suites array from NL80211_ATTR_AKM_SUITES. In addition, there was no explicit validation of the array length (we only have room for NL80211_MAX_NR_AKM_SUITES). This can result in a buffer write overflow for stack variables with arbitrary data from user space. The nl80211 commands using the affected functionality require GENL_ADMIN_PERM, so this is only exposed to admin users. Cc: stable@kernel.org Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-21 13:13:07 +00:00
if (settings->n_akm_suites > NL80211_MAX_NR_AKM_SUITES)
return -EINVAL;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
memcpy(settings->akm_suites, data, len);
}
cfg80211: support 4-way handshake offloading for WPA/WPA2-PSK Let drivers advertise support for station-mode 4-way handshake offloading with a new NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_PSK flag. Extend use of NL80211_ATTR_PMK attribute indicating it might be passed as part of NL80211_CMD_CONNECT command, and contain the PSK (which is the PMK, hence the name.) The driver/device is assumed to handle the 4-way handshake by itself in this case (including key derivations, etc.), instead of relying on the supplicant. This patch is somewhat based on this one (by Vladimir Kondratiev): https://patchwork.kernel.org/patch/1309561/. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [arend.vanspriel@broadcom.com rebase dealing with existing ATTR_PMK] Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [reword NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_PSK docs to indicate that this offload might be required] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-06-09 12:08:42 +00:00
if (info->attrs[NL80211_ATTR_PMK]) {
if (nla_len(info->attrs[NL80211_ATTR_PMK]) != WLAN_PMK_LEN)
return -EINVAL;
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_PSK))
return -EINVAL;
settings->psk = nla_data(info->attrs[NL80211_ATTR_PMK]);
}
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return 0;
}
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
static int nl80211_associate(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
struct ieee80211_channel *chan;
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
struct cfg80211_assoc_request req = {};
const u8 *bssid, *ssid;
int err, ssid_len = 0;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MAC] ||
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
!info->attrs[NL80211_ATTR_SSID] ||
!info->attrs[NL80211_ATTR_WIPHY_FREQ])
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->assoc)
return -EOPNOTSUPP;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: Clean up connect params and channel fetching Addition of the frequency hints showed up couple of places in cfg80211 where pointers could be marked const and a shared function could be used to fetch a valid channel. Signed-off-by: Jouni Malinen <j@w1.fi> [fix mwifiex] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:44 +00:00
chan = nl80211_get_valid_chan(&rdev->wiphy,
info->attrs[NL80211_ATTR_WIPHY_FREQ]);
if (!chan)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
if (info->attrs[NL80211_ATTR_IE]) {
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
req.ie = nla_data(info->attrs[NL80211_ATTR_IE]);
req.ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
nl80211 : Add support for configuring MFP NL80211_CMD_ASSOCIATE request must be able to indicate whether management frame protection (IEEE 802.11w) is being used. mac80211 was able to use MFP in client mode only with WEXT, but the new NL80211_ATTR_USE_MFP attribute will allow this to be done with nl80211, too. Since we are currently using nl80211 for MFP only with drivers that use user space SME, only MFP disabled and required values are used. However, the NL80211_ATTR_USE_MFP attribute is an enum that can be extended with MFP optional in the future, if that is needed with some drivers (e.g., if the RSN IE is generated by the driver). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-06 19:09:37 +00:00
if (info->attrs[NL80211_ATTR_USE_MFP]) {
cfg80211: fix MFP bug, sparse warnings sparse warns about a number of things, and one of them (use_mfp shadowed variable) actually is a bug, fix all of them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:10 +00:00
enum nl80211_mfp mfp =
nl80211 : Add support for configuring MFP NL80211_CMD_ASSOCIATE request must be able to indicate whether management frame protection (IEEE 802.11w) is being used. mac80211 was able to use MFP in client mode only with WEXT, but the new NL80211_ATTR_USE_MFP attribute will allow this to be done with nl80211, too. Since we are currently using nl80211 for MFP only with drivers that use user space SME, only MFP disabled and required values are used. However, the NL80211_ATTR_USE_MFP attribute is an enum that can be extended with MFP optional in the future, if that is needed with some drivers (e.g., if the RSN IE is generated by the driver). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-06 19:09:37 +00:00
nla_get_u32(info->attrs[NL80211_ATTR_USE_MFP]);
cfg80211: fix MFP bug, sparse warnings sparse warns about a number of things, and one of them (use_mfp shadowed variable) actually is a bug, fix all of them. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:10 +00:00
if (mfp == NL80211_MFP_REQUIRED)
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
req.use_mfp = true;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
else if (mfp != NL80211_MFP_NO)
return -EINVAL;
nl80211 : Add support for configuring MFP NL80211_CMD_ASSOCIATE request must be able to indicate whether management frame protection (IEEE 802.11w) is being used. mac80211 was able to use MFP in client mode only with WEXT, but the new NL80211_ATTR_USE_MFP attribute will allow this to be done with nl80211, too. Since we are currently using nl80211 for MFP only with drivers that use user space SME, only MFP disabled and required values are used. However, the NL80211_ATTR_USE_MFP attribute is an enum that can be extended with MFP optional in the future, if that is needed with some drivers (e.g., if the RSN IE is generated by the driver). Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-06 19:09:37 +00:00
}
cfg80211: let SME control reassociation vs. association Since we don't really know that well in the kernel, let's let the SME control whether it wants to use reassociation or not, by allowing it to give the previous BSSID in the associate() parameters. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 12:37:26 +00:00
if (info->attrs[NL80211_ATTR_PREV_BSSID])
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
req.prev_bssid = nla_data(info->attrs[NL80211_ATTR_PREV_BSSID]);
cfg80211: let SME control reassociation vs. association Since we don't really know that well in the kernel, let's let the SME control whether it wants to use reassociation or not, by allowing it to give the previous BSSID in the associate() parameters. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 12:37:26 +00:00
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_DISABLE_HT]))
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
req.flags |= ASSOC_REQ_DISABLE_HT;
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
if (info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK])
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
memcpy(&req.ht_capa_mask,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK]),
sizeof(req.ht_capa_mask));
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
if (info->attrs[NL80211_ATTR_HT_CAPABILITY]) {
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
if (!info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK])
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
return -EINVAL;
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
memcpy(&req.ht_capa,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY]),
sizeof(req.ht_capa));
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
}
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_DISABLE_VHT]))
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
req.flags |= ASSOC_REQ_DISABLE_VHT;
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK])
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
memcpy(&req.vht_capa_mask,
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK]),
sizeof(req.vht_capa_mask));
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY]) {
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
if (!info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK])
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
return -EINVAL;
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
memcpy(&req.vht_capa,
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY]),
sizeof(req.vht_capa));
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
}
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_USE_RRM])) {
cfg80211: Add global RRM capability Today, the supplicant will add the RRM capabilities Information Element in the association request only if Quiet period is supported (NL80211_FEATURE_QUIET). Quiet is one of many RRM features, and there are other RRM features that are not related to Quiet (e.g. neighbor report). Therefore, requiring Quiet to enable RRM is too restrictive. Some of the features, like neighbor report, can be supported by user space without any help from the kernel. Hence adding the RRM capabilities IE to association request should be the sole user space's decision. Removing the RRM dependency on Quiet in the driver solves this problem, but using an old driver with a user space tool that would not require Quiet feature would be problematic: the user space would add NL80211_ATTR_USE_RRM in the association request even if the kernel doesn't advertize NL80211_FEATURE_QUIET and the association would be denied by the kernel. This solution adds a global RRM capability, that tells user space that it can request RRM capabilities IE publishment without any specific feature support in the kernel. Signed-off-by: Beni Lev <beni.lev@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-02-17 18:30:00 +00:00
if (!((rdev->wiphy.features &
NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) &&
(rdev->wiphy.features & NL80211_FEATURE_QUIET)) &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_RRM))
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
return -EINVAL;
req.flags |= ASSOC_REQ_USE_RRM;
}
cfg80211: Add KEK/nonces for FILS association frames The new nl80211 attributes can be used to provide KEK and nonces to allow the driver to encrypt and decrypt FILS (Re)Association Request/Response frames in station mode. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 21:42:03 +00:00
if (info->attrs[NL80211_ATTR_FILS_KEK]) {
req.fils_kek = nla_data(info->attrs[NL80211_ATTR_FILS_KEK]);
req.fils_kek_len = nla_len(info->attrs[NL80211_ATTR_FILS_KEK]);
if (!info->attrs[NL80211_ATTR_FILS_NONCES])
return -EINVAL;
req.fils_nonces =
nla_data(info->attrs[NL80211_ATTR_FILS_NONCES]);
}
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
err = nl80211_crypto_settings(rdev, info, &req.crypto, 1);
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
if (!err) {
wdev_lock(dev->ieee80211_ptr);
cfg80211: NL80211_ATTR_SOCKET_OWNER support for CMD_CONNECT Disconnect or deauthenticate when the owning socket is closed if this flag is supplied to CMD_CONNECT or CMD_ASSOCIATE. This may be used to ensure userspace daemon doesn't leave an unmanaged connection behind. In some situations it would be possible to account for that, to some degree, in the deamon restart code or in the up/down scripts without the use of this attribute. But there will be systems where the daemon can go away for varying periods without a warning due to local resource management. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-06 21:33:43 +00:00
cfg80211: refactor association parameters cfg80211_mlme_assoc() has grown far too many arguments, make the caller build almost all of the driver struct and pass that to the function instead. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 19:09:09 +00:00
err = cfg80211_mlme_assoc(rdev, dev, chan, bssid,
ssid, ssid_len, &req);
cfg80211: NL80211_ATTR_SOCKET_OWNER support for CMD_CONNECT Disconnect or deauthenticate when the owning socket is closed if this flag is supplied to CMD_CONNECT or CMD_ASSOCIATE. This may be used to ensure userspace daemon doesn't leave an unmanaged connection behind. In some situations it would be possible to account for that, to some degree, in the deamon restart code or in the up/down scripts without the use of this attribute. But there will be systems where the daemon can go away for varying periods without a warning due to local resource management. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-06 21:33:43 +00:00
if (!err && info->attrs[NL80211_ATTR_SOCKET_OWNER]) {
dev->ieee80211_ptr->conn_owner_nlportid =
info->snd_portid;
memcpy(dev->ieee80211_ptr->disconnect_bssid,
bssid, ETH_ALEN);
}
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_unlock(dev->ieee80211_ptr);
}
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
return err;
}
static int nl80211_deauthenticate(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
const u8 *ie = NULL, *bssid;
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
int ie_len = 0, err;
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
u16 reason_code;
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
bool local_state_change;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_REASON_CODE])
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->deauth)
return -EOPNOTSUPP;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
reason_code = nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]);
if (reason_code == 0) {
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
/* Reason Code 0 is reserved */
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
nl80211: Add more through validation of MLME command parameters Check that the used authentication type and reason code are valid here so that drivers/mac80211 do not need to care about this. In addition, remove the unnecessary validation of SSID attribute length which is taken care of by netlink policy. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:17 +00:00
}
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
if (info->attrs[NL80211_ATTR_IE]) {
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
ie = nla_data(info->attrs[NL80211_ATTR_IE]);
ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE];
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_lock(dev->ieee80211_ptr);
err = cfg80211_mlme_deauth(rdev, dev, bssid, ie, ie_len, reason_code,
local_state_change);
wdev_unlock(dev->ieee80211_ptr);
return err;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
static int nl80211_disassociate(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
const u8 *ie = NULL, *bssid;
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
int ie_len = 0, err;
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
u16 reason_code;
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
bool local_state_change;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!info->attrs[NL80211_ATTR_REASON_CODE])
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->disassoc)
return -EOPNOTSUPP;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: Check iftype in cfg80211 code We do not want to require all the drivers using cfg80211 to need to do this. In addition, make the error values consistent by using EOPNOTSUPP instead of semi-random assortment of errno values. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:19 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
reason_code = nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]);
if (reason_code == 0) {
nl80211: validate some input better This patch changes nl80211 to: * validate that any IE input is a valid IE (stream) * move some validation code before locking * require that a reason code is given for both deauth/disassoc Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 11:40:28 +00:00
/* Reason Code 0 is reserved */
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
nl80211: Add more through validation of MLME command parameters Check that the used authentication type and reason code are valid here so that drivers/mac80211 do not need to care about this. In addition, remove the unnecessary validation of SSID attribute length which is taken care of by netlink policy. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-20 19:21:17 +00:00
}
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
if (info->attrs[NL80211_ATTR_IE]) {
cfg80211: keep track of BSSes In order to avoid problems with BSS structs going away while they're in use, I've long wanted to make cfg80211 keep track of them. Without the SME, that wasn't doable but now that we have the SME we can do this too. It can keep track of up to four separate authentications and one association, regardless of whether it's controlled by the cfg80211 SME or the userspace SME. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 15:20:43 +00:00
ie = nla_data(info->attrs[NL80211_ATTR_IE]);
ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
cfg80211: Add local-state-change-only auth/deauth/disassoc cfg80211 is quite strict on allowing authentication and association commands only in certain states. In order to meet these requirements, user space applications may need to clear authentication or association state in some cases. Currently, this can be done with deauth/disassoc command, but that ends up sending out Deauthentication or Disassociation frame unnecessarily. Add a new nl80211 attribute to allow this sending of the frame be skipped, but with all other deauth/disassoc operations being completed. Similar state change is also needed for IEEE 802.11r FT protocol in the FT-over-DS case which does not use Authentication frame exchange in a transition to another BSS. For this to work with cfg80211, an authentication entry needs to be created for the target BSS without sending out an Authentication frame. The nl80211 authentication command can be used for this purpose, too, with the new attribute to indicate that the command is only for changing local state. This enables wpa_supplicant to complete FT-over-DS transition successfully. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-04-04 06:37:19 +00:00
local_state_change = !!info->attrs[NL80211_ATTR_LOCAL_STATE_CHANGE];
cfg80211: remove some locked wrappers from mlme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_lock(dev->ieee80211_ptr);
err = cfg80211_mlme_disassoc(rdev, dev, bssid, ie, ie_len, reason_code,
local_state_change);
wdev_unlock(dev->ieee80211_ptr);
return err;
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
}
cfg80211/mac80211: improve ad-hoc multicast rate handling - store the multicast rate as an index instead of the rate value (reduces cpu overhead in a hotpath) - validate the rate values (must match a bitrate in at least one sband) Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-22 19:58:24 +00:00
static bool
nl80211_parse_mcast_rate(struct cfg80211_registered_device *rdev,
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
int mcast_rate[NUM_NL80211_BANDS],
cfg80211/mac80211: improve ad-hoc multicast rate handling - store the multicast rate as an index instead of the rate value (reduces cpu overhead in a hotpath) - validate the rate values (must match a bitrate in at least one sband) Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-22 19:58:24 +00:00
int rateval)
{
struct wiphy *wiphy = &rdev->wiphy;
bool found = false;
int band, i;
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
for (band = 0; band < NUM_NL80211_BANDS; band++) {
cfg80211/mac80211: improve ad-hoc multicast rate handling - store the multicast rate as an index instead of the rate value (reduces cpu overhead in a hotpath) - validate the rate values (must match a bitrate in at least one sband) Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-22 19:58:24 +00:00
struct ieee80211_supported_band *sband;
sband = wiphy->bands[band];
if (!sband)
continue;
for (i = 0; i < sband->n_bitrates; i++) {
if (sband->bitrates[i].bitrate == rateval) {
mcast_rate[band] = i + 1;
found = true;
break;
}
}
}
return found;
}
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
static int nl80211_join_ibss(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
struct cfg80211_ibss_params ibss;
struct wiphy *wiphy;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
struct cfg80211_cached_keys *connkeys = NULL;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
int err;
nl80211: allow configuring IBSS beacon interval Make the JOIN_IBSS command look at the beacon interval attribute to see if the user requested a specific beacon interval, if not default to 100 TU (wext too). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 15:45:38 +00:00
memset(&ibss, 0, sizeof(ibss));
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (!info->attrs[NL80211_ATTR_SSID] ||
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
!nla_len(info->attrs[NL80211_ATTR_SSID]))
return -EINVAL;
nl80211: allow configuring IBSS beacon interval Make the JOIN_IBSS command look at the beacon interval attribute to see if the user requested a specific beacon interval, if not default to 100 TU (wext too). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 15:45:38 +00:00
ibss.beacon_interval = 100;
cfg80211: identically validate beacon interval for AP/MESH/IBSS Beacon interval interface combinations validation was missing for MESH/IBSS join, add those. Johannes: also move the beacon interval check disallowing really tiny and really big intervals into the common function, which adds it for AP mode. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 09:44:02 +00:00
if (info->attrs[NL80211_ATTR_BEACON_INTERVAL])
nl80211: allow configuring IBSS beacon interval Make the JOIN_IBSS command look at the beacon interval attribute to see if the user requested a specific beacon interval, if not default to 100 TU (wext too). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 15:45:38 +00:00
ibss.beacon_interval =
nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]);
cfg80211: identically validate beacon interval for AP/MESH/IBSS Beacon interval interface combinations validation was missing for MESH/IBSS join, add those. Johannes: also move the beacon interval check disallowing really tiny and really big intervals into the common function, which adds it for AP mode. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 09:44:02 +00:00
cfg80211: support virtual interfaces with different beacon intervals This commit provides a mechanism for the host drivers to advertise the support for different beacon intervals among the respective interface combinations in a group, through NL80211_IFACE_COMB_BI_MIN_GCD (u32). This value will be compared against GCD of all beaconing interfaces of matching combinations. If the driver doesn't advertise this value, the old behaviour where all beacon intervals must be identical is retained. If it is specified, then any beacon interval for an interface in the interface combination as well as the GCD of all active beacon intervals in the combination must be greater or equal to this value. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [change commit message, some variable names, small other things] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-12 12:56:51 +00:00
err = cfg80211_validate_beacon_int(rdev, NL80211_IFTYPE_ADHOC,
ibss.beacon_interval);
cfg80211: identically validate beacon interval for AP/MESH/IBSS Beacon interval interface combinations validation was missing for MESH/IBSS join, add those. Johannes: also move the beacon interval check disallowing really tiny and really big intervals into the common function, which adds it for AP mode. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 09:44:02 +00:00
if (err)
return err;
nl80211: allow configuring IBSS beacon interval Make the JOIN_IBSS command look at the beacon interval attribute to see if the user requested a specific beacon interval, if not default to 100 TU (wext too). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 15:45:38 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->join_ibss)
return -EOPNOTSUPP;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC)
return -EOPNOTSUPP;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
wiphy = &rdev->wiphy;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
cfg80211: validate IBSS BSSID The IBSS BSSID is never validated, so an invalid one might end up being used. Fix this by rejecting invalid configuration. Reported-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 11:45:25 +00:00
if (info->attrs[NL80211_ATTR_MAC]) {
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
ibss.bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: validate IBSS BSSID The IBSS BSSID is never validated, so an invalid one might end up being used. Fix this by rejecting invalid configuration. Reported-by: Marek Lindner <lindner_marek@yahoo.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 11:45:25 +00:00
if (!is_valid_ether_addr(ibss.bssid))
return -EINVAL;
}
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
ibss.ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
ibss.ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]);
if (info->attrs[NL80211_ATTR_IE]) {
ibss.ie = nla_data(info->attrs[NL80211_ATTR_IE]);
ibss.ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
}
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
err = nl80211_parse_chandef(rdev, info, &ibss.chandef);
if (err)
return err;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
cfg80211: Enable GO operation on additional channels Allow GO operation on a channel marked with IEEE80211_CHAN_GO_CONCURRENT iff there is an active station interface that is associated to an AP operating on the same channel in the 2 GHz band or the same UNII band (in the 5 GHz band). This relaxation is not allowed if the channel is marked with IEEE80211_CHAN_RADAR. Note that this is a permissive approach to the FCC definitions, that require a clear assessment that the device operating the AP is an authorized master, i.e., with radar detection and DFS capabilities. It is assumed that such restrictions are enforced by user space. Furthermore, it is assumed, that if the conditions that allowed for the operation of the GO on such a channel change, i.e., the station interface disconnected from the AP, it is the responsibility of user space to evacuate the GO from the channel. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-23 07:13:01 +00:00
if (!cfg80211_reg_can_beacon(&rdev->wiphy, &ibss.chandef,
NL80211_IFTYPE_ADHOC))
nl80211: Parse channel type attribute in an ibss join request Prepare cfg80211 for IBSS HT: * extend cfg80211 ibss struct with channel_type * Check if extension channel can be used * Export can_beacon_sec_chan for use in mac80211 (will be called from ibss.c later). Signed-off-by: Alexander Simon <an.alexsimon@googlemail.com> [siwu@hrz.tu-chemnitz.de: Updates] * fix cfg80211_can_beacon_ext_chan comment * remove implicit channel_type enum assumptions * remove radar channel flags check * add HT IBSS feature flag * reword commit message Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-30 15:56:32 +00:00
return -EINVAL;
nl80211/cfg80211: add 5 and 10 MHz defines and wiphy flag Add defines for 5 and 10 MHz channel width and fix channel handling functions accordingly. Also check for and report the WIPHY_FLAG_SUPPORTS_5_10_MHZ capability. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix spelling in comment] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 11:00:28 +00:00
switch (ibss.chandef.width) {
nl80211: allow 5 and 10 MHz channels for IBSS Whether the wiphy supports it or not is already checked, so what is left is to enable these channel types. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-07-08 14:55:58 +00:00
case NL80211_CHAN_WIDTH_5:
case NL80211_CHAN_WIDTH_10:
nl80211/cfg80211: add 5 and 10 MHz defines and wiphy flag Add defines for 5 and 10 MHz channel width and fix channel handling functions accordingly. Also check for and report the WIPHY_FLAG_SUPPORTS_5_10_MHZ capability. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix spelling in comment] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 11:00:28 +00:00
case NL80211_CHAN_WIDTH_20_NOHT:
break;
case NL80211_CHAN_WIDTH_20:
case NL80211_CHAN_WIDTH_40:
cfg80211: add VHT support for IBSS Add NL80211_EXT_FEATURE_VHT_IBSS flag and VHT support for IBSS. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-21 15:52:39 +00:00
if (!(rdev->wiphy.features & NL80211_FEATURE_HT_IBSS))
return -EINVAL;
break;
case NL80211_CHAN_WIDTH_80:
case NL80211_CHAN_WIDTH_80P80:
case NL80211_CHAN_WIDTH_160:
if (!(rdev->wiphy.features & NL80211_FEATURE_HT_IBSS))
return -EINVAL;
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_VHT_IBSS))
return -EINVAL;
break;
nl80211/cfg80211: add 5 and 10 MHz defines and wiphy flag Add defines for 5 and 10 MHz channel width and fix channel handling functions accordingly. Also check for and report the WIPHY_FLAG_SUPPORTS_5_10_MHZ capability. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix spelling in comment] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 11:00:28 +00:00
default:
nl80211: Fix HT_IBSS feature check in ibss_join There is a standalone if, seems to be a regression of commit "nl80211/cfg80211: add VHT MCS support". Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-29 17:37:22 +00:00
return -EINVAL;
nl80211/cfg80211: add 5 and 10 MHz defines and wiphy flag Add defines for 5 and 10 MHz channel width and fix channel handling functions accordingly. Also check for and report the WIPHY_FLAG_SUPPORTS_5_10_MHZ capability. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix spelling in comment] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 11:00:28 +00:00
}
nl80211/cfg80211: add VHT MCS support Add support for reporting and calculating VHT MCSes. Note that I'm not completely sure that the bitrate calculations are correct, nor that they can't be simplified. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-09 13:56:41 +00:00
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
ibss.channel_fixed = !!info->attrs[NL80211_ATTR_FREQ_FIXED];
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
ibss.privacy = !!info->attrs[NL80211_ATTR_PRIVACY];
mac80211: Set basic rates while joining ibss network This patch adds support to nl80211 and mac80211 to set basic rates when joining/creating ibss network. Original patch was posted by Johannes Berg on the linux-wireless posting list. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-14 09:55:31 +00:00
if (info->attrs[NL80211_ATTR_BSS_BASIC_RATES]) {
u8 *rates =
nla_data(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
int n_rates =
nla_len(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
struct ieee80211_supported_band *sband =
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
wiphy->bands[ibss.chandef.chan->band];
mac80211: Set basic rates while joining ibss network This patch adds support to nl80211 and mac80211 to set basic rates when joining/creating ibss network. Original patch was posted by Johannes Berg on the linux-wireless posting list. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-14 09:55:31 +00:00
cfg80211: allow userspace to control supported rates in scan Some P2P scans are not allowed to advertise 11b rates, but that is a rather special case so instead of having that, allow userspace to request the rate sets (per band) that are advertised in scan probe request frames. Since it's needed in two places now, factor out some common code parsing a rate array. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-18 16:08:35 +00:00
err = ieee80211_get_ratemask(sband, rates, n_rates,
&ibss.basic_rates);
if (err)
return err;
mac80211: Set basic rates while joining ibss network This patch adds support to nl80211 and mac80211 to set basic rates when joining/creating ibss network. Original patch was posted by Johannes Berg on the linux-wireless posting list. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-14 09:55:31 +00:00
}
cfg80211/mac80211: improve ad-hoc multicast rate handling - store the multicast rate as an index instead of the rate value (reduces cpu overhead in a hotpath) - validate the rate values (must match a bitrate in at least one sband) Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-22 19:58:24 +00:00
nl80211: enable HT overrides for ibss Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 08:39:58 +00:00
if (info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK])
memcpy(&ibss.ht_capa_mask,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK]),
sizeof(ibss.ht_capa_mask));
if (info->attrs[NL80211_ATTR_HT_CAPABILITY]) {
if (!info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK])
return -EINVAL;
memcpy(&ibss.ht_capa,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY]),
sizeof(ibss.ht_capa));
}
cfg80211/mac80211: improve ad-hoc multicast rate handling - store the multicast rate as an index instead of the rate value (reduces cpu overhead in a hotpath) - validate the rate values (must match a bitrate in at least one sband) Signed-off-by: Felix Fietkau <nbd@openwrt.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-22 19:58:24 +00:00
if (info->attrs[NL80211_ATTR_MCAST_RATE] &&
!nl80211_parse_mcast_rate(rdev, ibss.mcast_rate,
nla_get_u32(info->attrs[NL80211_ATTR_MCAST_RATE])))
return -EINVAL;
mac80211: Set basic rates while joining ibss network This patch adds support to nl80211 and mac80211 to set basic rates when joining/creating ibss network. Original patch was posted by Johannes Berg on the linux-wireless posting list. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Teemu Paasikivi <ext-teemu.3.paasikivi@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-06-14 09:55:31 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (ibss.privacy && info->attrs[NL80211_ATTR_KEYS]) {
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
bool no_ht = false;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
connkeys = nl80211_parse_connkeys(rdev,
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
info->attrs[NL80211_ATTR_KEYS],
&no_ht);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (IS_ERR(connkeys))
return PTR_ERR(connkeys);
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
nl80211/cfg80211: support VHT channel configuration Change nl80211 to support specifying a VHT (or HT) using the control channel frequency (as before) and new attributes for the channel width and first and second center frequency. The old channel type is of course still supported for HT. Also change the cfg80211 channel definition struct to support these by adding the relevant fields to it (and removing the _type field.) This also adds new helper functions: - cfg80211_chandef_create to create a channel def struct given the control channel and channel type, - cfg80211_chandef_identical to check if two channel definitions are identical - cfg80211_chandef_compatible to check if the given channel definitions are compatible, and return the wider of the two This isn't entirely complete, but that doesn't matter until we have a driver using it. In particular, it's missing - regulatory checks on the usable bandwidth (if that even makes sense) - regulatory TX power (database can't deal with it) - a proper channel compatibility calculation for the new channel types Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 22:14:50 +00:00
if ((ibss.chandef.width != NL80211_CHAN_WIDTH_20_NOHT) &&
no_ht) {
nl80211: Zero out the connection keys memory when freeing them. The connection keys are zeroed out in all other cases except this one. Let's fix the last one as well. Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com> Reviewed-by: Julian Calaby <julian.calaby@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-02-11 00:00:22 +00:00
kzfree(connkeys);
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
return -EINVAL;
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
}
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
cfg80211/mac80211: userspace peer authorization in IBSS If the IBSS network is RSN-protected, let userspace authorize the stations instead of adding them as AUTHORIZED by default. Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-01-31 19:25:47 +00:00
ibss.control_port =
nla_get_flag(info->attrs[NL80211_ATTR_CONTROL_PORT]);
nl80211/cfg80211: enable DFS for IBSS mode To use DFS in IBSS mode, userspace is required to react to radar events. It can inform nl80211 that it is capable of doing so by adding a NL80211_ATTR_HANDLE_DFS attribute when joining the IBSS. This attribute is supplied to let the kernelspace know that the userspace application can and will handle radar events, e.g. by intiating channel switches to a valid channel. DFS channels may only be used if this attribute is supplied and the driver supports it. Driver support will be checked even if a channel without DFS will be initially joined, as a DFS channel may be chosen later. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> [fix attribute name in commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-07 16:41:05 +00:00
ibss.userspace_handles_dfs =
nla_get_flag(info->attrs[NL80211_ATTR_HANDLE_DFS]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
err = cfg80211_join_ibss(rdev, dev, &ibss, connkeys);
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (err)
cfg80211: clear connect keys when freeing them When freeing the connect keys, clear the memory to avoid having the key material stick around in memory "forever". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 10:39:54 +00:00
kzfree(connkeys);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
return err;
}
static int nl80211_leave_ibss(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->leave_ibss)
return -EOPNOTSUPP;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC)
return -EOPNOTSUPP;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return cfg80211_leave_ibss(rdev, dev, false);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
}
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
static int nl80211_set_mcast_rate(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
int mcast_rate[NUM_NL80211_BANDS];
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
u32 nla_rate;
int err;
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_ADHOC &&
nl80211: Allow setting multicast rate on OCB interfaces Allow setting multicast rate on OCB interfaces. Current behaviour results in EOPNOTSUPP when attempting this. Signed-off-by: Bertold Van den Bergh <bertold.vandenbergh@esat.kuleuven.be> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-05 14:02:21 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_MESH_POINT &&
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_OCB)
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
return -EOPNOTSUPP;
if (!rdev->ops->set_mcast_rate)
return -EOPNOTSUPP;
memset(mcast_rate, 0, sizeof(mcast_rate));
if (!info->attrs[NL80211_ATTR_MCAST_RATE])
return -EINVAL;
nla_rate = nla_get_u32(info->attrs[NL80211_ATTR_MCAST_RATE]);
if (!nl80211_parse_mcast_rate(rdev, mcast_rate, nla_rate))
return -EINVAL;
cfg80211: Add missing tracing to cfg80211 Add missing tracing for: 1. start_radar_detection() 2. set_mcast_rates() 3. set_coalesce() Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-22 19:27:46 +00:00
err = rdev_set_mcast_rate(rdev, dev, mcast_rate);
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
return err;
}
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
static struct sk_buff *
__cfg80211_alloc_vendor_skb(struct cfg80211_registered_device *rdev,
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
struct wireless_dev *wdev, int approxlen,
u32 portid, u32 seq, enum nl80211_commands cmd,
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
enum nl80211_attrs attr,
const struct nl80211_vendor_cmd_info *info,
gfp_t gfp)
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
{
struct sk_buff *skb;
void *hdr;
struct nlattr *data;
skb = nlmsg_new(approxlen + 100, gfp);
if (!skb)
return NULL;
hdr = nl80211hdr_put(skb, portid, seq, 0, cmd);
if (!hdr) {
kfree_skb(skb);
return NULL;
}
if (nla_put_u32(skb, NL80211_ATTR_WIPHY, rdev->wiphy_idx))
goto nla_put_failure;
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
if (info) {
if (nla_put_u32(skb, NL80211_ATTR_VENDOR_ID,
info->vendor_id))
goto nla_put_failure;
if (nla_put_u32(skb, NL80211_ATTR_VENDOR_SUBCMD,
info->subcmd))
goto nla_put_failure;
}
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
if (wdev) {
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(skb, NL80211_ATTR_WDEV,
wdev_id(wdev), NL80211_ATTR_PAD))
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
goto nla_put_failure;
if (wdev->netdev &&
nla_put_u32(skb, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex))
goto nla_put_failure;
}
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
data = nla_nest_start(skb, attr);
nl80211: always check nla_nest_start() return value If the message got full during nla_nest_start(), it can return NULL. None of the cases here seem like that can really happen, but check the return value nonetheless. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:55:57 +00:00
if (!data)
goto nla_put_failure;
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
((void **)skb->cb)[0] = rdev;
((void **)skb->cb)[1] = hdr;
((void **)skb->cb)[2] = data;
return skb;
nla_put_failure:
kfree_skb(skb);
return NULL;
}
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
nl80211: move vendor/testmode event skb functions out of ifdef The vendor/testmode event skb functions are needed outside the ifdef for vendor-specific events, so move them out. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-01 16:22:30 +00:00
struct sk_buff *__cfg80211_alloc_event_skb(struct wiphy *wiphy,
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
struct wireless_dev *wdev,
nl80211: move vendor/testmode event skb functions out of ifdef The vendor/testmode event skb functions are needed outside the ifdef for vendor-specific events, so move them out. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-01 16:22:30 +00:00
enum nl80211_commands cmd,
enum nl80211_attrs attr,
int vendor_event_idx,
int approxlen, gfp_t gfp)
{
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
nl80211: move vendor/testmode event skb functions out of ifdef The vendor/testmode event skb functions are needed outside the ifdef for vendor-specific events, so move them out. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-01 16:22:30 +00:00
const struct nl80211_vendor_cmd_info *info;
switch (cmd) {
case NL80211_CMD_TESTMODE:
if (WARN_ON(vendor_event_idx != -1))
return NULL;
info = NULL;
break;
case NL80211_CMD_VENDOR:
if (WARN_ON(vendor_event_idx < 0 ||
vendor_event_idx >= wiphy->n_vendor_events))
return NULL;
info = &wiphy->vendor_events[vendor_event_idx];
break;
default:
WARN_ON(1);
return NULL;
}
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
return __cfg80211_alloc_vendor_skb(rdev, wdev, approxlen, 0, 0,
nl80211: move vendor/testmode event skb functions out of ifdef The vendor/testmode event skb functions are needed outside the ifdef for vendor-specific events, so move them out. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-01 16:22:30 +00:00
cmd, attr, info, gfp);
}
EXPORT_SYMBOL(__cfg80211_alloc_event_skb);
void __cfg80211_send_event_skb(struct sk_buff *skb, gfp_t gfp)
{
struct cfg80211_registered_device *rdev = ((void **)skb->cb)[0];
void *hdr = ((void **)skb->cb)[1];
struct nlattr *data = ((void **)skb->cb)[2];
enum nl80211_multicast_groups mcgrp = NL80211_MCGRP_TESTMODE;
nl80211: clear skb cb before passing to netlink In testmode and vendor command reply/event SKBs we use the skb cb data to store nl80211 parameters between allocation and sending. This causes the code for CONFIG_NETLINK_MMAP to get confused, because it takes ownership of the skb cb data when the SKB is handed off to netlink, and it doesn't explicitly clear it. Clear the skb cb explicitly when we're done and before it gets passed to netlink to avoid this issue. Cc: stable@vger.kernel.org [this goes way back] Reported-by: Assaf Azulay <assaf.azulay@intel.com> Reported-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-07-30 12:55:26 +00:00
/* clear CB data for netlink core to own from now on */
memset(skb->cb, 0, sizeof(skb->cb));
nl80211: move vendor/testmode event skb functions out of ifdef The vendor/testmode event skb functions are needed outside the ifdef for vendor-specific events, so move them out. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-01 16:22:30 +00:00
nla_nest_end(skb, data);
genlmsg_end(skb, hdr);
if (data->nla_type == NL80211_ATTR_VENDOR_DATA)
mcgrp = NL80211_MCGRP_VENDOR;
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), skb, 0,
mcgrp, gfp);
}
EXPORT_SYMBOL(__cfg80211_send_event_skb);
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
#ifdef CONFIG_NL80211_TESTMODE
static int nl80211_testmode_do(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: add wdev to testmode cmd To allow drivers to implement per-interface testmode operations more easily, pass a wdev pointer if any identification for one was given from userspace. Clean up the code a bit while at it. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-31 15:04:15 +00:00
struct wireless_dev *wdev =
__cfg80211_wdev_from_attrs(genl_info_net(info), info->attrs);
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
int err;
cfg80211: add wdev to testmode cmd To allow drivers to implement per-interface testmode operations more easily, pass a wdev pointer if any identification for one was given from userspace. Clean up the code a bit while at it. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-31 15:04:15 +00:00
if (!rdev->ops->testmode_cmd)
return -EOPNOTSUPP;
if (IS_ERR(wdev)) {
err = PTR_ERR(wdev);
if (err != -EINVAL)
return err;
wdev = NULL;
} else if (wdev->wiphy != &rdev->wiphy) {
return -EINVAL;
}
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
if (!info->attrs[NL80211_ATTR_TESTDATA])
return -EINVAL;
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
rdev->cur_cmd_info = info;
cfg80211: add wdev to testmode cmd To allow drivers to implement per-interface testmode operations more easily, pass a wdev pointer if any identification for one was given from userspace. Clean up the code a bit while at it. Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-31 15:04:15 +00:00
err = rdev_testmode_cmd(rdev, wdev,
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
nla_data(info->attrs[NL80211_ATTR_TESTDATA]),
nla_len(info->attrs[NL80211_ATTR_TESTDATA]));
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
rdev->cur_cmd_info = NULL;
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
return err;
}
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
static int nl80211_testmode_dump(struct sk_buff *skb,
struct netlink_callback *cb)
{
nl80211: accept testmode dump with netdev All nl80211 commands that need only the wiphy still allow identifying it by giving an interface index, except, as Kenny pointed out, the testmode dump support. Fix this by looking up the wiphy via the ifidx in this case as well. Tested-by: Kenny Hsu <kenny.hsu@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-13 16:22:05 +00:00
struct cfg80211_registered_device *rdev;
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
int err;
long phy_idx;
void *data = NULL;
int data_len = 0;
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
rtnl_lock();
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
if (cb->args[0]) {
/*
* 0 is a valid index, but not valid for args[0],
* so we need to offset by 1.
*/
phy_idx = cb->args[0] - 1;
cfg80211: make rdev assignment clearer in nl80211_testmode_dump() Avoid assigning rdev to NULL when we already have it and getting it again from the wiphy index, by moving this code to relevant if block. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-07 20:13:56 +00:00
rdev = cfg80211_rdev_by_wiphy_idx(phy_idx);
if (!rdev) {
err = -ENOENT;
goto out_err;
}
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
} else {
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
struct nlattr **attrbuf = genl_family_attrbuf(&nl80211_fam);
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
err = nlmsg_parse(cb->nlh, GENL_HDRLEN + nl80211_fam.hdrsize,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
attrbuf, nl80211_fam.maxattr,
nl80211_policy, NULL);
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
if (err)
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
goto out_err;
nl80211: accept testmode dump with netdev All nl80211 commands that need only the wiphy still allow identifying it by giving an interface index, except, as Kenny pointed out, the testmode dump support. Fix this by looking up the wiphy via the ifidx in this case as well. Tested-by: Kenny Hsu <kenny.hsu@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-13 16:22:05 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
rdev = __cfg80211_rdev_from_attrs(sock_net(skb->sk), attrbuf);
nl80211: use __cfg80211_rdev_from_attrs for testmode To fix the testmode cross-namespace access problem, use __cfg80211_rdev_from_attrs() to get the device instead of open-coding similar functionality. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:23:16 +00:00
if (IS_ERR(rdev)) {
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
err = PTR_ERR(rdev);
goto out_err;
nl80211: accept testmode dump with netdev All nl80211 commands that need only the wiphy still allow identifying it by giving an interface index, except, as Kenny pointed out, the testmode dump support. Fix this by looking up the wiphy via the ifidx in this case as well. Tested-by: Kenny Hsu <kenny.hsu@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-13 16:22:05 +00:00
}
nl80211: use __cfg80211_rdev_from_attrs for testmode To fix the testmode cross-namespace access problem, use __cfg80211_rdev_from_attrs() to get the device instead of open-coding similar functionality. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:23:16 +00:00
phy_idx = rdev->wiphy_idx;
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
if (attrbuf[NL80211_ATTR_TESTDATA])
cb->args[1] = (long)attrbuf[NL80211_ATTR_TESTDATA];
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
}
if (cb->args[1]) {
data = nla_data((void *)cb->args[1]);
data_len = nla_len((void *)cb->args[1]);
}
nl80211: accept testmode dump with netdev All nl80211 commands that need only the wiphy still allow identifying it by giving an interface index, except, as Kenny pointed out, the testmode dump support. Fix this by looking up the wiphy via the ifidx in this case as well. Tested-by: Kenny Hsu <kenny.hsu@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-12-13 16:22:05 +00:00
if (!rdev->ops->testmode_dump) {
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
err = -EOPNOTSUPP;
goto out_err;
}
while (1) {
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
void *hdr = nl80211hdr_put(skb, NETLINK_CB(cb->skb).portid,
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
cb->nlh->nlmsg_seq, NLM_F_MULTI,
NL80211_CMD_TESTMODE);
struct nlattr *tmdata;
nl80211: nl80211hdr_put() doesn't return an ERR_PTR There are a few places which check nl80211hdr_put() for an ERR_PTR but actually it returns NULL on error and never error values. In nl80211_testmode_dump() the return wasn't checked at all so I have added one. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> [some whitespace changes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-14 11:50:01 +00:00
if (!hdr)
break;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(skb, NL80211_ATTR_WIPHY, phy_idx)) {
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
genlmsg_cancel(skb, hdr);
break;
}
tmdata = nla_nest_start(skb, NL80211_ATTR_TESTDATA);
if (!tmdata) {
genlmsg_cancel(skb, hdr);
break;
}
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_testmode_dump(rdev, skb, cb, data, data_len);
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
nla_nest_end(skb, tmdata);
if (err == -ENOBUFS || err == -ENOENT) {
genlmsg_cancel(skb, hdr);
break;
} else if (err) {
genlmsg_cancel(skb, hdr);
goto out_err;
}
genlmsg_end(skb, hdr);
}
err = skb->len;
/* see above */
cb->args[0] = phy_idx + 1;
out_err:
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
rtnl_unlock();
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
return err;
}
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
#endif
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
static int nl80211_connect(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
struct cfg80211_connect_params connect;
struct wiphy *wiphy;
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
struct cfg80211_cached_keys *connkeys = NULL;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
int err;
memset(&connect, 0, sizeof(connect));
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
if (!info->attrs[NL80211_ATTR_SSID] ||
!nla_len(info->attrs[NL80211_ATTR_SSID]))
return -EINVAL;
if (info->attrs[NL80211_ATTR_AUTH_TYPE]) {
connect.auth_type =
nla_get_u32(info->attrs[NL80211_ATTR_AUTH_TYPE]);
cfg80211: Allow user space to specify non-IEs to SAE Authentication SAE extends Authentication frames with fields that are not information elements. NL80211_ATTR_IE is not suitable for these, so introduce a new attribute that can be used to specify the fields needed for SAE in station mode. Signed-off-by: Jouni Malinen <j@w1.fi> [change to verify that SAE is only used with authenticate command] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-30 16:29:39 +00:00
if (!nl80211_valid_auth_type(rdev, connect.auth_type,
NL80211_CMD_CONNECT))
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return -EINVAL;
} else
connect.auth_type = NL80211_AUTHTYPE_AUTOMATIC;
connect.privacy = info->attrs[NL80211_ATTR_PRIVACY];
cfg80211: support 4-way handshake offloading for 802.1X Add API for setting the PMK to the driver. For FT support, allow setting also the PMK-R0 Name. This can be used by drivers that support 4-Way handshake offload while IEEE802.1X authentication is managed by upper layers. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> [arend.vanspriel@broadcom.com: add WANT_1X_4WAY_HS attribute] Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [reword NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X docs a bit to say that the device may require it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-06-09 12:08:43 +00:00
if (info->attrs[NL80211_ATTR_WANT_1X_4WAY_HS] &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X))
return -EINVAL;
connect.want_1x = info->attrs[NL80211_ATTR_WANT_1X_4WAY_HS];
cfg80211: allow changing port control protocol Some vendor specified mechanisms for 802.1X-style functionality use a different protocol than EAP (even if EAP is vendor-extensible). Allow setting the ethertype for the protocol when a driver has support for this. The default if unspecified is EAP, of course. Note: This is suitable only for station mode, not for AP implementation. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-27 11:26:53 +00:00
err = nl80211_crypto_settings(rdev, info, &connect.crypto,
nl80211: limit to one pairwise cipher for associate() In this case, only one cipher makes sense, unlike for connect() where it may be possible to have the card or driver select. No changes to mac80211 due to the way the structs are laid out -- but the loop in net/mac80211/cfg.c will degrade to just zero or one passes. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-02 19:36:37 +00:00
NL80211_MAX_NR_CIPHER_SUITES);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (err)
return err;
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: clean up naming once and for all We've named the registered devices 'drv' sometimes, thinking of "driver", which is not what it is, it's the internal representation of a wiphy, i.e. a device. Let's clean up the naming once and and use 'rdev' aka 'registered device' everywhere. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:12 +00:00
wiphy = &rdev->wiphy;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: Add background scan period attribute. Receive background scan period as part of connect command and pass the same to the driver. Signed-off-by: Bala Shanmugam <bkamatch@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-07 11:57:12 +00:00
connect.bg_scan_period = -1;
if (info->attrs[NL80211_ATTR_BG_SCAN_PERIOD] &&
(wiphy->flags & WIPHY_FLAG_SUPPORTS_FW_ROAM)) {
connect.bg_scan_period =
nla_get_u16(info->attrs[NL80211_ATTR_BG_SCAN_PERIOD]);
}
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (info->attrs[NL80211_ATTR_MAC])
connect.bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: Allow BSS hint to be provided for connect This clarifies the expected driver behavior on the older NL80211_ATTR_MAC and NL80211_ATTR_WIPHY_FREQ attributes and adds a new set of similar attributes with _HINT postfix to enable use of a recommendation of the initial BSS to choose. This can be helpful for some drivers that can avoid an additional full scan on connection request if the information is provided to them (user space tools like wpa_supplicant already has that information available based on earlier scans). In addition, this can be used to get more expected behavior for cases where a specific BSS should be picked first based on operations like Interworking network selection or WPS. These cases were already easily addressed with drivers that leave BSS selection to user space, but there was no convenient way to do this with drivers that take care of BSS selection internally without using the NL80211_ATTR_MAC which is not really desired since it is needed for other purposes to force the association to remain with the same BSS. Signed-off-by: Jouni Malinen <j@w1.fi> [add const, fix policy] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:00:47 +00:00
else if (info->attrs[NL80211_ATTR_MAC_HINT])
connect.bssid_hint =
nla_data(info->attrs[NL80211_ATTR_MAC_HINT]);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
connect.ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
connect.ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]);
if (info->attrs[NL80211_ATTR_IE]) {
connect.ie = nla_data(info->attrs[NL80211_ATTR_IE]);
connect.ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
}
cfg80211: Allow use_mfp to be specified with the connect command The NL80211_ATTR_USE_MFP attribute was originally added for NL80211_CMD_ASSOCIATE, but it is actually as useful (if not even more useful) with NL80211_CMD_CONNECT, so process that attribute with the connect command, too. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-15 15:15:57 +00:00
if (info->attrs[NL80211_ATTR_USE_MFP]) {
connect.mfp = nla_get_u32(info->attrs[NL80211_ATTR_USE_MFP]);
if (connect.mfp != NL80211_MFP_REQUIRED &&
connect.mfp != NL80211_MFP_NO)
return -EINVAL;
} else {
connect.mfp = NL80211_MFP_NO;
}
cfg80211: Add option to specify previous BSSID for Connect command This extends NL80211_CMD_CONNECT to allow the NL80211_ATTR_PREV_BSSID attribute to be used similarly to way this was already allowed with NL80211_CMD_ASSOCIATE. This allows user space to request reassociation (instead of association) when already connected to an AP. This provides an option to reassociate within an ESS without having to disconnect and associate with the AP. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-29 10:53:27 +00:00
if (info->attrs[NL80211_ATTR_PREV_BSSID])
connect.prev_bssid =
nla_data(info->attrs[NL80211_ATTR_PREV_BSSID]);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) {
cfg80211: Clean up connect params and channel fetching Addition of the frequency hints showed up couple of places in cfg80211 where pointers could be marked const and a shared function could be used to fetch a valid channel. Signed-off-by: Jouni Malinen <j@w1.fi> [fix mwifiex] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:44 +00:00
connect.channel = nl80211_get_valid_chan(
wiphy, info->attrs[NL80211_ATTR_WIPHY_FREQ]);
if (!connect.channel)
cfg80211: Allow BSS hint to be provided for connect This clarifies the expected driver behavior on the older NL80211_ATTR_MAC and NL80211_ATTR_WIPHY_FREQ attributes and adds a new set of similar attributes with _HINT postfix to enable use of a recommendation of the initial BSS to choose. This can be helpful for some drivers that can avoid an additional full scan on connection request if the information is provided to them (user space tools like wpa_supplicant already has that information available based on earlier scans). In addition, this can be used to get more expected behavior for cases where a specific BSS should be picked first based on operations like Interworking network selection or WPS. These cases were already easily addressed with drivers that leave BSS selection to user space, but there was no convenient way to do this with drivers that take care of BSS selection internally without using the NL80211_ATTR_MAC which is not really desired since it is needed for other purposes to force the association to remain with the same BSS. Signed-off-by: Jouni Malinen <j@w1.fi> [add const, fix policy] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:00:47 +00:00
return -EINVAL;
} else if (info->attrs[NL80211_ATTR_WIPHY_FREQ_HINT]) {
cfg80211: Clean up connect params and channel fetching Addition of the frequency hints showed up couple of places in cfg80211 where pointers could be marked const and a shared function could be used to fetch a valid channel. Signed-off-by: Jouni Malinen <j@w1.fi> [fix mwifiex] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-14 22:01:44 +00:00
connect.channel_hint = nl80211_get_valid_chan(
wiphy, info->attrs[NL80211_ATTR_WIPHY_FREQ_HINT]);
if (!connect.channel_hint)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EINVAL;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
}
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (connect.privacy && info->attrs[NL80211_ATTR_KEYS]) {
connkeys = nl80211_parse_connkeys(rdev,
cfg80211: Disallow HT/WEP in IBSS mode Currently, a user is allowed to choose a HT operating channel with WEP when creating an IBSS network. WEP is not allowed in HT configuration - this patch ensures that such requests are denied. Signed-off-by: Sujith Manoharan <c_manoha@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-18 04:49:28 +00:00
info->attrs[NL80211_ATTR_KEYS], NULL);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (IS_ERR(connkeys))
return PTR_ERR(connkeys);
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
}
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_DISABLE_HT]))
connect.flags |= ASSOC_REQ_DISABLE_HT;
if (info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK])
memcpy(&connect.ht_capa_mask,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK]),
sizeof(connect.ht_capa_mask));
if (info->attrs[NL80211_ATTR_HT_CAPABILITY]) {
nl80211: fix possible memory leak nl80211_connect() connkeys is malloced in nl80211_parse_connkeys() and should be freed in the error handling case, otherwise it will cause memory leak. spatch with a semantic match is used to found this problem. (http://coccinelle.lip6.fr/) Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-02 13:41:04 +00:00
if (!info->attrs[NL80211_ATTR_HT_CAPABILITY_MASK]) {
cfg80211: clear connect keys when freeing them When freeing the connect keys, clear the memory to avoid having the key material stick around in memory "forever". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 10:39:54 +00:00
kzfree(connkeys);
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
return -EINVAL;
nl80211: fix possible memory leak nl80211_connect() connkeys is malloced in nl80211_parse_connkeys() and should be freed in the error handling case, otherwise it will cause memory leak. spatch with a semantic match is used to found this problem. (http://coccinelle.lip6.fr/) Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-02 13:41:04 +00:00
}
wireless: Support ht-capabilities over-rides. This allows users to disable features such as HT, HT40, and to modify the MCS, AMPDU, and AMSDU settings for drivers that support it. The MCS, AMPDU, and AMSDU features that may be disabled are are reported in the phy-info netlink message as a mask. Attemping to disable features that are not supported will take no affect, but will not return errors. This is to aid backwards compatibility in user-space apps that may not be clever enough to deal with parsing the the capabilities mask. This patch only enables the infrastructure. An additional patch will enable the feature in mac80211. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 19:31:59 +00:00
memcpy(&connect.ht_capa,
nla_data(info->attrs[NL80211_ATTR_HT_CAPABILITY]),
sizeof(connect.ht_capa));
}
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_DISABLE_VHT]))
connect.flags |= ASSOC_REQ_DISABLE_VHT;
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK])
memcpy(&connect.vht_capa_mask,
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK]),
sizeof(connect.vht_capa_mask));
if (info->attrs[NL80211_ATTR_VHT_CAPABILITY]) {
if (!info->attrs[NL80211_ATTR_VHT_CAPABILITY_MASK]) {
cfg80211: clear connect keys when freeing them When freeing the connect keys, clear the memory to avoid having the key material stick around in memory "forever". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 10:39:54 +00:00
kzfree(connkeys);
cfg80211: add ability to override VHT capabilities For testing it's sometimes useful to be able to override certain VHT capability advertisement, add the ability to do that in cfg80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-21 16:36:01 +00:00
return -EINVAL;
}
memcpy(&connect.vht_capa,
nla_data(info->attrs[NL80211_ATTR_VHT_CAPABILITY]),
sizeof(connect.vht_capa));
}
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
if (nla_get_flag(info->attrs[NL80211_ATTR_USE_RRM])) {
cfg80211: Add global RRM capability Today, the supplicant will add the RRM capabilities Information Element in the association request only if Quiet period is supported (NL80211_FEATURE_QUIET). Quiet is one of many RRM features, and there are other RRM features that are not related to Quiet (e.g. neighbor report). Therefore, requiring Quiet to enable RRM is too restrictive. Some of the features, like neighbor report, can be supported by user space without any help from the kernel. Hence adding the RRM capabilities IE to association request should be the sole user space's decision. Removing the RRM dependency on Quiet in the driver solves this problem, but using an old driver with a user space tool that would not require Quiet feature would be problematic: the user space would add NL80211_ATTR_USE_RRM in the association request even if the kernel doesn't advertize NL80211_FEATURE_QUIET and the association would be denied by the kernel. This solution adds a global RRM capability, that tells user space that it can request RRM capabilities IE publishment without any specific feature support in the kernel. Signed-off-by: Beni Lev <beni.lev@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-02-17 18:30:00 +00:00
if (!((rdev->wiphy.features &
NL80211_FEATURE_DS_PARAM_SET_IE_IN_PROBES) &&
(rdev->wiphy.features & NL80211_FEATURE_QUIET)) &&
!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_RRM)) {
nl80211: Fix potential memory leak in nl80211_connect Free cached keys if the last early return path is taken. Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-12-11 20:04:52 +00:00
kzfree(connkeys);
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
return -EINVAL;
nl80211: Fix potential memory leak in nl80211_connect Free cached keys if the last early return path is taken. Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-12-11 20:04:52 +00:00
}
nl80211: Add flag attribute for RRM connections Add a flag attribute to use in associations, for tagging the target connection as supporting RRM. It is the responsibility of upper layers to set this flag only if both the underlying device, and the target network indeed support RRM. To be used in ASSOCIATE and CONNECT commands. Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-03 12:25:01 +00:00
connect.flags |= ASSOC_REQ_USE_RRM;
}
cfg80211: basic support for PBSS network type PBSS (Personal Basic Service Set) is a new BSS type for DMG networks. It is similar to infrastructure BSS, having an AP-like entity called PCP (PBSS Control Point), but it has few differences. PBSS support is mandatory for 11ad devices. Add support for PBSS by introducing a new PBSS flag attribute. The PBSS flag is used in the START_AP command to request starting a PCP instead of an AP, and in the CONNECT command to request connecting to a PCP instead of an AP. Signed-off-by: Lior David <liord@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-01-28 08:58:25 +00:00
connect.pbss = nla_get_flag(info->attrs[NL80211_ATTR_PBSS]);
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
if (connect.pbss && !rdev->wiphy.bands[NL80211_BAND_60GHZ]) {
cfg80211: basic support for PBSS network type PBSS (Personal Basic Service Set) is a new BSS type for DMG networks. It is similar to infrastructure BSS, having an AP-like entity called PCP (PBSS Control Point), but it has few differences. PBSS support is mandatory for 11ad devices. Add support for PBSS by introducing a new PBSS flag attribute. The PBSS flag is used in the START_AP command to request starting a PCP instead of an AP, and in the CONNECT command to request connecting to a PCP instead of an AP. Signed-off-by: Lior David <liord@codeaurora.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-01-28 08:58:25 +00:00
kzfree(connkeys);
return -EOPNOTSUPP;
}
nl80211: add feature for BSS selection support Introducing a new feature that the driver can use to indicate the driver/firmware supports configuration of BSS selection criteria upon CONNECT command. This can be useful when multiple BSS-es are found belonging to the same ESS, ie. Infra-BSS with same SSID. The criteria can then be used to offload selection of a preferred BSS. Reviewed-by: Hante Meuleman <meuleman@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Lei Zhang <leizh@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> [move wiphy support check into parse_bss_select()] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-02 19:37:18 +00:00
if (info->attrs[NL80211_ATTR_BSS_SELECT]) {
/* bss selection makes no sense if bssid is set */
if (connect.bssid) {
kzfree(connkeys);
return -EINVAL;
}
err = parse_bss_select(info->attrs[NL80211_ATTR_BSS_SELECT],
wiphy, &connect.bss_select);
if (err) {
kzfree(connkeys);
return err;
}
}
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
if (wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_FILS_SK_OFFLOAD) &&
info->attrs[NL80211_ATTR_FILS_ERP_USERNAME] &&
info->attrs[NL80211_ATTR_FILS_ERP_REALM] &&
info->attrs[NL80211_ATTR_FILS_ERP_NEXT_SEQ_NUM] &&
info->attrs[NL80211_ATTR_FILS_ERP_RRK]) {
connect.fils_erp_username =
nla_data(info->attrs[NL80211_ATTR_FILS_ERP_USERNAME]);
connect.fils_erp_username_len =
nla_len(info->attrs[NL80211_ATTR_FILS_ERP_USERNAME]);
connect.fils_erp_realm =
nla_data(info->attrs[NL80211_ATTR_FILS_ERP_REALM]);
connect.fils_erp_realm_len =
nla_len(info->attrs[NL80211_ATTR_FILS_ERP_REALM]);
connect.fils_erp_next_seq_num =
nla_get_u16(
info->attrs[NL80211_ATTR_FILS_ERP_NEXT_SEQ_NUM]);
connect.fils_erp_rrk =
nla_data(info->attrs[NL80211_ATTR_FILS_ERP_RRK]);
connect.fils_erp_rrk_len =
nla_len(info->attrs[NL80211_ATTR_FILS_ERP_RRK]);
} else if (info->attrs[NL80211_ATTR_FILS_ERP_USERNAME] ||
info->attrs[NL80211_ATTR_FILS_ERP_REALM] ||
info->attrs[NL80211_ATTR_FILS_ERP_NEXT_SEQ_NUM] ||
info->attrs[NL80211_ATTR_FILS_ERP_RRK]) {
kzfree(connkeys);
return -EINVAL;
}
cfg80211: remove some locked wrappers from sme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_lock(dev->ieee80211_ptr);
cfg80211: NL80211_ATTR_SOCKET_OWNER support for CMD_CONNECT Disconnect or deauthenticate when the owning socket is closed if this flag is supplied to CMD_CONNECT or CMD_ASSOCIATE. This may be used to ensure userspace daemon doesn't leave an unmanaged connection behind. In some situations it would be possible to account for that, to some degree, in the deamon restart code or in the up/down scripts without the use of this attribute. But there will be systems where the daemon can go away for varying periods without a warning due to local resource management. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-06 21:33:43 +00:00
cfg80211: Allow reassociation to be requested with internal SME If the user space issues a NL80211_CMD_CONNECT with NL80211_ATTR_PREV_BSSID when there is already a connection, allow this to proceed as a reassociation instead of rejecting the new connect command with EALREADY. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [validate prev_bssid] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-03-29 10:53:28 +00:00
err = cfg80211_connect(rdev, dev, &connect, connkeys,
connect.prev_bssid);
cfg80211: rework key operation This reworks the key operation in cfg80211, and now only allows, from userspace, configuring keys (via nl80211) after the connection has been established (in managed mode), the IBSS been joined (in IBSS mode), at any time (in AP[_VLAN] modes) or never for all the other modes. In order to do shared key authentication correctly, it is now possible to give a WEP key to the AUTH command. To configure static WEP keys, these are given to the CONNECT or IBSS_JOIN command directly, for a userspace SME it is assumed it will configure it properly after the connection has been established. Since mac80211 used to check the default key in IBSS mode to see whether or not the network is protected, it needs an update in that area, as well as an update to make use of the WEP key passed to auth() for shared key authentication. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-08 12:22:54 +00:00
if (err)
cfg80211: clear connect keys when freeing them When freeing the connect keys, clear the memory to avoid having the key material stick around in memory "forever". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 10:39:54 +00:00
kzfree(connkeys);
cfg80211: NL80211_ATTR_SOCKET_OWNER support for CMD_CONNECT Disconnect or deauthenticate when the owning socket is closed if this flag is supplied to CMD_CONNECT or CMD_ASSOCIATE. This may be used to ensure userspace daemon doesn't leave an unmanaged connection behind. In some situations it would be possible to account for that, to some degree, in the deamon restart code or in the up/down scripts without the use of this attribute. But there will be systems where the daemon can go away for varying periods without a warning due to local resource management. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-06 21:33:43 +00:00
if (!err && info->attrs[NL80211_ATTR_SOCKET_OWNER]) {
dev->ieee80211_ptr->conn_owner_nlportid = info->snd_portid;
if (connect.bssid)
memcpy(dev->ieee80211_ptr->disconnect_bssid,
connect.bssid, ETH_ALEN);
else
memset(dev->ieee80211_ptr->disconnect_bssid,
0, ETH_ALEN);
}
wdev_unlock(dev->ieee80211_ptr);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return err;
}
cfg80211: Add support to update connection parameters Add functionality to update the connection parameters when in connected state, so that driver/firmware uses the updated parameters for subsequent roaming. This is for drivers that support internal BSS selection and roaming. The new command does not change the current association state, i.e., it can be used to update IE contents for future (re)associations without causing an immediate disassociation or reassociation with the current BSS. This commit implements the required functionality for updating IEs for (Re)Association Request frame only. Other parameters can be added in future when required. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-27 13:51:11 +00:00
static int nl80211_update_connect_params(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_connect_params connect = {};
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
u32 changed = 0;
int ret;
if (!rdev->ops->update_connect_params)
return -EOPNOTSUPP;
if (info->attrs[NL80211_ATTR_IE]) {
if (!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
connect.ie = nla_data(info->attrs[NL80211_ATTR_IE]);
connect.ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
changed |= UPDATE_ASSOC_IES;
}
wdev_lock(dev->ieee80211_ptr);
if (!wdev->current_bss)
ret = -ENOLINK;
else
ret = rdev_update_connect_params(rdev, dev, &connect, changed);
wdev_unlock(dev->ieee80211_ptr);
return ret;
}
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
static int nl80211_disconnect(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
u16 reason;
cfg80211: remove some locked wrappers from sme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
int ret;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (!info->attrs[NL80211_ATTR_REASON_CODE])
reason = WLAN_REASON_DEAUTH_LEAVING;
else
reason = nla_get_u16(info->attrs[NL80211_ATTR_REASON_CODE]);
if (reason == 0)
return -EINVAL;
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: remove some locked wrappers from sme API By making all the API functions require wdev locking we can clean up the API a bit, getting rid of the locking version of each function. This also decreases the size of cfg80211 by a small amount. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 15:44:01 +00:00
wdev_lock(dev->ieee80211_ptr);
ret = cfg80211_disconnect(rdev, dev, reason, true);
wdev_unlock(dev->ieee80211_ptr);
return ret;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
}
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
static int nl80211_wiphy_netns(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
struct net *net;
int err;
nl80211: Allow set network namespace by fd Added new NL80211_ATTR_NETNS_FD which allows to set namespace via nl80211 by fd. Signed-off-by: Vadim Kochan <vadim4j@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-12 14:34:05 +00:00
if (info->attrs[NL80211_ATTR_PID]) {
u32 pid = nla_get_u32(info->attrs[NL80211_ATTR_PID]);
net = get_net_ns_by_pid(pid);
} else if (info->attrs[NL80211_ATTR_NETNS_FD]) {
u32 fd = nla_get_u32(info->attrs[NL80211_ATTR_NETNS_FD]);
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
nl80211: Allow set network namespace by fd Added new NL80211_ATTR_NETNS_FD which allows to set namespace via nl80211 by fd. Signed-off-by: Vadim Kochan <vadim4j@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-12 14:34:05 +00:00
net = get_net_ns_by_fd(fd);
} else {
return -EINVAL;
}
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (IS_ERR(net))
return PTR_ERR(net);
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
err = 0;
/* check if anything to do */
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!net_eq(wiphy_net(&rdev->wiphy), net))
err = cfg80211_switch_netns(rdev, net);
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
put_net(net);
return err;
}
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
static int nl80211_setdel_pmksa(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
int (*rdev_ops)(struct wiphy *wiphy, struct net_device *dev,
struct cfg80211_pmksa *pmksa) = NULL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
struct cfg80211_pmksa pmksa;
memset(&pmksa, 0, sizeof(struct cfg80211_pmksa));
if (!info->attrs[NL80211_ATTR_PMKID])
return -EINVAL;
pmksa.pmkid = nla_data(info->attrs[NL80211_ATTR_PMKID]);
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
if (info->attrs[NL80211_ATTR_MAC]) {
pmksa.bssid = nla_data(info->attrs[NL80211_ATTR_MAC]);
} else if (info->attrs[NL80211_ATTR_SSID] &&
info->attrs[NL80211_ATTR_FILS_CACHE_ID] &&
(info->genlhdr->cmd == NL80211_CMD_DEL_PMKSA ||
info->attrs[NL80211_ATTR_PMK])) {
pmksa.ssid = nla_data(info->attrs[NL80211_ATTR_SSID]);
pmksa.ssid_len = nla_len(info->attrs[NL80211_ATTR_SSID]);
pmksa.cache_id =
nla_data(info->attrs[NL80211_ATTR_FILS_CACHE_ID]);
} else {
return -EINVAL;
}
if (info->attrs[NL80211_ATTR_PMK]) {
pmksa.pmk = nla_data(info->attrs[NL80211_ATTR_PMK]);
pmksa.pmk_len = nla_len(info->attrs[NL80211_ATTR_PMK]);
}
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
switch (info->genlhdr->cmd) {
case NL80211_CMD_SET_PMKSA:
rdev_ops = rdev->ops->set_pmksa;
break;
case NL80211_CMD_DEL_PMKSA:
rdev_ops = rdev->ops->del_pmksa;
break;
default:
WARN_ON(1);
break;
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev_ops)
return -EOPNOTSUPP;
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return rdev_ops(&rdev->wiphy, dev, &pmksa);
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
}
static int nl80211_flush_pmksa(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (dev->ieee80211_ptr->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
dev->ieee80211_ptr->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->flush_pmksa)
return -EOPNOTSUPP;
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_flush_pmksa(rdev, dev);
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
}
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
static int nl80211_tdls_mgmt(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
u8 action_code, dialog_token;
cfg80211: Pass TDLS peer capability information in tdls_mgmt While framing the TDLS Setup Confirmation frame, the driver needs to know if the TDLS peer is VHT/HT/WMM capable and thus shall construct the VHT/HT operation / WMM parameter elements accordingly. Supplicant determines if the TDLS peer is VHT/HT/WMM capable based on the presence of the respective IEs in the received TDLS Setup Response frame. The host driver should not need to parse the received TDLS Response frame and thus, should be able to rely on the supplicant to indicate the capability of the peer through additional flags while transmitting the TDLS Setup Confirmation frame through tdls_mgmt operations. Signed-off-by: Sunil Dutt Undekari <usdutt@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-20 10:52:09 +00:00
u32 peer_capability = 0;
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
u16 status_code;
u8 *peer;
cfg80211: pass TDLS initiator in tdls_mgmt operations The TDLS initiator is set once during link setup. If determines the address ordering in the link identifier IE. Fix dependent drivers - mwifiex and mac80211. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-11 14:18:21 +00:00
bool initiator;
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
if (!(rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_TDLS) ||
!rdev->ops->tdls_mgmt)
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_TDLS_ACTION] ||
!info->attrs[NL80211_ATTR_STATUS_CODE] ||
!info->attrs[NL80211_ATTR_TDLS_DIALOG_TOKEN] ||
!info->attrs[NL80211_ATTR_IE] ||
!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
peer = nla_data(info->attrs[NL80211_ATTR_MAC]);
action_code = nla_get_u8(info->attrs[NL80211_ATTR_TDLS_ACTION]);
status_code = nla_get_u16(info->attrs[NL80211_ATTR_STATUS_CODE]);
dialog_token = nla_get_u8(info->attrs[NL80211_ATTR_TDLS_DIALOG_TOKEN]);
cfg80211: pass TDLS initiator in tdls_mgmt operations The TDLS initiator is set once during link setup. If determines the address ordering in the link identifier IE. Fix dependent drivers - mwifiex and mac80211. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-11 14:18:21 +00:00
initiator = nla_get_flag(info->attrs[NL80211_ATTR_TDLS_INITIATOR]);
cfg80211: Pass TDLS peer capability information in tdls_mgmt While framing the TDLS Setup Confirmation frame, the driver needs to know if the TDLS peer is VHT/HT/WMM capable and thus shall construct the VHT/HT operation / WMM parameter elements accordingly. Supplicant determines if the TDLS peer is VHT/HT/WMM capable based on the presence of the respective IEs in the received TDLS Setup Response frame. The host driver should not need to parse the received TDLS Response frame and thus, should be able to rely on the supplicant to indicate the capability of the peer through additional flags while transmitting the TDLS Setup Confirmation frame through tdls_mgmt operations. Signed-off-by: Sunil Dutt Undekari <usdutt@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-20 10:52:09 +00:00
if (info->attrs[NL80211_ATTR_TDLS_PEER_CAPABILITY])
peer_capability =
nla_get_u32(info->attrs[NL80211_ATTR_TDLS_PEER_CAPABILITY]);
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_tdls_mgmt(rdev, dev, peer, action_code,
cfg80211: Pass TDLS peer capability information in tdls_mgmt While framing the TDLS Setup Confirmation frame, the driver needs to know if the TDLS peer is VHT/HT/WMM capable and thus shall construct the VHT/HT operation / WMM parameter elements accordingly. Supplicant determines if the TDLS peer is VHT/HT/WMM capable based on the presence of the respective IEs in the received TDLS Setup Response frame. The host driver should not need to parse the received TDLS Response frame and thus, should be able to rely on the supplicant to indicate the capability of the peer through additional flags while transmitting the TDLS Setup Confirmation frame through tdls_mgmt operations. Signed-off-by: Sunil Dutt Undekari <usdutt@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-20 10:52:09 +00:00
dialog_token, status_code, peer_capability,
cfg80211: pass TDLS initiator in tdls_mgmt operations The TDLS initiator is set once during link setup. If determines the address ordering in the link identifier IE. Fix dependent drivers - mwifiex and mac80211. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-06-11 14:18:21 +00:00
initiator,
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
nla_data(info->attrs[NL80211_ATTR_IE]),
nla_len(info->attrs[NL80211_ATTR_IE]));
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
}
static int nl80211_tdls_oper(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
enum nl80211_tdls_operation operation;
u8 *peer;
if (!(rdev->wiphy.flags & WIPHY_FLAG_SUPPORTS_TDLS) ||
!rdev->ops->tdls_oper)
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_TDLS_OPERATION] ||
!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
operation = nla_get_u8(info->attrs[NL80211_ATTR_TDLS_OPERATION]);
peer = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_tdls_oper(rdev, dev, peer, operation);
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
}
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
static int nl80211_remain_on_channel(struct sk_buff *skb,
struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
struct cfg80211_chan_def chandef;
cfg80211: Disallow moving out of operating DFS channel in non-ETSI For non-ETSI regulatory domain, CAC result on DFS channel may not be valid once moving out of that channel (as done during remain-on-channel, scannning and off-channel tx). Running CAC on an operating DFS channel after every off-channel operation will only add complexity and disturb the current link. Better do not allow any off-channel switch from a DFS operating channel in non-ETSI domain. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-27 11:34:34 +00:00
const struct cfg80211_chan_def *compat_chandef;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
struct sk_buff *msg;
void *hdr;
u64 cookie;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
u32 duration;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
int err;
if (!info->attrs[NL80211_ATTR_WIPHY_FREQ] ||
!info->attrs[NL80211_ATTR_DURATION])
return -EINVAL;
duration = nla_get_u32(info->attrs[NL80211_ATTR_DURATION]);
cfg80211: validate remain-on-channel time better The remain-on-channel time validation shouldn't depend on the value of HZ, as it does now with the check against jiffies, since then you might use a value that works on one system but not on another. Fix it by checking against a minimum that's fixed. Also add validation of the wait duration for a management frame TX since this also translates into remain-on-channel internally. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-01 10:50:54 +00:00
if (!rdev->ops->remain_on_channel ||
!(rdev->wiphy.flags & WIPHY_FLAG_HAS_REMAIN_ON_CHANNEL))
return -EOPNOTSUPP;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
/*
cfg80211: validate remain-on-channel time better The remain-on-channel time validation shouldn't depend on the value of HZ, as it does now with the check against jiffies, since then you might use a value that works on one system but not on another. Fix it by checking against a minimum that's fixed. Also add validation of the wait duration for a management frame TX since this also translates into remain-on-channel internally. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-01 10:50:54 +00:00
* We should be on that channel for at least a minimum amount of
* time (10ms) but no longer than the driver supports.
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
*/
cfg80211: validate remain-on-channel time better The remain-on-channel time validation shouldn't depend on the value of HZ, as it does now with the check against jiffies, since then you might use a value that works on one system but not on another. Fix it by checking against a minimum that's fixed. Also add validation of the wait duration for a management frame TX since this also translates into remain-on-channel internally. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-01 10:50:54 +00:00
if (duration < NL80211_MIN_REMAIN_ON_CHANNEL_TIME ||
nl80211: advertise maximum remain-on-channel duration With the upcoming hardware offload implementation, some devices will have a different maximum duration for the remain-on-channel command. Advertise the maximum duration in mac80211, and make mac80211 set it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-14 16:54:28 +00:00
duration > rdev->wiphy.max_remain_on_channel_duration)
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
return -EINVAL;
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
err = nl80211_parse_chandef(rdev, info, &chandef);
if (err)
return err;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
cfg80211: Disallow moving out of operating DFS channel in non-ETSI For non-ETSI regulatory domain, CAC result on DFS channel may not be valid once moving out of that channel (as done during remain-on-channel, scannning and off-channel tx). Running CAC on an operating DFS channel after every off-channel operation will only add complexity and disturb the current link. Better do not allow any off-channel switch from a DFS operating channel in non-ETSI domain. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-27 11:34:34 +00:00
wdev_lock(wdev);
if (!cfg80211_off_channel_oper_allowed(wdev) &&
!cfg80211_chandef_identical(&wdev->chandef, &chandef)) {
compat_chandef = cfg80211_chandef_compatible(&wdev->chandef,
&chandef);
if (compat_chandef != &chandef) {
wdev_unlock(wdev);
return -EBUSY;
}
}
wdev_unlock(wdev);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!msg)
return -ENOMEM;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
NL80211_CMD_REMAIN_ON_CHANNEL);
nl80211: nl80211hdr_put() doesn't return an ERR_PTR There are a few places which check nl80211hdr_put() for an ERR_PTR but actually it returns NULL on error and never error values. In nl80211_testmode_dump() the return wasn't checked at all so I have added one. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> [some whitespace changes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-14 11:50:01 +00:00
if (!hdr) {
err = -ENOBUFS;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
goto free_msg;
}
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
err = rdev_remain_on_channel(rdev, wdev, chandef.chan,
duration, &cookie);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
if (err)
goto free_msg;
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
genlmsg_end(msg, hdr);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
nla_put_failure:
err = -ENOBUFS;
free_msg:
nlmsg_free(msg);
return err;
}
static int nl80211_cancel_remain_on_channel(struct sk_buff *skb,
struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
u64 cookie;
if (!info->attrs[NL80211_ATTR_COOKIE])
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->cancel_remain_on_channel)
return -EOPNOTSUPP;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_cancel_remain_on_channel(rdev, wdev, cookie);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
}
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
static int nl80211_set_tx_bitrate_mask(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_bitrate_mask mask;
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
int err;
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->set_bitrate_mask)
return -EOPNOTSUPP;
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
cfg80211: Add support to configure a beacon data rate This allows an option to configure a single beacon tx rate for an AP. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 12:08:44 +00:00
err = nl80211_parse_tx_bitrate_mask(info, &mask);
if (err)
return err;
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_set_bitrate_mask(rdev, dev, NULL, &mask);
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
}
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
static int nl80211_register_mgmt(struct sk_buff *skb, struct genl_info *info)
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
u16 frame_type = IEEE80211_FTYPE_MGMT | IEEE80211_STYPE_ACTION;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (!info->attrs[NL80211_ATTR_FRAME_MATCH])
return -EINVAL;
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
if (info->attrs[NL80211_ATTR_FRAME_TYPE])
frame_type = nla_get_u16(info->attrs[NL80211_ATTR_FRAME_TYPE]);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_P2P_CLIENT:
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_MESH_POINT:
case NL80211_IFTYPE_P2P_GO:
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
case NL80211_IFTYPE_P2P_DEVICE:
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
break;
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
case NL80211_IFTYPE_NAN:
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
default:
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EOPNOTSUPP;
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
}
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
/* not much point in registering if we can't reply */
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->mgmt_tx)
return -EOPNOTSUPP;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
return cfg80211_mlme_register_mgmt(wdev, info->snd_portid, frame_type,
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nla_data(info->attrs[NL80211_ATTR_FRAME_MATCH]),
nla_len(info->attrs[NL80211_ATTR_FRAME_MATCH]));
}
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
static int nl80211_tx_mgmt(struct sk_buff *skb, struct genl_info *info)
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
struct cfg80211_chan_def chandef;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
int err;
nl80211: fix compiler warning John reported the following warning: net/wireless/nl80211.c: In function ‘nl80211_tx_mgmt’: net/wireless/nl80211.c:5286:8: warning: ‘hdr’ may be used uninitialized in this function Evidently, his version of gcc isn't able to see that when "msg" is initialized, "hdr" must also be. My gcc, 4.6.1, can actually see that and doesn't warn. Simply initialize the variable to NULL. That means if the compiler was ever right we'll crash though so isn't really optimal since it may hide warnings from the compiler when somebody modifies this code in the future. Reported-by: John Linville <linville@tuxdriver.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-10 08:44:46 +00:00
void *hdr = NULL;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
u64 cookie;
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
struct sk_buff *msg = NULL;
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
struct cfg80211_mgmt_tx_params params = {
.dont_wait_for_ack =
info->attrs[NL80211_ATTR_DONT_WAIT_FOR_ACK],
};
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (!info->attrs[NL80211_ATTR_FRAME])
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
return -EINVAL;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->mgmt_tx)
return -EOPNOTSUPP;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
switch (wdev->iftype) {
nl80211: allow sending CMD_FRAME without specifying any frequency Users may want to send a frame on the current channel without specifying it. This is particularly useful for the correct implementation of the IBSS/RSN support in wpa_supplicant which requires to receive and send AUTH frames. Make mgmt_tx pass a NULL channel to the driver if none has been specified by the user. Signed-off-by: Antonio Quartulli <antonio@open-mesh.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-11 12:20:03 +00:00
case NL80211_IFTYPE_P2P_DEVICE:
if (!info->attrs[NL80211_ATTR_WIPHY_FREQ])
return -EINVAL;
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_P2P_CLIENT:
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_MESH_POINT:
case NL80211_IFTYPE_P2P_GO:
break;
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
case NL80211_IFTYPE_NAN:
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
default:
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return -EOPNOTSUPP;
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
}
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
if (info->attrs[NL80211_ATTR_DURATION]) {
cfg80211: add flags for off-channel capabilities Currently mac80211 implements these for all devices, but given restrictions of some devices that isn't really true, so prepare for being able to remove the capability for some mac80211 devices. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 14:33:48 +00:00
if (!(rdev->wiphy.flags & WIPHY_FLAG_OFFCHAN_TX))
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
return -EINVAL;
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
params.wait = nla_get_u32(info->attrs[NL80211_ATTR_DURATION]);
cfg80211: validate remain-on-channel time better The remain-on-channel time validation shouldn't depend on the value of HZ, as it does now with the check against jiffies, since then you might use a value that works on one system but not on another. Fix it by checking against a minimum that's fixed. Also add validation of the wait duration for a management frame TX since this also translates into remain-on-channel internally. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-01 10:50:54 +00:00
/*
* We should wait on the channel for at least a minimum amount
* of time (10ms) but no longer than the driver supports.
*/
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
if (params.wait < NL80211_MIN_REMAIN_ON_CHANNEL_TIME ||
params.wait > rdev->wiphy.max_remain_on_channel_duration)
cfg80211: validate remain-on-channel time better The remain-on-channel time validation shouldn't depend on the value of HZ, as it does now with the check against jiffies, since then you might use a value that works on one system but not on another. Fix it by checking against a minimum that's fixed. Also add validation of the wait duration for a management frame TX since this also translates into remain-on-channel internally. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-06-01 10:50:54 +00:00
return -EINVAL;
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
}
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
params.offchan = info->attrs[NL80211_ATTR_OFFCHANNEL_TX_OK];
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
if (params.offchan && !(rdev->wiphy.flags & WIPHY_FLAG_OFFCHAN_TX))
cfg80211: add flags for off-channel capabilities Currently mac80211 implements these for all devices, but given restrictions of some devices that isn't really true, so prepare for being able to remove the capability for some mac80211 devices. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 14:33:48 +00:00
return -EINVAL;
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
params.no_cck = nla_get_flag(info->attrs[NL80211_ATTR_TX_NO_CCK_RATE]);
nl80211/cfg80211: Add support to disable CCK rate for management frame Add a new nl80211 attribute to specify whether to send the management frames in CCK rate or not. As of now the wpa_supplicant is disabling CCK rate at P2P init itself. So this patch helps to send P2P probe request/probe response/action frames being sent at non CCK rate in 2GHz without disabling 11b rates. This attribute is used with NL80211_CMD_TRIGGER_SCAN and NL80211_CMD_FRAME commands to disable CCK rate for management frame transmission. Cc: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Rajkumar Manoharan <rmanohar@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-25 09:23:30 +00:00
nl80211: allow sending CMD_FRAME without specifying any frequency Users may want to send a frame on the current channel without specifying it. This is particularly useful for the correct implementation of the IBSS/RSN support in wpa_supplicant which requires to receive and send AUTH frames. Make mgmt_tx pass a NULL channel to the driver if none has been specified by the user. Signed-off-by: Antonio Quartulli <antonio@open-mesh.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-11 12:20:03 +00:00
/* get the channel if any has been specified, otherwise pass NULL to
* the driver. The latter will use the current one
*/
chandef.chan = NULL;
if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) {
err = nl80211_parse_chandef(rdev, info, &chandef);
if (err)
return err;
}
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
if (!chandef.chan && params.offchan)
nl80211: allow sending CMD_FRAME without specifying any frequency Users may want to send a frame on the current channel without specifying it. This is particularly useful for the correct implementation of the IBSS/RSN support in wpa_supplicant which requires to receive and send AUTH frames. Make mgmt_tx pass a NULL channel to the driver if none has been specified by the user. Signed-off-by: Antonio Quartulli <antonio@open-mesh.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-11 12:20:03 +00:00
return -EINVAL;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211: Disallow moving out of operating DFS channel in non-ETSI For non-ETSI regulatory domain, CAC result on DFS channel may not be valid once moving out of that channel (as done during remain-on-channel, scannning and off-channel tx). Running CAC on an operating DFS channel after every off-channel operation will only add complexity and disturb the current link. Better do not allow any off-channel switch from a DFS operating channel in non-ETSI domain. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-27 11:34:34 +00:00
wdev_lock(wdev);
if (params.offchan && !cfg80211_off_channel_oper_allowed(wdev)) {
wdev_unlock(wdev);
return -EBUSY;
}
wdev_unlock(wdev);
cfg80211: Add API to update CSA counters in mgmt frames Add NL80211_ATTR_CSA_C_OFFSETS_TX which holds an array of offsets to the CSA counters which should be updated when sending a management frames with NL80211_CMD_FRAME. This API should be used by the drivers that wish to keep the CSA counter updated in probe responses, but do not implement probe response offloading and so, do not use ieee80211_proberesp_get function. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-09 11:11:44 +00:00
params.buf = nla_data(info->attrs[NL80211_ATTR_FRAME]);
params.len = nla_len(info->attrs[NL80211_ATTR_FRAME]);
if (info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]) {
int len = nla_len(info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]);
int i;
if (len % sizeof(u16))
return -EINVAL;
params.n_csa_offsets = len / sizeof(u16);
params.csa_offsets =
nla_data(info->attrs[NL80211_ATTR_CSA_C_OFFSETS_TX]);
/* check that all the offsets fit the frame */
for (i = 0; i < params.n_csa_offsets; i++) {
if (params.csa_offsets[i] >= params.len)
return -EINVAL;
}
}
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
if (!params.dont_wait_for_ack) {
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
NL80211_CMD_FRAME);
nl80211: nl80211hdr_put() doesn't return an ERR_PTR There are a few places which check nl80211hdr_put() for an ERR_PTR but actually it returns NULL on error and never error values. In nl80211_testmode_dump() the return wasn't checked at all so I have added one. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> [some whitespace changes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-14 11:50:01 +00:00
if (!hdr) {
err = -ENOBUFS;
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
goto free_msg;
}
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
}
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
cfg80211: aggregate mgmt_tx parameters into a struct Change cfg80211 and mac80211 to use cfg80211_mgmt_tx_params struct to aggregate parameters for mgmt_tx functions. This makes the functions' signatures less clumsy and allows less painful parameters extension. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> [fix all other drivers] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-18 17:06:49 +00:00
params.chan = chandef.chan;
err = cfg80211_mlme_mgmt_tx(rdev, wdev, &params, &cookie);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (err)
goto free_msg;
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
if (msg) {
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211/mac80211: allow management TX to not wait for ACK For probe responses it can be useful to not wait for ACK to avoid retransmissions if the station that sent the probe is already on the next channel, so allow userspace to request not caring about the ACK with a new nl80211 flag. Since mac80211 needs to be updated for the new function prototype anyway implement it right away -- it's just a few lines of code. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:21 +00:00
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
}
return 0;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nla_put_failure:
err = -ENOBUFS;
free_msg:
nlmsg_free(msg);
return err;
}
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
static int nl80211_tx_mgmt_cancel_wait(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev = info->user_ptr[1];
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
u64 cookie;
if (!info->attrs[NL80211_ATTR_COOKIE])
return -EINVAL;
if (!rdev->ops->mgmt_tx_cancel_wait)
return -EOPNOTSUPP;
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_ADHOC:
case NL80211_IFTYPE_P2P_CLIENT:
case NL80211_IFTYPE_AP:
case NL80211_IFTYPE_AP_VLAN:
case NL80211_IFTYPE_P2P_GO:
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
case NL80211_IFTYPE_P2P_DEVICE:
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
break;
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
case NL80211_IFTYPE_NAN:
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
default:
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
return -EOPNOTSUPP;
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
}
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_mgmt_tx_cancel_wait(rdev, wdev, cookie);
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
}
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
static int nl80211_set_power_save(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
struct wireless_dev *wdev;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
u8 ps_state;
bool state;
int err;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!info->attrs[NL80211_ATTR_PS_STATE])
return -EINVAL;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
ps_state = nla_get_u32(info->attrs[NL80211_ATTR_PS_STATE]);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (ps_state != NL80211_PS_DISABLED && ps_state != NL80211_PS_ENABLED)
return -EINVAL;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
wdev = dev->ieee80211_ptr;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->set_power_mgmt)
return -EOPNOTSUPP;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
state = (ps_state == NL80211_PS_ENABLED) ? true : false;
if (state == wdev->ps)
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return 0;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_set_power_mgmt(rdev, dev, state, wdev->ps_timeout);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!err)
wdev->ps = state;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
return err;
}
static int nl80211_get_power_save(struct sk_buff *skb, struct genl_info *info)
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
enum nl80211_ps_state ps_state;
struct wireless_dev *wdev;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev = info->user_ptr[1];
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
struct sk_buff *msg;
void *hdr;
int err;
wdev = dev->ieee80211_ptr;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!rdev->ops->set_power_mgmt)
return -EOPNOTSUPP;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (!msg)
return -ENOMEM;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
NL80211_CMD_GET_POWER_SAVE);
if (!hdr) {
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
err = -ENOBUFS;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
goto free_msg;
}
if (wdev->ps)
ps_state = NL80211_PS_ENABLED;
else
ps_state = NL80211_PS_DISABLED;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_PS_STATE, ps_state))
goto nla_put_failure;
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
genlmsg_end(msg, hdr);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
return genlmsg_reply(msg, info);
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
nla_put_failure:
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
err = -ENOBUFS;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
free_msg:
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
nlmsg_free(msg);
return err;
}
nl80211: make netlink attribute policies const There's no reason for netlink attribute policies to be __read_mostly, they can just be const. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-20 22:58:15 +00:00
static const struct nla_policy
nl80211_attr_cqm_policy[NL80211_ATTR_CQM_MAX + 1] = {
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
[NL80211_ATTR_CQM_RSSI_THOLD] = { .type = NLA_BINARY },
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
[NL80211_ATTR_CQM_RSSI_HYST] = { .type = NLA_U32 },
[NL80211_ATTR_CQM_RSSI_THRESHOLD_EVENT] = { .type = NLA_U32 },
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
[NL80211_ATTR_CQM_TXE_RATE] = { .type = NLA_U32 },
[NL80211_ATTR_CQM_TXE_PKTS] = { .type = NLA_U32 },
[NL80211_ATTR_CQM_TXE_INTVL] = { .type = NLA_U32 },
cfg80211: Pass new RSSI level in CQM RSSI notification Update the drivers to pass the RSSI level as a cfg80211_cqm_rssi_notify parameter and pass this value to userspace in a new nl80211 attribute. This helps both userspace and also helps in the implementation of the multiple RSSI thresholds CQM mechanism. Note for marvell/mwifiex I pass 0 for the RSSI value because the new RSSI value is not available to the driver at the time of the cfg80211_cqm_rssi_notify call, but the driver queries the new value immediately after that, so it is actually available just a moment later if we wanted to defer caling cfg80211_cqm_rssi_notify until that moment. Without this, the new cfg80211 code (patch 3) will call .get_station which will send a duplicate HostCmd_CMD_RSSI_INFO command to the hardware. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-25 11:43:41 +00:00
[NL80211_ATTR_CQM_RSSI_LEVEL] = { .type = NLA_S32 },
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
};
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
static int nl80211_set_cqm_txe(struct genl_info *info,
nl80211: remove unnecessary checks The CQM TX-error rate/interval can't be less than zero since they're unsigned values, remove checks. Also fix indentation of the function. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-26 11:51:52 +00:00
u32 rate, u32 pkts, u32 intvl)
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
nl80211: remove unnecessary checks The CQM TX-error rate/interval can't be less than zero since they're unsigned values, remove checks. Also fix indentation of the function. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-26 11:51:52 +00:00
if (rate > 100 || intvl > NL80211_CQM_TXE_MAX_INTVL)
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
return -EINVAL;
if (!rdev->ops->set_cqm_txe_config)
return -EOPNOTSUPP;
if (wdev->iftype != NL80211_IFTYPE_STATION &&
wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
return rdev_set_cqm_txe_config(rdev, dev, rate, pkts, intvl);
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
}
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
static int cfg80211_cqm_rssi_update(struct cfg80211_registered_device *rdev,
struct net_device *dev)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
s32 last, low, high;
u32 hyst;
int i, n;
int err;
/* RSSI reporting disabled? */
if (!wdev->cqm_config)
return rdev_set_cqm_rssi_range_config(rdev, dev, 0, 0);
/*
* Obtain current RSSI value if possible, if not and no RSSI threshold
* event has been received yet, we should receive an event after a
* connection is established and enough beacons received to calculate
* the average.
*/
if (!wdev->cqm_config->last_rssi_event_value && wdev->current_bss &&
rdev->ops->get_station) {
struct station_info sinfo;
u8 *mac_addr;
mac_addr = wdev->current_bss->pub.bssid;
err = rdev_get_station(rdev, dev, mac_addr, &sinfo);
if (err)
return err;
if (sinfo.filled & BIT(NL80211_STA_INFO_BEACON_SIGNAL_AVG))
wdev->cqm_config->last_rssi_event_value =
(s8) sinfo.rx_beacon_signal_avg;
}
last = wdev->cqm_config->last_rssi_event_value;
hyst = wdev->cqm_config->rssi_hyst;
n = wdev->cqm_config->n_rssi_thresholds;
for (i = 0; i < n; i++)
if (last < wdev->cqm_config->rssi_thresholds[i])
break;
low = i > 0 ?
(wdev->cqm_config->rssi_thresholds[i - 1] - hyst) : S32_MIN;
high = i < n ?
(wdev->cqm_config->rssi_thresholds[i] + hyst - 1) : S32_MAX;
return rdev_set_cqm_rssi_range_config(rdev, dev, low, high);
}
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
static int nl80211_set_cqm_rssi(struct genl_info *info,
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
const s32 *thresholds, int n_thresholds,
u32 hysteresis)
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
{
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
int i, err;
s32 prev = S32_MIN;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
/* Check all values negative and sorted */
for (i = 0; i < n_thresholds; i++) {
if (thresholds[i] > 0 || thresholds[i] <= prev)
return -EINVAL;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
prev = thresholds[i];
}
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211/nl80211: introduce p2p device types This adds P2P-STA and P2P-GO as device types so we can distinguish between those and normal STA or AP (respectively) type interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-09-16 12:58:22 +00:00
if (wdev->iftype != NL80211_IFTYPE_STATION &&
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
wdev_lock(wdev);
cfg80211_cqm_config_free(wdev);
wdev_unlock(wdev);
if (n_thresholds <= 1 && rdev->ops->set_cqm_rssi_config) {
if (n_thresholds == 0 || thresholds[0] == 0) /* Disabling */
return rdev_set_cqm_rssi_config(rdev, dev, 0, 0);
return rdev_set_cqm_rssi_config(rdev, dev,
thresholds[0], hysteresis);
}
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_CQM_RSSI_LIST))
return -EOPNOTSUPP;
if (n_thresholds == 1 && thresholds[0] == 0) /* Disabling */
n_thresholds = 0;
wdev_lock(wdev);
if (n_thresholds) {
struct cfg80211_cqm_config *cqm_config;
cqm_config = kzalloc(sizeof(struct cfg80211_cqm_config) +
n_thresholds * sizeof(s32), GFP_KERNEL);
if (!cqm_config) {
err = -ENOMEM;
goto unlock;
}
cqm_config->rssi_hyst = hysteresis;
cqm_config->n_rssi_thresholds = n_thresholds;
memcpy(cqm_config->rssi_thresholds, thresholds,
n_thresholds * sizeof(s32));
wdev->cqm_config = cqm_config;
}
err = cfg80211_cqm_rssi_update(rdev, dev);
unlock:
wdev_unlock(wdev);
return err;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
}
static int nl80211_set_cqm(struct sk_buff *skb, struct genl_info *info)
{
struct nlattr *attrs[NL80211_ATTR_CQM_MAX + 1];
struct nlattr *cqm;
int err;
cqm = info->attrs[NL80211_ATTR_CQM];
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
if (!cqm)
return -EINVAL;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
err = nla_parse_nested(attrs, NL80211_ATTR_CQM_MAX, cqm,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_attr_cqm_policy, info->extack);
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
if (err)
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
return err;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
if (attrs[NL80211_ATTR_CQM_RSSI_THOLD] &&
attrs[NL80211_ATTR_CQM_RSSI_HYST]) {
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
const s32 *thresholds =
nla_data(attrs[NL80211_ATTR_CQM_RSSI_THOLD]);
int len = nla_len(attrs[NL80211_ATTR_CQM_RSSI_THOLD]);
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
u32 hysteresis = nla_get_u32(attrs[NL80211_ATTR_CQM_RSSI_HYST]);
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
if (len % 4)
return -EINVAL;
return nl80211_set_cqm_rssi(info, thresholds, len / 4,
hysteresis);
nl80211: clean up CQM settings code Clean up the CQM settings code a bit and while at it enforce that when setting the threshold to 0 (disable) the hysteresis is also set to 0 to avoid confusion. As we haven't enforce it, simply override userspace. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-06 12:10:48 +00:00
}
if (attrs[NL80211_ATTR_CQM_TXE_RATE] &&
attrs[NL80211_ATTR_CQM_TXE_PKTS] &&
attrs[NL80211_ATTR_CQM_TXE_INTVL]) {
u32 rate = nla_get_u32(attrs[NL80211_ATTR_CQM_TXE_RATE]);
u32 pkts = nla_get_u32(attrs[NL80211_ATTR_CQM_TXE_PKTS]);
u32 intvl = nla_get_u32(attrs[NL80211_ATTR_CQM_TXE_INTVL]);
return nl80211_set_cqm_txe(info, rate, pkts, intvl);
}
return -EINVAL;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
}
cfg80211: 802.11p OCB mode handling This patch adds new iface type (NL80211_IFTYPE_OCB) representing the OCB (Outside the Context of a BSS) mode. When establishing a connection to the network a cfg80211_join_ocb function is called (particular nl80211_command is added as well). A mandatory parameters during the ocb_join operation are 'center frequency' and 'channel width (5/10 MHz)'. Changes done in mac80211 are minimal possible required to avoid many warnings (warning: enumeration value 'NL80211_IFTYPE_OCB' not handled in switch) during compilation. Full functionality (where needed) is added in the following patch. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-03 09:33:18 +00:00
static int nl80211_join_ocb(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct ocb_setup setup = {};
int err;
err = nl80211_parse_chandef(rdev, info, &setup.chandef);
if (err)
return err;
return cfg80211_join_ocb(rdev, dev, &setup);
}
static int nl80211_leave_ocb(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
return cfg80211_leave_ocb(rdev, dev);
}
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
static int nl80211_join_mesh(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct mesh_config cfg;
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
struct mesh_setup setup;
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
int err;
/* start with default */
memcpy(&cfg, &default_mesh_config, sizeof(cfg));
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
memcpy(&setup, &default_mesh_setup, sizeof(setup));
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
if (info->attrs[NL80211_ATTR_MESH_CONFIG]) {
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
/* and parse parameters if given */
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
err = nl80211_parse_mesh_config(info, &cfg, NULL);
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
if (err)
return err;
}
if (!info->attrs[NL80211_ATTR_MESH_ID] ||
!nla_len(info->attrs[NL80211_ATTR_MESH_ID]))
return -EINVAL;
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
setup.mesh_id = nla_data(info->attrs[NL80211_ATTR_MESH_ID]);
setup.mesh_id_len = nla_len(info->attrs[NL80211_ATTR_MESH_ID]);
{nl,cfg,mac}80211: Allow Setting Multicast Rate in Mesh Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@gmail.com> Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-25 01:15:20 +00:00
if (info->attrs[NL80211_ATTR_MCAST_RATE] &&
!nl80211_parse_mcast_rate(rdev, setup.mcast_rate,
nla_get_u32(info->attrs[NL80211_ATTR_MCAST_RATE])))
return -EINVAL;
{cfg,nl,mac}80211: set beacon interval and DTIM period on mesh join Move the default mesh beacon interval and DTIM period to cfg80211 and make them accessible to nl80211. This enables setting both values when joining an MBSS. Previously the DTIM parameter was not set by mac80211 so the driver's default value was used. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:51 +00:00
if (info->attrs[NL80211_ATTR_BEACON_INTERVAL]) {
setup.beacon_interval =
nla_get_u32(info->attrs[NL80211_ATTR_BEACON_INTERVAL]);
cfg80211: identically validate beacon interval for AP/MESH/IBSS Beacon interval interface combinations validation was missing for MESH/IBSS join, add those. Johannes: also move the beacon interval check disallowing really tiny and really big intervals into the common function, which adds it for AP mode. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 09:44:02 +00:00
cfg80211: support virtual interfaces with different beacon intervals This commit provides a mechanism for the host drivers to advertise the support for different beacon intervals among the respective interface combinations in a group, through NL80211_IFACE_COMB_BI_MIN_GCD (u32). This value will be compared against GCD of all beaconing interfaces of matching combinations. If the driver doesn't advertise this value, the old behaviour where all beacon intervals must be identical is retained. If it is specified, then any beacon interval for an interface in the interface combination as well as the GCD of all active beacon intervals in the combination must be greater or equal to this value. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> [change commit message, some variable names, small other things] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-12 12:56:51 +00:00
err = cfg80211_validate_beacon_int(rdev,
NL80211_IFTYPE_MESH_POINT,
setup.beacon_interval);
cfg80211: identically validate beacon interval for AP/MESH/IBSS Beacon interval interface combinations validation was missing for MESH/IBSS join, add those. Johannes: also move the beacon interval check disallowing really tiny and really big intervals into the common function, which adds it for AP mode. Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-11 09:44:02 +00:00
if (err)
return err;
{cfg,nl,mac}80211: set beacon interval and DTIM period on mesh join Move the default mesh beacon interval and DTIM period to cfg80211 and make them accessible to nl80211. This enables setting both values when joining an MBSS. Previously the DTIM parameter was not set by mac80211 so the driver's default value was used. Signed-off-by: Marco Porsch <marco@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-07 15:04:51 +00:00
}
if (info->attrs[NL80211_ATTR_DTIM_PERIOD]) {
setup.dtim_period =
nla_get_u32(info->attrs[NL80211_ATTR_DTIM_PERIOD]);
if (setup.dtim_period < 1 || setup.dtim_period > 100)
return -EINVAL;
}
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
if (info->attrs[NL80211_ATTR_MESH_SETUP]) {
/* parse additional setup parameters if given */
err = nl80211_parse_mesh_setup(info, &setup);
if (err)
return err;
}
nl80211: user_mpm overrides auto_open_plinks If the user requested a userspace MPM, automatically disable auto_open_plinks to fully disable the kernel MPM. Signed-off-by: Thomas Pedersen <thomas@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-04 21:06:13 +00:00
if (setup.user_mpm)
cfg.auto_open_plinks = false;
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
if (info->attrs[NL80211_ATTR_WIPHY_FREQ]) {
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
err = nl80211_parse_chandef(rdev, info, &setup.chandef);
if (err)
return err;
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
} else {
/* cfg80211_join_mesh() will sort it out */
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
setup.chandef.chan = NULL;
cfg80211: provide channel to join_mesh function Just like the AP mode patch, instead of setting the channel and then joining the mesh network, provide the channel to join the network on to the join_mesh() function. Like in AP mode, you can also give the channel to the join-mesh nl80211 command now. Unlike AP mode, it picks a default channel if none was given. As libertas uses mesh mode interfaces but has no join_mesh callback and we can't simply break it, keep some compatibility code for that case and configure the channel directly for it. In the non-libertas case, where we store the channel until join, allow setting it while the interface is down. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-05-16 21:50:20 +00:00
}
{nl,mac,cfg}80211: Allow user to configure basic rates for mesh Currently mesh uses mandatory rates as the default basic rates. Allow basic rates to be configured during mesh join. Basic rates are applied only if channel is also provided with mesh join command. Signed-off-by: Ashok Nagarajan <ashok@cozybit.com> [some whitespace fixes, refuse basic rates w/o channel] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 17:33:36 +00:00
if (info->attrs[NL80211_ATTR_BSS_BASIC_RATES]) {
u8 *rates = nla_data(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
int n_rates =
nla_len(info->attrs[NL80211_ATTR_BSS_BASIC_RATES]);
struct ieee80211_supported_band *sband;
if (!setup.chandef.chan)
return -EINVAL;
sband = rdev->wiphy.bands[setup.chandef.chan->band];
err = ieee80211_get_ratemask(sband, rates, n_rates,
&setup.basic_rates);
if (err)
return err;
}
cfg80211: add checks for beacon rate, extend to mesh The previous commit added support for specifying the beacon rate for AP mode. Add features checks to this, and extend it to also support the rate configuration for mesh networks. For IBSS it's not as simple due to joining etc., so that's not yet supported. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-19 07:44:44 +00:00
if (info->attrs[NL80211_ATTR_TX_RATES]) {
err = nl80211_parse_tx_bitrate_mask(info, &setup.beacon_rate);
if (err)
return err;
err = validate_beacon_tx_rate(rdev, setup.chandef.chan->band,
&setup.beacon_rate);
if (err)
return err;
}
wireless: Only join DFS channels in mesh mode if userspace flags support When joining a mesh network it is not guaranteed that userspace has a daemon listening for radar events. This is however required for channels requiring DFS. To flag that userspace will handle radar events, it needs to set NL80211_ATTR_HANDLE_DFS. This matches the current mechanism used for IBSS mode. Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-05-16 09:23:11 +00:00
setup.userspace_handles_dfs =
nla_get_flag(info->attrs[NL80211_ATTR_HANDLE_DFS]);
mac80211: Let userspace enable and configure vendor specific path selection. Userspace will now be allowed to toggle between the default path selection algorithm (HWMP, implemented in the kernel), and a vendor specific alternative. Also in the same patch, allow userspace to add information elements to mesh beacons. This is accordance with the Extensible Path Selection Framework specified in version 7.0 of the 802.11s draft. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:49 +00:00
return cfg80211_join_mesh(rdev, dev, &setup, &cfg);
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
}
static int nl80211_leave_mesh(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
return cfg80211_leave_mesh(rdev, dev);
}
cfg80211: don't allow WoWLAN support without CONFIG_PM When CONFIG_PM is disabled, no device can possibly support WoWLAN since it can't go to sleep to start with. Due to this, mac80211 had even rejected the hardware registration. By making all the code and data for WoWLAN depend on CONFIG_PM we can promote this runtime error to a compile-time error. Add #ifdef around all WoWLAN code to remove it in systems that don't need it as they never suspend. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Acked-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 07:23:48 +00:00
#ifdef CONFIG_PM
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
static int nl80211_send_wowlan_patterns(struct sk_buff *msg,
struct cfg80211_registered_device *rdev)
{
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
struct cfg80211_wowlan *wowlan = rdev->wiphy.wowlan_config;
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
struct nlattr *nl_pats, *nl_pat;
int i, pat_len;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if (!wowlan->n_patterns)
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
return 0;
nl_pats = nla_nest_start(msg, NL80211_WOWLAN_TRIG_PKT_PATTERN);
if (!nl_pats)
return -ENOBUFS;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
for (i = 0; i < wowlan->n_patterns; i++) {
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
nl_pat = nla_nest_start(msg, i + 1);
if (!nl_pat)
return -ENOBUFS;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
pat_len = wowlan->patterns[i].pattern_len;
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
if (nla_put(msg, NL80211_PKTPAT_MASK, DIV_ROUND_UP(pat_len, 8),
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
wowlan->patterns[i].mask) ||
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
nla_put(msg, NL80211_PKTPAT_PATTERN, pat_len,
wowlan->patterns[i].pattern) ||
nla_put_u32(msg, NL80211_PKTPAT_OFFSET,
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
wowlan->patterns[i].pkt_offset))
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
return -ENOBUFS;
nla_nest_end(msg, nl_pat);
}
nla_nest_end(msg, nl_pats);
return 0;
}
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
static int nl80211_send_wowlan_tcp(struct sk_buff *msg,
struct cfg80211_wowlan_tcp *tcp)
{
struct nlattr *nl_tcp;
if (!tcp)
return 0;
nl_tcp = nla_nest_start(msg, NL80211_WOWLAN_TRIG_TCP_CONNECTION);
if (!nl_tcp)
return -ENOBUFS;
netlink: implement nla_put_in_addr and nla_put_in6_addr IP addresses are often stored in netlink attributes. Add generic functions to do that. For nla_put_in_addr, it would be nicer to pass struct in_addr but this is not used universally throughout the kernel, in way too many places __be32 is used to store IPv4 address. Signed-off-by: Jiri Benc <jbenc@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-29 14:59:25 +00:00
if (nla_put_in_addr(msg, NL80211_WOWLAN_TCP_SRC_IPV4, tcp->src) ||
nla_put_in_addr(msg, NL80211_WOWLAN_TCP_DST_IPV4, tcp->dst) ||
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
nla_put(msg, NL80211_WOWLAN_TCP_DST_MAC, ETH_ALEN, tcp->dst_mac) ||
nla_put_u16(msg, NL80211_WOWLAN_TCP_SRC_PORT, tcp->src_port) ||
nla_put_u16(msg, NL80211_WOWLAN_TCP_DST_PORT, tcp->dst_port) ||
nla_put(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD,
tcp->payload_len, tcp->payload) ||
nla_put_u32(msg, NL80211_WOWLAN_TCP_DATA_INTERVAL,
tcp->data_interval) ||
nla_put(msg, NL80211_WOWLAN_TCP_WAKE_PAYLOAD,
tcp->wake_len, tcp->wake_data) ||
nla_put(msg, NL80211_WOWLAN_TCP_WAKE_MASK,
DIV_ROUND_UP(tcp->wake_len, 8), tcp->wake_mask))
return -ENOBUFS;
if (tcp->payload_seq.len &&
nla_put(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD_SEQ,
sizeof(tcp->payload_seq), &tcp->payload_seq))
return -ENOBUFS;
if (tcp->payload_tok.len &&
nla_put(msg, NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN,
sizeof(tcp->payload_tok) + tcp->tokens_size,
&tcp->payload_tok))
return -ENOBUFS;
cfg80211: fix sending WoWLAN TCP wakeup settings The code sending the current WoWLAN TCP wakeup settings in nl80211_send_wowlan_tcp() is not closing the nested attribute, thus causing the parser to get confused on the receiver side in userspace (iw). Fix this. Cc: stable@vger.kernel.org [3.9] Reported-by: Deepak Arora <deepakx.arora@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-16 08:24:28 +00:00
nla_nest_end(msg, nl_tcp);
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return 0;
}
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
static int nl80211_send_wowlan_nd(struct sk_buff *msg,
struct cfg80211_sched_scan_request *req)
{
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
struct nlattr *nd, *freqs, *matches, *match, *scan_plans, *scan_plan;
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
int i;
if (!req)
return 0;
nd = nla_nest_start(msg, NL80211_WOWLAN_TRIG_NET_DETECT);
if (!nd)
return -ENOBUFS;
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
if (req->n_scan_plans == 1 &&
nla_put_u32(msg, NL80211_ATTR_SCHED_SCAN_INTERVAL,
req->scan_plans[0].interval * 1000))
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
return -ENOBUFS;
nl80211: add net-detect delay to wowlan info Pass the initial net-detect delay (NL80211_ATTR_SCHED_SCAN_DELAY) attribute in the WoWLAN info response. Additionally, remove a bogus TODO comment. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-17 14:36:01 +00:00
if (nla_put_u32(msg, NL80211_ATTR_SCHED_SCAN_DELAY, req->delay))
return -ENOBUFS;
cfg80211: Add support to sched scan to report better BSSs Enhance sched scan to support option of finding a better BSS while in connected state. Firmware scans the medium and reports when it finds a known BSS which has better RSSI than the current connected BSS. New attributes to specify the relative RSSI (compared to the current BSS) are added to the sched scan to implement this. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:20 +00:00
if (req->relative_rssi_set) {
struct nl80211_bss_select_rssi_adjust rssi_adjust;
if (nla_put_s8(msg, NL80211_ATTR_SCHED_SCAN_RELATIVE_RSSI,
req->relative_rssi))
return -ENOBUFS;
rssi_adjust.band = req->rssi_adjust.band;
rssi_adjust.delta = req->rssi_adjust.delta;
if (nla_put(msg, NL80211_ATTR_SCHED_SCAN_RSSI_ADJUST,
sizeof(rssi_adjust), &rssi_adjust))
return -ENOBUFS;
}
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
freqs = nla_nest_start(msg, NL80211_ATTR_SCAN_FREQUENCIES);
if (!freqs)
return -ENOBUFS;
nl80211: always check nla_put* return values A few instances were found where we didn't check them, add the missing checks even though they'll probably never trigger as the message should be large enough here. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:59:21 +00:00
for (i = 0; i < req->n_channels; i++) {
if (nla_put_u32(msg, i, req->channels[i]->center_freq))
return -ENOBUFS;
}
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
nla_nest_end(msg, freqs);
if (req->n_match_sets) {
matches = nla_nest_start(msg, NL80211_ATTR_SCHED_SCAN_MATCH);
nl80211: always check nla_nest_start() return value If the message got full during nla_nest_start(), it can return NULL. None of the cases here seem like that can really happen, but check the return value nonetheless. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:55:57 +00:00
if (!matches)
return -ENOBUFS;
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
for (i = 0; i < req->n_match_sets; i++) {
match = nla_nest_start(msg, i);
nl80211: always check nla_nest_start() return value If the message got full during nla_nest_start(), it can return NULL. None of the cases here seem like that can really happen, but check the return value nonetheless. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:55:57 +00:00
if (!match)
return -ENOBUFS;
nl80211: always check nla_put* return values A few instances were found where we didn't check them, add the missing checks even though they'll probably never trigger as the message should be large enough here. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:59:21 +00:00
if (nla_put(msg, NL80211_SCHED_SCAN_MATCH_ATTR_SSID,
req->match_sets[i].ssid.ssid_len,
req->match_sets[i].ssid.ssid))
return -ENOBUFS;
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
nla_nest_end(msg, match);
}
nla_nest_end(msg, matches);
}
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
scan_plans = nla_nest_start(msg, NL80211_ATTR_SCHED_SCAN_PLANS);
if (!scan_plans)
return -ENOBUFS;
for (i = 0; i < req->n_scan_plans; i++) {
scan_plan = nla_nest_start(msg, i + 1);
nl80211: always check nla_nest_start() return value If the message got full during nla_nest_start(), it can return NULL. None of the cases here seem like that can really happen, but check the return value nonetheless. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-14 07:55:57 +00:00
if (!scan_plan)
return -ENOBUFS;
cfg80211: Add multiple scan plans for scheduled scan Add the option to configure multiple 'scan plans' for scheduled scan. Each 'scan plan' defines the number of scan cycles and the interval between scans. The scan plans are executed in the order they were configured. The last scan plan will always run infinitely and thus defines only the interval between scans. The maximum number of scan plans supported by the device and the maximum number of iterations in a single scan plan are advertised to userspace so it can configure the scan plans appropriately. When scheduled scan results are received there is no way to know which scan plan is being currently executed, so there is no way to know when the next scan iteration will start. This is not a problem, however. The scan start timestamp is only used for flushing old scan results, and there is no difference between flushing all results received until the end of the previous iteration or the start of the current one, since no results will be received in between. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-12 06:51:34 +00:00
if (!scan_plan ||
nla_put_u32(msg, NL80211_SCHED_SCAN_PLAN_INTERVAL,
req->scan_plans[i].interval) ||
(req->scan_plans[i].iterations &&
nla_put_u32(msg, NL80211_SCHED_SCAN_PLAN_ITERATIONS,
req->scan_plans[i].iterations)))
return -ENOBUFS;
nla_nest_end(msg, scan_plan);
}
nla_nest_end(msg, scan_plans);
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
nla_nest_end(msg, nd);
return 0;
}
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
static int nl80211_get_wowlan(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct sk_buff *msg;
void *hdr;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
u32 size = NLMSG_DEFAULT_SIZE;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (!rdev->wiphy.wowlan)
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
return -EOPNOTSUPP;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if (rdev->wiphy.wowlan_config && rdev->wiphy.wowlan_config->tcp) {
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
/* adjust size to have room for all the data */
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
size += rdev->wiphy.wowlan_config->tcp->tokens_size +
rdev->wiphy.wowlan_config->tcp->payload_len +
rdev->wiphy.wowlan_config->tcp->wake_len +
rdev->wiphy.wowlan_config->tcp->wake_len / 8;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
}
msg = nlmsg_new(size, GFP_KERNEL);
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
if (!msg)
return -ENOMEM;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
NL80211_CMD_GET_WOWLAN);
if (!hdr)
goto nla_put_failure;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if (rdev->wiphy.wowlan_config) {
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
struct nlattr *nl_wowlan;
nl_wowlan = nla_nest_start(msg, NL80211_ATTR_WOWLAN_TRIGGERS);
if (!nl_wowlan)
goto nla_put_failure;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if ((rdev->wiphy.wowlan_config->any &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_ANY)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->disconnect &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_DISCONNECT)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->magic_pkt &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_MAGIC_PKT)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->gtk_rekey_failure &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_GTK_REKEY_FAILURE)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->eap_identity_req &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_EAP_IDENT_REQUEST)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->four_way_handshake &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_4WAY_HANDSHAKE)) ||
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
(rdev->wiphy.wowlan_config->rfkill_release &&
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_flag(msg, NL80211_WOWLAN_TRIG_RFKILL_RELEASE)))
goto nla_put_failure;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
if (nl80211_send_wowlan_patterns(msg, rdev))
goto nla_put_failure;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if (nl80211_send_wowlan_tcp(msg,
rdev->wiphy.wowlan_config->tcp))
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
goto nla_put_failure;
nl80211: send netdetect configuration info in NL80211_CMD_GET_WOWLAN Send the netdetect configuration information in the response to NL8021_CMD_GET_WOWLAN commands. This includes the scan interval, SSIDs to match and frequencies to scan. Additionally, add the NL80211_WOWLAN_TRIG_NET_DETECT with NL80211_ATTR_WOWLAN_TRIGGERS_SUPPORTED. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-09 12:06:37 +00:00
if (nl80211_send_wowlan_nd(
msg,
rdev->wiphy.wowlan_config->nd_config))
goto nla_put_failure;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nla_nest_end(msg, nl_wowlan);
}
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
nla_put_failure:
nlmsg_free(msg);
return -ENOBUFS;
}
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
static int nl80211_parse_wowlan_tcp(struct cfg80211_registered_device *rdev,
struct nlattr *attr,
struct cfg80211_wowlan *trig)
{
struct nlattr *tb[NUM_NL80211_WOWLAN_TCP];
struct cfg80211_wowlan_tcp *cfg;
struct nl80211_wowlan_tcp_data_token *tok = NULL;
struct nl80211_wowlan_tcp_data_seq *seq = NULL;
u32 size;
u32 data_size, wake_size, tokens_size = 0, wake_mask_size;
int err, port;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (!rdev->wiphy.wowlan->tcp)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, MAX_NL80211_WOWLAN_TCP, attr,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_wowlan_tcp_policy, NULL);
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
if (err)
return err;
if (!tb[NL80211_WOWLAN_TCP_SRC_IPV4] ||
!tb[NL80211_WOWLAN_TCP_DST_IPV4] ||
!tb[NL80211_WOWLAN_TCP_DST_MAC] ||
!tb[NL80211_WOWLAN_TCP_DST_PORT] ||
!tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD] ||
!tb[NL80211_WOWLAN_TCP_DATA_INTERVAL] ||
!tb[NL80211_WOWLAN_TCP_WAKE_PAYLOAD] ||
!tb[NL80211_WOWLAN_TCP_WAKE_MASK])
return -EINVAL;
data_size = nla_len(tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD]);
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (data_size > rdev->wiphy.wowlan->tcp->data_payload_max)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
if (nla_get_u32(tb[NL80211_WOWLAN_TCP_DATA_INTERVAL]) >
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
rdev->wiphy.wowlan->tcp->data_interval_max ||
cfg80211: prohibit zero keepalive interval It's not useful to specify a 0 keepalive interval, this would send too much data. Prohibit this to also avoid device issues. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-26 12:56:40 +00:00
nla_get_u32(tb[NL80211_WOWLAN_TCP_DATA_INTERVAL]) == 0)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
wake_size = nla_len(tb[NL80211_WOWLAN_TCP_WAKE_PAYLOAD]);
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (wake_size > rdev->wiphy.wowlan->tcp->wake_payload_max)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
wake_mask_size = nla_len(tb[NL80211_WOWLAN_TCP_WAKE_MASK]);
if (wake_mask_size != DIV_ROUND_UP(wake_size, 8))
return -EINVAL;
if (tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN]) {
u32 tokln = nla_len(tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN]);
tok = nla_data(tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD_TOKEN]);
tokens_size = tokln - sizeof(*tok);
if (!tok->len || tokens_size % tok->len)
return -EINVAL;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (!rdev->wiphy.wowlan->tcp->tok)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (tok->len > rdev->wiphy.wowlan->tcp->tok->max_len)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (tok->len < rdev->wiphy.wowlan->tcp->tok->min_len)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (tokens_size > rdev->wiphy.wowlan->tcp->tok->bufsize)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
if (tok->offset + tok->len > data_size)
return -EINVAL;
}
if (tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD_SEQ]) {
seq = nla_data(tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD_SEQ]);
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (!rdev->wiphy.wowlan->tcp->seq)
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
return -EINVAL;
if (seq->len == 0 || seq->len > 4)
return -EINVAL;
if (seq->len + seq->offset > data_size)
return -EINVAL;
}
size = sizeof(*cfg);
size += data_size;
size += wake_size + wake_mask_size;
size += tokens_size;
cfg = kzalloc(size, GFP_KERNEL);
if (!cfg)
return -ENOMEM;
netlink: implement nla_get_in_addr and nla_get_in6_addr Those are counterparts to nla_put_in_addr and nla_put_in6_addr. Signed-off-by: Jiri Benc <jbenc@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-03-29 14:59:26 +00:00
cfg->src = nla_get_in_addr(tb[NL80211_WOWLAN_TCP_SRC_IPV4]);
cfg->dst = nla_get_in_addr(tb[NL80211_WOWLAN_TCP_DST_IPV4]);
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
memcpy(cfg->dst_mac, nla_data(tb[NL80211_WOWLAN_TCP_DST_MAC]),
ETH_ALEN);
if (tb[NL80211_WOWLAN_TCP_SRC_PORT])
port = nla_get_u16(tb[NL80211_WOWLAN_TCP_SRC_PORT]);
else
port = 0;
#ifdef CONFIG_INET
/* allocate a socket and port for it and use it */
err = __sock_create(wiphy_net(&rdev->wiphy), PF_INET, SOCK_STREAM,
IPPROTO_TCP, &cfg->sock, 1);
if (err) {
kfree(cfg);
return err;
}
if (inet_csk_get_port(cfg->sock->sk, port)) {
sock_release(cfg->sock);
kfree(cfg);
return -EADDRINUSE;
}
cfg->src_port = inet_sk(cfg->sock->sk)->inet_num;
#else
if (!port) {
kfree(cfg);
return -EINVAL;
}
cfg->src_port = port;
#endif
cfg->dst_port = nla_get_u16(tb[NL80211_WOWLAN_TCP_DST_PORT]);
cfg->payload_len = data_size;
cfg->payload = (u8 *)cfg + sizeof(*cfg) + tokens_size;
memcpy((void *)cfg->payload,
nla_data(tb[NL80211_WOWLAN_TCP_DATA_PAYLOAD]),
data_size);
if (seq)
cfg->payload_seq = *seq;
cfg->data_interval = nla_get_u32(tb[NL80211_WOWLAN_TCP_DATA_INTERVAL]);
cfg->wake_len = wake_size;
cfg->wake_data = (u8 *)cfg + sizeof(*cfg) + tokens_size + data_size;
memcpy((void *)cfg->wake_data,
nla_data(tb[NL80211_WOWLAN_TCP_WAKE_PAYLOAD]),
wake_size);
cfg->wake_mask = (u8 *)cfg + sizeof(*cfg) + tokens_size +
data_size + wake_size;
memcpy((void *)cfg->wake_mask,
nla_data(tb[NL80211_WOWLAN_TCP_WAKE_MASK]),
wake_mask_size);
if (tok) {
cfg->tokens_size = tokens_size;
memcpy(&cfg->payload_tok, tok, sizeof(*tok) + tokens_size);
}
trig->tcp = cfg;
return 0;
}
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
static int nl80211_parse_wowlan_nd(struct cfg80211_registered_device *rdev,
const struct wiphy_wowlan_support *wowlan,
struct nlattr *attr,
struct cfg80211_wowlan *trig)
{
struct nlattr **tb;
int err;
tb = kzalloc(NUM_NL80211_ATTR * sizeof(*tb), GFP_KERNEL);
if (!tb)
return -ENOMEM;
if (!(wowlan->flags & WIPHY_WOWLAN_NET_DETECT)) {
err = -EOPNOTSUPP;
goto out;
}
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
err = nla_parse_nested(tb, NL80211_ATTR_MAX, attr, nl80211_policy,
NULL);
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
if (err)
goto out;
nl80211: fix validation of scheduled scan info for wowlan netdetect For wowlan netdetect a separate limit is defined for the number of matchsets. Currently, this limit is ignored and the regular limit for scheduled scan matchsets, ie. struct wiphy::max_match_sets, is used for the net-detect case as well. Cc: Johannes Berg <johannes@sipsolutions.net> Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-27 12:27:44 +00:00
trig->nd_config = nl80211_parse_sched_scan(&rdev->wiphy, NULL, tb,
wowlan->max_nd_match_sets);
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
err = PTR_ERR_OR_ZERO(trig->nd_config);
if (err)
trig->nd_config = NULL;
out:
kfree(tb);
return err;
}
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
static int nl80211_set_wowlan(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct nlattr *tb[NUM_NL80211_WOWLAN_TRIG];
struct cfg80211_wowlan new_triggers = {};
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
struct cfg80211_wowlan *ntrig;
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
const struct wiphy_wowlan_support *wowlan = rdev->wiphy.wowlan;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
int err, i;
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
bool prev_enabled = rdev->wiphy.wowlan_config;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
bool regular = false;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
cfg80211: take WoWLAN support information out of wiphy struct There's no need to take up the space for devices that don't support WoWLAN, and most drivers can even make the support data static const (except where it's modified at runtime.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-03 15:25:34 +00:00
if (!wowlan)
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
return -EOPNOTSUPP;
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
if (!info->attrs[NL80211_ATTR_WOWLAN_TRIGGERS]) {
cfg80211_rdev_free_wowlan(rdev);
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
rdev->wiphy.wowlan_config = NULL;
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
goto set_wakeup;
}
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, MAX_NL80211_WOWLAN_TRIG,
info->attrs[NL80211_ATTR_WOWLAN_TRIGGERS],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_wowlan_policy, info->extack);
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
if (err)
return err;
if (tb[NL80211_WOWLAN_TRIG_ANY]) {
if (!(wowlan->flags & WIPHY_WOWLAN_ANY))
return -EINVAL;
new_triggers.any = true;
}
if (tb[NL80211_WOWLAN_TRIG_DISCONNECT]) {
if (!(wowlan->flags & WIPHY_WOWLAN_DISCONNECT))
return -EINVAL;
new_triggers.disconnect = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
}
if (tb[NL80211_WOWLAN_TRIG_MAGIC_PKT]) {
if (!(wowlan->flags & WIPHY_WOWLAN_MAGIC_PKT))
return -EINVAL;
new_triggers.magic_pkt = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
}
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
if (tb[NL80211_WOWLAN_TRIG_GTK_REKEY_SUPPORTED])
return -EINVAL;
if (tb[NL80211_WOWLAN_TRIG_GTK_REKEY_FAILURE]) {
if (!(wowlan->flags & WIPHY_WOWLAN_GTK_REKEY_FAILURE))
return -EINVAL;
new_triggers.gtk_rekey_failure = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
}
if (tb[NL80211_WOWLAN_TRIG_EAP_IDENT_REQUEST]) {
if (!(wowlan->flags & WIPHY_WOWLAN_EAP_IDENTITY_REQ))
return -EINVAL;
new_triggers.eap_identity_req = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
}
if (tb[NL80211_WOWLAN_TRIG_4WAY_HANDSHAKE]) {
if (!(wowlan->flags & WIPHY_WOWLAN_4WAY_HANDSHAKE))
return -EINVAL;
new_triggers.four_way_handshake = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
}
if (tb[NL80211_WOWLAN_TRIG_RFKILL_RELEASE]) {
if (!(wowlan->flags & WIPHY_WOWLAN_RFKILL_RELEASE))
return -EINVAL;
new_triggers.rfkill_release = true;
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211: advertise GTK rekey support, new triggers Since we now have the necessary API in place to support GTK rekeying, applications will need to know whether it is supported by a device. Add a pseudo-trigger that is used only to advertise that capability. Also, add some new triggers that match what iwlagn devices can do. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-13 08:48:55 +00:00
}
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
if (tb[NL80211_WOWLAN_TRIG_PKT_PATTERN]) {
struct nlattr *pat;
int n_patterns = 0;
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
int rem, pat_len, mask_len, pkt_offset;
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
struct nlattr *pat_tb[NUM_NL80211_PKTPAT];
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nla_for_each_nested(pat, tb[NL80211_WOWLAN_TRIG_PKT_PATTERN],
rem)
n_patterns++;
if (n_patterns > wowlan->n_patterns)
return -EINVAL;
new_triggers.patterns = kcalloc(n_patterns,
sizeof(new_triggers.patterns[0]),
GFP_KERNEL);
if (!new_triggers.patterns)
return -ENOMEM;
new_triggers.n_patterns = n_patterns;
i = 0;
nla_for_each_nested(pat, tb[NL80211_WOWLAN_TRIG_PKT_PATTERN],
rem) {
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
u8 *mask_pat;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
nla_parse_nested(pat_tb, MAX_NL80211_PKTPAT, pat,
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
NULL, info->extack);
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
err = -EINVAL;
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
if (!pat_tb[NL80211_PKTPAT_MASK] ||
!pat_tb[NL80211_PKTPAT_PATTERN])
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
goto error;
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
pat_len = nla_len(pat_tb[NL80211_PKTPAT_PATTERN]);
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
mask_len = DIV_ROUND_UP(pat_len, 8);
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
if (nla_len(pat_tb[NL80211_PKTPAT_MASK]) != mask_len)
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
goto error;
if (pat_len > wowlan->pattern_max_len ||
pat_len < wowlan->pattern_min_len)
goto error;
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
if (!pat_tb[NL80211_PKTPAT_OFFSET])
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
pkt_offset = 0;
else
pkt_offset = nla_get_u32(
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
pat_tb[NL80211_PKTPAT_OFFSET]);
nl80211: add packet offset information for wowlan pattern If user knows the location of a wowlan pattern to be matched in Rx packet, he can provide an offset with the pattern. This will help drivers to ignore initial bytes and match the pattern efficiently. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [refactor pattern sending] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-12 20:16:26 +00:00
if (pkt_offset > wowlan->max_pkt_offset)
goto error;
new_triggers.patterns[i].pkt_offset = pkt_offset;
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
mask_pat = kmalloc(mask_len + pat_len, GFP_KERNEL);
if (!mask_pat) {
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
err = -ENOMEM;
goto error;
}
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
new_triggers.patterns[i].mask = mask_pat;
memcpy(mask_pat, nla_data(pat_tb[NL80211_PKTPAT_MASK]),
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
mask_len);
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
mask_pat += mask_len;
new_triggers.patterns[i].pattern = mask_pat;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
new_triggers.patterns[i].pattern_len = pat_len;
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
memcpy(mask_pat,
cfg80211/nl80211: rename packet pattern related structures and enums Currently packet patterns and it's enum/structures are used only for WoWLAN feature. As we intend to reuse them for new feature packet coalesce, they are renamed in this patch. Older names are kept for backward compatibility purpose. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-26 02:03:56 +00:00
nla_data(pat_tb[NL80211_PKTPAT_PATTERN]),
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
pat_len);
i++;
}
}
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
if (tb[NL80211_WOWLAN_TRIG_TCP_CONNECTION]) {
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
err = nl80211_parse_wowlan_tcp(
rdev, tb[NL80211_WOWLAN_TRIG_TCP_CONNECTION],
&new_triggers);
if (err)
goto error;
}
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
if (tb[NL80211_WOWLAN_TRIG_NET_DETECT]) {
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
regular = true;
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
err = nl80211_parse_wowlan_nd(
rdev, wowlan, tb[NL80211_WOWLAN_TRIG_NET_DETECT],
&new_triggers);
if (err)
goto error;
}
nl80211: prohibit mixing 'any' and regular wowlan triggers If the device supports waking up on 'any' signal - i.e. it continues operating as usual and wakes up the host on pretty much anything that happens, then it makes no sense to also configure the more restricted WoWLAN mode where the device operates more autonomously but also in a more restricted fashion. Currently only cw2100 supports both 'any' and other triggers, but it seems to be broken as it doesn't configure anything to the device, so we can't currently get into a situation where both even can correctly be configured. This is about to change (Intel devices are going to support both and have different behaviour depending on configuration) so make sure the conflicting modes cannot be configured. (It seems that cw2100 advertises 'any' and 'disconnect' as a means of saying that's what it will always do, but that isn't really the way this API was meant to be used nor does it actually mean anything as 'any' always implies 'disconnect' already, and the driver doesn't change device configuration in any way depending on the settings.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-01 07:10:13 +00:00
/* The 'any' trigger means the device continues operating more or less
* as in its normal operation mode and wakes up the host on most of the
* normal interrupts (like packet RX, ...)
* It therefore makes little sense to combine with the more constrained
* wakeup trigger modes.
*/
if (new_triggers.any && regular) {
err = -EINVAL;
goto error;
}
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
ntrig = kmemdup(&new_triggers, sizeof(new_triggers), GFP_KERNEL);
if (!ntrig) {
err = -ENOMEM;
goto error;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
}
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
cfg80211_rdev_free_wowlan(rdev);
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
rdev->wiphy.wowlan_config = ntrig;
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
nl80211: allow enabling WoWLAN without triggers It may be desirable to use WoWLAN without triggers to keep the connection alive to the AP while suspended. Allow this use by enabling WoWLAN without triggers if no triggers were requested. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 14:25:02 +00:00
set_wakeup:
cfg80211: make WoWLAN configuration available to drivers Make the current WoWLAN configuration available to drivers at runtime. This isn't really useful for the normal WoWLAN behaviour and accessing it can also be racy, but drivers may use it for testing the WoWLAN device behaviour while the host stays up & running to observe the device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 07:30:07 +00:00
if (rdev->ops->set_wakeup &&
prev_enabled != !!rdev->wiphy.wowlan_config)
rdev_set_wakeup(rdev, rdev->wiphy.wowlan_config);
cfg80211/mac80211: enable proper device_set_wakeup_enable handling In WoWLAN, we only get the triggers when we actually get to suspend. As a consequence, drivers currently don't know that the device should enable wakeup. However, the device_set_wakeup_enable() API is intended to be called when the wakeup is enabled, not later when needed. Add a new set_wakeup() call to cfg80211 and mac80211 to allow drivers to properly call device_set_wakeup_enable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-04 13:05:25 +00:00
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
return 0;
error:
for (i = 0; i < new_triggers.n_patterns; i++)
kfree(new_triggers.patterns[i].mask);
kfree(new_triggers.patterns);
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
if (new_triggers.tcp && new_triggers.tcp->sock)
sock_release(new_triggers.tcp->sock);
kfree(new_triggers.tcp);
nl80211: Fix potential memory leak in nl80211_set_wowlan Compared to cfg80211_rdev_free_wowlan in core.h, the error goto label lacks the freeing of nd_config. Fix that. Signed-off-by: Ola Olsson <ola.olsson@sonymobile.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-12-12 22:17:17 +00:00
kfree(new_triggers.nd_config);
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
return err;
}
cfg80211: don't allow WoWLAN support without CONFIG_PM When CONFIG_PM is disabled, no device can possibly support WoWLAN since it can't go to sleep to start with. Due to this, mac80211 had even rejected the hardware registration. By making all the code and data for WoWLAN depend on CONFIG_PM we can promote this runtime error to a compile-time error. Add #ifdef around all WoWLAN code to remove it in systems that don't need it as they never suspend. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Acked-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 07:23:48 +00:00
#endif
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
static int nl80211_send_coalesce_rules(struct sk_buff *msg,
struct cfg80211_registered_device *rdev)
{
struct nlattr *nl_pats, *nl_pat, *nl_rule, *nl_rules;
int i, j, pat_len;
struct cfg80211_coalesce_rules *rule;
if (!rdev->coalesce->n_rules)
return 0;
nl_rules = nla_nest_start(msg, NL80211_ATTR_COALESCE_RULE);
if (!nl_rules)
return -ENOBUFS;
for (i = 0; i < rdev->coalesce->n_rules; i++) {
nl_rule = nla_nest_start(msg, i + 1);
if (!nl_rule)
return -ENOBUFS;
rule = &rdev->coalesce->rules[i];
if (nla_put_u32(msg, NL80211_ATTR_COALESCE_RULE_DELAY,
rule->delay))
return -ENOBUFS;
if (nla_put_u32(msg, NL80211_ATTR_COALESCE_RULE_CONDITION,
rule->condition))
return -ENOBUFS;
nl_pats = nla_nest_start(msg,
NL80211_ATTR_COALESCE_RULE_PKT_PATTERN);
if (!nl_pats)
return -ENOBUFS;
for (j = 0; j < rule->n_patterns; j++) {
nl_pat = nla_nest_start(msg, j + 1);
if (!nl_pat)
return -ENOBUFS;
pat_len = rule->patterns[j].pattern_len;
if (nla_put(msg, NL80211_PKTPAT_MASK,
DIV_ROUND_UP(pat_len, 8),
rule->patterns[j].mask) ||
nla_put(msg, NL80211_PKTPAT_PATTERN, pat_len,
rule->patterns[j].pattern) ||
nla_put_u32(msg, NL80211_PKTPAT_OFFSET,
rule->patterns[j].pkt_offset))
return -ENOBUFS;
nla_nest_end(msg, nl_pat);
}
nla_nest_end(msg, nl_pats);
nla_nest_end(msg, nl_rule);
}
nla_nest_end(msg, nl_rules);
return 0;
}
static int nl80211_get_coalesce(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct sk_buff *msg;
void *hdr;
if (!rdev->wiphy.coalesce)
return -EOPNOTSUPP;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
NL80211_CMD_GET_COALESCE);
if (!hdr)
goto nla_put_failure;
if (rdev->coalesce && nl80211_send_coalesce_rules(msg, rdev))
goto nla_put_failure;
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
nla_put_failure:
nlmsg_free(msg);
return -ENOBUFS;
}
void cfg80211_rdev_free_coalesce(struct cfg80211_registered_device *rdev)
{
struct cfg80211_coalesce *coalesce = rdev->coalesce;
int i, j;
struct cfg80211_coalesce_rules *rule;
if (!coalesce)
return;
for (i = 0; i < coalesce->n_rules; i++) {
rule = &coalesce->rules[i];
for (j = 0; j < rule->n_patterns; j++)
kfree(rule->patterns[j].mask);
kfree(rule->patterns);
}
kfree(coalesce->rules);
kfree(coalesce);
rdev->coalesce = NULL;
}
static int nl80211_parse_coalesce_rule(struct cfg80211_registered_device *rdev,
struct nlattr *rule,
struct cfg80211_coalesce_rules *new_rule)
{
int err, i;
const struct wiphy_coalesce_support *coalesce = rdev->wiphy.coalesce;
struct nlattr *tb[NUM_NL80211_ATTR_COALESCE_RULE], *pat;
int rem, pat_len, mask_len, pkt_offset, n_patterns = 0;
struct nlattr *pat_tb[NUM_NL80211_PKTPAT];
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, NL80211_ATTR_COALESCE_RULE_MAX, rule,
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nl80211_coalesce_policy, NULL);
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
if (err)
return err;
if (tb[NL80211_ATTR_COALESCE_RULE_DELAY])
new_rule->delay =
nla_get_u32(tb[NL80211_ATTR_COALESCE_RULE_DELAY]);
if (new_rule->delay > coalesce->max_delay)
return -EINVAL;
if (tb[NL80211_ATTR_COALESCE_RULE_CONDITION])
new_rule->condition =
nla_get_u32(tb[NL80211_ATTR_COALESCE_RULE_CONDITION]);
if (new_rule->condition != NL80211_COALESCE_CONDITION_MATCH &&
new_rule->condition != NL80211_COALESCE_CONDITION_NO_MATCH)
return -EINVAL;
if (!tb[NL80211_ATTR_COALESCE_RULE_PKT_PATTERN])
return -EINVAL;
nla_for_each_nested(pat, tb[NL80211_ATTR_COALESCE_RULE_PKT_PATTERN],
rem)
n_patterns++;
if (n_patterns > coalesce->n_patterns)
return -EINVAL;
new_rule->patterns = kcalloc(n_patterns, sizeof(new_rule->patterns[0]),
GFP_KERNEL);
if (!new_rule->patterns)
return -ENOMEM;
new_rule->n_patterns = n_patterns;
i = 0;
nla_for_each_nested(pat, tb[NL80211_ATTR_COALESCE_RULE_PKT_PATTERN],
rem) {
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
u8 *mask_pat;
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
nla_parse_nested(pat_tb, MAX_NL80211_PKTPAT, pat, NULL, NULL);
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
if (!pat_tb[NL80211_PKTPAT_MASK] ||
!pat_tb[NL80211_PKTPAT_PATTERN])
return -EINVAL;
pat_len = nla_len(pat_tb[NL80211_PKTPAT_PATTERN]);
mask_len = DIV_ROUND_UP(pat_len, 8);
if (nla_len(pat_tb[NL80211_PKTPAT_MASK]) != mask_len)
return -EINVAL;
if (pat_len > coalesce->pattern_max_len ||
pat_len < coalesce->pattern_min_len)
return -EINVAL;
if (!pat_tb[NL80211_PKTPAT_OFFSET])
pkt_offset = 0;
else
pkt_offset = nla_get_u32(pat_tb[NL80211_PKTPAT_OFFSET]);
if (pkt_offset > coalesce->max_pkt_offset)
return -EINVAL;
new_rule->patterns[i].pkt_offset = pkt_offset;
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
mask_pat = kmalloc(mask_len + pat_len, GFP_KERNEL);
if (!mask_pat)
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
return -ENOMEM;
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
new_rule->patterns[i].mask = mask_pat;
memcpy(mask_pat, nla_data(pat_tb[NL80211_PKTPAT_MASK]),
mask_len);
mask_pat += mask_len;
new_rule->patterns[i].pattern = mask_pat;
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
new_rule->patterns[i].pattern_len = pat_len;
cfg80211: constify wowlan/coalesce mask/pattern pointers This requires changing the nl80211 parsing code a bit to use intermediate pointers for the allocation, but clarifies the API towards the drivers. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-19 15:59:50 +00:00
memcpy(mask_pat, nla_data(pat_tb[NL80211_PKTPAT_PATTERN]),
pat_len);
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
i++;
}
return 0;
}
static int nl80211_set_coalesce(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
const struct wiphy_coalesce_support *coalesce = rdev->wiphy.coalesce;
struct cfg80211_coalesce new_coalesce = {};
struct cfg80211_coalesce *n_coalesce;
int err, rem_rule, n_rules = 0, i, j;
struct nlattr *rule;
struct cfg80211_coalesce_rules *tmp_rule;
if (!rdev->wiphy.coalesce || !rdev->ops->set_coalesce)
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_COALESCE_RULE]) {
cfg80211_rdev_free_coalesce(rdev);
cfg80211: Add missing tracing to cfg80211 Add missing tracing for: 1. start_radar_detection() 2. set_mcast_rates() 3. set_coalesce() Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-22 19:27:46 +00:00
rdev_set_coalesce(rdev, NULL);
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
return 0;
}
nla_for_each_nested(rule, info->attrs[NL80211_ATTR_COALESCE_RULE],
rem_rule)
n_rules++;
if (n_rules > coalesce->n_rules)
return -EINVAL;
new_coalesce.rules = kcalloc(n_rules, sizeof(new_coalesce.rules[0]),
GFP_KERNEL);
if (!new_coalesce.rules)
return -ENOMEM;
new_coalesce.n_rules = n_rules;
i = 0;
nla_for_each_nested(rule, info->attrs[NL80211_ATTR_COALESCE_RULE],
rem_rule) {
err = nl80211_parse_coalesce_rule(rdev, rule,
&new_coalesce.rules[i]);
if (err)
goto error;
i++;
}
cfg80211: Add missing tracing to cfg80211 Add missing tracing for: 1. start_radar_detection() 2. set_mcast_rates() 3. set_coalesce() Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-22 19:27:46 +00:00
err = rdev_set_coalesce(rdev, &new_coalesce);
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
if (err)
goto error;
n_coalesce = kmemdup(&new_coalesce, sizeof(new_coalesce), GFP_KERNEL);
if (!n_coalesce) {
err = -ENOMEM;
goto error;
}
cfg80211_rdev_free_coalesce(rdev);
rdev->coalesce = n_coalesce;
return 0;
error:
for (i = 0; i < new_coalesce.n_rules; i++) {
tmp_rule = &new_coalesce.rules[i];
for (j = 0; j < tmp_rule->n_patterns; j++)
kfree(tmp_rule->patterns[j].mask);
kfree(tmp_rule->patterns);
}
kfree(new_coalesce.rules);
return err;
}
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
static int nl80211_set_rekey_data(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct nlattr *tb[NUM_NL80211_REKEY_DATA];
struct cfg80211_gtk_rekey_data rekey_data;
int err;
if (!info->attrs[NL80211_ATTR_REKEY_DATA])
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, MAX_NL80211_REKEY_DATA,
info->attrs[NL80211_ATTR_REKEY_DATA],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_rekey_policy, info->extack);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
if (err)
return err;
if (nla_len(tb[NL80211_REKEY_DATA_REPLAY_CTR]) != NL80211_REPLAY_CTR_LEN)
return -ERANGE;
if (nla_len(tb[NL80211_REKEY_DATA_KEK]) != NL80211_KEK_LEN)
return -ERANGE;
if (nla_len(tb[NL80211_REKEY_DATA_KCK]) != NL80211_KCK_LEN)
return -ERANGE;
cfg80211: don't put kek/kck/replay counter on the stack There's no need to put the values on the stack, just pass a pointer to the data in the nl80211 message. This reduces stack usage and avoids potential issues with putting sensitive data on the stack. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 19:28:06 +00:00
rekey_data.kek = nla_data(tb[NL80211_REKEY_DATA_KEK]);
rekey_data.kck = nla_data(tb[NL80211_REKEY_DATA_KCK]);
rekey_data.replay_ctr = nla_data(tb[NL80211_REKEY_DATA_REPLAY_CTR]);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
wdev_lock(wdev);
if (!wdev->current_bss) {
err = -ENOTCONN;
goto out;
}
if (!rdev->ops->set_rekey_data) {
err = -EOPNOTSUPP;
goto out;
}
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_set_rekey_data(rdev, dev, &rekey_data);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
out:
wdev_unlock(wdev);
return err;
}
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
static int nl80211_register_unexpected_frame(struct sk_buff *skb,
struct genl_info *info)
{
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
if (wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO)
return -EINVAL;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (wdev->ap_unexpected_nlportid)
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
return -EBUSY;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
wdev->ap_unexpected_nlportid = info->snd_portid;
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
return 0;
}
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
static int nl80211_probe_client(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct sk_buff *msg;
void *hdr;
const u8 *addr;
u64 cookie;
int err;
if (wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
if (!rdev->ops->probe_client)
return -EOPNOTSUPP;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
NL80211_CMD_PROBE_CLIENT);
nl80211: nl80211hdr_put() doesn't return an ERR_PTR There are a few places which check nl80211hdr_put() for an ERR_PTR but actually it returns NULL on error and never error values. In nl80211_testmode_dump() the return wasn't checked at all so I have added one. Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com> [some whitespace changes] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-14 11:50:01 +00:00
if (!hdr) {
err = -ENOBUFS;
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
goto free_msg;
}
addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
cfg80211: add wrappers for registered_device_ops This will allow adding central tracing like in mac80211. Signed-off-by: Hila Gonen <hila.gonen@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 14:19:42 +00:00
err = rdev_probe_client(rdev, dev, addr, &cookie);
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
if (err)
goto free_msg;
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
nla_put_failure:
err = -ENOBUFS;
free_msg:
nlmsg_free(msg);
return err;
}
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
static int nl80211_register_beacons(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
struct cfg80211_beacon_registration *reg, *nreg;
int rv;
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
if (!(rdev->wiphy.flags & WIPHY_FLAG_REPORTS_OBSS))
return -EOPNOTSUPP;
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
nreg = kzalloc(sizeof(*nreg), GFP_KERNEL);
if (!nreg)
return -ENOMEM;
/* First, check if already registered. */
spin_lock_bh(&rdev->beacon_registrations_lock);
list_for_each_entry(reg, &rdev->beacon_registrations, list) {
if (reg->nlportid == info->snd_portid) {
rv = -EALREADY;
goto out_err;
}
}
/* Add it to the list */
nreg->nlportid = info->snd_portid;
list_add(&nreg->list, &rdev->beacon_registrations);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
spin_unlock_bh(&rdev->beacon_registrations_lock);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
return 0;
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
out_err:
spin_unlock_bh(&rdev->beacon_registrations_lock);
kfree(nreg);
return rv;
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
}
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
static int nl80211_start_p2p_device(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
int err;
if (!rdev->ops->start_p2p_device)
return -EOPNOTSUPP;
if (wdev->iftype != NL80211_IFTYPE_P2P_DEVICE)
return -EOPNOTSUPP;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (wdev_running(wdev))
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
return 0;
cfg80211/mac80211: move more combination checks to mac80211 Get rid of the cfg80211_can_add_interface() and cfg80211_can_change_interface() functions by moving that functionality to mac80211. With this patch all interface combination checks are now out of cfg80211 (except for the channel switch case which will be addressed in a future commit). Additionally, modify the ieee80211_check_combinations() function so that an undefined chandef can be passed, in order to use it before a channel is defined. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-02-27 09:07:21 +00:00
if (rfkill_blocked(rdev->rfkill))
return -ERFKILL;
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
cfg80211: add tracing for P2P Device start/stop These were missed due to the tracing work having started on a kernel that didn't have P2P Device yet, implement them now. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-23 13:16:50 +00:00
err = rdev_start_p2p_device(rdev, wdev);
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
if (err)
return err;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
wdev->is_running = true;
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
rdev->opencount++;
return 0;
}
static int nl80211_stop_p2p_device(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
if (wdev->iftype != NL80211_IFTYPE_P2P_DEVICE)
return -EOPNOTSUPP;
if (!rdev->ops->stop_p2p_device)
return -EOPNOTSUPP;
cfg80211: always check for scan end on P2P device If a P2P device wdev is removed while it has a scan, then the scan completion might crash later as it is already freed by that time. To avoid the crash always check the scan completion when the P2P device is being removed for some reason. If the driver already canceled it, don't want and free it, otherwise warn and leak it to avoid later crashes. In order to do this, locking needs to be changed away from the rdev mutex (which can't always be guaranteed). For now, use the sched_scan_mtx instead, I'll rename it to just scan_mtx in a later patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-03-19 14:04:07 +00:00
cfg80211_stop_p2p_device(rdev, wdev);
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
return 0;
}
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
static int nl80211_start_nan(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
struct cfg80211_nan_conf conf = {};
int err;
if (wdev->iftype != NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
nl80211: fix logic inversion in start_nan() Arend inadvertently inverted the logic while converting to wdev_running(), fix that. Fixes: 73c7da3dae1e ("cfg80211: add generic helper to check interface is running") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-11-21 12:55:48 +00:00
if (wdev_running(wdev))
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
return -EEXIST;
if (rfkill_blocked(rdev->rfkill))
return -ERFKILL;
if (!info->attrs[NL80211_ATTR_NAN_MASTER_PREF])
return -EINVAL;
conf.master_pref =
nla_get_u8(info->attrs[NL80211_ATTR_NAN_MASTER_PREF]);
if (!conf.master_pref)
return -EINVAL;
cfg80211: fix NAN bands definition The nl80211_nan_dual_band_conf enumeration doesn't make much sense. The default value is assigned to a bit, which makes it weird if the default bit and other bits are set at the same time. To improve this, get rid of NL80211_NAN_BAND_DEFAULT and add a wiphy configuration to let the drivers define which bands are supported. This is exposed to the userspace, which then can make a decision on which band(s) to use. Additionally, rename all "dual_band" elements to "bands", to make things clearer. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-08 13:00:34 +00:00
if (info->attrs[NL80211_ATTR_BANDS]) {
u32 bands = nla_get_u32(info->attrs[NL80211_ATTR_BANDS]);
if (bands & ~(u32)wdev->wiphy->nan_supported_bands)
return -EOPNOTSUPP;
if (bands && !(bands & BIT(NL80211_BAND_2GHZ)))
return -EINVAL;
conf.bands = bands;
}
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
err = rdev_start_nan(rdev, wdev, &conf);
if (err)
return err;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
wdev->is_running = true;
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
rdev->opencount++;
return 0;
}
static int nl80211_stop_nan(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
if (wdev->iftype != NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
cfg80211_stop_nan(rdev, wdev);
return 0;
}
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
static int validate_nan_filter(struct nlattr *filter_attr)
{
struct nlattr *attr;
int len = 0, n_entries = 0, rem;
nla_for_each_nested(attr, filter_attr, rem) {
len += nla_len(attr);
n_entries++;
}
if (len >= U8_MAX)
return -EINVAL;
return n_entries;
}
static int handle_nan_filter(struct nlattr *attr_filter,
struct cfg80211_nan_func *func,
bool tx)
{
struct nlattr *attr;
int n_entries, rem, i;
struct cfg80211_nan_func_filter *filter;
n_entries = validate_nan_filter(attr_filter);
if (n_entries < 0)
return n_entries;
BUILD_BUG_ON(sizeof(*func->rx_filters) != sizeof(*func->tx_filters));
filter = kcalloc(n_entries, sizeof(*func->rx_filters), GFP_KERNEL);
if (!filter)
return -ENOMEM;
i = 0;
nla_for_each_nested(attr, attr_filter, rem) {
netlink: Add nla_memdup() to wrap kmemdup() use on nlattr Wrap several common instances of: kmemdup(nla_data(attr), nla_len(attr), GFP_KERNEL); Signed-off-by: Thomas Graf <tgraf@suug.ch> Acked-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Daniel Borkmann <daniel@iogearbox.net> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-26 08:53:16 +00:00
filter[i].filter = nla_memdup(attr, GFP_KERNEL);
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
filter[i].len = nla_len(attr);
i++;
}
if (tx) {
func->num_tx_filters = n_entries;
func->tx_filters = filter;
} else {
func->num_rx_filters = n_entries;
func->rx_filters = filter;
}
return 0;
}
static int nl80211_nan_add_func(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
struct nlattr *tb[NUM_NL80211_NAN_FUNC_ATTR], *func_attr;
struct cfg80211_nan_func *func;
struct sk_buff *msg = NULL;
void *hdr = NULL;
int err = 0;
if (wdev->iftype != NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (!wdev_running(wdev))
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
return -ENOTCONN;
if (!info->attrs[NL80211_ATTR_NAN_FUNC])
return -EINVAL;
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(tb, NL80211_NAN_FUNC_ATTR_MAX,
info->attrs[NL80211_ATTR_NAN_FUNC],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_nan_func_policy, info->extack);
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
if (err)
return err;
func = kzalloc(sizeof(*func), GFP_KERNEL);
if (!func)
return -ENOMEM;
func->cookie = wdev->wiphy->cookie_counter++;
if (!tb[NL80211_NAN_FUNC_TYPE] ||
nla_get_u8(tb[NL80211_NAN_FUNC_TYPE]) > NL80211_NAN_FUNC_MAX_TYPE) {
err = -EINVAL;
goto out;
}
func->type = nla_get_u8(tb[NL80211_NAN_FUNC_TYPE]);
if (!tb[NL80211_NAN_FUNC_SERVICE_ID]) {
err = -EINVAL;
goto out;
}
memcpy(func->service_id, nla_data(tb[NL80211_NAN_FUNC_SERVICE_ID]),
sizeof(func->service_id));
func->close_range =
nla_get_flag(tb[NL80211_NAN_FUNC_CLOSE_RANGE]);
if (tb[NL80211_NAN_FUNC_SERVICE_INFO]) {
func->serv_spec_info_len =
nla_len(tb[NL80211_NAN_FUNC_SERVICE_INFO]);
func->serv_spec_info =
kmemdup(nla_data(tb[NL80211_NAN_FUNC_SERVICE_INFO]),
func->serv_spec_info_len,
GFP_KERNEL);
if (!func->serv_spec_info) {
err = -ENOMEM;
goto out;
}
}
if (tb[NL80211_NAN_FUNC_TTL])
func->ttl = nla_get_u32(tb[NL80211_NAN_FUNC_TTL]);
switch (func->type) {
case NL80211_NAN_FUNC_PUBLISH:
if (!tb[NL80211_NAN_FUNC_PUBLISH_TYPE]) {
err = -EINVAL;
goto out;
}
func->publish_type =
nla_get_u8(tb[NL80211_NAN_FUNC_PUBLISH_TYPE]);
func->publish_bcast =
nla_get_flag(tb[NL80211_NAN_FUNC_PUBLISH_BCAST]);
if ((!(func->publish_type & NL80211_NAN_SOLICITED_PUBLISH)) &&
func->publish_bcast) {
err = -EINVAL;
goto out;
}
break;
case NL80211_NAN_FUNC_SUBSCRIBE:
func->subscribe_active =
nla_get_flag(tb[NL80211_NAN_FUNC_SUBSCRIBE_ACTIVE]);
break;
case NL80211_NAN_FUNC_FOLLOW_UP:
if (!tb[NL80211_NAN_FUNC_FOLLOW_UP_ID] ||
!tb[NL80211_NAN_FUNC_FOLLOW_UP_REQ_ID]) {
err = -EINVAL;
goto out;
}
func->followup_id =
nla_get_u8(tb[NL80211_NAN_FUNC_FOLLOW_UP_ID]);
func->followup_reqid =
nla_get_u8(tb[NL80211_NAN_FUNC_FOLLOW_UP_REQ_ID]);
memcpy(func->followup_dest.addr,
nla_data(tb[NL80211_NAN_FUNC_FOLLOW_UP_DEST]),
sizeof(func->followup_dest.addr));
if (func->ttl) {
err = -EINVAL;
goto out;
}
break;
default:
err = -EINVAL;
goto out;
}
if (tb[NL80211_NAN_FUNC_SRF]) {
struct nlattr *srf_tb[NUM_NL80211_NAN_SRF_ATTR];
nl80211: use nla_parse_nested() instead of nla_parse() It's just an inline doing the same thing, but the code is nicer with it. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-26 12:42:21 +00:00
err = nla_parse_nested(srf_tb, NL80211_NAN_SRF_ATTR_MAX,
tb[NL80211_NAN_FUNC_SRF],
netlink: pass extended ACK struct where available This is an add-on to the previous patch that passes the extended ACK structure where it's already available by existing genl_info or extack function arguments. This was done with this spatch (with some manual adjustment of indentation): @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, info->extack) ... } @@ expression A, B, C, D, E; identifier fn, info; @@ fn(..., struct genl_info *info, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, info->extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse(A, B, C, D, E, NULL) +nla_parse(A, B, C, D, E, extack) ...> } @@ expression A, B, C, D, E; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { ... -nlmsg_parse(A, B, C, D, E, NULL) +nlmsg_parse(A, B, C, D, E, extack) ... } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_parse_nested(A, B, C, D, NULL) +nla_parse_nested(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nlmsg_validate(A, B, C, D, NULL) +nlmsg_validate(A, B, C, D, extack) ...> } @@ expression A, B, C, D; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate(A, B, C, D, NULL) +nla_validate(A, B, C, D, extack) ...> } @@ expression A, B, C; identifier fn, extack; @@ fn(..., struct netlink_ext_ack *extack, ...) { <... -nla_validate_nested(A, B, C, NULL) +nla_validate_nested(A, B, C, extack) ...> } Signed-off-by: Johannes Berg <johannes.berg@intel.com> Reviewed-by: Jiri Pirko <jiri@mellanox.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:08 +00:00
nl80211_nan_srf_policy, info->extack);
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
if (err)
goto out;
func->srf_include =
nla_get_flag(srf_tb[NL80211_NAN_SRF_INCLUDE]);
if (srf_tb[NL80211_NAN_SRF_BF]) {
if (srf_tb[NL80211_NAN_SRF_MAC_ADDRS] ||
!srf_tb[NL80211_NAN_SRF_BF_IDX]) {
err = -EINVAL;
goto out;
}
func->srf_bf_len =
nla_len(srf_tb[NL80211_NAN_SRF_BF]);
func->srf_bf =
kmemdup(nla_data(srf_tb[NL80211_NAN_SRF_BF]),
func->srf_bf_len, GFP_KERNEL);
if (!func->srf_bf) {
err = -ENOMEM;
goto out;
}
func->srf_bf_idx =
nla_get_u8(srf_tb[NL80211_NAN_SRF_BF_IDX]);
} else {
struct nlattr *attr, *mac_attr =
srf_tb[NL80211_NAN_SRF_MAC_ADDRS];
int n_entries, rem, i = 0;
if (!mac_attr) {
err = -EINVAL;
goto out;
}
n_entries = validate_acl_mac_addrs(mac_attr);
if (n_entries <= 0) {
err = -EINVAL;
goto out;
}
func->srf_num_macs = n_entries;
func->srf_macs =
kzalloc(sizeof(*func->srf_macs) * n_entries,
GFP_KERNEL);
if (!func->srf_macs) {
err = -ENOMEM;
goto out;
}
nla_for_each_nested(attr, mac_attr, rem)
memcpy(func->srf_macs[i++].addr, nla_data(attr),
sizeof(*func->srf_macs));
}
}
if (tb[NL80211_NAN_FUNC_TX_MATCH_FILTER]) {
err = handle_nan_filter(tb[NL80211_NAN_FUNC_TX_MATCH_FILTER],
func, true);
if (err)
goto out;
}
if (tb[NL80211_NAN_FUNC_RX_MATCH_FILTER]) {
err = handle_nan_filter(tb[NL80211_NAN_FUNC_RX_MATCH_FILTER],
func, false);
if (err)
goto out;
}
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg) {
err = -ENOMEM;
goto out;
}
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
NL80211_CMD_ADD_NAN_FUNCTION);
/* This can't really happen - we just allocated 4KB */
if (WARN_ON(!hdr)) {
err = -ENOMEM;
goto out;
}
err = rdev_add_nan_func(rdev, wdev, func);
out:
if (err < 0) {
cfg80211_free_nan_func(func);
nlmsg_free(msg);
return err;
}
/* propagate the instance id and cookie to userspace */
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, func->cookie,
NL80211_ATTR_PAD))
goto nla_put_failure;
func_attr = nla_nest_start(msg, NL80211_ATTR_NAN_FUNC);
if (!func_attr)
goto nla_put_failure;
if (nla_put_u8(msg, NL80211_NAN_FUNC_INSTANCE_ID,
func->instance_id))
goto nla_put_failure;
nla_nest_end(msg, func_attr);
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
nla_put_failure:
nlmsg_free(msg);
return -ENOBUFS;
}
static int nl80211_nan_del_func(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
u64 cookie;
if (wdev->iftype != NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (!wdev_running(wdev))
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
return -ENOTCONN;
if (!info->attrs[NL80211_ATTR_COOKIE])
return -EINVAL;
cookie = nla_get_u64(info->attrs[NL80211_ATTR_COOKIE]);
rdev_del_nan_func(rdev, wdev, cookie);
return 0;
}
cfg80211: allow the user space to change current NAN configuration Some NAN configuration paramaters may change during the operation of the NAN device. For example, a user may want to update master preference value when the device gets plugged/unplugged to the power. Add API that allows to do so. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:16 +00:00
static int nl80211_nan_change_config(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
struct cfg80211_nan_conf conf = {};
u32 changed = 0;
if (wdev->iftype != NL80211_IFTYPE_NAN)
return -EOPNOTSUPP;
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (!wdev_running(wdev))
cfg80211: allow the user space to change current NAN configuration Some NAN configuration paramaters may change during the operation of the NAN device. For example, a user may want to update master preference value when the device gets plugged/unplugged to the power. Add API that allows to do so. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:16 +00:00
return -ENOTCONN;
if (info->attrs[NL80211_ATTR_NAN_MASTER_PREF]) {
conf.master_pref =
nla_get_u8(info->attrs[NL80211_ATTR_NAN_MASTER_PREF]);
if (conf.master_pref <= 1 || conf.master_pref == 255)
return -EINVAL;
changed |= CFG80211_NAN_CONF_CHANGED_PREF;
}
cfg80211: fix NAN bands definition The nl80211_nan_dual_band_conf enumeration doesn't make much sense. The default value is assigned to a bit, which makes it weird if the default bit and other bits are set at the same time. To improve this, get rid of NL80211_NAN_BAND_DEFAULT and add a wiphy configuration to let the drivers define which bands are supported. This is exposed to the userspace, which then can make a decision on which band(s) to use. Additionally, rename all "dual_band" elements to "bands", to make things clearer. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-08 13:00:34 +00:00
if (info->attrs[NL80211_ATTR_BANDS]) {
u32 bands = nla_get_u32(info->attrs[NL80211_ATTR_BANDS]);
if (bands & ~(u32)wdev->wiphy->nan_supported_bands)
return -EOPNOTSUPP;
if (bands && !(bands & BIT(NL80211_BAND_2GHZ)))
return -EINVAL;
conf.bands = bands;
changed |= CFG80211_NAN_CONF_CHANGED_BANDS;
cfg80211: allow the user space to change current NAN configuration Some NAN configuration paramaters may change during the operation of the NAN device. For example, a user may want to update master preference value when the device gets plugged/unplugged to the power. Add API that allows to do so. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:16 +00:00
}
if (!changed)
return -EINVAL;
return rdev_nan_change_conf(rdev, wdev, &conf, changed);
}
cfg80211: provide a function to report a match for NAN Provide a function the driver can call to report a match. This will send the event to the user space. If the NAN instance is tied to the owner, the notifications will be sent to the socket that started the NAN interface only. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:17 +00:00
void cfg80211_nan_match(struct wireless_dev *wdev,
struct cfg80211_nan_match_params *match, gfp_t gfp)
{
struct wiphy *wiphy = wdev->wiphy;
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
struct nlattr *match_attr, *local_func_attr, *peer_func_attr;
struct sk_buff *msg;
void *hdr;
if (WARN_ON(!match->inst_id || !match->peer_inst_id || !match->addr))
return;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_NAN_MATCH);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
(wdev->netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex)) ||
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
goto nla_put_failure;
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, match->cookie,
NL80211_ATTR_PAD) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, match->addr))
goto nla_put_failure;
match_attr = nla_nest_start(msg, NL80211_ATTR_NAN_MATCH);
if (!match_attr)
goto nla_put_failure;
local_func_attr = nla_nest_start(msg, NL80211_NAN_MATCH_FUNC_LOCAL);
if (!local_func_attr)
goto nla_put_failure;
if (nla_put_u8(msg, NL80211_NAN_FUNC_INSTANCE_ID, match->inst_id))
goto nla_put_failure;
nla_nest_end(msg, local_func_attr);
peer_func_attr = nla_nest_start(msg, NL80211_NAN_MATCH_FUNC_PEER);
if (!peer_func_attr)
goto nla_put_failure;
if (nla_put_u8(msg, NL80211_NAN_FUNC_TYPE, match->type) ||
nla_put_u8(msg, NL80211_NAN_FUNC_INSTANCE_ID, match->peer_inst_id))
goto nla_put_failure;
if (match->info && match->info_len &&
nla_put(msg, NL80211_NAN_FUNC_SERVICE_INFO, match->info_len,
match->info))
goto nla_put_failure;
nla_nest_end(msg, peer_func_attr);
nla_nest_end(msg, match_attr);
genlmsg_end(msg, hdr);
if (!wdev->owner_nlportid)
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy),
msg, 0, NL80211_MCGRP_NAN, gfp);
else
genlmsg_unicast(wiphy_net(&rdev->wiphy), msg,
wdev->owner_nlportid);
return;
nla_put_failure:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_nan_match);
cfg80211: Provide an API to report NAN function termination Provide a function that reports NAN DE function termination. The function may be terminated due to one of the following reasons: user request, ttl expiration or failure. If the NAN instance is tied to the owner, the notification will be sent to the socket that started the NAN interface only Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:18 +00:00
void cfg80211_nan_func_terminated(struct wireless_dev *wdev,
u8 inst_id,
enum nl80211_nan_func_term_reason reason,
u64 cookie, gfp_t gfp)
{
struct wiphy *wiphy = wdev->wiphy;
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
struct sk_buff *msg;
struct nlattr *func_attr;
void *hdr;
if (WARN_ON(!inst_id))
return;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_DEL_NAN_FUNCTION);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
(wdev->netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex)) ||
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
goto nla_put_failure;
if (nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD))
goto nla_put_failure;
func_attr = nla_nest_start(msg, NL80211_ATTR_NAN_FUNC);
if (!func_attr)
goto nla_put_failure;
if (nla_put_u8(msg, NL80211_NAN_FUNC_INSTANCE_ID, inst_id) ||
nla_put_u8(msg, NL80211_NAN_FUNC_TERM_REASON, reason))
goto nla_put_failure;
nla_nest_end(msg, func_attr);
genlmsg_end(msg, hdr);
if (!wdev->owner_nlportid)
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy),
msg, 0, NL80211_MCGRP_NAN, gfp);
else
genlmsg_unicast(wiphy_net(&rdev->wiphy), msg,
wdev->owner_nlportid);
return;
nla_put_failure:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_nan_func_terminated);
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
static int nl80211_get_protocol_features(struct sk_buff *skb,
struct genl_info *info)
{
void *hdr;
struct sk_buff *msg;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return -ENOMEM;
hdr = nl80211hdr_put(msg, info->snd_portid, info->snd_seq, 0,
NL80211_CMD_GET_PROTOCOL_FEATURES);
if (!hdr)
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_ATTR_PROTOCOL_FEATURES,
NL80211_PROTOCOL_FEATURE_SPLIT_WIPHY_DUMP))
goto nla_put_failure;
genlmsg_end(msg, hdr);
return genlmsg_reply(msg, info);
nla_put_failure:
kfree_skb(msg);
return -ENOBUFS;
}
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
static int nl80211_update_ft_ies(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct cfg80211_update_ft_ies_params ft_params;
struct net_device *dev = info->user_ptr[1];
if (!rdev->ops->update_ft_ies)
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_MDID] ||
!is_valid_ie_attr(info->attrs[NL80211_ATTR_IE]))
return -EINVAL;
memset(&ft_params, 0, sizeof(ft_params));
ft_params.md = nla_get_u16(info->attrs[NL80211_ATTR_MDID]);
ft_params.ie = nla_data(info->attrs[NL80211_ATTR_IE]);
ft_params.ie_len = nla_len(info->attrs[NL80211_ATTR_IE]);
return rdev_update_ft_ies(rdev, dev, &ft_params);
}
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
static int nl80211_crit_protocol_start(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
enum nl80211_crit_proto_id proto = NL80211_CRIT_PROTO_UNSPEC;
u16 duration;
int ret;
if (!rdev->ops->crit_proto_start)
return -EOPNOTSUPP;
if (WARN_ON(!rdev->ops->crit_proto_stop))
return -EINVAL;
if (rdev->crit_proto_nlportid)
return -EBUSY;
/* determine protocol if provided */
if (info->attrs[NL80211_ATTR_CRIT_PROT_ID])
proto = nla_get_u16(info->attrs[NL80211_ATTR_CRIT_PROT_ID]);
if (proto >= NUM_NL80211_CRIT_PROTO)
return -EINVAL;
/* timeout must be provided */
if (!info->attrs[NL80211_ATTR_MAX_CRIT_PROT_DURATION])
return -EINVAL;
duration =
nla_get_u16(info->attrs[NL80211_ATTR_MAX_CRIT_PROT_DURATION]);
if (duration > NL80211_CRIT_PROTO_MAX_DURATION)
return -ERANGE;
ret = rdev_crit_proto_start(rdev, wdev, proto, duration);
if (!ret)
rdev->crit_proto_nlportid = info->snd_portid;
return ret;
}
static int nl80211_crit_protocol_stop(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev = info->user_ptr[1];
if (!rdev->ops->crit_proto_stop)
return -EOPNOTSUPP;
if (rdev->crit_proto_nlportid) {
rdev->crit_proto_nlportid = 0;
rdev_crit_proto_stop(rdev, wdev);
}
return 0;
}
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
static int nl80211_vendor_cmd(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct wireless_dev *wdev =
__cfg80211_wdev_from_attrs(genl_info_net(info), info->attrs);
int i, err;
u32 vid, subcmd;
if (!rdev->wiphy.vendor_commands)
return -EOPNOTSUPP;
if (IS_ERR(wdev)) {
err = PTR_ERR(wdev);
if (err != -EINVAL)
return err;
wdev = NULL;
} else if (wdev->wiphy != &rdev->wiphy) {
return -EINVAL;
}
if (!info->attrs[NL80211_ATTR_VENDOR_ID] ||
!info->attrs[NL80211_ATTR_VENDOR_SUBCMD])
return -EINVAL;
vid = nla_get_u32(info->attrs[NL80211_ATTR_VENDOR_ID]);
subcmd = nla_get_u32(info->attrs[NL80211_ATTR_VENDOR_SUBCMD]);
for (i = 0; i < rdev->wiphy.n_vendor_commands; i++) {
const struct wiphy_vendor_command *vcmd;
void *data = NULL;
int len = 0;
vcmd = &rdev->wiphy.vendor_commands[i];
if (vcmd->info.vendor_id != vid || vcmd->info.subcmd != subcmd)
continue;
if (vcmd->flags & (WIPHY_VENDOR_CMD_NEED_WDEV |
WIPHY_VENDOR_CMD_NEED_NETDEV)) {
if (!wdev)
return -EINVAL;
if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_NETDEV &&
!wdev->netdev)
return -EINVAL;
if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_RUNNING) {
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (!wdev_running(wdev))
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
return -ENETDOWN;
}
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (!vcmd->doit)
return -EOPNOTSUPP;
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
} else {
wdev = NULL;
}
if (info->attrs[NL80211_ATTR_VENDOR_DATA]) {
data = nla_data(info->attrs[NL80211_ATTR_VENDOR_DATA]);
len = nla_len(info->attrs[NL80211_ATTR_VENDOR_DATA]);
}
rdev->cur_cmd_info = info;
err = rdev->wiphy.vendor_commands[i].doit(&rdev->wiphy, wdev,
data, len);
rdev->cur_cmd_info = NULL;
return err;
}
return -EOPNOTSUPP;
}
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
static int nl80211_prepare_vendor_dump(struct sk_buff *skb,
struct netlink_callback *cb,
struct cfg80211_registered_device **rdev,
struct wireless_dev **wdev)
{
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
struct nlattr **attrbuf = genl_family_attrbuf(&nl80211_fam);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
u32 vid, subcmd;
unsigned int i;
int vcmd_idx = -1;
int err;
void *data = NULL;
unsigned int data_len = 0;
if (cb->args[0]) {
/* subtract the 1 again here */
struct wiphy *wiphy = wiphy_idx_to_wiphy(cb->args[0] - 1);
struct wireless_dev *tmp;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!wiphy)
return -ENODEV;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
*rdev = wiphy_to_rdev(wiphy);
*wdev = NULL;
if (cb->args[1]) {
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry(tmp, &wiphy->wdev_list, list) {
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (tmp->identifier == cb->args[1] - 1) {
*wdev = tmp;
break;
}
}
}
/* keep rtnl locked in successful case */
return 0;
}
netlink: pass extended ACK struct to parsing functions Pass the new extended ACK reporting struct to all of the generic netlink parsing functions. For now, pass NULL in almost all callers (except for some in the core.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-04-12 12:34:07 +00:00
err = nlmsg_parse(cb->nlh, GENL_HDRLEN + nl80211_fam.hdrsize, attrbuf,
nl80211_fam.maxattr, nl80211_policy, NULL);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
return err;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
if (!attrbuf[NL80211_ATTR_VENDOR_ID] ||
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
!attrbuf[NL80211_ATTR_VENDOR_SUBCMD])
return -EINVAL;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
*wdev = __cfg80211_wdev_from_attrs(sock_net(skb->sk), attrbuf);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (IS_ERR(*wdev))
*wdev = NULL;
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
*rdev = __cfg80211_rdev_from_attrs(sock_net(skb->sk), attrbuf);
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (IS_ERR(*rdev))
return PTR_ERR(*rdev);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
vid = nla_get_u32(attrbuf[NL80211_ATTR_VENDOR_ID]);
subcmd = nla_get_u32(attrbuf[NL80211_ATTR_VENDOR_SUBCMD]);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
for (i = 0; i < (*rdev)->wiphy.n_vendor_commands; i++) {
const struct wiphy_vendor_command *vcmd;
vcmd = &(*rdev)->wiphy.vendor_commands[i];
if (vcmd->info.vendor_id != vid || vcmd->info.subcmd != subcmd)
continue;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!vcmd->dumpit)
return -EOPNOTSUPP;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
vcmd_idx = i;
break;
}
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (vcmd_idx < 0)
return -EOPNOTSUPP;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
genetlink: introduce and use genl_family_attrbuf() This helper function allows family implementations to access their family's attrbuf. This gets rid of the attrbuf usage in families, and also adds locking validation, since it's not valid to use the attrbuf with parallel_ops or outside of the dumpit callback. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:01 +00:00
if (attrbuf[NL80211_ATTR_VENDOR_DATA]) {
data = nla_data(attrbuf[NL80211_ATTR_VENDOR_DATA]);
data_len = nla_len(attrbuf[NL80211_ATTR_VENDOR_DATA]);
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
}
/* 0 is the first index - add 1 to parse only once */
cb->args[0] = (*rdev)->wiphy_idx + 1;
/* add 1 to know if it was NULL */
cb->args[1] = *wdev ? (*wdev)->identifier + 1 : 0;
cb->args[2] = vcmd_idx;
cb->args[3] = (unsigned long)data;
cb->args[4] = data_len;
/* keep rtnl locked in successful case */
return 0;
}
static int nl80211_vendor_cmd_dump(struct sk_buff *skb,
struct netlink_callback *cb)
{
struct cfg80211_registered_device *rdev;
struct wireless_dev *wdev;
unsigned int vcmd_idx;
const struct wiphy_vendor_command *vcmd;
void *data;
int data_len;
int err;
struct nlattr *vendor_data;
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
rtnl_lock();
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
err = nl80211_prepare_vendor_dump(skb, cb, &rdev, &wdev);
if (err)
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
goto out;
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
vcmd_idx = cb->args[2];
data = (void *)cb->args[3];
data_len = cb->args[4];
vcmd = &rdev->wiphy.vendor_commands[vcmd_idx];
if (vcmd->flags & (WIPHY_VENDOR_CMD_NEED_WDEV |
WIPHY_VENDOR_CMD_NEED_NETDEV)) {
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!wdev) {
err = -EINVAL;
goto out;
}
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_NETDEV &&
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
!wdev->netdev) {
err = -EINVAL;
goto out;
}
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
if (vcmd->flags & WIPHY_VENDOR_CMD_NEED_RUNNING) {
nl80211: fix dumpit error path RTNL deadlocks Sowmini pointed out Dmitry's RTNL deadlock report to me, and it turns out to be perfectly accurate - there are various error paths that miss unlock of the RTNL. To fix those, change the locking a bit to not be conditional in all those nl80211_prepare_*_dump() functions, but make those require the RTNL to start with, and fix the buggy error paths. This also let me use sparse (by appropriately overriding the rtnl_lock/rtnl_unlock functions) to validate the changes. Cc: stable@vger.kernel.org Reported-by: Sowmini Varadhan <sowmini.varadhan@oracle.com> Reported-by: Dmitry Vyukov <dvyukov@google.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-15 13:26:04 +00:00
if (!wdev_running(wdev)) {
err = -ENETDOWN;
goto out;
}
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
}
}
while (1) {
void *hdr = nl80211hdr_put(skb, NETLINK_CB(cb->skb).portid,
cb->nlh->nlmsg_seq, NLM_F_MULTI,
NL80211_CMD_VENDOR);
if (!hdr)
break;
if (nla_put_u32(skb, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
(wdev && nla_put_u64_64bit(skb, NL80211_ATTR_WDEV,
wdev_id(wdev),
NL80211_ATTR_PAD))) {
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
genlmsg_cancel(skb, hdr);
break;
}
vendor_data = nla_nest_start(skb, NL80211_ATTR_VENDOR_DATA);
if (!vendor_data) {
genlmsg_cancel(skb, hdr);
break;
}
err = vcmd->dumpit(&rdev->wiphy, wdev, skb, data, data_len,
(unsigned long *)&cb->args[5]);
nla_nest_end(skb, vendor_data);
if (err == -ENOBUFS || err == -ENOENT) {
genlmsg_cancel(skb, hdr);
break;
} else if (err) {
genlmsg_cancel(skb, hdr);
goto out;
}
genlmsg_end(skb, hdr);
}
err = skb->len;
out:
rtnl_unlock();
return err;
}
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
struct sk_buff *__cfg80211_alloc_reply_skb(struct wiphy *wiphy,
enum nl80211_commands cmd,
enum nl80211_attrs attr,
int approxlen)
{
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
if (WARN_ON(!rdev->cur_cmd_info))
return NULL;
cfg80211: Allow NL80211_ATTR_IFINDEX to be added to vendor events This modifies cfg80211_vendor_event_alloc() with an additional argument struct wireless_dev *wdev. __cfg80211_alloc_event_skb() is modified to take in *wdev argument, if wdev != NULL, both the NL80211_ATTR_IFINDEX and wdev identifier are added to the vendor event. These changes make it easier for drivers to add ifindex indication in vendor events cleanly. This also updates all existing users of cfg80211_vendor_event_alloc() and __cfg80211_alloc_event_skb() in the kernel tree. Signed-off-by: Ahmad Kholaif <akholaif@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-26 13:26:53 +00:00
return __cfg80211_alloc_vendor_skb(rdev, NULL, approxlen,
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
rdev->cur_cmd_info->snd_portid,
rdev->cur_cmd_info->snd_seq,
nl80211: support vendor-specific events In addition to vendor-specific commands, also support vendor-specific events. These must be registered with cfg80211 before they can be used. They're also advertised in nl80211 in the wiphy information so that userspace knows can be expected. The events themselves are sent on a new multicast group called "vendor". Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-18 13:43:31 +00:00
cmd, attr, NULL, GFP_KERNEL);
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
}
EXPORT_SYMBOL(__cfg80211_alloc_reply_skb);
int cfg80211_vendor_cmd_reply(struct sk_buff *skb)
{
struct cfg80211_registered_device *rdev = ((void **)skb->cb)[0];
void *hdr = ((void **)skb->cb)[1];
struct nlattr *data = ((void **)skb->cb)[2];
nl80211: clear skb cb before passing to netlink In testmode and vendor command reply/event SKBs we use the skb cb data to store nl80211 parameters between allocation and sending. This causes the code for CONFIG_NETLINK_MMAP to get confused, because it takes ownership of the skb cb data when the SKB is handed off to netlink, and it doesn't explicitly clear it. Clear the skb cb explicitly when we're done and before it gets passed to netlink to avoid this issue. Cc: stable@vger.kernel.org [this goes way back] Reported-by: Assaf Azulay <assaf.azulay@intel.com> Reported-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-07-30 12:55:26 +00:00
/* clear CB data for netlink core to own from now on */
memset(skb->cb, 0, sizeof(skb->cb));
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
if (WARN_ON(!rdev->cur_cmd_info)) {
kfree_skb(skb);
return -EINVAL;
}
nla_nest_end(skb, data);
genlmsg_end(skb, hdr);
return genlmsg_reply(skb, rdev->cur_cmd_info);
}
EXPORT_SYMBOL_GPL(cfg80211_vendor_cmd_reply);
cfg80211: Add support for QoS mapping This allows QoS mapping from external networks to be implemented as defined in IEEE Std 802.11-2012, 10.24.9. APs can use this to advertise DSCP ranges and exceptions for mapping frames to a specific UP over Wi-Fi. The payload of the QoS Map Set element (IEEE Std 802.11-2012, 8.4.2.97) is sent to the driver through the new NL80211_ATTR_QOS_MAP attribute to configure the local behavior either on the AP (based on local configuration) or on a station (based on information received from the AP). Signed-off-by: Kyeyoon Park <kyeyoonp@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-17 07:01:30 +00:00
static int nl80211_set_qos_map(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct cfg80211_qos_map *qos_map = NULL;
struct net_device *dev = info->user_ptr[1];
u8 *pos, len, num_des, des_len, des;
int ret;
if (!rdev->ops->set_qos_map)
return -EOPNOTSUPP;
if (info->attrs[NL80211_ATTR_QOS_MAP]) {
pos = nla_data(info->attrs[NL80211_ATTR_QOS_MAP]);
len = nla_len(info->attrs[NL80211_ATTR_QOS_MAP]);
if (len % 2 || len < IEEE80211_QOS_MAP_LEN_MIN ||
len > IEEE80211_QOS_MAP_LEN_MAX)
return -EINVAL;
qos_map = kzalloc(sizeof(struct cfg80211_qos_map), GFP_KERNEL);
if (!qos_map)
return -ENOMEM;
num_des = (len - IEEE80211_QOS_MAP_LEN_MIN) >> 1;
if (num_des) {
des_len = num_des *
sizeof(struct cfg80211_dscp_exception);
memcpy(qos_map->dscp_exception, pos, des_len);
qos_map->num_des = num_des;
for (des = 0; des < num_des; des++) {
if (qos_map->dscp_exception[des].up > 7) {
kfree(qos_map);
return -EINVAL;
}
}
pos += des_len;
}
memcpy(qos_map->up, pos, IEEE80211_QOS_MAP_LEN_MIN);
}
wdev_lock(dev->ieee80211_ptr);
ret = nl80211_key_allowed(dev->ieee80211_ptr);
if (!ret)
ret = rdev_set_qos_map(rdev, dev, qos_map);
wdev_unlock(dev->ieee80211_ptr);
kfree(qos_map);
return ret;
}
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
static int nl80211_add_tx_ts(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
const u8 *peer;
u8 tsid, up;
u16 admitted_time = 0;
int err;
cfg80211: make WMM TSPEC support flag an nl80211 feature flag During the review of the corresponding wpa_supplicant patches we noticed that the only way for it to detect that this functionality is supported currently is to check for the command support. This can be misleading though, as the command was also designed to, in the future, support pure 802.11 TSPECs. Expose the WMM-TSPEC feature flag to nl80211 so later we can also expose an 802.11-TSPEC feature flag (if needed) to differentiate the two cases. Note: this change isn't needed in 3.18 as there's no driver there yet that supports the functionality at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-22 07:25:06 +00:00
if (!(rdev->wiphy.features & NL80211_FEATURE_SUPPORTS_WMM_ADMISSION))
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_TSID] || !info->attrs[NL80211_ATTR_MAC] ||
!info->attrs[NL80211_ATTR_USER_PRIO])
return -EINVAL;
tsid = nla_get_u8(info->attrs[NL80211_ATTR_TSID]);
if (tsid >= IEEE80211_NUM_TIDS)
return -EINVAL;
up = nla_get_u8(info->attrs[NL80211_ATTR_USER_PRIO]);
if (up >= IEEE80211_NUM_UPS)
return -EINVAL;
/* WMM uses TIDs 0-7 even for TSPEC */
cfg80211: make WMM TSPEC support flag an nl80211 feature flag During the review of the corresponding wpa_supplicant patches we noticed that the only way for it to detect that this functionality is supported currently is to check for the command support. This can be misleading though, as the command was also designed to, in the future, support pure 802.11 TSPECs. Expose the WMM-TSPEC feature flag to nl80211 so later we can also expose an 802.11-TSPEC feature flag (if needed) to differentiate the two cases. Note: this change isn't needed in 3.18 as there's no driver there yet that supports the functionality at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-22 07:25:06 +00:00
if (tsid >= IEEE80211_FIRST_TSPEC_TSID) {
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
/* TODO: handle 802.11 TSPEC/admission control
cfg80211: make WMM TSPEC support flag an nl80211 feature flag During the review of the corresponding wpa_supplicant patches we noticed that the only way for it to detect that this functionality is supported currently is to check for the command support. This can be misleading though, as the command was also designed to, in the future, support pure 802.11 TSPECs. Expose the WMM-TSPEC feature flag to nl80211 so later we can also expose an 802.11-TSPEC feature flag (if needed) to differentiate the two cases. Note: this change isn't needed in 3.18 as there's no driver there yet that supports the functionality at all. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-10-22 07:25:06 +00:00
* need more attributes for that (e.g. BA session requirement);
* change the WMM adminssion test above to allow both then
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
*/
return -EINVAL;
}
peer = nla_data(info->attrs[NL80211_ATTR_MAC]);
if (info->attrs[NL80211_ATTR_ADMITTED_TIME]) {
admitted_time =
nla_get_u16(info->attrs[NL80211_ATTR_ADMITTED_TIME]);
if (!admitted_time)
return -EINVAL;
}
wdev_lock(wdev);
switch (wdev->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_P2P_CLIENT:
if (wdev->current_bss)
break;
err = -ENOTCONN;
goto out;
default:
err = -EOPNOTSUPP;
goto out;
}
err = rdev_add_tx_ts(rdev, dev, tsid, peer, up, admitted_time);
out:
wdev_unlock(wdev);
return err;
}
static int nl80211_del_tx_ts(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
const u8 *peer;
u8 tsid;
int err;
if (!info->attrs[NL80211_ATTR_TSID] || !info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
tsid = nla_get_u8(info->attrs[NL80211_ATTR_TSID]);
peer = nla_data(info->attrs[NL80211_ATTR_MAC]);
wdev_lock(wdev);
err = rdev_del_tx_ts(rdev, dev, tsid, peer);
wdev_unlock(wdev);
return err;
}
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
static int nl80211_tdls_channel_switch(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_chan_def chandef = {};
const u8 *addr;
u8 oper_class;
int err;
if (!rdev->ops->tdls_channel_switch ||
!(rdev->wiphy.features & NL80211_FEATURE_TDLS_CHANNEL_SWITCH))
return -EOPNOTSUPP;
switch (dev->ieee80211_ptr->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_P2P_CLIENT:
break;
default:
return -EOPNOTSUPP;
}
if (!info->attrs[NL80211_ATTR_MAC] ||
!info->attrs[NL80211_ATTR_OPER_CLASS])
return -EINVAL;
err = nl80211_parse_chandef(rdev, info, &chandef);
if (err)
return err;
/*
* Don't allow wide channels on the 2.4Ghz band, as per IEEE802.11-2012
* section 10.22.6.2.1. Disallow 5/10Mhz channels as well for now, the
* specification is not defined for them.
*/
cfg80211: remove enum ieee80211_band This enum is already perfectly aliased to enum nl80211_band, and the only reason for it is that we get IEEE80211_NUM_BANDS out of it. There's no really good reason to not declare the number of bands in nl80211 though, so do that and remove the cfg80211 one. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-12 13:56:15 +00:00
if (chandef.chan->band == NL80211_BAND_2GHZ &&
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
chandef.width != NL80211_CHAN_WIDTH_20_NOHT &&
chandef.width != NL80211_CHAN_WIDTH_20)
return -EINVAL;
/* we will be active on the TDLS link */
cfg80211: use RTNL locked reg_can_beacon for IR-relaxation The RTNL is required to check for IR-relaxation conditions that allow more channels to beacon. Export an RTNL locked version of reg_can_beacon and use it where possible in AP/STA interface type flows, where IR-relaxation may be applicable. Fixes: 06f207fc5418 ("cfg80211: change GO_CONCURRENT to IR_CONCURRENT for STA") Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-07-08 12:41:44 +00:00
if (!cfg80211_reg_can_beacon_relax(&rdev->wiphy, &chandef,
wdev->iftype))
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
return -EINVAL;
/* don't allow switching to DFS channels */
if (cfg80211_chandef_dfs_required(wdev->wiphy, &chandef, wdev->iftype))
return -EINVAL;
addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
oper_class = nla_get_u8(info->attrs[NL80211_ATTR_OPER_CLASS]);
wdev_lock(wdev);
err = rdev_tdls_channel_switch(rdev, dev, addr, oper_class, &chandef);
wdev_unlock(wdev);
return err;
}
static int nl80211_tdls_cancel_channel_switch(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
const u8 *addr;
if (!rdev->ops->tdls_channel_switch ||
!rdev->ops->tdls_cancel_channel_switch ||
!(rdev->wiphy.features & NL80211_FEATURE_TDLS_CHANNEL_SWITCH))
return -EOPNOTSUPP;
switch (dev->ieee80211_ptr->iftype) {
case NL80211_IFTYPE_STATION:
case NL80211_IFTYPE_P2P_CLIENT:
break;
default:
return -EOPNOTSUPP;
}
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
addr = nla_data(info->attrs[NL80211_ATTR_MAC]);
wdev_lock(wdev);
rdev_tdls_cancel_channel_switch(rdev, dev, addr);
wdev_unlock(wdev);
return 0;
}
cfg80211: configure multicast to unicast for AP interfaces Add the ability to configure if an AP (and associated VLANs) will do multicast-to-unicast conversion for ARP, IPv4 and IPv6 frames (possibly within 802.1Q). If enabled, such frames are to be sent to each station separately, with the DA replaced by their own MAC address rather than the group address. Note that this may break certain expectations of the receiver, such as the ability to drop unicast IP packets received within multicast L2 frames, or the ability to not send ICMP destination unreachable messages for packets received in L2 multicast (which is required, but the receiver can't tell the difference if this new option is enabled.) This also doesn't implement the 802.11 DMS (directed multicast service). Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [fix disabling, add better documentation & commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-10 17:12:22 +00:00
static int nl80211_set_multicast_to_unicast(struct sk_buff *skb,
struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
const struct nlattr *nla;
bool enabled;
if (!rdev->ops->set_multicast_to_unicast)
return -EOPNOTSUPP;
if (wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO)
return -EOPNOTSUPP;
nla = info->attrs[NL80211_ATTR_MULTICAST_TO_UNICAST_ENABLED];
enabled = nla_get_flag(nla);
return rdev_set_multicast_to_unicast(rdev, dev, enabled);
}
cfg80211: support 4-way handshake offloading for 802.1X Add API for setting the PMK to the driver. For FT support, allow setting also the PMK-R0 Name. This can be used by drivers that support 4-Way handshake offload while IEEE802.1X authentication is managed by upper layers. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> [arend.vanspriel@broadcom.com: add WANT_1X_4WAY_HS attribute] Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [reword NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X docs a bit to say that the device may require it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-06-09 12:08:43 +00:00
static int nl80211_set_pmk(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_pmk_conf pmk_conf = {};
int ret;
if (wdev->iftype != NL80211_IFTYPE_STATION &&
wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X))
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_MAC] || !info->attrs[NL80211_ATTR_PMK])
return -EINVAL;
wdev_lock(wdev);
if (!wdev->current_bss) {
ret = -ENOTCONN;
goto out;
}
pmk_conf.aa = nla_data(info->attrs[NL80211_ATTR_MAC]);
if (memcmp(pmk_conf.aa, wdev->current_bss->pub.bssid, ETH_ALEN)) {
ret = -EINVAL;
goto out;
}
pmk_conf.pmk = nla_data(info->attrs[NL80211_ATTR_PMK]);
pmk_conf.pmk_len = nla_len(info->attrs[NL80211_ATTR_PMK]);
if (pmk_conf.pmk_len != WLAN_PMK_LEN &&
pmk_conf.pmk_len != WLAN_PMK_LEN_SUITE_B_192) {
ret = -EINVAL;
goto out;
}
if (info->attrs[NL80211_ATTR_PMKR0_NAME]) {
int r0_name_len = nla_len(info->attrs[NL80211_ATTR_PMKR0_NAME]);
if (r0_name_len != WLAN_PMK_NAME_LEN) {
ret = -EINVAL;
goto out;
}
pmk_conf.pmk_r0_name =
nla_data(info->attrs[NL80211_ATTR_PMKR0_NAME]);
}
ret = rdev_set_pmk(rdev, dev, &pmk_conf);
out:
wdev_unlock(wdev);
return ret;
}
static int nl80211_del_pmk(struct sk_buff *skb, struct genl_info *info)
{
struct cfg80211_registered_device *rdev = info->user_ptr[0];
struct net_device *dev = info->user_ptr[1];
struct wireless_dev *wdev = dev->ieee80211_ptr;
const u8 *aa;
int ret;
if (wdev->iftype != NL80211_IFTYPE_STATION &&
wdev->iftype != NL80211_IFTYPE_P2P_CLIENT)
return -EOPNOTSUPP;
if (!wiphy_ext_feature_isset(&rdev->wiphy,
NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X))
return -EOPNOTSUPP;
if (!info->attrs[NL80211_ATTR_MAC])
return -EINVAL;
wdev_lock(wdev);
aa = nla_data(info->attrs[NL80211_ATTR_MAC]);
ret = rdev_del_pmk(rdev, dev, aa);
wdev_unlock(wdev);
return ret;
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
#define NL80211_FLAG_NEED_WIPHY 0x01
#define NL80211_FLAG_NEED_NETDEV 0x02
#define NL80211_FLAG_NEED_RTNL 0x04
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
#define NL80211_FLAG_CHECK_NETDEV_UP 0x08
#define NL80211_FLAG_NEED_NETDEV_UP (NL80211_FLAG_NEED_NETDEV |\
NL80211_FLAG_CHECK_NETDEV_UP)
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
#define NL80211_FLAG_NEED_WDEV 0x10
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
/* If a netdev is associated, it must be UP, P2P must be started */
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
#define NL80211_FLAG_NEED_WDEV_UP (NL80211_FLAG_NEED_WDEV |\
NL80211_FLAG_CHECK_NETDEV_UP)
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
#define NL80211_FLAG_CLEAR_SKB 0x20
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
genetlink: allow making ops const Allow making the ops array const by not modifying the ops flags on registration but rather only when ops are sent out in the family information. No users are updated yet except for the pre_doit/post_doit calls in wireless (the only ones that exist now.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-14 16:14:45 +00:00
static int nl80211_pre_doit(const struct genl_ops *ops, struct sk_buff *skb,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct genl_info *info)
{
struct cfg80211_registered_device *rdev;
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
struct wireless_dev *wdev;
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct net_device *dev;
bool rtnl = ops->internal_flags & NL80211_FLAG_NEED_RTNL;
if (rtnl)
rtnl_lock();
if (ops->internal_flags & NL80211_FLAG_NEED_WIPHY) {
nl80211: fix netns separation There are currently a few ways to "escape" the network namespace and access a wiphy that belongs to another namespace. Add a netns argument to the relevant functions to fix this. One remaining issue with testmode will be fixed in a follow-up patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:14:22 +00:00
rdev = cfg80211_get_dev_from_info(genl_info_net(info), info);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (IS_ERR(rdev)) {
if (rtnl)
rtnl_unlock();
return PTR_ERR(rdev);
}
info->user_ptr[0] = rdev;
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
} else if (ops->internal_flags & NL80211_FLAG_NEED_NETDEV ||
ops->internal_flags & NL80211_FLAG_NEED_WDEV) {
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
ASSERT_RTNL();
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
wdev = __cfg80211_wdev_from_attrs(genl_info_net(info),
info->attrs);
if (IS_ERR(wdev)) {
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (rtnl)
rtnl_unlock();
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
return PTR_ERR(wdev);
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
}
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
dev = wdev->netdev;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
rdev = wiphy_to_rdev(wdev->wiphy);
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
if (ops->internal_flags & NL80211_FLAG_NEED_NETDEV) {
if (!dev) {
if (rtnl)
rtnl_unlock();
return -EINVAL;
}
info->user_ptr[1] = dev;
} else {
info->user_ptr[1] = wdev;
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
}
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (ops->internal_flags & NL80211_FLAG_CHECK_NETDEV_UP &&
!wdev_running(wdev)) {
if (rtnl)
rtnl_unlock();
return -ENETDOWN;
}
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
cfg80211: add generic helper to check interface is running Add a helper using wdev to check if interface is running. This deals with both non-netdev and netdev interfaces. In struct wireless_dev replace 'p2p_started' and 'nan_started' by 'is_running' as those are mutually exclusive anyway, and unify all the code to use wdev_running(). Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-20 19:08:22 +00:00
if (dev)
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
dev_hold(dev);
nl80211: prepare for non-netdev wireless devs In order to support a P2P device abstraction and Bluetooth high-speed AMPs, we need to have a way to identify virtual interfaces that don't have a netdev associated. Do this by adding a NL80211_ATTR_WDEV attribute to identify a wdev which may or may not also be a netdev. To simplify things, use a 64-bit value with the high 32 bits being the wiphy index for this new wdev identifier in the nl80211 API. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 12:33:17 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
info->user_ptr[0] = rdev;
}
return 0;
}
genetlink: allow making ops const Allow making the ops array const by not modifying the ops flags on registration but rather only when ops are sent out in the family information. No users are updated yet except for the pre_doit/post_doit calls in wireless (the only ones that exist now.) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-14 16:14:45 +00:00
static void nl80211_post_doit(const struct genl_ops *ops, struct sk_buff *skb,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
struct genl_info *info)
{
nl80211: add NL80211_FLAG_NEED_WDEV Some nl80211 callbacks will soon need the wdev instead of the netdev, so add NL80211_FLAG_NEED_WDEV to allow them to request that. Add NL80211_FLAG_NEED_WDEV_UP as well which checks the netdev is UP if one exists. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:23:36 +00:00
if (info->user_ptr[1]) {
if (ops->internal_flags & NL80211_FLAG_NEED_WDEV) {
struct wireless_dev *wdev = info->user_ptr[1];
if (wdev->netdev)
dev_put(wdev->netdev);
} else {
dev_put(info->user_ptr[1]);
}
}
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
if (ops->internal_flags & NL80211_FLAG_NEED_RTNL)
rtnl_unlock();
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
/* If needed, clear the netlink message payload from the SKB
* as it might contain key data that shouldn't stick around on
* the heap after the SKB is freed. The netlink message header
* is still needed for further processing, so leave it intact.
*/
if (ops->internal_flags & NL80211_FLAG_CLEAR_SKB) {
struct nlmsghdr *nlh = nlmsg_hdr(skb);
memset(nlmsg_data(nlh), 0, nlmsg_len(nlh));
}
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
}
genetlink: make all genl_ops users const Now that genl_ops are no longer modified in place when registering, they can be made const. This patch was done mostly with spatch: @@ identifier ops; @@ +const struct genl_ops ops[] = { ... }; (except the struct thing in net/openvswitch/datapath.c) Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-14 16:14:46 +00:00
static const struct genl_ops nl80211_ops[] = {
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
.cmd = NL80211_CMD_GET_WIPHY,
.doit = nl80211_get_wiphy,
.dumpit = nl80211_dump_wiphy,
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
.done = nl80211_dump_wiphy_done,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
.policy = nl80211_policy,
/* can be retrieved by unprivileged users */
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
{
.cmd = NL80211_CMD_SET_WIPHY,
.doit = nl80211_set_wiphy,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
{
.cmd = NL80211_CMD_GET_INTERFACE,
.doit = nl80211_get_interface,
.dumpit = nl80211_dump_interface,
.policy = nl80211_policy,
/* can be retrieved by unprivileged users */
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV |
NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
{
.cmd = NL80211_CMD_SET_INTERFACE,
.doit = nl80211_set_interface,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
{
.cmd = NL80211_CMD_NEW_INTERFACE,
.doit = nl80211_new_interface,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
{
.cmd = NL80211_CMD_DEL_INTERFACE,
.doit = nl80211_del_interface,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wireless_dev for interface management In order to be able to create P2P Device wdevs, move the virtual interface management over to wireless_dev structures. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:00:26 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
},
{
.cmd = NL80211_CMD_GET_KEY,
.doit = nl80211_get_key,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
},
{
.cmd = NL80211_CMD_SET_KEY,
.doit = nl80211_set_key,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
NL80211_FLAG_NEED_RTNL |
NL80211_FLAG_CLEAR_SKB,
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
},
{
.cmd = NL80211_CMD_NEW_KEY,
.doit = nl80211_new_key,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
NL80211_FLAG_NEED_RTNL |
NL80211_FLAG_CLEAR_SKB,
cfg80211/nl80211: introduce key handling This introduces key handling to cfg80211/nl80211. Default and group keys can be added, changed and removed; sequence counters for each key can be retrieved. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:29 +00:00
},
{
.cmd = NL80211_CMD_DEL_KEY,
.doit = nl80211_del_key,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
},
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
{
.cmd = NL80211_CMD_SET_BEACON,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
.doit = nl80211_set_beacon,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
},
{
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
.cmd = NL80211_CMD_START_AP,
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
.doit = nl80211_start_ap,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
},
{
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
.cmd = NL80211_CMD_STOP_AP,
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: restructure AP/GO mode API The AP/GO mode API isn't very clearly defined, it has "set beacon" and "new beacon" etc. Modify the API to the following: * start AP -- all settings * change beacon -- new beacon data * stop AP -- stop AP mode operation This also reflects in the nl80211 API, rename the commands there correspondingly (but keep the old names for compatibility.) Overall, this makes it much clearer what's going on in the API. Kalle developed the ath6kl changes, I created the rest of the patch. Signed-off-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-02-13 14:17:18 +00:00
.doit = nl80211_stop_ap,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: add beacon settings This adds the necessary API to cfg80211/nl80211 to allow changing beaconing settings. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:32 +00:00
},
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
{
.cmd = NL80211_CMD_GET_STATION,
.doit = nl80211_get_station,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
.dumpit = nl80211_dump_station,
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
.policy = nl80211_policy,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
},
{
.cmd = NL80211_CMD_SET_STATION,
.doit = nl80211_set_station,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
},
{
.cmd = NL80211_CMD_NEW_STATION,
.doit = nl80211_new_station,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: station handling This patch adds station handling to cfg80211/nl80211. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-12-19 01:03:34 +00:00
},
{
.cmd = NL80211_CMD_DEL_STATION,
.doit = nl80211_del_station,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
},
{
.cmd = NL80211_CMD_GET_MPATH,
.doit = nl80211_get_mpath,
.dumpit = nl80211_dump_mpath,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
},
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
{
.cmd = NL80211_CMD_GET_MPP,
.doit = nl80211_get_mpp,
.dumpit = nl80211_dump_mpp,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: add ops to query mesh proxy path table Add two new cfg80211 operations for querying a table with proxied mesh paths. Signed-off-by: Henning Rogge <henning.rogge@fkie.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-12 06:58:49 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
{
.cmd = NL80211_CMD_SET_MPATH,
.doit = nl80211_set_mpath,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
},
{
.cmd = NL80211_CMD_NEW_MPATH,
.doit = nl80211_new_mpath,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211/cfg80211: support for mesh, sta dumping Added support for mesh id and mesh path operation as well as station structure dumping. Signed-off-by: Luis Carlos Cobo <luisca@cozybit.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-02-23 14:17:06 +00:00
},
{
.cmd = NL80211_CMD_DEL_MPATH,
.doit = nl80211_del_mpath,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
mac80211/cfg80211: Add BSS configuration options for AP mode This change adds a new cfg80211 command, NL80211_CMD_SET_BSS, to allow AP mode BSS parameters to be changed from user space (e.g., hostapd). The drivers using mac80211 are expected to be modified with separate changes to use the new BSS info parameter for short slot time in the bss_info_changed() handler. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-08-07 17:07:01 +00:00
},
{
.cmd = NL80211_CMD_SET_BSS,
.doit = nl80211_set_bss,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
},
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
{
.cmd = NL80211_CMD_GET_REG,
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
.doit = nl80211_get_reg_do,
.dumpit = nl80211_get_reg_dump,
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
.policy = nl80211_policy,
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
.internal_flags = NL80211_FLAG_NEED_RTNL,
cfg80211: add get reg command This lets userspace request to get the currently set regulatory domain. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-30 17:26:42 +00:00
/* can be retrieved by unprivileged users */
},
cfg80211: reg: make CRDA support optional If there's a built-in regulatory database, there may be little point in also calling out to CRDA and failing if the system is configured that way. Allow removing CRDA support to save ~1K kernel size. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-15 07:25:18 +00:00
#ifdef CONFIG_CFG80211_CRDA_SUPPORT
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
{
.cmd = NL80211_CMD_SET_REG,
.doit = nl80211_set_reg,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
cfg80211: vastly simplify locking Virtually all code paths in cfg80211 already (need to) hold the RTNL. As such, there's little point in having another four mutexes for various parts of the code, they just cause lock ordering issues (and much of the time, the RTNL and a few of the others need thus be held.) Simplify all this by getting rid of the extra four mutexes and just use the RTNL throughout. Only a few code changes were needed to do this and we can get rid of a work struct for bonus points. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-08 19:45:15 +00:00
.internal_flags = NL80211_FLAG_NEED_RTNL,
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
},
cfg80211: reg: make CRDA support optional If there's a built-in regulatory database, there may be little point in also calling out to CRDA and failing if the system is configured that way. Allow removing CRDA support to save ~1K kernel size. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-15 07:25:18 +00:00
#endif
cfg80211: Add new wireless regulatory infrastructure This adds the new wireless regulatory infrastructure. The main motiviation behind this was to centralize regulatory code as each driver was implementing their own regulatory solution, and to replace the initial centralized code we have where: * only 3 regulatory domains are supported: US, JP and EU * regulatory domains can only be changed through module parameter * all rules were built statically in the kernel We now have support for regulatory domains for many countries and regulatory domains are now queried through a userspace agent through udev allowing distributions to update regulatory rules without updating the kernel. Each driver can regulatory_hint() a regulatory domain based on either their EEPROM mapped regulatory domain value to a respective ISO/IEC 3166-1 country code or pass an internally built regulatory domain. We also add support to let the user set the regulatory domain through userspace in case of faulty EEPROMs to further help compliance. Support for world roaming will be added soon for cards capable of this. For more information see: http://wireless.kernel.org/en/developers/Regulatory/CRDA For now we leave an option to enable the old module parameter, ieee80211_regdom, and to build the 3 old regdomains statically (US, JP and EU). This option is CONFIG_WIRELESS_OLD_REGULATORY. These old static definitions and the module parameter is being scheduled for removal for 2.6.29. Note that if you use this you won't make use of a world regulatory domain as its pointless. If you leave this option enabled and if CRDA is present and you use US or JP we will try to ask CRDA to update us a regulatory domain for us. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-09-10 06:19:48 +00:00
{
.cmd = NL80211_CMD_REQ_SET_REG,
.doit = nl80211_req_set_reg,
.policy = nl80211_policy,
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
.flags = GENL_ADMIN_PERM,
},
{
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
.cmd = NL80211_CMD_GET_MESH_CONFIG,
.doit = nl80211_get_mesh_config,
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
.policy = nl80211_policy,
/* can be retrieved by unprivileged users */
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
},
{
mac80211: Rename mesh_params to mesh_config to prepare for mesh_setup Mesh parameters can be to setup a mesh or to configure it. This patch renames the ambiguous name mesh_params to mesh_config in preparation for mesh_setup. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-17 01:37:48 +00:00
.cmd = NL80211_CMD_SET_MESH_CONFIG,
.doit = nl80211_update_mesh_config,
Add nl80211 commands to get and set o11s mesh networking parameters The two new commands are NL80211_CMD_GET_MESH_PARAMS and NL80211_CMD_SET_MESH_PARAMS. There is a new attribute enum, NL80211_ATTR_MESH_PARAMS, which enumerates the various mesh configuration parameters. Moved struct mesh_config from mac80211/ieee80211_i.h to net/cfg80211.h. nl80211_get_mesh_params and nl80211_set_mesh_params unpack the netlink messages and ask the driver to get or set the configuration. This is done via two new function stubs, get_mesh_params and set_mesh_params, in struct cfg80211_ops. Signed-off-by: Colin McCabe <colin@cozybit.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2008-10-21 19:03:48 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: New command for adding extra IE(s) into management frames A new nl80211 command, NL80211_CMD_SET_MGMT_EXTRA_IE, can be used to add arbitrary IE data into the end of management frames. The interface allows extra IEs to be configured for each management frame subtype, but only some of them (ProbeReq, ProbeResp, Auth, (Re)AssocReq, Deauth, Disassoc) are currently accepted in mac80211 implementation. This makes it easier to implement IEEE 802.11 extensions like WPS and FT that add IE(s) into some management frames. In addition, this can be useful for testing and experimentation purposes. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-01-13 14:03:29 +00:00
},
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
.cmd = NL80211_CMD_TRIGGER_SCAN,
.doit = nl80211_trigger_scan,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
},
cfg80211: Add support for aborting an ongoing scan Implement new functionality for aborting an ongoing scan. Add NL80211_CMD_ABORT_SCAN to the nl80211 interface. After aborting the scan, driver shall provide the scan status by calling cfg80211_scan_done(). Reviewed-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Sunil Dutt <usdutt@qti.qualcomm.com> [change command to take wdev instead of netdev so that it can be used on p2p-device scans] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-30 13:44:49 +00:00
{
.cmd = NL80211_CMD_ABORT_SCAN,
.doit = nl80211_abort_scan,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: Add support for aborting an ongoing scan Implement new functionality for aborting an ongoing scan. Add NL80211_CMD_ABORT_SCAN to the nl80211 interface. After aborting the scan, driver shall provide the scan status by calling cfg80211_scan_done(). Reviewed-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Sunil Dutt <usdutt@qti.qualcomm.com> [change command to take wdev instead of netdev so that it can be used on p2p-device scans] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-10-30 13:44:49 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
.cmd = NL80211_CMD_GET_SCAN,
.policy = nl80211_policy,
.dumpit = nl80211_dump_scan,
},
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
{
.cmd = NL80211_CMD_START_SCHED_SCAN,
.doit = nl80211_start_sched_scan,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_STOP_SCHED_SCAN,
.doit = nl80211_stop_sched_scan,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
{
.cmd = NL80211_CMD_AUTHENTICATE,
.doit = nl80211_authenticate,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
NL80211_FLAG_NEED_RTNL |
NL80211_FLAG_CLEAR_SKB,
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
},
{
.cmd = NL80211_CMD_ASSOCIATE,
.doit = nl80211_associate,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
},
{
.cmd = NL80211_CMD_DEAUTHENTICATE,
.doit = nl80211_deauthenticate,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
},
{
.cmd = NL80211_CMD_DISASSOCIATE,
.doit = nl80211_disassociate,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: Add MLME primitives to support external SME This patch adds new nl80211 commands to allow user space to request authentication and association (and also deauthentication and disassociation). The commands are structured to allow separate authentication and association steps, i.e., the interface between kernel and user space is similar to the MLME SAP interface in IEEE 802.11 standard and an user space application takes the role of the SME. The patch introduces MLME-AUTHENTICATE.request, MLME-{,RE}ASSOCIATE.request, MLME-DEAUTHENTICATE.request, and MLME-DISASSOCIATE.request primitives. The authentication and association commands request the actual operations in two steps (assuming the driver supports this; if not, separate authentication step is skipped; this could end up being a separate "connect" command). The initial implementation for mac80211 uses the current net/mac80211/mlme.c for actual sending and processing of management frames and the new nl80211 commands will just stop the current state machine from moving automatically from authentication to association. Future cleanup may move more of the MLME operations into cfg80211. The goal of this design is to provide more control of authentication and association process to user space without having to move the full MLME implementation. This should be enough to allow IEEE 802.11r FT protocol and 802.11s SAE authentication to be implemented. Obviously, this will also bring the extra benefit of not having to use WEXT for association requests with mac80211. An example implementation of a user space SME using the new nl80211 commands is available for wpa_supplicant. This patch is enough to get IEEE 802.11r FT protocol working with over-the-air mechanism (over-the-DS will need additional MLME primitives for handling the FT Action frames). Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:22 +00:00
},
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
{
.cmd = NL80211_CMD_JOIN_IBSS,
.doit = nl80211_join_ibss,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
},
{
.cmd = NL80211_CMD_LEAVE_IBSS,
.doit = nl80211_leave_ibss,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
},
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
#ifdef CONFIG_NL80211_TESTMODE
{
.cmd = NL80211_CMD_TESTMODE,
.doit = nl80211_testmode_do,
nl80211: add testmode dump support This adds dump support to testmode. The testmode dump support in nl80211 requires using two of the six cb->args, the rest can be used by the driver to figure out where the dump position is at or to store other data across invocations. Signed-off-by: Wey-Yi Guy <wey-yi.w.guy@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-20 16:05:54 +00:00
.dumpit = nl80211_testmode_dump,
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
cfg80211: introduce nl80211 testmode command This introduces a new NL80211_CMD_TESTMODE for testing and calibration use with nl80211. There's no multiplexing like like iwpriv had, and the command is not available by default, it needs to be explicitly enabled in Kconfig and shouldn't be enabled in most kernels. The command requires a wiphy index or interface index to identify the device to operate on, and the new TESTDATA attribute. There also is API for sending replies to the command, and testmode multicast messages (on a testmode multicast group). I've also updated mac80211 to be able to pass through the command to the driver, since it itself doesn't implement the testmode command. Additionally, to give people an idea of how to use the command, I've added a little code to hwsim that makes use of the new command to set the powersave mode, this is currently done via debugfs and should remain there, and the testmode command only serves as an example of how to use this best -- with nested netlink attributes in the TESTDATA attribute. A hwsim testmode tool can be found at http://git.sipsolutions.net/hwsim.git/. This tool is BSD licensed so people can easily use it as a basis for their own internal fabrication and validation tools. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:51 +00:00
},
#endif
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
.cmd = NL80211_CMD_CONNECT,
.doit = nl80211_connect,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
},
cfg80211: Add support to update connection parameters Add functionality to update the connection parameters when in connected state, so that driver/firmware uses the updated parameters for subsequent roaming. This is for drivers that support internal BSS selection and roaming. The new command does not change the current association state, i.e., it can be used to update IE contents for future (re)associations without causing an immediate disassociation or reassociation with the current BSS. This commit implements the required functionality for updating IEs for (Re)Association Request frame only. Other parameters can be added in future when required. Signed-off-by: vamsi krishna <vamsin@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-27 13:51:11 +00:00
{
.cmd = NL80211_CMD_UPDATE_CONNECT_PARAMS,
.doit = nl80211_update_connect_params,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
.cmd = NL80211_CMD_DISCONNECT,
.doit = nl80211_disconnect,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use generic check for netif_running Use a new flag that requires the netdev to be UP and use it to check instead of coding the check into all functions that require it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:14:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
},
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
{
.cmd = NL80211_CMD_SET_WIPHY_NETNS,
.doit = nl80211_wiphy_netns,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
},
cfg80211: return channel noise via survey API This patch implements the NL80211_CMD_GET_SURVEY command and an get_survey() ops that a driver can implement. The goal of this command is to allow a drivers to report channel survey data (e.g. channel noise, channel occupation). For now, only the mechanism to report back channel noise has been implemented. In future, there will either be a survey-trigger command --- or the existing scan-trigger command will be enhanced. This will allow user-space to request survey for arbitrary channels. Note: any driver that cannot report channel noise should not report any value at all, e.g. made-up -92 dBm. Signed-off-by: Holger Schurig <holgerschurig@gmail.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-11 11:25:40 +00:00
{
.cmd = NL80211_CMD_GET_SURVEY,
.policy = nl80211_policy,
.dumpit = nl80211_dump_survey,
},
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
{
.cmd = NL80211_CMD_SET_PMKSA,
.doit = nl80211_setdel_pmksa,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
},
{
.cmd = NL80211_CMD_DEL_PMKSA,
.doit = nl80211_setdel_pmksa,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
},
{
.cmd = NL80211_CMD_FLUSH_PMKSA,
.doit = nl80211_flush_pmksa,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
nl80211: PMKSA caching support This is an interface to set, delete and flush PMKIDs through nl80211. Main users would be fullmac devices which firmwares are capable of generating the RSN IEs for the re-association requests, e.g. iwmc3200wifi. Signed-off-by: Samuel Ortiz <sameo@linux.intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-11-24 22:59:15 +00:00
},
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
{
.cmd = NL80211_CMD_REMAIN_ON_CHANNEL,
.doit = nl80211_remain_on_channel,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
},
{
.cmd = NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL,
.doit = nl80211_cancel_remain_on_channel,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
},
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
{
.cmd = NL80211_CMD_SET_TX_BITRATE_MASK,
.doit = nl80211_set_tx_bitrate_mask,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
nl80211: New command for setting TX rate mask for rate control Add a new NL80211_CMD_SET_TX_BITRATE_MASK command and related attributes to provide support for setting TX rate mask for rate control. This uses the existing cfg80211 set_bitrate_mask operation that was previously used only with WEXT compat code (SIOCSIWRATE). The nl80211 command allows more generic configuration of allowed rates as a mask instead of fixed/max rate. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-29 10:59:45 +00:00
},
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
{
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
.cmd = NL80211_CMD_REGISTER_FRAME,
.doit = nl80211_register_mgmt,
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
},
{
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
.cmd = NL80211_CMD_FRAME,
.doit = nl80211_tx_mgmt,
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
nl80211/cfg80211: extend mgmt-tx API for off-channel With p2p, it is sometimes necessary to transmit a frame (typically an action frame) on another channel than the current channel. Enable this through the CMD_FRAME API, and allow it to wait for a response. A new command allows that wait to be aborted. However, allow userspace to specify whether or not it wants to allow off-channel TX, it may actually want to use the same channel only. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-11-25 09:02:29 +00:00
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_FRAME_WAIT_CANCEL,
.doit = nl80211_tx_mgmt_cancel_wait,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
NL80211_FLAG_NEED_RTNL,
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
},
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
{
.cmd = NL80211_CMD_SET_POWER_SAVE,
.doit = nl80211_set_power_save,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
},
{
.cmd = NL80211_CMD_GET_POWER_SAVE,
.doit = nl80211_get_power_save,
.policy = nl80211_policy,
/* can be retrieved by unprivileged users */
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
nl80211: add power save commands The most needed command from nl80211, which Wireless Extensions had, is support for power save mode. Add a simple command to make it possible to enable and disable power save via nl80211. I was also planning about extending the interface, for example adding the timeout value, but after thinking more about this I decided not to do it. Basically there were three reasons: Firstly, the parameters for power save are very much hardware dependent. Trying to find a unified interface which would work with all hardware, and still make sense to users, will be very difficult. Secondly, IEEE 802.11 power save implementation in Linux is still in state of flux. We have a long way to still to go and there is no way to predict what kind of implementation we will have after few years. And because we need to support nl80211 interface a long time, practically forever, adding now parameters to nl80211 might create maintenance problems later on. Third issue are the users. Power save parameters are mostly used for debugging, so debugfs is better, more flexible, interface for this. For example, wpa_supplicant currently doesn't configure anything related to power save mode. It's better to strive that kernel can automatically optimise the power save parameters, like with help of pm qos network and other traffic parameters. Later on, when we have better understanding of power save, we can extend this command with more features, if there's a need for that. Signed-off-by: Kalle Valo <kalle.valo@nokia.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-17 15:58:10 +00:00
},
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
{
.cmd = NL80211_CMD_SET_CQM,
.doit = nl80211_set_cqm,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
},
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
{
.cmd = NL80211_CMD_SET_CHANNEL,
.doit = nl80211_set_channel,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use the new genetlink pre/post_doit hooks This makes nl80211 use the new genetlink pre_doit/post_doit hooks for locking and checking the interface/wiphy index. This significantly reduces the code size and the likelihood of locking errors. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-04 19:36:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
cfg80211/mac80211: better channel handling Currently (all tested with hwsim) you can do stupid things like setting up an AP on a certain channel, then adding another virtual interface and making that associate on another channel -- this will make the beaconing to move channel but obviously without the necessary IEs data update. In order to improve this situation, first make the configuration APIs (cfg80211 and nl80211) aware of multi-channel operation -- we'll eventually need that in the future anyway. There's one userland API change and one API addition. The API change is that now SET_WIPHY must be called with virtual interface index rather than only wiphy index in order to take effect for that interface -- luckily all current users (hostapd) do that. For monitor interfaces, the old setting is preserved, but monitors are always slaved to other devices anyway so no guarantees. The second userland API change is the introduction of a per virtual interface SET_CHANNEL command, that hostapd should use going forward to make it easier to understand what's going on (it can automatically detect a kernel with this command). Other than mac80211, no existing cfg80211 drivers are affected by this change because they only allow a single virtual interface. mac80211, however, now needs to be aware that the channel settings are per interface now, and needs to disallow (for now) real multi-channel operation, which is another important part of this patch. One of the immediate benefits is that you can now start hostapd to operate on a hardware that already has a connection on another virtual interface, as long as you specify the same channel. Note that two things are left unhandled (this is an improvement -- not a complete fix): * different HT/no-HT modes currently you could start an HT AP and then connect to a non-HT network on the same channel which would configure the hardware for no HT; that can be fixed fairly easily * CSA An AP we're connected to on a virtual interface might indicate switching channels, and in that case we would follow it, regardless of how many other interfaces are operating; this requires more effort to fix but is pretty rare after all Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-05-05 13:25:02 +00:00
},
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
{
.cmd = NL80211_CMD_SET_WDS_PEER,
.doit = nl80211_set_wds_peer,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: use new genl helpers for WDS Bill Jordan's patch to allow setting the WDS peer crossed with my patch removing all the boilerplate code in nl80211, and consequently he didn't make use of it yet. Fix that. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-07 11:10:30 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
cfg80211: patches to allow setting the WDS peer Added a nl interface to set the peer bssid of a WDS interface. Signed-off-by: Bill Jordan <bjordan@rajant.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-10-01 17:54:28 +00:00
},
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
{
.cmd = NL80211_CMD_JOIN_MESH,
.doit = nl80211_join_mesh,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_LEAVE_MESH,
.doit = nl80211_leave_mesh,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/mac80211: add mesh join/leave commands Instead of tying mesh activity to interface up, add join and leave commands for mesh. Since we must be backward compatible, let cfg80211 handle joining a mesh if a mesh ID was pre-configured when the device goes up. Note that this therefore must modify mac80211 as well since mac80211 needs to lose the logic to start the mesh on interface up. We now allow querying mesh parameters before the mesh is connected, which simply returns defaults. Setting them (internally renamed to "update") is only allowed while connected. Specify them with the new mesh join command instead where needed. In mac80211, beaconing must now also follow the mesh enabled/not enabled state, which is done by testing the mesh ID. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-03 08:20:44 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: 802.11p OCB mode handling This patch adds new iface type (NL80211_IFTYPE_OCB) representing the OCB (Outside the Context of a BSS) mode. When establishing a connection to the network a cfg80211_join_ocb function is called (particular nl80211_command is added as well). A mandatory parameters during the ocb_join operation are 'center frequency' and 'channel width (5/10 MHz)'. Changes done in mac80211 are minimal possible required to avoid many warnings (warning: enumeration value 'NL80211_IFTYPE_OCB' not handled in switch) during compilation. Full functionality (where needed) is added in the following patch. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-03 09:33:18 +00:00
{
.cmd = NL80211_CMD_JOIN_OCB,
.doit = nl80211_join_ocb,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: 802.11p OCB mode handling This patch adds new iface type (NL80211_IFTYPE_OCB) representing the OCB (Outside the Context of a BSS) mode. When establishing a connection to the network a cfg80211_join_ocb function is called (particular nl80211_command is added as well). A mandatory parameters during the ocb_join operation are 'center frequency' and 'channel width (5/10 MHz)'. Changes done in mac80211 are minimal possible required to avoid many warnings (warning: enumeration value 'NL80211_IFTYPE_OCB' not handled in switch) during compilation. Full functionality (where needed) is added in the following patch. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-03 09:33:18 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_LEAVE_OCB,
.doit = nl80211_leave_ocb,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: 802.11p OCB mode handling This patch adds new iface type (NL80211_IFTYPE_OCB) representing the OCB (Outside the Context of a BSS) mode. When establishing a connection to the network a cfg80211_join_ocb function is called (particular nl80211_command is added as well). A mandatory parameters during the ocb_join operation are 'center frequency' and 'channel width (5/10 MHz)'. Changes done in mac80211 are minimal possible required to avoid many warnings (warning: enumeration value 'NL80211_IFTYPE_OCB' not handled in switch) during compilation. Full functionality (where needed) is added in the following patch. Signed-off-by: Rostislav Lisovy <rostislav.lisovy@fel.cvut.cz> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-03 09:33:18 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: don't allow WoWLAN support without CONFIG_PM When CONFIG_PM is disabled, no device can possibly support WoWLAN since it can't go to sleep to start with. Due to this, mac80211 had even rejected the hardware registration. By making all the code and data for WoWLAN depend on CONFIG_PM we can promote this runtime error to a compile-time error. Add #ifdef around all WoWLAN code to remove it in systems that don't need it as they never suspend. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Acked-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 07:23:48 +00:00
#ifdef CONFIG_PM
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
{
.cmd = NL80211_CMD_GET_WOWLAN,
.doit = nl80211_get_wowlan,
.policy = nl80211_policy,
/* can be retrieved by unprivileged users */
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_SET_WOWLAN,
.doit = nl80211_set_wowlan,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211/cfg80211: WoWLAN support This is based on (but now quite far from) the original work from Luis and Eliad. Add support for configuring WoWLAN triggers, and getting the configuration out again. Changes from the original patchset are too numerous to list, but one important change needs highlighting: the suspend() callback is passed NULL for the trigger configuration if userspace has not configured WoWLAN at all. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: Eliad Peller <eliad@wizery.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-04 13:37:28 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: don't allow WoWLAN support without CONFIG_PM When CONFIG_PM is disabled, no device can possibly support WoWLAN since it can't go to sleep to start with. Due to this, mac80211 had even rejected the hardware registration. By making all the code and data for WoWLAN depend on CONFIG_PM we can promote this runtime error to a compile-time error. Add #ifdef around all WoWLAN code to remove it in systems that don't need it as they never suspend. Cc: Kalle Valo <kvalo@qca.qualcomm.com> Acked-by: Luciano Coelho <coelho@ti.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-27 07:23:48 +00:00
#endif
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
{
.cmd = NL80211_CMD_SET_REKEY_OFFLOAD,
.doit = nl80211_set_rekey_data,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
cfg80211: clear nl80211 messages carrying keys after processing Clear any nl80211 messages that might contain keys after processing them to avoid leaving their data in memory "forever" after they've been freed. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-10 12:00:16 +00:00
NL80211_FLAG_NEED_RTNL |
NL80211_FLAG_CLEAR_SKB,
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
},
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
{
.cmd = NL80211_CMD_TDLS_MGMT,
.doit = nl80211_tdls_mgmt,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_TDLS_OPER,
.doit = nl80211_tdls_oper,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: support sending TDLS commands/frames Add support for sending high-level TDLS commands and TDLS frames via NL80211_CMD_TDLS_OPER and NL80211_CMD_TDLS_MGMT, respectively. Add appropriate cfg80211 callbacks for lower level drivers. Add wiphy capability flags for TDLS support and advertise them via nl80211. Signed-off-by: Arik Nemtsov <arik@wizery.com> Cc: Kalyan C Gaddam <chakkal@iit.edu> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-28 11:12:50 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
{
.cmd = NL80211_CMD_UNEXPECTED_FRAME,
.doit = nl80211_register_unexpected_frame,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
},
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
{
.cmd = NL80211_CMD_PROBE_CLIENT,
.doit = nl80211_probe_client,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: ensure interface is up in various APIs The nl80211 handling code should ensure as much as it can that the interface is in a valid state, it can certainly ensure the interface is running. Not doing so can cause calls through mac80211 into the driver that result in warnings and unspecified behaviour in the driver. Cc: stable@vger.kernel.org Reported-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-02 08:51:55 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
NL80211_FLAG_NEED_RTNL,
},
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
{
.cmd = NL80211_CMD_REGISTER_BEACONS,
.doit = nl80211_register_beacons,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
},
wireless: Add NoAck per tid support This patch contains the configuration changes in nl80211/cfg80211. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 13:20:43 +00:00
{
.cmd = NL80211_CMD_SET_NOACK_MAP,
.doit = nl80211_set_noack_map,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
wireless: Add NoAck per tid support This patch contains the configuration changes in nl80211/cfg80211. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-18 13:20:43 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
{
.cmd = NL80211_CMD_START_P2P_DEVICE,
.doit = nl80211_start_p2p_device,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_STOP_P2P_DEVICE,
.doit = nl80211_stop_p2p_device,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
cfg80211: add start / stop NAN commands This allows user space to start/stop NAN interface. A NAN interface is like P2P device in a few aspects: it doesn't have a netdev associated to it. Add the new interface type and prevent operations that can't be executed on NAN interface like scan. Define several attributes that may be configured by user space when starting NAN functionality (master preference and dual band operation) Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:13 +00:00
},
{
.cmd = NL80211_CMD_START_NAN,
.doit = nl80211_start_nan,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_WDEV |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_STOP_NAN,
.doit = nl80211_stop_nan,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
cfg80211: add add_nan_func / del_nan_func A NAN function can be either publish, subscribe or follow up. Make all the necessary verifications and just pass the request to the driver. Allow the user space application that starts NAN to forbid any other socket to add or remove functions. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Ayala Beker <ayala.beker@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:15 +00:00
},
{
.cmd = NL80211_CMD_ADD_NAN_FUNCTION,
.doit = nl80211_nan_add_func,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_DEL_NAN_FUNCTION,
.doit = nl80211_nan_del_func,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
cfg80211: allow the user space to change current NAN configuration Some NAN configuration paramaters may change during the operation of the NAN device. For example, a user may want to update master preference value when the device gets plugged/unplugged to the power. Add API that allows to do so. Signed-off-by: Andrei Otcheretianski <andrei.otcheretianski@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-09-20 14:31:16 +00:00
},
{
.cmd = NL80211_CMD_CHANGE_NAN_CONFIG,
.doit = nl80211_nan_change_config,
.policy = nl80211_policy,
.flags = GENL_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
cfg80211: add P2P Device abstraction In order to support using a different MAC address for the P2P Device address we must first have a P2P Device abstraction that can be assigned a MAC address. This abstraction will also be useful to support offloading P2P operations to the device, e.g. periodic listen for discoverability. Currently, the driver is responsible for assigning a MAC address to the P2P Device, but this could be changed by allowing a MAC address to be given to the NEW_INTERFACE command. As it has no associated netdev, a P2P Device can only be identified by its wdev identifier but the previous patches allowed using the wdev identifier in various APIs, e.g. remain-on-channel. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 22:19:54 +00:00
},
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
{
.cmd = NL80211_CMD_SET_MCAST_RATE,
.doit = nl80211_set_mcast_rate,
cfg80211/nl80211: add API for MAC address ACLs Add API to enable drivers to implement MAC address based access control in AP/P2P GO mode. Capable drivers advertise this capability by setting the maximum number of MAC addresses in such a list in wiphy->max_acl_mac_addrs. An initial ACL may be given to the NL80211_CMD_START_AP command and/or changed later with NL80211_CMD_SET_MAC_ACL. Black- and whitelists are supported, but not simultaneously. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> [rewrite commit log, many cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-18 05:48:45 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/nl80211: add API for MAC address ACLs Add API to enable drivers to implement MAC address based access control in AP/P2P GO mode. Capable drivers advertise this capability by setting the maximum number of MAC addresses in such a list in wiphy->max_acl_mac_addrs. An initial ACL may be given to the NL80211_CMD_START_AP command and/or changed later with NL80211_CMD_SET_MAC_ACL. Black- and whitelists are supported, but not simultaneously. Signed-off-by: Vasanthakumar Thiagarajan <vthiagar@qca.qualcomm.com> [rewrite commit log, many cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-18 05:48:45 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_SET_MAC_ACL,
.doit = nl80211_set_mac_acl,
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl/cfg80211: add the NL80211_CMD_SET_MCAST_RATE command This command triggers a new callback: set_mcast_rate(). It enables the user to change the rate used to send multicast frames for vif configured as IBSS or MESH_POINT Signed-off-by: Antonio Quartulli <ordex@autistici.org> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-02 12:27:48 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
},
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
{
.cmd = NL80211_CMD_RADAR_DETECT,
.doit = nl80211_start_radar_detection,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
nl80211: allow splitting wiphy information in dumps The per-wiphy information is getting large, to the point where with more than the typical number of channels it's too large and overflows, and userspace can't get any of the information at all. To address this (in a way that doesn't require making all messages bigger) allow userspace to specify that it can deal with wiphy information split across multiple parts of the dump, and if it can split up the data. This also splits up each channel separately so an arbitrary number of channels can be supported. Additionally, since GET_WIPHY has the same problem, add support for filtering the wiphy dump and get information for a single wiphy only, this allows userspace apps to use dump in this case to retrieve all data from a single device. As userspace needs to know if all this this is supported, add a global nl80211 feature set and include a bit for this behaviour in it. Cc: Dennis H Jensen <dennis.h.jensen@siemens.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-14 15:19:38 +00:00
{
.cmd = NL80211_CMD_GET_PROTOCOL_FEATURES,
.doit = nl80211_get_protocol_features,
.policy = nl80211_policy,
},
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
{
.cmd = NL80211_CMD_UPDATE_FT_IES,
.doit = nl80211_update_ft_ies,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
{
.cmd = NL80211_CMD_CRIT_PROTOCOL_START,
.doit = nl80211_crit_protocol_start,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_CRIT_PROTOCOL_STOP,
.doit = nl80211_crit_protocol_stop,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
.internal_flags = NL80211_FLAG_NEED_WDEV_UP |
NL80211_FLAG_NEED_RTNL,
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
},
{
.cmd = NL80211_CMD_GET_COALESCE,
.doit = nl80211_get_coalesce,
.policy = nl80211_policy,
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_SET_COALESCE,
.doit = nl80211_set_coalesce,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211/nl80211: Add packet coalesce support In most cases, host that receives IPv4 and IPv6 multicast/broadcast packets does not do anything with these packets. Therefore the reception of these unwanted packets causes unnecessary processing and power consumption. Packet coalesce feature helps to reduce number of received interrupts to host by buffering these packets in firmware/hardware for some predefined time. Received interrupt will be generated when one of the following events occur. a) Expiration of hardware timer whose expiration time is set to maximum coalescing delay of matching coalesce rule. b) Coalescing buffer in hardware reaches it's limit. c) Packet doesn't match any of the configured coalesce rules. This patch adds set/get configuration support for packet coalesce. User needs to configure following parameters for creating a coalesce rule. a) Maximum coalescing delay b) List of packet patterns which needs to be matched c) Condition for coalescence. pattern 'match' or 'no match' Multiple such rules can be created. This feature needs to be advertised during driver initialization. Drivers are supposed to do required firmware/hardware settings based on user configuration. Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> [fix kernel-doc, change free function, fix copy/paste error] Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
2013-06-28 18:51:26 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
},
{
.cmd = NL80211_CMD_CHANNEL_SWITCH,
.doit = nl80211_channel_switch,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211/cfg80211: add channel switch command To allow channel switch announcements within beacons, add the channel switch command to nl80211/cfg80211. This is implementation is intended for AP and (later) IBSS mode. Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> Signed-off-by: Mathias Kretschmer <mathias.kretschmer@fokus.fraunhofer.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-07-11 14:09:05 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
{
.cmd = NL80211_CMD_VENDOR,
.doit = nl80211_vendor_cmd,
nl80211: support vendor dumpit commands In order to transfer many items in vendor commands, support the dumpit netlink method for them. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-08-15 19:39:49 +00:00
.dumpit = nl80211_vendor_cmd_dump,
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
nl80211: vendor command support Add support for vendor-specific commands to nl80211. This is intended to be used for really vendor-specific functionality that can't be implemented in a generic fashion for any reason. It's *NOT* intended to be used for any normal/generic feature or any optimisations that could be implemented across drivers. Currently, only vendor commands (with replies) are supported, no dump operations or vendor-specific notifications. Also add a function wdev_to_ieee80211_vif() to mac80211 which is needed for mac80211-based drivers wanting to implement any vendor commands. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-13 12:37:47 +00:00
.internal_flags = NL80211_FLAG_NEED_WIPHY |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: Add support for QoS mapping This allows QoS mapping from external networks to be implemented as defined in IEEE Std 802.11-2012, 10.24.9. APs can use this to advertise DSCP ranges and exceptions for mapping frames to a specific UP over Wi-Fi. The payload of the QoS Map Set element (IEEE Std 802.11-2012, 8.4.2.97) is sent to the driver through the new NL80211_ATTR_QOS_MAP attribute to configure the local behavior either on the AP (based on local configuration) or on a station (based on information received from the AP). Signed-off-by: Kyeyoon Park <kyeyoonp@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-17 07:01:30 +00:00
{
.cmd = NL80211_CMD_SET_QOS_MAP,
.doit = nl80211_set_qos_map,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: Add support for QoS mapping This allows QoS mapping from external networks to be implemented as defined in IEEE Std 802.11-2012, 10.24.9. APs can use this to advertise DSCP ranges and exceptions for mapping frames to a specific UP over Wi-Fi. The payload of the QoS Map Set element (IEEE Std 802.11-2012, 8.4.2.97) is sent to the driver through the new NL80211_ATTR_QOS_MAP attribute to configure the local behavior either on the AP (based on local configuration) or on a station (based on information received from the AP). Signed-off-by: Kyeyoon Park <kyeyoonp@qca.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-12-17 07:01:30 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
{
.cmd = NL80211_CMD_ADD_TX_TS,
.doit = nl80211_add_tx_ts,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_DEL_TX_TS,
.doit = nl80211_del_tx_ts,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: add WMM traffic stream API Add nl80211 and driver API to validate, add and delete traffic streams with appropriate settings. The API calls for userspace doing the action frame handshake with the peer, and then allows only to set up the parameters in the driver. To avoid setting up a session only to tear it down again, the validate API is provided, but the real usage later can still fail so userspace must be prepared for that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 19:55:35 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
{
.cmd = NL80211_CMD_TDLS_CHANNEL_SWITCH,
.doit = nl80211_tdls_channel_switch,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_TDLS_CANCEL_CHANNEL_SWITCH,
.doit = nl80211_tdls_cancel_channel_switch,
.policy = nl80211_policy,
nl80211: Allow privileged operations from user namespaces While a wiphy can be transferred to network namespaces, a process having CAP_NET_ADMIN in a non-initial user namespace can not administrate such devices due to the genetlink GENL_ADMIN_PERM restrictions. For openvswitch having the same issue, a new GENL_UNS_ADMIN_PERM flag has been introduced, commit 4a92602aa1cd ("openvswitch: allow management from inside user namespaces"). This patch changes all privileged operations operating on a wiphy, dev or wdev to allow their administration using the same mechanism. All operations use either NEED_WIPHY, NEED_WDEV or NEED_NETDEV, which implies a namespace aware lookup of the device. The only exception is NL80211_CMD_SET_WIPHY, which explicitly uses a namespace aware phy lookup. Signed-off-by: Martin Willi <martin@strongswan.org> [also allow cancel scan, for completeness] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-09 16:33:58 +00:00
.flags = GENL_UNS_ADMIN_PERM,
cfg80211: introduce TDLS channel switch commands Introduce commands to initiate and cancel TDLS channel-switching. Once TDLS channel-switching is started, the lower level driver is responsible for continually initiating channel-switch operations and returning to the base (AP) channel to listen for beacons from time to time. Upon cancellation of the channel-switch all communication between the relevant TDLS peers will continue on the base channel. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-19 10:54:26 +00:00
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: configure multicast to unicast for AP interfaces Add the ability to configure if an AP (and associated VLANs) will do multicast-to-unicast conversion for ARP, IPv4 and IPv6 frames (possibly within 802.1Q). If enabled, such frames are to be sent to each station separately, with the DA replaced by their own MAC address rather than the group address. Note that this may break certain expectations of the receiver, such as the ability to drop unicast IP packets received within multicast L2 frames, or the ability to not send ICMP destination unreachable messages for packets received in L2 multicast (which is required, but the receiver can't tell the difference if this new option is enabled.) This also doesn't implement the 802.11 DMS (directed multicast service). Signed-off-by: Michael Braun <michael-dev@fami-braun.de> [fix disabling, add better documentation & commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-10-10 17:12:22 +00:00
{
.cmd = NL80211_CMD_SET_MULTICAST_TO_UNICAST,
.doit = nl80211_set_multicast_to_unicast,
.policy = nl80211_policy,
.flags = GENL_UNS_ADMIN_PERM,
.internal_flags = NL80211_FLAG_NEED_NETDEV |
NL80211_FLAG_NEED_RTNL,
},
cfg80211: support 4-way handshake offloading for 802.1X Add API for setting the PMK to the driver. For FT support, allow setting also the PMK-R0 Name. This can be used by drivers that support 4-Way handshake offload while IEEE802.1X authentication is managed by upper layers. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com> [arend.vanspriel@broadcom.com: add WANT_1X_4WAY_HS attribute] Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [reword NL80211_EXT_FEATURE_4WAY_HANDSHAKE_STA_1X docs a bit to say that the device may require it] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-06-09 12:08:43 +00:00
{
.cmd = NL80211_CMD_SET_PMK,
.doit = nl80211_set_pmk,
.policy = nl80211_policy,
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
{
.cmd = NL80211_CMD_DEL_PMK,
.doit = nl80211_del_pmk,
.policy = nl80211_policy,
.internal_flags = NL80211_FLAG_NEED_NETDEV_UP |
NL80211_FLAG_NEED_RTNL,
},
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
};
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
genetlink: mark families as __ro_after_init Now genl_register_family() is the only thing (other than the users themselves, perhaps, but I didn't find any doing that) writing to the family struct. In all families that I found, genl_register_family() is only called from __init functions (some indirectly, in which case I've add __init annotations to clarifly things), so all can actually be marked __ro_after_init. This protects the data structure from accidental corruption. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:05 +00:00
static struct genl_family nl80211_fam __ro_after_init = {
genetlink: statically initialize families Instead of providing macros/inline functions to initialize the families, make all users initialize them statically and get rid of the macros. This reduces the kernel code size by about 1.6k on x86-64 (with allyesconfig). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:03 +00:00
.name = NL80211_GENL_NAME, /* have users key off the name instead */
.hdrsize = 0, /* no private header */
.version = 1, /* no particular meaning now */
.maxattr = NL80211_ATTR_MAX,
.netnsok = true,
.pre_doit = nl80211_pre_doit,
.post_doit = nl80211_post_doit,
.module = THIS_MODULE,
.ops = nl80211_ops,
.n_ops = ARRAY_SIZE(nl80211_ops),
.mcgrps = nl80211_mcgrps,
.n_mcgrps = ARRAY_SIZE(nl80211_mcgrps),
};
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* notification functions */
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
void nl80211_notify_wiphy(struct cfg80211_registered_device *rdev,
enum nl80211_commands cmd)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
struct sk_buff *msg;
nl80211: use small state buffer for wiphy_dump Avoid parsing the original dump message again and again by allocating a small state struct that is used by the functions involved in the dump, storing this struct in cb->args[0]. This reduces the memory allocation size as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-19 08:57:22 +00:00
struct nl80211_dump_wiphy_state state = {};
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
WARN_ON(cmd != NL80211_CMD_NEW_WIPHY &&
cmd != NL80211_CMD_DEL_WIPHY);
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (!msg)
return;
cfg80211: send events when devices are added/removed We're currently sending NEW_WIPHY events for renames (which is a bit odd, but now can't be changed), but also send them for really new devices that register. Also send DEL_WIPHY events when a device is removed, the event ID for this was already reserved. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-05-26 11:52:25 +00:00
if (nl80211_send_wiphy(rdev, cmd, msg, 0, 0, 0, &state) < 0) {
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
nlmsg_free(msg);
return;
}
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_CONFIG, GFP_KERNEL);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
}
cfg80211: always notify userspace of new wireless netdevs This change alters the semantics of NL80211_CMD_NEW_INTERFACE events by always sending this event whenever a new net_device object associated with a wdev is registered. Prior to this change, this event was only sent as a result of NL80211_CMD_NEW_INTERFACE command sent from userspace. This allows userspace to reliably detect new wireless interfaces (e.g. due to hardware hot-plug events, etc). For wdevs created without an associated net_device object (e.g. NL80211_IFTYPE_P2P_DEVICE), the NL80211_CMD_NEW_INTERFACE event is still generated inside the relevant nl80211 command handler. Signed-off-by: Denis Kenzior <denkenz@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-08-03 21:58:33 +00:00
void nl80211_notify_iface(struct cfg80211_registered_device *rdev,
struct wireless_dev *wdev,
enum nl80211_commands cmd)
{
struct sk_buff *msg;
WARN_ON(cmd != NL80211_CMD_NEW_INTERFACE &&
cmd != NL80211_CMD_DEL_INTERFACE);
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return;
if (nl80211_send_iface(msg, 0, 0, 0, rdev, wdev,
cmd == NL80211_CMD_DEL_INTERFACE) < 0) {
nlmsg_free(msg);
return;
}
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
NL80211_MCGRP_CONFIG, GFP_KERNEL);
}
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
static int nl80211_add_scan_req(struct sk_buff *msg,
struct cfg80211_registered_device *rdev)
{
struct cfg80211_scan_request *req = rdev->scan_req;
struct nlattr *nest;
int i;
if (WARN_ON(!req))
return 0;
nest = nla_nest_start(msg, NL80211_ATTR_SCAN_SSIDS);
if (!nest)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
for (i = 0; i < req->n_ssids; i++) {
if (nla_put(msg, i, req->ssids[i].ssid_len, req->ssids[i].ssid))
goto nla_put_failure;
}
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
nla_nest_end(msg, nest);
nest = nla_nest_start(msg, NL80211_ATTR_SCAN_FREQUENCIES);
if (!nest)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
for (i = 0; i < req->n_channels; i++) {
if (nla_put_u32(msg, i, req->channels[i]->center_freq))
goto nla_put_failure;
}
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
nla_nest_end(msg, nest);
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (req->ie &&
nla_put(msg, NL80211_ATTR_IE, req->ie_len, req->ie))
goto nla_put_failure;
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (req->flags &&
nla_put_u32(msg, NL80211_ATTR_SCAN_FLAGS, req->flags))
goto nla_put_failure;
{nl,cfg}80211: add a flags word to scan requests Add a flags word to direct and scheduled scan requests; it will be used for control of optional behaviours such as flushing the bss cache prior to doing a scan. Signed-off-by: Sam Leffler <sleffler@chromium.org> Tested-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Amitkumar Karwar <akarwar@marvell.com> Signed-off-by: Bing Zhao <bzhao@marvell.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-12 04:03:31 +00:00
nl80211: support beacon report scanning Beacon report radio measurement requires reporting observed BSSs on the channels specified in the beacon request. If the measurement mode is set to passive or active, it requires actually performing a scan (passive or active, accordingly), and reporting the time that the scan was started and the time each beacon/probe was received (both in terms of TSF of the BSS of the requesting AP). If the request mode is table, this information is optional. In addition, the radio measurement request specifies the channel dwell time for the measurement. In order to use scan for beacon report when the mode is active or passive, add a parameter to scan request that specifies the channel dwell time, and add scan start time and beacon received time to scan results information. Supporting beacon report is required for Multi Band Operation (MBO). Signed-off-by: Assaf Krauss <assaf.krauss@intel.com> Signed-off-by: David Spinadel <david.spinadel@intel.com> Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-07-05 14:10:13 +00:00
if (req->info.scan_start_tsf &&
(nla_put_u64_64bit(msg, NL80211_ATTR_SCAN_START_TIME_TSF,
req->info.scan_start_tsf, NL80211_BSS_PAD) ||
nla_put(msg, NL80211_ATTR_SCAN_START_TIME_TSF_BSSID, ETH_ALEN,
req->info.tsf_bssid)))
goto nla_put_failure;
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
return 0;
nla_put_failure:
return -ENOBUFS;
}
nl80211: rework {sched_,}scan event related functions A couple of functions used with scan events were named with term "send" although they were only preparing the the event message so renamed those. Also remove nl80211_send_sched_scan_results() in favor of just calling nl80211_send_sched_scan() with the right value. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [mention nl80211_send_sched_scan_results() in the commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-16 11:21:54 +00:00
static int nl80211_prep_scan_msg(struct sk_buff *msg,
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
struct cfg80211_registered_device *rdev,
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
struct wireless_dev *wdev,
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
u32 portid, u32 seq, int flags,
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
u32 cmd)
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
void *hdr;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
hdr = nl80211hdr_put(msg, portid, seq, flags, cmd);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (!hdr)
return -1;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
(wdev->netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex)) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: bounce scan request back to userspace When a scan finishes only the program that asked for it knows what kind of scan it was; let's tell everybody else about the scan parameters as well so they can evaluate the result of the scan better. Also helps with debugging. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-24 14:43:15 +00:00
/* ignore errors and send incomplete event anyway */
nl80211_add_scan_req(msg, rdev);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
static int
nl80211: rework {sched_,}scan event related functions A couple of functions used with scan events were named with term "send" although they were only preparing the the event message so renamed those. Also remove nl80211_send_sched_scan_results() in favor of just calling nl80211_send_sched_scan() with the right value. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [mention nl80211_send_sched_scan_results() in the commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-16 11:21:54 +00:00
nl80211_prep_sched_scan_msg(struct sk_buff *msg,
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
struct cfg80211_sched_scan_request *req, u32 cmd)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
{
void *hdr;
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
if (!hdr)
return -1;
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY,
wiphy_to_rdev(req->wiphy)->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, req->dev->ifindex) ||
nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, req->reqid,
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
netlink: make nlmsg_end() and genlmsg_end() void Contrary to common expectations for an "int" return, these functions return only a positive value -- if used correctly they cannot even return 0 because the message header will necessarily be in the skb. This makes the very common pattern of if (genlmsg_end(...) < 0) { ... } be a whole bunch of dead code. Many places also simply do return nlmsg_end(...); and the caller is expected to deal with it. This also commonly (at least for me) causes errors, because it is very common to write if (my_function(...)) /* error condition */ and if my_function() does "return nlmsg_end()" this is of course wrong. Additionally, there's not a single place in the kernel that actually needs the message length returned, and if anyone needs it later then it'll be very easy to just use skb->len there. Remove this, and make the functions void. This removes a bunch of dead code as described above. The patch adds lines because I did - return nlmsg_end(...); + nlmsg_end(...); + return 0; I could have preserved all the function's return values by returning skb->len, but instead I've audited all the places calling the affected functions and found that none cared. A few places actually compared the return value with <= 0 in dump functionality, but that could just be changed to < 0 with no change in behaviour, so I opted for the more efficient version. One instance of the error I've made numerous times now is also present in net/phonet/pn_netlink.c in the route_dumpit() function - it didn't check for <0 or <=0 and thus broke out of the loop every single time. I've preserved this since it will (I think) have caused the messages to userspace to be formatted differently with just a single message for every SKB returned to userspace. It's possible that this isn't needed for the tools that actually use this, but I don't even know what they are so couldn't test that changing this behaviour would be acceptable. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2015-01-16 21:09:00 +00:00
genlmsg_end(msg, hdr);
return 0;
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
return -EMSGSIZE;
}
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
void nl80211_send_scan_start(struct cfg80211_registered_device *rdev,
nl80211: move scan API to wdev The new P2P Device will have to be able to scan for P2P search, so move scanning to use struct wireless_dev instead of struct net_device. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-18 17:17:03 +00:00
struct wireless_dev *wdev)
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
{
struct sk_buff *msg;
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
if (!msg)
return;
nl80211: rework {sched_,}scan event related functions A couple of functions used with scan events were named with term "send" although they were only preparing the the event message so renamed those. Also remove nl80211_send_sched_scan_results() in favor of just calling nl80211_send_sched_scan() with the right value. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [mention nl80211_send_sched_scan_results() in the commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-16 11:21:54 +00:00
if (nl80211_prep_scan_msg(msg, rdev, wdev, 0, 0, 0,
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
NL80211_CMD_TRIGGER_SCAN) < 0) {
nlmsg_free(msg);
return;
}
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_SCAN, GFP_KERNEL);
cfg80211: issue netlink notification when scan starts To ease multiple apps working together smoothly, send a notification when a scan is started. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-06-16 17:56:42 +00:00
}
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
struct sk_buff *nl80211_build_scan_msg(struct cfg80211_registered_device *rdev,
struct wireless_dev *wdev, bool aborted)
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
{
struct sk_buff *msg;
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
if (!msg)
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
return NULL;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nl80211: rework {sched_,}scan event related functions A couple of functions used with scan events were named with term "send" although they were only preparing the the event message so renamed those. Also remove nl80211_send_sched_scan_results() in favor of just calling nl80211_send_sched_scan() with the right value. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [mention nl80211_send_sched_scan_results() in the commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-16 11:21:54 +00:00
if (nl80211_prep_scan_msg(msg, rdev, wdev, 0, 0, 0,
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
aborted ? NL80211_CMD_SCAN_ABORTED :
NL80211_CMD_NEW_SCAN_RESULTS) < 0) {
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
nlmsg_free(msg);
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
return NULL;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
}
cfg80211: send scan results from work queue Due to the previous commit, when a scan finishes, it is in theory possible to hit the following sequence: 1. interface starts being removed 2. scan is cancelled by driver and cfg80211 is notified 3. scan done work is scheduled 4. interface is removed completely, rdev->scan_req is freed, event sent to userspace but scan done work remains pending 5. new scan is requested on another virtual interface 6. scan done work runs, freeing the still-running scan To fix this situation, hang on to the scan done message and block new scans while that is the case, and only send the message from the work function, regardless of whether the scan_req is already freed from interface removal. This makes step 5 above impossible and changes step 6 to be 5. scan done work runs, sending the scan done message As this can't work for wext, so we send the message immediately, but this shouldn't be an issue since we still return -EBUSY. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-22 09:14:19 +00:00
return msg;
cfg80211/nl80211: scanning (and mac80211 update to use it) This patch adds basic scan capability to cfg80211/nl80211 and changes mac80211 to use it. The BSS list that cfg80211 maintains is made driver-accessible with a private area in each BSS struct, but mac80211 doesn't yet use it. That's another large project. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-02-10 20:25:55 +00:00
}
nl80211: rework {sched_,}scan event related functions A couple of functions used with scan events were named with term "send" although they were only preparing the the event message so renamed those. Also remove nl80211_send_sched_scan_results() in favor of just calling nl80211_send_sched_scan() with the right value. Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [mention nl80211_send_sched_scan_results() in the commit log] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-12-16 11:21:54 +00:00
/* send message created by nl80211_build_scan_msg() */
void nl80211_send_scan_msg(struct cfg80211_registered_device *rdev,
struct sk_buff *msg)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
{
if (!msg)
return;
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_SCAN, GFP_KERNEL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
}
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
void nl80211_send_sched_scan(struct cfg80211_sched_scan_request *req, u32 cmd)
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
{
struct sk_buff *msg;
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
if (!msg)
return;
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
if (nl80211_prep_sched_scan_msg(msg, req, cmd) < 0) {
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
nlmsg_free(msg);
return;
}
nl80211: add request id in scheduled scan event messages For multi-scheduled scan support in subsequent patch a request id will be added. This patch add this request id to the scheduled scan event messages. For now the request id will always be zero. With multi-scheduled scan its value will inform user-space to which scan the event relates. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-13 12:06:27 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(req->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_SCAN, GFP_KERNEL);
cfg80211/nl80211: add support for scheduled scans Implement new functionality for scheduled scan offload. With this feature we can scan automatically at certain intervals. The idea is that the hardware can perform scan automatically and filter on desired results without waking up the host unnecessarily. Add NL80211_CMD_START_SCHED_SCAN and NL80211_CMD_STOP_SCHED_SCAN commands to the nl80211 interface. When results are available they are reported by NL80211_CMD_SCHED_SCAN_RESULTS events. The userspace is informed when the scheduled scan has stopped with a NL80211_CMD_SCHED_SCAN_STOPPED event, which can be triggered either by the driver or by a call to NL80211_CMD_STOP_SCHED_SCAN. Signed-off-by: Luciano Coelho <coelho@ti.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-05-11 14:09:35 +00:00
}
cfg80211: allow wiphy specific regdomain management Add a new regulatory flag that allows a driver to manage regdomain changes/updates for its own wiphy. A self-managed wiphys only employs regulatory information obtained from the FW and driver and does not use other cfg80211 sources like beacon-hints, country-code IEs and hints from other devices on the same system. Conversely, a self-managed wiphy does not share its regulatory hints with other devices in the system. If a system contains several devices, one or more of which are self-managed, there might be contradictory regulatory settings between them. Usage of flag is generally discouraged. Only use it if the FW/driver is incompatible with non-locally originated hints. A new API lets the driver send a complete regdomain, to be applied on its wiphy only. After a wiphy-specific regdomain change takes place, usermode will get a new type of change notification. The regulatory core also takes care enforce regulatory restrictions, in case some interfaces are on forbidden channels. Signed-off-by: Jonathan Doron <jonathanx.doron@intel.com> Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:00 +00:00
static bool nl80211_reg_change_event_fill(struct sk_buff *msg,
struct regulatory_request *request)
cfg80211: add regulatory netlink multicast group This allows us to send to userspace "regulatory" events. For now we just send an event when we change regulatory domains. We also notify userspace when devices are using their own custom world roaming regulatory domains. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-10 02:07:42 +00:00
{
/* Userspace can always count this one always being set */
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u8(msg, NL80211_ATTR_REG_INITIATOR, request->initiator))
goto nla_put_failure;
if (request->alpha2[0] == '0' && request->alpha2[1] == '0') {
if (nla_put_u8(msg, NL80211_ATTR_REG_TYPE,
NL80211_REGDOM_TYPE_WORLD))
goto nla_put_failure;
} else if (request->alpha2[0] == '9' && request->alpha2[1] == '9') {
if (nla_put_u8(msg, NL80211_ATTR_REG_TYPE,
NL80211_REGDOM_TYPE_CUSTOM_WORLD))
goto nla_put_failure;
} else if ((request->alpha2[0] == '9' && request->alpha2[1] == '8') ||
request->intersect) {
if (nla_put_u8(msg, NL80211_ATTR_REG_TYPE,
NL80211_REGDOM_TYPE_INTERSECTION))
goto nla_put_failure;
} else {
if (nla_put_u8(msg, NL80211_ATTR_REG_TYPE,
NL80211_REGDOM_TYPE_COUNTRY) ||
nla_put_string(msg, NL80211_ATTR_REG_ALPHA2,
request->alpha2))
goto nla_put_failure;
}
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
if (request->wiphy_idx != WIPHY_IDX_INVALID) {
struct wiphy *wiphy = wiphy_idx_to_wiphy(request->wiphy_idx);
if (wiphy &&
nla_put_u32(msg, NL80211_ATTR_WIPHY, request->wiphy_idx))
goto nla_put_failure;
cfg80211: return private regdom for self-managed devices If a device has self-managed regulatory, insist on returning the wiphy specific regdomain if a wiphy-idx is specified. The global regdomain is meaningless for such devices. Also add an attribute for self-managed devices, so usermode can distinguish them as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:01 +00:00
if (wiphy &&
wiphy->regulatory_flags & REGULATORY_WIPHY_SELF_MANAGED &&
nla_put_flag(msg, NL80211_ATTR_WIPHY_SELF_MANAGED_REG))
goto nla_put_failure;
cfg80211: allow usermode to query wiphy specific regdom If a wiphy-idx is specified, the kernel will return the wiphy specific regdomain, if such exists. Otherwise return the global regdom. When no wiphy-idx is specified, return the global regdomain as well as all wiphy-specific regulatory domains in the system, via a new nested list of attributes. Add a new attribute for each wiphy-specific regdomain, for usermode to identify it as such. Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:25:59 +00:00
}
cfg80211: add regulatory netlink multicast group This allows us to send to userspace "regulatory" events. For now we just send an event when we change regulatory domains. We also notify userspace when devices are using their own custom world roaming regulatory domains. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-10 02:07:42 +00:00
cfg80211: allow wiphy specific regdomain management Add a new regulatory flag that allows a driver to manage regdomain changes/updates for its own wiphy. A self-managed wiphys only employs regulatory information obtained from the FW and driver and does not use other cfg80211 sources like beacon-hints, country-code IEs and hints from other devices on the same system. Conversely, a self-managed wiphy does not share its regulatory hints with other devices in the system. If a system contains several devices, one or more of which are self-managed, there might be contradictory regulatory settings between them. Usage of flag is generally discouraged. Only use it if the FW/driver is incompatible with non-locally originated hints. A new API lets the driver send a complete regdomain, to be applied on its wiphy only. After a wiphy-specific regdomain change takes place, usermode will get a new type of change notification. The regulatory core also takes care enforce regulatory restrictions, in case some interfaces are on forbidden channels. Signed-off-by: Jonathan Doron <jonathanx.doron@intel.com> Signed-off-by: Arik Nemtsov <arikx.nemtsov@intel.com> Reviewed-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-12-15 17:26:00 +00:00
return true;
nla_put_failure:
return false;
}
/*
* This can happen on global regulatory changes or device specific settings
* based on custom regulatory domains.
*/
void nl80211_common_reg_change_event(enum nl80211_commands cmd_id,
struct regulatory_request *request)
{
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd_id);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nl80211_reg_change_event_fill(msg, request) == false)
goto nla_put_failure;
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: add regulatory netlink multicast group This allows us to send to userspace "regulatory" events. For now we just send an event when we change regulatory domains. We also notify userspace when devices are using their own custom world roaming regulatory domains. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-10 02:07:42 +00:00
cfg80211: fix circular lock dependency (1) Luis reported this lockdep complaint, that he had also reported earlier but when trying to analyse I had been locking at the wrong code, and never saw the problem: (slightly abridged) ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.31-rc4-wl #6 ------------------------------------------------------- wpa_supplicant/3799 is trying to acquire lock: (cfg80211_mutex){+.+.+.}, at: [<ffffffffa009246a>] cfg80211_get_dev_from_ifindex+0x1a/0x90 [cfg80211] but task is already holding lock: (rtnl_mutex){+.+.+.}, at: [<ffffffff81400ff2>] rtnl_lock+0x12/0x20 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (rtnl_mutex){+.+.+.}: [<ffffffff810857b6>] __lock_acquire+0xd76/0x12b0 [<ffffffff81085dd3>] lock_acquire+0xe3/0x120 [<ffffffff814ee7a4>] mutex_lock_nested+0x44/0x350 [<ffffffff81400ff2>] rtnl_lock+0x12/0x20 [<ffffffffa009f6a5>] nl80211_send_reg_change_event+0x1f5/0x2a0 [cfg80211] [<ffffffffa009529e>] set_regdom+0x28e/0x4c0 [cfg80211] -> #0 (cfg80211_mutex){+.+.+.}: [<ffffffff8108587b>] __lock_acquire+0xe3b/0x12b0 [<ffffffff81085dd3>] lock_acquire+0xe3/0x120 [<ffffffff814ee7a4>] mutex_lock_nested+0x44/0x350 [<ffffffffa009246a>] cfg80211_get_dev_from_ifindex+0x1a/0x90 [cfg80211] [<ffffffffa009813f>] get_rdev_dev_by_info_ifindex+0x6f/0xa0 [cfg80211] [<ffffffffa009b12b>] nl80211_set_interface+0x3b/0x260 [cfg80211] When looking at the correct code, the problem is quite obvious. I'm not entirely sure which code paths lead here, so until I can analyse it better let's just use RCU to avoid the problem. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-25 08:54:13 +00:00
rcu_read_lock();
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_allns(&nl80211_fam, msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
cfg80211: fix circular lock dependency (1) Luis reported this lockdep complaint, that he had also reported earlier but when trying to analyse I had been locking at the wrong code, and never saw the problem: (slightly abridged) ======================================================= [ INFO: possible circular locking dependency detected ] 2.6.31-rc4-wl #6 ------------------------------------------------------- wpa_supplicant/3799 is trying to acquire lock: (cfg80211_mutex){+.+.+.}, at: [<ffffffffa009246a>] cfg80211_get_dev_from_ifindex+0x1a/0x90 [cfg80211] but task is already holding lock: (rtnl_mutex){+.+.+.}, at: [<ffffffff81400ff2>] rtnl_lock+0x12/0x20 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (rtnl_mutex){+.+.+.}: [<ffffffff810857b6>] __lock_acquire+0xd76/0x12b0 [<ffffffff81085dd3>] lock_acquire+0xe3/0x120 [<ffffffff814ee7a4>] mutex_lock_nested+0x44/0x350 [<ffffffff81400ff2>] rtnl_lock+0x12/0x20 [<ffffffffa009f6a5>] nl80211_send_reg_change_event+0x1f5/0x2a0 [cfg80211] [<ffffffffa009529e>] set_regdom+0x28e/0x4c0 [cfg80211] -> #0 (cfg80211_mutex){+.+.+.}: [<ffffffff8108587b>] __lock_acquire+0xe3b/0x12b0 [<ffffffff81085dd3>] lock_acquire+0xe3/0x120 [<ffffffff814ee7a4>] mutex_lock_nested+0x44/0x350 [<ffffffffa009246a>] cfg80211_get_dev_from_ifindex+0x1a/0x90 [cfg80211] [<ffffffffa009813f>] get_rdev_dev_by_info_ifindex+0x6f/0xa0 [cfg80211] [<ffffffffa009b12b>] nl80211_set_interface+0x3b/0x260 [cfg80211] When looking at the correct code, the problem is quite obvious. I'm not entirely sure which code paths lead here, so until I can analyse it better let's just use RCU to avoid the problem. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-25 08:54:13 +00:00
rcu_read_unlock();
cfg80211: add regulatory netlink multicast group This allows us to send to userspace "regulatory" events. For now we just send an event when we change regulatory domains. We also notify userspace when devices are using their own custom world roaming regulatory domains. Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-10 02:07:42 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
static void nl80211_send_mlme_event(struct cfg80211_registered_device *rdev,
struct net_device *netdev,
const u8 *buf, size_t len,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
enum nl80211_commands cmd, gfp_t gfp,
int uapsd_queues)
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + len, gfp);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put(msg, NL80211_ATTR_FRAME, len, buf))
goto nla_put_failure;
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
if (uapsd_queues >= 0) {
struct nlattr *nla_wmm =
nla_nest_start(msg, NL80211_ATTR_STA_WME);
if (!nla_wmm)
goto nla_put_failure;
if (nla_put_u8(msg, NL80211_STA_WME_UAPSD_QUEUES,
uapsd_queues))
goto nla_put_failure;
nla_nest_end(msg, nla_wmm);
}
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
void nl80211_send_rx_auth(struct cfg80211_registered_device *rdev,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
struct net_device *netdev, const u8 *buf,
size_t len, gfp_t gfp)
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
{
nl80211_send_mlme_event(rdev, netdev, buf, len,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
NL80211_CMD_AUTHENTICATE, gfp, -1);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
}
void nl80211_send_rx_assoc(struct cfg80211_registered_device *rdev,
struct net_device *netdev, const u8 *buf,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
size_t len, gfp_t gfp, int uapsd_queues)
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
{
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
nl80211_send_mlme_event(rdev, netdev, buf, len,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
NL80211_CMD_ASSOCIATE, gfp, uapsd_queues);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
}
nl80211: Generate deauth/disassoc event for locally generated frames Previously, nl80211 mlme events were generated only for received deauthentication and disassociation frames. We need to do the same for locally generated ones in order to let applications know that we disconnected (e.g., when AP does not reply to a probe). Rename the nl80211 and cfg80211 functions (s/rx_//) to make it clearer that they are used for both received and locally generated frames. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 18:53:56 +00:00
void nl80211_send_deauth(struct cfg80211_registered_device *rdev,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
struct net_device *netdev, const u8 *buf,
size_t len, gfp_t gfp)
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
{
nl80211_send_mlme_event(rdev, netdev, buf, len,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
NL80211_CMD_DEAUTHENTICATE, gfp, -1);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
}
nl80211: Generate deauth/disassoc event for locally generated frames Previously, nl80211 mlme events were generated only for received deauthentication and disassociation frames. We need to do the same for locally generated ones in order to let applications know that we disconnected (e.g., when AP does not reply to a probe). Rename the nl80211 and cfg80211 functions (s/rx_//) to make it clearer that they are used for both received and locally generated frames. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 18:53:56 +00:00
void nl80211_send_disassoc(struct cfg80211_registered_device *rdev,
struct net_device *netdev, const u8 *buf,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
size_t len, gfp_t gfp)
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
{
nl80211_send_mlme_event(rdev, netdev, buf, len,
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
NL80211_CMD_DISASSOCIATE, gfp, -1);
nl80211: Event notifications for MLME events Add new nl80211 event notifications (and a new multicast group, "mlme") for informing user space about received and processed Authentication, (Re)Association Response, Deauthentication, and Disassociation frames in station and IBSS modes (i.e., MLME SAP interface primitives MLME-AUTHENTICATE.confirm, MLME-ASSOCIATE.confirm, MLME-REASSOCIATE.confirm, MLME-DEAUTHENTICATE.indicate, and MLME-DISASSOCIATE.indication). The event data is encapsulated as the 802.11 management frame since we already have the frame in that format and it includes all the needed information. This is the initial step in providing MLME SAP interface for authentication and association with nl80211. In other words, kernel code will act as the MLME and a user space application can control it as the SME. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-19 11:39:21 +00:00
}
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
void cfg80211_rx_unprot_mlme_mgmt(struct net_device *dev, const u8 *buf,
size_t len)
nl80211: Add notification for dropped Deauth/Disassoc Add a new notification to indicate that a received, unprotected Deauthentication or Disassociation frame was dropped due to management frame protection being in use. This notification is needed to allow user space (e.g., wpa_supplicant) to implement SA Query procedure to recover from association state mismatch between an AP and STA. This is needed to avoid getting stuck in non-working state when MFP (IEEE 802.11w) is used and a protected Deauthentication or Disassociation frame is dropped for any reason. After that, the station would silently discard any unprotected Deauthentication or Disassociation frame that could be indicating that the AP does not have association for the STA (when the Reason Code would be 6 or 7). IEEE Std 802.11w-2009, 11.13 describes this recovery mechanism. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-15 22:52:40 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
const struct ieee80211_mgmt *mgmt = (void *)buf;
u32 cmd;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
if (WARN_ON(len < 2))
return;
nl80211: Add notification for dropped Deauth/Disassoc Add a new notification to indicate that a received, unprotected Deauthentication or Disassociation frame was dropped due to management frame protection being in use. This notification is needed to allow user space (e.g., wpa_supplicant) to implement SA Query procedure to recover from association state mismatch between an AP and STA. This is needed to avoid getting stuck in non-working state when MFP (IEEE 802.11w) is used and a protected Deauthentication or Disassociation frame is dropped for any reason. After that, the station would silently discard any unprotected Deauthentication or Disassociation frame that could be indicating that the AP does not have association for the STA (when the Reason Code would be 6 or 7). IEEE Std 802.11w-2009, 11.13 describes this recovery mechanism. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-15 22:52:40 +00:00
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
if (ieee80211_is_deauth(mgmt->frame_control))
cmd = NL80211_CMD_UNPROT_DEAUTHENTICATE;
else
cmd = NL80211_CMD_UNPROT_DISASSOCIATE;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
trace_cfg80211_rx_unprot_mlme_mgmt(dev, buf, len);
cfg80211/mac80211: add wmm info to assoc event Userspace might need to know what queues are configured for uapsd (e.g. for setting proper default values in tspecs). Add this bitmap to the association event (inside wmm nested attribute) Add additional parameter to cfg80211_rx_assoc_resp, and update its callers. Signed-off-by: Eliad Peller <eliadx.peller@intel.com> Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-09 14:09:45 +00:00
nl80211_send_mlme_event(rdev, dev, buf, len, cmd, GFP_ATOMIC, -1);
nl80211: Add notification for dropped Deauth/Disassoc Add a new notification to indicate that a received, unprotected Deauthentication or Disassociation frame was dropped due to management frame protection being in use. This notification is needed to allow user space (e.g., wpa_supplicant) to implement SA Query procedure to recover from association state mismatch between an AP and STA. This is needed to avoid getting stuck in non-working state when MFP (IEEE 802.11w) is used and a protected Deauthentication or Disassociation frame is dropped for any reason. After that, the station would silently discard any unprotected Deauthentication or Disassociation frame that could be indicating that the AP does not have association for the STA (when the Reason Code would be 6 or 7). IEEE Std 802.11w-2009, 11.13 describes this recovery mechanism. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-15 22:52:40 +00:00
}
cfg80211/mac80211: clean up cfg80211 SME APIs Do some cleanups in the cfg80211 SME APIs, which are only used by mac80211. Most of these functions get a frame passed, and there isn't really any reason to export multiple functions as cfg80211 can check the frame type instead, do that. Additionally, the API functions have confusing names like cfg80211_send_...() which was meant to indicate that it sends an event to userspace, but gets a bit confusing when there's both TX and RX and they're not all clearly labeled. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-05-15 22:55:00 +00:00
EXPORT_SYMBOL(cfg80211_rx_unprot_mlme_mgmt);
nl80211: Add notification for dropped Deauth/Disassoc Add a new notification to indicate that a received, unprotected Deauthentication or Disassociation frame was dropped due to management frame protection being in use. This notification is needed to allow user space (e.g., wpa_supplicant) to implement SA Query procedure to recover from association state mismatch between an AP and STA. This is needed to avoid getting stuck in non-working state when MFP (IEEE 802.11w) is used and a protected Deauthentication or Disassociation frame is dropped for any reason. After that, the station would silently discard any unprotected Deauthentication or Disassociation frame that could be indicating that the AP does not have association for the STA (when the Reason Code would be 6 or 7). IEEE Std 802.11w-2009, 11.13 describes this recovery mechanism. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-12-15 22:52:40 +00:00
cfg80211: make nl80211_send_mlme_timeout() static Fixes sparse complaint: CHECK net/wireless/nl80211.c net/wireless/nl80211.c:3694:6: warning: symbol 'nl80211_send_mlme_timeout' was not declared. Should it be static? Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-05-02 04:34:48 +00:00
static void nl80211_send_mlme_timeout(struct cfg80211_registered_device *rdev,
struct net_device *netdev, int cmd,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
const u8 *addr, gfp_t gfp)
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put_flag(msg, NL80211_ATTR_TIMED_OUT) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr))
goto nla_put_failure;
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
void nl80211_send_auth_timeout(struct cfg80211_registered_device *rdev,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
struct net_device *netdev, const u8 *addr,
gfp_t gfp)
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
{
nl80211_send_mlme_timeout(rdev, netdev, NL80211_CMD_AUTHENTICATE,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
addr, gfp);
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
}
void nl80211_send_assoc_timeout(struct cfg80211_registered_device *rdev,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
struct net_device *netdev, const u8 *addr,
gfp_t gfp)
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
{
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
nl80211_send_mlme_timeout(rdev, netdev, NL80211_CMD_ASSOCIATE,
addr, gfp);
nl80211: Add event for authentication/association timeout SME needs to be notified when the authentication or association attempt times out and MLME has stopped processing in order to allow the SME to decide what to do next. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-22 18:38:25 +00:00
}
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
void nl80211_send_connect_result(struct cfg80211_registered_device *rdev,
cfg80211: Use a structure to pass connect response params Currently the connect event from driver takes all the connection response parameters as arguments. With support for new features these response parameters can grow. Use a structure to pass these parameters rather than passing them as function arguments. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [add to documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:33 +00:00
struct net_device *netdev,
struct cfg80211_connect_resp_params *cr,
cfg80211: Specify the reason for connect timeout This enhances the connect timeout API to also carry the reason for the timeout. These reason codes for the connect time out are represented by enum nl80211_timeout_reason and are passed to user space through a new attribute NL80211_ATTR_TIMEOUT_REASON (u32). Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [keep gfp_t argument last] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:21 +00:00
gfp_t gfp)
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
msg = nlmsg_new(100 + cr->req_ie_len + cr->resp_ie_len +
cr->fils_kek_len + cr->pmk_len +
(cr->pmkid ? WLAN_PMKID_LEN : 0), gfp);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_CONNECT);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
cfg80211: Use a structure to pass connect response params Currently the connect event from driver takes all the connection response parameters as arguments. With support for new features these response parameters can grow. Use a structure to pass these parameters rather than passing them as function arguments. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [add to documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:33 +00:00
(cr->bssid &&
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, cr->bssid)) ||
cfg80211: Allow cfg80211_connect_result() errors to be distinguished Previously, the status parameter to cfg80211_connect_result() was documented as using WLAN_STATUS_UNSPECIFIED_FAILURE (1) when the real status code for the failure is not known. This value can be used by an AP (and often is) and as such, user space cannot distinguish between explicitly rejected authentication/association and not being able to even try to associate or not receiving a response from the AP. Add a new inline function, cfg80211_connect_timeout(), to be used when the driver knows that the connection attempt failed due to a reason where connection could not be attempt or no response was received from the AP. The internal functions now allow a negative status value (-1) to be used as an indication of this special case. This results in the NL80211_ATTR_TIMED_OUT to be added to the NL80211_CMD_CONNECT event to allow user space to determine this case was hit. For backwards compatibility, NL80211_STATUS_CODE with the value WLAN_STATUS_UNSPECIFIED_FAILURE is still indicated in the event in such a case. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [johannes: fix cfg80211_connect_bss() prototype to use int for status, add cfg80211_connect_timeout() to docbook, fix docbook] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-30 21:16:50 +00:00
nla_put_u16(msg, NL80211_ATTR_STATUS_CODE,
cfg80211: Use a structure to pass connect response params Currently the connect event from driver takes all the connection response parameters as arguments. With support for new features these response parameters can grow. Use a structure to pass these parameters rather than passing them as function arguments. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [add to documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:33 +00:00
cr->status < 0 ? WLAN_STATUS_UNSPECIFIED_FAILURE :
cr->status) ||
(cr->status < 0 &&
cfg80211: Specify the reason for connect timeout This enhances the connect timeout API to also carry the reason for the timeout. These reason codes for the connect time out are represented by enum nl80211_timeout_reason and are passed to user space through a new attribute NL80211_ATTR_TIMEOUT_REASON (u32). Signed-off-by: Purushottam Kushwaha <pkushwah@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [keep gfp_t argument last] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-12 23:12:21 +00:00
(nla_put_flag(msg, NL80211_ATTR_TIMED_OUT) ||
cfg80211: Use a structure to pass connect response params Currently the connect event from driver takes all the connection response parameters as arguments. With support for new features these response parameters can grow. Use a structure to pass these parameters rather than passing them as function arguments. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> [add to documentation] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:33 +00:00
nla_put_u32(msg, NL80211_ATTR_TIMEOUT_REASON,
cr->timeout_reason))) ||
(cr->req_ie &&
nla_put(msg, NL80211_ATTR_REQ_IE, cr->req_ie_len, cr->req_ie)) ||
(cr->resp_ie &&
nla_put(msg, NL80211_ATTR_RESP_IE, cr->resp_ie_len,
cfg80211: Add support for FILS shared key authentication offload Enhance nl80211 and cfg80211 connect request and response APIs to support FILS shared key authentication offload. The new nl80211 attributes can be used to provide additional information to the driver to establish a FILS connection. Also enhance the set/del PMKSA to allow support for adding and deleting PMKSA based on FILS cache identifier. Add a new feature flag that drivers can use to advertize support for FILS shared key authentication and association in station mode when using their own SME. Signed-off-by: Vidyullatha Kanchanapally <vkanchan@qti.qualcomm.com> Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-03-30 21:22:34 +00:00
cr->resp_ie)) ||
(cr->update_erp_next_seq_num &&
nla_put_u16(msg, NL80211_ATTR_FILS_ERP_NEXT_SEQ_NUM,
cr->fils_erp_next_seq_num)) ||
(cr->status == WLAN_STATUS_SUCCESS &&
((cr->fils_kek &&
nla_put(msg, NL80211_ATTR_FILS_KEK, cr->fils_kek_len,
cr->fils_kek)) ||
(cr->pmk &&
nla_put(msg, NL80211_ATTR_PMK, cr->pmk_len, cr->pmk)) ||
(cr->pmkid &&
nla_put(msg, NL80211_ATTR_PMKID, WLAN_PMKID_LEN, cr->pmkid)))))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
void nl80211_send_roamed(struct cfg80211_registered_device *rdev,
cfg80211: unify cfg80211_roamed() and cfg80211_roamed_bss() cfg80211_roamed() and cfg80211_roamed_bss() take the same arguments except that cfg80211_roamed() requires the BSSID and cfg80211_roamed_bss() requires the bss entry. Unify the two functions by using a struct for driver initiated roaming information so that either the BSSID or the bss entry can be passed as an argument to the unified function. Signed-off-by: Avraham Stern <avraham.stern@intel.com> [modified the ath6k, brcm80211, rndis and wlan-ng drivers accordingly] Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [modify brcmfmac to remove the useless cast, spotted by Arend] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 07:58:49 +00:00
struct net_device *netdev,
struct cfg80211_roam_info *info, gfp_t gfp)
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: unify cfg80211_roamed() and cfg80211_roamed_bss() cfg80211_roamed() and cfg80211_roamed_bss() take the same arguments except that cfg80211_roamed() requires the BSSID and cfg80211_roamed_bss() requires the bss entry. Unify the two functions by using a struct for driver initiated roaming information so that either the BSSID or the bss entry can be passed as an argument to the unified function. Signed-off-by: Avraham Stern <avraham.stern@intel.com> [modified the ath6k, brcm80211, rndis and wlan-ng drivers accordingly] Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [modify brcmfmac to remove the useless cast, spotted by Arend] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 07:58:49 +00:00
const u8 *bssid = info->bss ? info->bss->bssid : info->bssid;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
cfg80211: unify cfg80211_roamed() and cfg80211_roamed_bss() cfg80211_roamed() and cfg80211_roamed_bss() take the same arguments except that cfg80211_roamed() requires the BSSID and cfg80211_roamed_bss() requires the bss entry. Unify the two functions by using a struct for driver initiated roaming information so that either the BSSID or the bss entry can be passed as an argument to the unified function. Signed-off-by: Avraham Stern <avraham.stern@intel.com> [modified the ath6k, brcm80211, rndis and wlan-ng drivers accordingly] Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [modify brcmfmac to remove the useless cast, spotted by Arend] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 07:58:49 +00:00
msg = nlmsg_new(100 + info->req_ie_len + info->resp_ie_len, gfp);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_ROAM);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid) ||
cfg80211: unify cfg80211_roamed() and cfg80211_roamed_bss() cfg80211_roamed() and cfg80211_roamed_bss() take the same arguments except that cfg80211_roamed() requires the BSSID and cfg80211_roamed_bss() requires the bss entry. Unify the two functions by using a struct for driver initiated roaming information so that either the BSSID or the bss entry can be passed as an argument to the unified function. Signed-off-by: Avraham Stern <avraham.stern@intel.com> [modified the ath6k, brcm80211, rndis and wlan-ng drivers accordingly] Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [modify brcmfmac to remove the useless cast, spotted by Arend] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 07:58:49 +00:00
(info->req_ie &&
nla_put(msg, NL80211_ATTR_REQ_IE, info->req_ie_len,
info->req_ie)) ||
(info->resp_ie &&
nla_put(msg, NL80211_ATTR_RESP_IE, info->resp_ie_len,
nl80211: add authorized flag to ROAM event Drivers that initiate roaming while being connected to a network that uses 802.1X authentication need to inform user space if 802.1X authentication is further required after roaming. For example, when using the Fast transition protocol, roaming within the mobility domain does not require new 802.1X authentication, but roaming to another mobility domain does. In addition, some drivers may not support 802.1X authentication (so it has to be done in user space), while other drivers do. Add a flag to the roaming notification to indicate if user space is required to do 802.1X authentication after the roaming or not. This flag will only be used for networks that use 802.1X authentication. For networks that do not use 802.1X authentication it is assumed that no further action is required from user space after the roaming notification. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Luca Coelho <luciano.coelho@intel.com> [arend.vanspriel@broadcom.com reuse NL80211_ATTR_PORT_AUTHORIZED] Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [rebase to apply w/o the flag in CONNECT] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-06-09 12:08:45 +00:00
info->resp_ie)) ||
(info->authorized &&
nla_put_flag(msg, NL80211_ATTR_PORT_AUTHORIZED)))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
void nl80211_send_disconnected(struct cfg80211_registered_device *rdev,
struct net_device *netdev, u16 reason,
cfg80211: fix locking Over time, a lot of locking issues have crept into the smarts of cfg80211, so e.g. scan completion can race against a new scan, IBSS join can race against leaving an IBSS, etc. Introduce a new per-interface lock that protects most of the per-interface data that we need to keep track of, and sprinkle assertions about that lock everywhere. Some things now need to be offloaded to work structs so that we don't require being able to sleep in functions the drivers call. The exception to that are the MLME callbacks (rx_auth etc.) that currently only mac80211 calls because it was easier to do that there instead of in cfg80211, and future drivers implementing those calls will, if they ever exist, probably need to use a similar scheme like mac80211 anyway... In order to be able to handle _deauth and _disassoc properly, introduce a cookie passed to it that will determine locking requirements. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-07 01:56:11 +00:00
const u8 *ie, size_t ie_len, bool from_ap)
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + ie_len, GFP_KERNEL);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_DISCONNECT);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
(from_ap && reason &&
nla_put_u16(msg, NL80211_ATTR_REASON_CODE, reason)) ||
(from_ap &&
nla_put_flag(msg, NL80211_ATTR_DISCONNECTED_BY_AP)) ||
(ie && nla_put(msg, NL80211_ATTR_IE, ie_len, ie)))
goto nla_put_failure;
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, GFP_KERNEL);
cfg80211: connect/disconnect API This patch introduces the cfg80211 connect/disconnect API. The goal here is to run the AUTH and ASSOC steps in one call. This is needed for some fullmac cards that run both steps directly from the target, after the host driver sends a connect command. Additionally, all the new crypto parameters for connect() are now also valid for associate() -- although associate requires the IEs to be used, the information can be useful for drivers and should be given. Signed-off-by: Samuel Ortiz <samuel.ortiz@intel.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:54 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
void nl80211_send_ibss_bssid(struct cfg80211_registered_device *rdev,
struct net_device *netdev, const u8 *bssid,
gfp_t gfp)
{
struct sk_buff *msg;
void *hdr;
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_JOIN_IBSS);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid))
goto nla_put_failure;
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/nl80211: add IBSS API This adds IBSS API along with (preliminary) wext handlers. The wext handlers can only do IBSS so you need to call them from your own wext handlers if the mode is IBSS. The nl80211 API requires * an SSID * a channel (frequency) for the case that a new IBSS has to be created It optionally supports * a flag to fix the channel * a fixed BSSID The cfg80211 code also takes care to leave the IBSS before the netdev is set down. If wireless extensions are used, it also caches values when the interface is down and instructs the driver to join when the interface is set up. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-19 19:24:32 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_notify_new_peer_candidate(struct net_device *dev, const u8 *addr,
const u8* ie, u8 ie_len, gfp_t gfp)
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
struct sk_buff *msg;
void *hdr;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
if (WARN_ON(wdev->iftype != NL80211_IFTYPE_MESH_POINT))
return;
trace_cfg80211_notify_new_peer_candidate(dev, addr);
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + ie_len, gfp);
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_NEW_PEER_CANDIDATE);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
(ie_len && ie &&
nla_put(msg, NL80211_ATTR_IE, ie_len , ie)))
goto nla_put_failure;
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_notify_new_peer_candidate);
nl80211: New notification to discover mesh peer candidates. Notify userspace when a beacon/presp is received from a suitable mesh peer candidate for whom no sta information exists. Userspace can then decide to create a sta info for the candidate. If userspace is not ready to authenticate the peer right away, it can create the sta info with the authenticated flag unset and set it later. Signed-off-by: Javier Cardona <javier@cozybit.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-04-07 22:08:34 +00:00
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
void nl80211_michael_mic_failure(struct cfg80211_registered_device *rdev,
struct net_device *netdev, const u8 *addr,
enum nl80211_key_type key_type, int key_id,
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
const u8 *tsc, gfp_t gfp)
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
{
struct sk_buff *msg;
void *hdr;
cfg80211: use proper allocation flags Instead of hardcoding GFP_ATOMIC everywhere, add a new function parameter that gets the flags from the caller. Obviously then I need to update all callers (all of them in mac80211), and it turns out that now it's ok to use GFP_KERNEL in almost all places. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-01 19:26:47 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_MICHAEL_MIC_FAILURE);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
(addr && nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr)) ||
nla_put_u32(msg, NL80211_ATTR_KEY_TYPE, key_type) ||
(key_id != -1 &&
nla_put_u8(msg, NL80211_ATTR_KEY_IDX, key_id)) ||
(tsc && nla_put(msg, NL80211_ATTR_KEY_SEQ, 6, tsc)))
goto nla_put_failure;
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211: Add Michael MIC failure event Define a new nl80211 event, NL80211_CMD_MICHAEL_MIC_FAILURE, to be used to notify user space about locally detected Michael MIC failures. This matches with the MLME-MICHAELMICFAILURE.indication() primitive. Since we do not actually have TSC in the skb anymore when mac80211_ev_michael_mic_failure() is called, that function is changed to take in the TSC as an optional parameter instead of as a requirement to include the TSC after the hdr field (which we did not really follow). For now, TSC is not included in the events from mac80211, but it could be added at some point. Signed-off-by: Jouni Malinen <j@w1.fi> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-03-27 19:59:49 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
void nl80211_send_beacon_hint_event(struct wiphy *wiphy,
struct ieee80211_channel *channel_before,
struct ieee80211_channel *channel_after)
{
struct sk_buff *msg;
void *hdr;
struct nlattr *nl_freq;
net: use NLMSG_DEFAULT_SIZE in nlmsg_new() allocations nlmsg_new() adds the size of the netlink header to the value that has been passed as parameter. If NLMSG_GOODSIZE is selected, we request an allocation of one memory page plus the size of the header. Instead, NLMSG_DEFAULT_SIZE should be used since it already substracts the size of the Netlink header. I have the impression that the similar naming in both constant is error prone when using it with nlmsg_new(). This is already documented in include/net/netlink.h Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-19 22:27:55 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_ATOMIC);
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_REG_BEACON_HINT);
if (!hdr) {
nlmsg_free(msg);
return;
}
/*
* Since we are applying the beacon hint to a wiphy we know its
* wiphy_idx is valid
*/
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, get_wiphy_idx(wiphy)))
goto nla_put_failure;
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
/* Before */
nl_freq = nla_nest_start(msg, NL80211_ATTR_FREQ_BEFORE);
if (!nl_freq)
goto nla_put_failure;
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
if (nl80211_msg_put_channel(msg, channel_before, false))
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
goto nla_put_failure;
nla_nest_end(msg, nl_freq);
/* After */
nl_freq = nla_nest_start(msg, NL80211_ATTR_FREQ_AFTER);
if (!nl_freq)
goto nla_put_failure;
nl80211: conditionally add back radar information If userspace is updated to deal with large split wiphy information dumps, add back the radar information that could otherwise push the data over the limit of the netlink dump messages. Cc: Simon Wunderlich <simon.wunderlich@s2003.tu-chemnitz.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-18 22:54:36 +00:00
if (nl80211_msg_put_channel(msg, channel_after, false))
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
goto nla_put_failure;
nla_nest_end(msg, nl_freq);
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
rcu_read_lock();
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_allns(&nl80211_fam, msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_REGULATORY, GFP_ATOMIC);
cfg80211: make aware of net namespaces In order to make cfg80211/nl80211 aware of network namespaces, we have to do the following things: * del_virtual_intf method takes an interface index rather than a netdev pointer - simply change this * nl80211 uses init_net a lot, it changes to use the sender's network namespace * scan requests use the interface index, hold a netdev pointer and reference instead * we want a wiphy and its associated virtual interfaces to be in one netns together, so - we need to be able to change ns for a given interface, so export dev_change_net_namespace() - for each virtual interface set the NETIF_F_NETNS_LOCAL flag, and clear that flag only when the wiphy changes ns, to disallow breaking this invariant * when a network namespace goes away, we need to reparent the wiphy to init_net * cfg80211 users that support creating virtual interfaces must create them in the wiphy's namespace, currently this affects only mac80211 The end result is that you can now switch an entire wiphy into a different network namespace with the new command iw phy#<idx> set netns <pid> and all virtual interfaces will follow (or the operation fails). Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-07-13 22:33:35 +00:00
rcu_read_unlock();
cfg80211: send regulatory beacon hint events to userspace This informs userspace when a change has occured on a world roaming wiphy's channel which has lifted some restrictions due to a regulatory beacon hint. Because this is now sent to userspace through the regulatory multicast group we remove the debug prints we used to use as they are no longer necessary. Acked-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Luis R. Rodriguez <lrodriguez@atheros.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-04-02 18:08:09 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
static void nl80211_send_remain_on_chan_event(
int cmd, struct cfg80211_registered_device *rdev,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct wireless_dev *wdev, u64 cookie,
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
struct ieee80211_channel *chan,
unsigned int duration, gfp_t gfp)
{
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
(wdev->netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex)) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_FREQ, chan->center_freq) ||
cfg80211: remove remain-on-channel channel type As mwifiex (and mac80211 in the software case) are the only drivers actually implementing remain-on-channel with channel type, userspace can't be relying on it. This is the case, as it's used only for P2P operations right now. Rather than adding a flag to tell userspace whether or not it can actually rely on it, simplify all the code by removing the ability to use different channel types. Leave only the validation of the attribute, so that if we extend it again later (with the needed capability flag), it can't break userspace sending invalid data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 17:31:02 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_CHANNEL_TYPE,
NL80211_CHAN_NO_HT) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (cmd == NL80211_CMD_REMAIN_ON_CHANNEL &&
nla_put_u32(msg, NL80211_ATTR_DURATION, duration))
goto nla_put_failure;
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_ready_on_channel(struct wireless_dev *wdev, u64 cookie,
struct ieee80211_channel *chan,
unsigned int duration, gfp_t gfp)
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_ready_on_channel(wdev, cookie, chan, duration);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
nl80211_send_remain_on_chan_event(NL80211_CMD_REMAIN_ON_CHANNEL,
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
rdev, wdev, cookie, chan,
cfg80211: remove remain-on-channel channel type As mwifiex (and mac80211 in the software case) are the only drivers actually implementing remain-on-channel with channel type, userspace can't be relying on it. This is the case, as it's used only for P2P operations right now. Rather than adding a flag to tell userspace whether or not it can actually rely on it, simplify all the code by removing the ability to use different channel types. Leave only the validation of the attribute, so that if we extend it again later (with the needed capability flag), it can't break userspace sending invalid data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 17:31:02 +00:00
duration, gfp);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_ready_on_channel);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_remain_on_channel_expired(struct wireless_dev *wdev, u64 cookie,
struct ieee80211_channel *chan,
gfp_t gfp)
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_ready_on_channel_expired(wdev, cookie, chan);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
nl80211_send_remain_on_chan_event(NL80211_CMD_CANCEL_REMAIN_ON_CHANNEL,
cfg80211: remove remain-on-channel channel type As mwifiex (and mac80211 in the software case) are the only drivers actually implementing remain-on-channel with channel type, userspace can't be relying on it. This is the case, as it's used only for P2P operations right now. Rather than adding a flag to tell userspace whether or not it can actually rely on it, simplify all the code by removing the ability to use different channel types. Leave only the validation of the attribute, so that if we extend it again later (with the needed capability flag), it can't break userspace sending invalid data. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 17:31:02 +00:00
rdev, wdev, cookie, chan, 0, gfp);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_remain_on_channel_expired);
cfg80211: add remain-on-channel command Add new commands for requesting the driver to remain awake on a specified channel for the specified amount of time (and another command to cancel such an operation). This can be used to implement userspace-controlled off-channel operations, like Public Action frame exchange on another channel than the operation channel. The off-channel operation should behave similarly to scan, i.e. the local station (if associated) moves into power save mode to request the AP to buffer frames for it and then moves to the other channel to allow the off-channel operation to be completed. The duration parameter can be used to request enough time to receive a response from the target station. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:41 +00:00
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_new_sta(struct net_device *dev, const u8 *mac_addr,
struct station_info *sinfo, gfp_t gfp)
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = dev->ieee80211_ptr->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
struct sk_buff *msg;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_new_sta(dev, mac_addr, sinfo);
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
if (!msg)
return;
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
if (nl80211_send_station(msg, NL80211_CMD_NEW_STATION, 0, 0, 0,
cfg80211: allow CFG80211_SIGNAL_TYPE_UNSPEC in station_info The station_info struct had demanded dBm signal values, but the cfg80211 wireless extensions implementation was also accepting "unspecified" (i.e. RSSI) unit values while the nl80211 code was completely unaware of them. Resolve this by formally allowing the "unspecified" units while making nl80211 ignore them. Signed-off-by: John W. Linville <linville@tuxdriver.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net>
2012-03-15 17:25:41 +00:00
rdev, dev, mac_addr, sinfo) < 0) {
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
nlmsg_free(msg);
return;
}
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_new_sta);
mac80211/cfg80211: add station events When, for instance, a new IBSS peer is found, userspace wants to be notified. Add events for all new stations that mac80211 learns about. Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2009-12-23 12:15:44 +00:00
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
void cfg80211_del_sta_sinfo(struct net_device *dev, const u8 *mac_addr,
struct station_info *sinfo, gfp_t gfp)
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = dev->ieee80211_ptr->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
struct sk_buff *msg;
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
struct station_info empty_sinfo = {};
if (!sinfo)
sinfo = &empty_sinfo;
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_del_sta(dev, mac_addr);
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
if (!msg)
return;
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
if (nl80211_send_station(msg, NL80211_CMD_DEL_STATION, 0, 0, 0,
cfg80211: fix checking nl80211_send_station() return value The return value from nl80211_send_station() is the length of the skb, or a negative error, so abort sending the message only when the return value was negative. This fixes the ibss_rsn wpa_supplicant test case. Reported-by: Jouni Malinen <j@w1.fi> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-01-16 20:05:02 +00:00
rdev, dev, mac_addr, sinfo) < 0) {
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
nlmsg_free(msg);
return;
}
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
}
cfg80211: allow including station info in delete event When a station is removed, its statistics may be interesting to userspace, for example for further aggregation of statistics of all stations that ever connected to an AP. Introduce a new cfg80211_del_sta_sinfo() function (and make the cfg80211_del_sta() a static inline calling it) to allow passing a struct station_info along with this, and send the data in the nl80211 event message. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-14 16:14:00 +00:00
EXPORT_SYMBOL(cfg80211_del_sta_sinfo);
cfg80211: Add nl80211 event for deletion of a station entry Indicate an NL80211_CMD_DEL_STATION event when a station entry in mac80211 is deleted to match with the NL80211_CMD_NEW_STATION event that is used when the entry was added. This is needed, e.g., to allow user space to remove a peer from RSN IBSS Authenticator state machine to avoid re-authentication and re-keying delays when the peer is not reachable anymore. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-03-23 13:29:52 +00:00
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_conn_failed(struct net_device *dev, const u8 *mac_addr,
enum nl80211_connect_failed_reason reason,
gfp_t gfp)
cfg80211/nl80211: Notify connection request failure in AP mode In AP mode, when a station requests connection to an AP and if the request is failed for particular reason, userspace is notified about the failure through NL80211_CMD_CONN_FAILED command. Reason for the failure is sent through the attribute NL80211_ATTR_CONN_FAILED_REASON. Signed-off-by: Pandiyarajan Pitchaimuthu <c_ppitch@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-18 11:20:49 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = dev->ieee80211_ptr->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211/nl80211: Notify connection request failure in AP mode In AP mode, when a station requests connection to an AP and if the request is failed for particular reason, userspace is notified about the failure through NL80211_CMD_CONN_FAILED command. Reason for the failure is sent through the attribute NL80211_ATTR_CONN_FAILED_REASON. Signed-off-by: Pandiyarajan Pitchaimuthu <c_ppitch@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-18 11:20:49 +00:00
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_GOODSIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_CONN_FAILED);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, mac_addr) ||
nla_put_u32(msg, NL80211_ATTR_CONN_FAILED_REASON, reason))
goto nla_put_failure;
genlmsg_end(msg, hdr);
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/nl80211: Notify connection request failure in AP mode In AP mode, when a station requests connection to an AP and if the request is failed for particular reason, userspace is notified about the failure through NL80211_CMD_CONN_FAILED command. Reason for the failure is sent through the attribute NL80211_ATTR_CONN_FAILED_REASON. Signed-off-by: Pandiyarajan Pitchaimuthu <c_ppitch@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-18 11:20:49 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_conn_failed);
cfg80211/nl80211: Notify connection request failure in AP mode In AP mode, when a station requests connection to an AP and if the request is failed for particular reason, userspace is notified about the failure through NL80211_CMD_CONN_FAILED command. Reason for the failure is sent through the attribute NL80211_ATTR_CONN_FAILED_REASON. Signed-off-by: Pandiyarajan Pitchaimuthu <c_ppitch@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-09-18 11:20:49 +00:00
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
static bool __nl80211_unexpected_frame(struct net_device *dev, u8 cmd,
const u8 *addr, gfp_t gfp)
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
struct sk_buff *msg;
void *hdr;
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
u32 nlportid = ACCESS_ONCE(wdev->ap_unexpected_nlportid);
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
if (!nlportid)
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
return false;
msg = nlmsg_new(100, gfp);
if (!msg)
return true;
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, cmd);
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
if (!hdr) {
nlmsg_free(msg);
return true;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr))
goto nla_put_failure;
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
genlmsg_unicast(wiphy_net(&rdev->wiphy), msg, nlportid);
nl80211: allow subscribing to unexpected class3 frames To implement AP mode without monitor interfaces we need to be able to send a deauth to stations that send frames without being associated. Enable this by adding a new nl80211 event for such frames that an application can subscribe to. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:12 +00:00
return true;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
return true;
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
bool cfg80211_rx_spurious_frame(struct net_device *dev,
const u8 *addr, gfp_t gfp)
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
bool ret;
trace_cfg80211_rx_spurious_frame(dev, addr);
if (WARN_ON(wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO)) {
trace_cfg80211_return_bool(false);
return false;
}
ret = __nl80211_unexpected_frame(dev, NL80211_CMD_UNEXPECTED_FRAME,
addr, gfp);
trace_cfg80211_return_bool(ret);
return ret;
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_rx_spurious_frame);
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
bool cfg80211_rx_unexpected_4addr_frame(struct net_device *dev,
const u8 *addr, gfp_t gfp)
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
bool ret;
trace_cfg80211_rx_unexpected_4addr_frame(dev, addr);
if (WARN_ON(wdev->iftype != NL80211_IFTYPE_AP &&
wdev->iftype != NL80211_IFTYPE_P2P_GO &&
wdev->iftype != NL80211_IFTYPE_AP_VLAN)) {
trace_cfg80211_return_bool(false);
return false;
}
ret = __nl80211_unexpected_frame(dev,
NL80211_CMD_UNEXPECTED_4ADDR_FRAME,
addr, gfp);
trace_cfg80211_return_bool(ret);
return ret;
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_rx_unexpected_4addr_frame);
cfg80211: add event for unexpected 4addr frames The frames are used by AP/STA WDS mode, and hostapd needs to know when such a frame was received to set up the VLAN appropriately to allow using it. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:19 +00:00
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
int nl80211_send_mgmt(struct cfg80211_registered_device *rdev,
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
struct wireless_dev *wdev, u32 nlportid,
cfg80211/mac80211: report signal strength for mgmt frames Add the signal strength (in dBm only for now) to frames that are received via nl80211's various frame APIs. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Acked-by: Kalle Valo <kvalo@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-03-05 21:18:41 +00:00
int freq, int sig_dbm,
cfg80211: add flags to cfg80211_rx_mgmt() Add flags intended to report various auxiliary information and introduce the NL80211_RXMGMT_FLAG_ANSWERED flag to report that the frame was already answered by the device. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [REPLIED->ANSWERED, reword commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-15 11:51:28 +00:00
const u8 *buf, size_t len, u32 flags, gfp_t gfp)
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
{
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct net_device *netdev = wdev->netdev;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
struct sk_buff *msg;
void *hdr;
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + len, gfp);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (!msg)
return -ENOMEM;
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (!hdr) {
nlmsg_free(msg);
return -ENOMEM;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
(netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
netdev->ifindex)) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put_u32(msg, NL80211_ATTR_WIPHY_FREQ, freq) ||
(sig_dbm &&
nla_put_u32(msg, NL80211_ATTR_RX_SIGNAL_DBM, sig_dbm)) ||
cfg80211: add flags to cfg80211_rx_mgmt() Add flags intended to report various auxiliary information and introduce the NL80211_RXMGMT_FLAG_ANSWERED flag to report that the frame was already answered by the device. Signed-off-by: Vladimir Kondratiev <qca_vkondrat@qca.qualcomm.com> [REPLIED->ANSWERED, reword commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-08-15 11:51:28 +00:00
nla_put(msg, NL80211_ATTR_FRAME, len, buf) ||
(flags &&
nla_put_u32(msg, NL80211_ATTR_RXMGMT_FLAGS, flags)))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
return genlmsg_unicast(wiphy_net(&rdev->wiphy), msg, nlportid);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
return -ENOBUFS;
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_mgmt_tx_status(struct wireless_dev *wdev, u64 cookie,
const u8 *buf, size_t len, bool ack, gfp_t gfp)
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
struct net_device *netdev = wdev->netdev;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
struct sk_buff *msg;
void *hdr;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_mgmt_tx_status(wdev, cookie, ack);
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + len, gfp);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (!msg)
return;
cfg80211/mac80211: extensible frame processing Allow userspace to register for more than just action frames by giving the frame subtype, and make it possible to use this in various modes as well. With some tweaks and some added functionality this will, in the future, also be usable in AP mode and be able to replace the cooked monitor interface currently used in that case. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-08-12 13:38:38 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME_TX_STATUS);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
cfg80211: use wdev in mgmt-tx/ROC APIs The management frame and remain-on-channel APIs will be needed in the P2P device abstraction, so move them over to the new wdev-based APIs. Userspace can still use both the interface index and wdev identifier for them so it's backward compatible, but for the P2P Device wdev it will be able to use the wdev identifier only. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-06-15 13:30:18 +00:00
(netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
netdev->ifindex)) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
nla_put(msg, NL80211_ATTR_FRAME, len, buf) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
(ack && nla_put_flag(msg, NL80211_ATTR_ACK)))
goto nla_put_failure;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_mgmt_tx_status);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
static struct sk_buff *cfg80211_prepare_cqm(struct net_device *dev,
const char *mac, gfp_t gfp)
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
struct sk_buff *msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
void **cb;
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
if (!msg)
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
return NULL;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cb = (void **)msg->cb;
cb[0] = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_NOTIFY_CQM);
if (!cb[0]) {
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
nlmsg_free(msg);
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
return NULL;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex))
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
if (mac && nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, mac))
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
goto nla_put_failure;
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cb[1] = nla_nest_start(msg, NL80211_ATTR_CQM);
if (!cb[1])
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
goto nla_put_failure;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cb[2] = rdev;
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
return msg;
nla_put_failure:
nlmsg_free(msg);
return NULL;
}
static void cfg80211_send_cqm(struct sk_buff *msg, gfp_t gfp)
{
void **cb = (void **)msg->cb;
struct cfg80211_registered_device *rdev = cb[2];
nla_nest_end(msg, cb[1]);
genlmsg_end(msg, cb[0]);
memset(msg->cb, 0, sizeof(msg->cb));
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
}
void cfg80211_cqm_rssi_notify(struct net_device *dev,
enum nl80211_cqm_rssi_threshold_event rssi_event,
cfg80211: Pass new RSSI level in CQM RSSI notification Update the drivers to pass the RSSI level as a cfg80211_cqm_rssi_notify parameter and pass this value to userspace in a new nl80211 attribute. This helps both userspace and also helps in the implementation of the multiple RSSI thresholds CQM mechanism. Note for marvell/mwifiex I pass 0 for the RSSI value because the new RSSI value is not available to the driver at the time of the cfg80211_cqm_rssi_notify call, but the driver queries the new value immediately after that, so it is actually available just a moment later if we wanted to defer caling cfg80211_cqm_rssi_notify until that moment. Without this, the new cfg80211 code (patch 3) will call .get_station which will send a duplicate HostCmd_CMD_RSSI_INFO command to the hardware. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-25 11:43:41 +00:00
s32 rssi_level, gfp_t gfp)
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
{
struct sk_buff *msg;
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cfg80211: Pass new RSSI level in CQM RSSI notification Update the drivers to pass the RSSI level as a cfg80211_cqm_rssi_notify parameter and pass this value to userspace in a new nl80211 attribute. This helps both userspace and also helps in the implementation of the multiple RSSI thresholds CQM mechanism. Note for marvell/mwifiex I pass 0 for the RSSI value because the new RSSI value is not available to the driver at the time of the cfg80211_cqm_rssi_notify call, but the driver queries the new value immediately after that, so it is actually available just a moment later if we wanted to defer caling cfg80211_cqm_rssi_notify until that moment. Without this, the new cfg80211 code (patch 3) will call .get_station which will send a duplicate HostCmd_CMD_RSSI_INFO command to the hardware. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-25 11:43:41 +00:00
trace_cfg80211_cqm_rssi_notify(dev, rssi_event, rssi_level);
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cfg80211: clean up beacon loss CQM event Having it as a sub-event for RSSI thresholds is very ugly, but luckily no userspace actually uses the events yet. Move the event to its own function call internally and to its own event attribute in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:42:02 +00:00
if (WARN_ON(rssi_event != NL80211_CQM_RSSI_THRESHOLD_EVENT_LOW &&
rssi_event != NL80211_CQM_RSSI_THRESHOLD_EVENT_HIGH))
return;
cfg80211: Accept multiple RSSI thresholds for CQM Change the SET CQM command's RSSI threshold attribute to accept any number of thresholds as a sorted array. The API should be backwards compatible so that if one s32 threshold value is passed, the old mechanism is enabled. The netlink event generated is the same in both cases. cfg80211 handles an arbitrary number of RSSI thresholds but drivers have to provide a method (set_cqm_rssi_range_config) that configures a range set by a high and a low value. Drivers have to call back when the RSSI goes out of that range and there's no additional event for each time the range is reconfigured as there was with the current one-threshold API. This method doesn't have a hysteresis parameter because there's no benefit to the cfg80211 code from having the hysteresis be handled by hardware/driver in terms of the number of wakeups. At the same time it would likely be less consistent between drivers if offloaded or done in the drivers. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-02-10 09:02:31 +00:00
if (wdev->cqm_config) {
wdev->cqm_config->last_rssi_event_value = rssi_level;
cfg80211_cqm_rssi_update(rdev, dev);
if (rssi_level == 0)
rssi_level = wdev->cqm_config->last_rssi_event_value;
}
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
msg = cfg80211_prepare_cqm(dev, NULL, gfp);
if (!msg)
return;
if (nla_put_u32(msg, NL80211_ATTR_CQM_RSSI_THRESHOLD_EVENT,
rssi_event))
goto nla_put_failure;
cfg80211: Pass new RSSI level in CQM RSSI notification Update the drivers to pass the RSSI level as a cfg80211_cqm_rssi_notify parameter and pass this value to userspace in a new nl80211 attribute. This helps both userspace and also helps in the implementation of the multiple RSSI thresholds CQM mechanism. Note for marvell/mwifiex I pass 0 for the RSSI value because the new RSSI value is not available to the driver at the time of the cfg80211_cqm_rssi_notify call, but the driver queries the new value immediately after that, so it is actually available just a moment later if we wanted to defer caling cfg80211_cqm_rssi_notify until that moment. Without this, the new cfg80211 code (patch 3) will call .get_station which will send a duplicate HostCmd_CMD_RSSI_INFO command to the hardware. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-25 11:43:41 +00:00
if (rssi_level && nla_put_s32(msg, NL80211_ATTR_CQM_RSSI_LEVEL,
rssi_level))
goto nla_put_failure;
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
cfg80211_send_cqm(msg, gfp);
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
return;
nla_put_failure:
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
EXPORT_SYMBOL(cfg80211_cqm_rssi_notify);
cfg80211: Add connection quality monitoring support to nl80211 Add support for basic configuration of a connection quality monitoring to the nl80211 interface, and basic support for notifying about triggered monitoring events. Via this interface a user-space connection manager may configure and receive pre-warning events of deteriorating WLAN connection quality, and start preparing for roaming in advance, before the connection is already lost. An example usage of such a trigger is starting scanning for nearby AP's in an attempt to find one with better connection quality, and associate to it before the connection characteristics of the existing connection become too bad or the association is even lost, leading in a prolonged delay in connectivity. The interface currently supports only RSSI, but it could be later extended to include other parameters, such as signal-to-noise ratio, if need for that arises. Signed-off-by: Juuso Oikarinen <juuso.oikarinen@nokia.com> Reviewed-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-03-23 07:02:33 +00:00
cfg80211: refactor the various CQM event sending code Much of the code can be shared by moving it into helper functions for the CQM event sending. Also move the code closer together, even in the header file. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:37:43 +00:00
void cfg80211_cqm_txe_notify(struct net_device *dev,
const u8 *peer, u32 num_packets,
u32 rate, u32 intvl, gfp_t gfp)
{
struct sk_buff *msg;
msg = cfg80211_prepare_cqm(dev, peer, gfp);
if (!msg)
return;
if (nla_put_u32(msg, NL80211_ATTR_CQM_TXE_PKTS, num_packets))
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_ATTR_CQM_TXE_RATE, rate))
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_ATTR_CQM_TXE_INTVL, intvl))
goto nla_put_failure;
cfg80211_send_cqm(msg, gfp);
return;
nla_put_failure:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_cqm_txe_notify);
void cfg80211_cqm_pktloss_notify(struct net_device *dev,
const u8 *peer, u32 num_packets, gfp_t gfp)
{
struct sk_buff *msg;
trace_cfg80211_cqm_pktloss_notify(dev, peer, num_packets);
msg = cfg80211_prepare_cqm(dev, peer, gfp);
if (!msg)
return;
if (nla_put_u32(msg, NL80211_ATTR_CQM_PKT_LOSS_EVENT, num_packets))
goto nla_put_failure;
cfg80211_send_cqm(msg, gfp);
return;
nla_put_failure:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_cqm_pktloss_notify);
cfg80211: clean up beacon loss CQM event Having it as a sub-event for RSSI thresholds is very ugly, but luckily no userspace actually uses the events yet. Move the event to its own function call internally and to its own event attribute in nl80211. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-26 11:42:02 +00:00
void cfg80211_cqm_beacon_loss_notify(struct net_device *dev, gfp_t gfp)
{
struct sk_buff *msg;
msg = cfg80211_prepare_cqm(dev, NULL, gfp);
if (!msg)
return;
if (nla_put_flag(msg, NL80211_ATTR_CQM_BEACON_LOSS_EVENT))
goto nla_put_failure;
cfg80211_send_cqm(msg, gfp);
return;
nla_put_failure:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_cqm_beacon_loss_notify);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
static void nl80211_gtk_rekey_notify(struct cfg80211_registered_device *rdev,
struct net_device *netdev, const u8 *bssid,
const u8 *replay_ctr, gfp_t gfp)
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
{
struct sk_buff *msg;
struct nlattr *rekey_attr;
void *hdr;
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_SET_REKEY_OFFLOAD);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, bssid))
goto nla_put_failure;
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
rekey_attr = nla_nest_start(msg, NL80211_ATTR_REKEY_DATA);
if (!rekey_attr)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put(msg, NL80211_REKEY_DATA_REPLAY_CTR,
NL80211_REPLAY_CTR_LEN, replay_ctr))
goto nla_put_failure;
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
nla_nest_end(msg, rekey_attr);
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/nl80211: support GTK rekey offload In certain circumstances, like WoWLAN scenarios, devices may implement (partial) GTK rekeying on the device to avoid waking up the host for it. In order to successfully go through GTK rekeying, the KEK, KCK and the replay counter are required. Add API to let the supplicant hand the parameters to the driver which may store it for future GTK rekey operations. Note that, of course, if GTK rekeying is done by the device, the EAP frame must not be passed up to userspace, instead a rekey event needs to be sent to let userspace update its replay counter. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-07-05 14:35:40 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_gtk_rekey_notify(struct net_device *dev, const u8 *bssid,
const u8 *replay_ctr, gfp_t gfp)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_gtk_rekey_notify(dev, bssid);
nl80211_gtk_rekey_notify(rdev, dev, bssid, replay_ctr, gfp);
}
EXPORT_SYMBOL(cfg80211_gtk_rekey_notify);
static void
nl80211_pmksa_candidate_notify(struct cfg80211_registered_device *rdev,
struct net_device *netdev, int index,
const u8 *bssid, bool preauth, gfp_t gfp)
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
{
struct sk_buff *msg;
struct nlattr *attr;
void *hdr;
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_PMKSA_CANDIDATE);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex))
goto nla_put_failure;
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
attr = nla_nest_start(msg, NL80211_ATTR_PMKSA_CANDIDATE);
if (!attr)
goto nla_put_failure;
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_PMKSA_CANDIDATE_INDEX, index) ||
nla_put(msg, NL80211_PMKSA_CANDIDATE_BSSID, ETH_ALEN, bssid) ||
(preauth &&
nla_put_flag(msg, NL80211_PMKSA_CANDIDATE_PREAUTH)))
goto nla_put_failure;
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
nla_nest_end(msg, attr);
nl80211: clean up genlmsg_end uses genlmsg_end() cannot fail, it just returns the length of the message. Thus, error handling for it is useless. While removing it, I also noticed a useless variable and removed this it as well. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-10-22 17:05:51 +00:00
genlmsg_end(msg, hdr);
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/nl80211: Add PMKSA caching candidate event When the driver (or most likely firmware) decides which AP to use for roaming based on internal scan result processing, user space needs to be notified of PMKSA caching candidates to allow RSN pre-authentication to be used. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-09-16 15:56:23 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_pmksa_candidate_notify(struct net_device *dev, int index,
const u8 *bssid, bool preauth, gfp_t gfp)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
trace_cfg80211_pmksa_candidate_notify(dev, index, bssid, preauth);
nl80211_pmksa_candidate_notify(rdev, dev, index, bssid, preauth, gfp);
}
EXPORT_SYMBOL(cfg80211_pmksa_candidate_notify);
static void nl80211_ch_switch_notify(struct cfg80211_registered_device *rdev,
struct net_device *netdev,
struct cfg80211_chan_def *chandef,
cfg80211: add channel switch started notification Add a new NL80211_CH_SWITCH_STARTED_NOTIFY message that can be sent to the userspace when a channel switch process has started. This allows userspace to take action, for instance, by requesting other interfaces to switch channel as necessary. This patch introduces a function that allows the drivers to send this notification. It should be used when the driver starts processing a channel switch initiated by a remote device (eg. when a STA receives a CSA from the AP) and when it successfully starts a userspace-triggered channel switch (eg. when hostapd triggers a channel swith in the AP). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-07 12:31:35 +00:00
gfp_t gfp,
enum nl80211_commands notif,
u8 count)
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
{
struct sk_buff *msg;
void *hdr;
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
if (!msg)
return;
cfg80211: add channel switch started notification Add a new NL80211_CH_SWITCH_STARTED_NOTIFY message that can be sent to the userspace when a channel switch process has started. This allows userspace to take action, for instance, by requesting other interfaces to switch channel as necessary. This patch introduces a function that allows the drivers to send this notification. It should be used when the driver starts processing a channel switch initiated by a remote device (eg. when a STA receives a CSA from the AP) and when it successfully starts a userspace-triggered channel switch (eg. when hostapd triggers a channel swith in the AP). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-07 12:31:35 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, notif);
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
if (!hdr) {
nlmsg_free(msg);
return;
}
cfg80211: pass a channel definition struct Instead of passing a channel pointer and channel type to all functions and driver methods, pass a new channel definition struct. Right now, this struct contains just the control channel and channel type, but for VHT this will change. Also, add a small inline cfg80211_get_chandef_type() so that drivers don't need to use the _type field of the new structure all the time, which will change. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-08 20:25:48 +00:00
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex))
goto nla_put_failure;
if (nl80211_send_chandef(msg, chandef))
Merge branch 'master' into for-davem Conflicts: drivers/net/wireless/iwlwifi/iwl-testmode.c net/wireless/nl80211.c
2012-04-12 18:25:14 +00:00
goto nla_put_failure;
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
cfg80211: add channel switch started notification Add a new NL80211_CH_SWITCH_STARTED_NOTIFY message that can be sent to the userspace when a channel switch process has started. This allows userspace to take action, for instance, by requesting other interfaces to switch channel as necessary. This patch introduces a function that allows the drivers to send this notification. It should be used when the driver starts processing a channel switch initiated by a remote device (eg. when a STA receives a CSA from the AP) and when it successfully starts a userspace-triggered channel switch (eg. when hostapd triggers a channel swith in the AP). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-07 12:31:35 +00:00
if ((notif == NL80211_CMD_CH_SWITCH_STARTED_NOTIFY) &&
(nla_put_u32(msg, NL80211_ATTR_CH_SWITCH_COUNT, count)))
goto nla_put_failure;
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
genlmsg_end(msg, hdr);
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: add channel switch notify event The firmware may decide to switch channels while already beaconing, e.g. in response to a cfg80211 connect request on a different vif. Add this event to notify userspace when an AP or GO interface has successfully migrated to a new channel, so it can update its configuration accordingly. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2012-04-06 20:35:47 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
void cfg80211_ch_switch_notify(struct net_device *dev,
struct cfg80211_chan_def *chandef)
cfg80211: support TX error rate CQM Let the user configure serveral TX error conection quality monitoring parameters: % error rate, survey interval, and # of attempted packets. On exceeding the TX failure rate over the given interval, the driver will send a CQM notify event with the actual TX failure rate and packets attempted. Signed-off-by: Thomas Pedersen <c_tpeder@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-07-12 23:17:33 +00:00
{
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211/mac80211/ath6kl: acquire wdev lock outside ch_switch_notify The channel switch notification should be sent under the wdev/sdata-lock, preferably in the same moment as the channel change happens, to avoid races by other callers (e.g. start/stop_ap). This also adds the previously missing sdata_lock protection in csa_finalize_work. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:51 +00:00
ASSERT_WDEV_LOCK(wdev);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211/mac80211/ath6kl: acquire wdev lock outside ch_switch_notify The channel switch notification should be sent under the wdev/sdata-lock, preferably in the same moment as the channel change happens, to avoid races by other callers (e.g. start/stop_ap). This also adds the previously missing sdata_lock protection in csa_finalize_work. Reported-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: Simon Wunderlich <sw@simonwunderlich.de> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-21 17:19:51 +00:00
trace_cfg80211_ch_switch_notify(dev, chandef);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
cfg80211: consider existing DFS interfaces It was possible to break interface combinations in the following way: combo 1: iftype = AP, num_ifaces = 2, num_chans = 2, combo 2: iftype = AP, num_ifaces = 1, num_chans = 1, radar = HT20 With the above interface combinations it was possible to: step 1. start AP on DFS channel by matching combo 2 step 2. start AP on non-DFS channel by matching combo 1 This was possible beacuse (step 2) did not consider if other interfaces require radar detection. The patch changes how cfg80211 tracks channels - instead of channel itself now a complete chandef is stored. Signed-off-by: Michal Kazior <michal.kazior@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-29 13:22:27 +00:00
wdev->chandef = *chandef;
cfg80211: set preset_chandef after channel switch Set preset_chandef in channel switch notification. In other case we will have old preset_chandef. Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 13:29:21 +00:00
wdev->preset_chandef = *chandef;
cfg80211: add channel switch started notification Add a new NL80211_CH_SWITCH_STARTED_NOTIFY message that can be sent to the userspace when a channel switch process has started. This allows userspace to take action, for instance, by requesting other interfaces to switch channel as necessary. This patch introduces a function that allows the drivers to send this notification. It should be used when the driver starts processing a channel switch initiated by a remote device (eg. when a STA receives a CSA from the AP) and when it successfully starts a userspace-triggered channel switch (eg. when hostapd triggers a channel swith in the AP). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-07 12:31:35 +00:00
nl80211_ch_switch_notify(rdev, dev, chandef, GFP_KERNEL,
NL80211_CMD_CH_SWITCH_NOTIFY, 0);
cfg80211: move exported event functions into nl80211 This is the sort of thing gcc's LTO could do, but since we don't have that yet we can also do it manually. The advantage is reduced code, both source and binary, e.g. on x86-64 text data bss dec hex filename 442825 56230 776 499831 7a077 cfg80211.ko (before) 441585 56230 776 498591 79b9f cfg80211.ko (after) a reduction of ~1k. But in order to not complicate the code move only those functions that are simple wrappers, not those that have functionality of their own. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-22 21:05:20 +00:00
}
EXPORT_SYMBOL(cfg80211_ch_switch_notify);
cfg80211: add channel switch started notification Add a new NL80211_CH_SWITCH_STARTED_NOTIFY message that can be sent to the userspace when a channel switch process has started. This allows userspace to take action, for instance, by requesting other interfaces to switch channel as necessary. This patch introduces a function that allows the drivers to send this notification. It should be used when the driver starts processing a channel switch initiated by a remote device (eg. when a STA receives a CSA from the AP) and when it successfully starts a userspace-triggered channel switch (eg. when hostapd triggers a channel swith in the AP). Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-11-07 12:31:35 +00:00
void cfg80211_ch_switch_started_notify(struct net_device *dev,
struct cfg80211_chan_def *chandef,
u8 count)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
struct wiphy *wiphy = wdev->wiphy;
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
trace_cfg80211_ch_switch_started_notify(dev, chandef);
nl80211_ch_switch_notify(rdev, dev, chandef, GFP_KERNEL,
NL80211_CMD_CH_SWITCH_STARTED_NOTIFY, count);
}
EXPORT_SYMBOL(cfg80211_ch_switch_started_notify);
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
void
nl80211_radar_notify(struct cfg80211_registered_device *rdev,
cfg80211/mac80211: DFS setup chandef for cac event To report channel width correctly we have to send correct channel parameters from mac80211 when calling cfg80211_cac_event(). This is required in case of using channel width higher than 20MHz and we have to set correct dfs channel state after CAC (NL80211_DFS_AVAILABLE). Signed-off-by: Janusz Dziedzic <janusz.dziedzic@tieto.com> Reviewed-by: Luis R. Rodriguez <mcgrof@do-not-panic.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-11-06 12:55:51 +00:00
const struct cfg80211_chan_def *chandef,
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
enum nl80211_radar_event event,
struct net_device *netdev, gfp_t gfp)
{
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_RADAR_DETECT);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx))
goto nla_put_failure;
/* NOP and radar events don't need a netdev parameter */
if (netdev) {
struct wireless_dev *wdev = netdev->ieee80211_ptr;
if (nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
goto nla_put_failure;
}
if (nla_put_u32(msg, NL80211_ATTR_RADAR_EVENT, event))
goto nla_put_failure;
if (nl80211_send_chandef(msg, chandef))
goto nla_put_failure;
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211/cfg80211: add radar detection command/event Add new NL80211_CMD_RADAR_DETECT, which starts the Channel Availability Check (CAC). This command will also notify the usermode about events (CAC finished, CAC aborted, radar detected, NOP finished). Once radar detection has started it should continuously monitor for radars as long as the channel is active. This patch enables DFS for AP mode in nl80211/cfg80211. Based on original patch by Victor Goldenshtein <victorg@ti.com> Signed-off-by: Simon Wunderlich <siwu@hrz.tu-chemnitz.de> [remove WIPHY_FLAG_HAS_RADAR_DETECT again -- my mistake] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-08 17:16:19 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
void cfg80211_probe_status(struct net_device *dev, const u8 *addr,
u64 cookie, bool acked, gfp_t gfp)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
struct sk_buff *msg;
void *hdr;
cfg80211: add cfg80211 exported function tracing Also add tracing to the API functions that drivers (and mac80211) can call in cfg80211. Signed-off-by: Beni Lev <beni.lev@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-08-27 09:49:39 +00:00
trace_cfg80211_probe_status(dev, addr, cookie, acked);
net: Use NLMSG_DEFAULT_SIZE in combination with nlmsg_new() Using NLMSG_GOODSIZE results in multiple pages being used as nlmsg_new() will automatically add the size of the netlink header to the payload thus exceeding the page limit. NLMSG_DEFAULT_SIZE takes this into account. Signed-off-by: Thomas Graf <tgraf@suug.ch> Cc: Jiri Pirko <jpirko@redhat.com> Cc: Dmitry Eremin-Solenikov <dbaryshkov@gmail.com> Cc: Sergey Lapin <slapin@ossfans.org> Cc: Johannes Berg <johannes@sipsolutions.net> Cc: Lauro Ramos Venancio <lauro.venancio@openbossa.org> Cc: Aloisio Almeida Jr <aloisio.almeida@openbossa.org> Cc: Samuel Ortiz <sameo@linux.intel.com> Reviewed-by: Jiri Pirko <jpirko@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-06-28 03:57:45 +00:00
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
cfg80211: add cfg80211 exported function tracing Also add tracing to the API functions that drivers (and mac80211) can call in cfg80211. Signed-off-by: Beni Lev <beni.lev@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-08-27 09:49:39 +00:00
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_PROBE_CLIENT);
if (!hdr) {
nlmsg_free(msg);
return;
}
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, addr) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_COOKIE, cookie,
NL80211_ATTR_PAD) ||
wireless: Stop using NLA_PUT*(). These macros contain a hidden goto, and are thus extremely error prone and make code hard to audit. Signed-off-by: David S. Miller <davem@davemloft.net>
2012-03-29 08:41:26 +00:00
(acked && nla_put_flag(msg, NL80211_ATTR_ACK)))
goto nla_put_failure;
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
nl80211: add API to probe a client When the AP SME in hostapd is used it wants to probe the clients when they have been idle for some time. Add explicit API to support this. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:15 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_probe_status);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
void cfg80211_report_obss_beacon(struct wiphy *wiphy,
const u8 *frame, size_t len,
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
int freq, int sig_dbm)
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
{
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
struct sk_buff *msg;
void *hdr;
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
struct cfg80211_beacon_registration *reg;
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: add cfg80211 exported function tracing Also add tracing to the API functions that drivers (and mac80211) can call in cfg80211. Signed-off-by: Beni Lev <beni.lev@intel.com> Reviewed-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Reviewed-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-08-27 09:49:39 +00:00
trace_cfg80211_report_obss_beacon(wiphy, frame, len, freq, sig_dbm);
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
spin_lock_bh(&rdev->beacon_registrations_lock);
list_for_each_entry(reg, &rdev->beacon_registrations, list) {
msg = nlmsg_new(len + 100, GFP_ATOMIC);
if (!msg) {
spin_unlock_bh(&rdev->beacon_registrations_lock);
return;
}
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FRAME);
if (!hdr)
goto nla_put_failure;
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
(freq &&
nla_put_u32(msg, NL80211_ATTR_WIPHY_FREQ, freq)) ||
(sig_dbm &&
nla_put_u32(msg, NL80211_ATTR_RX_SIGNAL_DBM, sig_dbm)) ||
nla_put(msg, NL80211_ATTR_FRAME, len, frame))
goto nla_put_failure;
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
genlmsg_end(msg, hdr);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
genlmsg_unicast(wiphy_net(&rdev->wiphy), msg, reg->nlportid);
}
spin_unlock_bh(&rdev->beacon_registrations_lock);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
return;
nla_put_failure:
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
spin_unlock_bh(&rdev->beacon_registrations_lock);
if (hdr)
genlmsg_cancel(msg, hdr);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_report_obss_beacon);
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
#ifdef CONFIG_PM
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
static int cfg80211_net_detect_results(struct sk_buff *msg,
struct cfg80211_wowlan_wakeup *wakeup)
{
struct cfg80211_wowlan_nd_info *nd = wakeup->net_detect;
struct nlattr *nl_results, *nl_match, *nl_freqs;
int i, j;
nl_results = nla_nest_start(
msg, NL80211_WOWLAN_TRIG_NET_DETECT_RESULTS);
if (!nl_results)
return -EMSGSIZE;
for (i = 0; i < nd->n_matches; i++) {
struct cfg80211_wowlan_nd_match *match = nd->matches[i];
nl_match = nla_nest_start(msg, i);
if (!nl_match)
break;
/* The SSID attribute is optional in nl80211, but for
* simplicity reasons it's always present in the
* cfg80211 structure. If a driver can't pass the
* SSID, that needs to be changed. A zero length SSID
* is still a valid SSID (wildcard), so it cannot be
* used for this purpose.
*/
if (nla_put(msg, NL80211_ATTR_SSID, match->ssid.ssid_len,
match->ssid.ssid)) {
nla_nest_cancel(msg, nl_match);
goto out;
}
if (match->n_channels) {
nl_freqs = nla_nest_start(
msg, NL80211_ATTR_SCAN_FREQUENCIES);
if (!nl_freqs) {
nla_nest_cancel(msg, nl_match);
goto out;
}
for (j = 0; j < match->n_channels; j++) {
nl80211: use loop index as type for net detect frequency results We currently add nested members of the NL80211_ATTR_SCAN_FREQUENCIES as NLA_U32 attributes of type NL80211_ATTR_WIPHY_FREQ in cfg80211_net_detect_results. However, since there can be an arbitrary number of frequency results, we should use the loop index of the loop used to add the frequency results to NL80211_ATTR_SCAN_FREQUENCIES as the type (i.e. nla_type) for each result attribute, rather than a fixed type. This change is in line with how nested members are added to NL80211_ATTR_SCAN_FREQUENCIES in the functions nl80211_send_wowlan_nd and nl80211_add_scan_req. Signed-off-by: Samuel Tan <samueltan@chromium.org> Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-02-09 19:29:15 +00:00
if (nla_put_u32(msg, j, match->channels[j])) {
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
nla_nest_cancel(msg, nl_freqs);
nla_nest_cancel(msg, nl_match);
goto out;
}
}
nla_nest_end(msg, nl_freqs);
}
nla_nest_end(msg, nl_match);
}
out:
nla_nest_end(msg, nl_results);
return 0;
}
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
void cfg80211_report_wowlan_wakeup(struct wireless_dev *wdev,
struct cfg80211_wowlan_wakeup *wakeup,
gfp_t gfp)
{
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
struct sk_buff *msg;
void *hdr;
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
int size = 200;
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
trace_cfg80211_report_wowlan_wakeup(wdev->wiphy, wdev, wakeup);
if (wakeup)
size += wakeup->packet_present_len;
msg = nlmsg_new(size, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_SET_WOWLAN);
if (!hdr)
goto free_msg;
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
goto free_msg;
if (wdev->netdev && nla_put_u32(msg, NL80211_ATTR_IFINDEX,
wdev->netdev->ifindex))
goto free_msg;
if (wakeup) {
struct nlattr *reasons;
reasons = nla_nest_start(msg, NL80211_ATTR_WOWLAN_TRIGGERS);
nl80211: check nla_nest_start() return value Coverity pointed out that we might dereference NULL later if nla_nest_start() returns a failure. This isn't really true since we'd bomb out before, but we should check the return value directly, so do that. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:16:58 +00:00
if (!reasons)
goto free_msg;
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
if (wakeup->disconnect &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_DISCONNECT))
goto free_msg;
if (wakeup->magic_pkt &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_MAGIC_PKT))
goto free_msg;
if (wakeup->gtk_rekey_failure &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_GTK_REKEY_FAILURE))
goto free_msg;
if (wakeup->eap_identity_req &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_EAP_IDENT_REQUEST))
goto free_msg;
if (wakeup->four_way_handshake &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_4WAY_HANDSHAKE))
goto free_msg;
if (wakeup->rfkill_release &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_RFKILL_RELEASE))
goto free_msg;
if (wakeup->pattern_idx >= 0 &&
nla_put_u32(msg, NL80211_WOWLAN_TRIG_PKT_PATTERN,
wakeup->pattern_idx))
goto free_msg;
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (wakeup->tcp_match &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_WAKEUP_TCP_MATCH))
goto free_msg;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (wakeup->tcp_connlost &&
nla_put_flag(msg, NL80211_WOWLAN_TRIG_WAKEUP_TCP_CONNLOST))
goto free_msg;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (wakeup->tcp_nomoretokens &&
nla_put_flag(msg,
NL80211_WOWLAN_TRIG_WAKEUP_TCP_NOMORETOKENS))
goto free_msg;
cfg80211: configuration for WoWLAN over TCP Intel Wireless devices are able to make a TCP connection after suspending, sending some data and waking up when the connection receives wakeup data (or breaks). Add the WoWLAN configuration and feature advertising API for it. Acked-by: David S. Miller <davem@davemloft.net> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-23 21:57:40 +00:00
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
if (wakeup->packet) {
u32 pkt_attr = NL80211_WOWLAN_TRIG_WAKEUP_PKT_80211;
u32 len_attr = NL80211_WOWLAN_TRIG_WAKEUP_PKT_80211_LEN;
if (!wakeup->packet_80211) {
pkt_attr =
NL80211_WOWLAN_TRIG_WAKEUP_PKT_8023;
len_attr =
NL80211_WOWLAN_TRIG_WAKEUP_PKT_8023_LEN;
}
if (wakeup->packet_len &&
nla_put_u32(msg, len_attr, wakeup->packet_len))
goto free_msg;
if (nla_put(msg, pkt_attr, wakeup->packet_present_len,
wakeup->packet))
goto free_msg;
}
cfg80211: add wowlan net-detect support Add a new WoWLAN API to enable net-detect as a wake up trigger. Net-detect allows the device to scan in the background while the host is asleep to wake up the host system when a matching network is found. Reuse the scheduled scan attributes to specify how the scan is performed while suspended and the matches that will trigger a wake event. Signed-off-by: Luciano Coelho <luciano.coelho@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-09-17 08:55:28 +00:00
if (wakeup->net_detect &&
cfg80211_net_detect_results(msg, wakeup))
goto free_msg;
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
nla_nest_end(msg, reasons);
}
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211/mac80211: support reporting wakeup reason When waking up from WoWLAN, it is useful to know what triggered the wakeup. Support reporting the wakeup reason(s) in cfg80211 (and a pass-through in mac80211) to allow userspace to know. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-01-22 11:34:29 +00:00
return;
free_msg:
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_report_wowlan_wakeup);
#endif
cfg80211: Add TDLS event to allow drivers to request operations The NL80211_CMD_TDLS_OPER command was previously used only for userspace request for the kernel code to perform TDLS operations. However, there are also cases where the driver may need to request operations from userspace, e.g., when using security on the AP path. Add a new cfg80211 function for generating a TDLS operation event for drivers to request a new link to be set up (NL80211_TDLS_SETUP) or an existing link to be torn down (NL80211_TDLS_TEARDOWN). Drivers can optionally use these events, e.g., based on noticing data traffic being sent to a peer station that is seen with good signal strength. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-16 20:49:57 +00:00
void cfg80211_tdls_oper_request(struct net_device *dev, const u8 *peer,
enum nl80211_tdls_operation oper,
u16 reason_code, gfp_t gfp)
{
struct wireless_dev *wdev = dev->ieee80211_ptr;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
cfg80211: Add TDLS event to allow drivers to request operations The NL80211_CMD_TDLS_OPER command was previously used only for userspace request for the kernel code to perform TDLS operations. However, there are also cases where the driver may need to request operations from userspace, e.g., when using security on the AP path. Add a new cfg80211 function for generating a TDLS operation event for drivers to request a new link to be set up (NL80211_TDLS_SETUP) or an existing link to be torn down (NL80211_TDLS_TEARDOWN). Drivers can optionally use these events, e.g., based on noticing data traffic being sent to a peer station that is seen with good signal strength. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-16 20:49:57 +00:00
struct sk_buff *msg;
void *hdr;
trace_cfg80211_tdls_oper_request(wdev->wiphy, dev, peer, oper,
reason_code);
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_TDLS_OPER);
if (!hdr) {
nlmsg_free(msg);
return;
}
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, dev->ifindex) ||
nla_put_u8(msg, NL80211_ATTR_TDLS_OPERATION, oper) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, peer) ||
(reason_code > 0 &&
nla_put_u16(msg, NL80211_ATTR_REASON_CODE, reason_code)))
goto nla_put_failure;
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
cfg80211: Add TDLS event to allow drivers to request operations The NL80211_CMD_TDLS_OPER command was previously used only for userspace request for the kernel code to perform TDLS operations. However, there are also cases where the driver may need to request operations from userspace, e.g., when using security on the AP path. Add a new cfg80211 function for generating a TDLS operation event for drivers to request a new link to be set up (NL80211_TDLS_SETUP) or an existing link to be torn down (NL80211_TDLS_TEARDOWN). Drivers can optionally use these events, e.g., based on noticing data traffic being sent to a peer station that is seen with good signal strength. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-16 20:49:57 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, gfp);
cfg80211: Add TDLS event to allow drivers to request operations The NL80211_CMD_TDLS_OPER command was previously used only for userspace request for the kernel code to perform TDLS operations. However, there are also cases where the driver may need to request operations from userspace, e.g., when using security on the AP path. Add a new cfg80211 function for generating a TDLS operation event for drivers to request a new link to be set up (NL80211_TDLS_SETUP) or an existing link to be torn down (NL80211_TDLS_TEARDOWN). Drivers can optionally use these events, e.g., based on noticing data traffic being sent to a peer station that is seen with good signal strength. Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-11-16 20:49:57 +00:00
return;
nla_put_failure:
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_tdls_oper_request);
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
static int nl80211_netlink_notify(struct notifier_block * nb,
unsigned long state,
void *_notify)
{
struct netlink_notify *notify = _notify;
struct cfg80211_registered_device *rdev;
struct wireless_dev *wdev;
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
struct cfg80211_beacon_registration *reg, *tmp;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
nl80211: check netlink protocol in socket release notification A non-privileged user can create a netlink socket with the same port_id as used by an existing open nl80211 netlink socket (e.g. as used by a hostapd process) with a different protocol number. Closing this socket will then lead to the notification going to nl80211's socket release notification handler, and possibly cause an action such as removing a virtual interface. Fix this issue by checking that the netlink protocol is NETLINK_GENERIC. Since generic netlink has no notifier chain of its own, we can't fix the problem more generically. Fixes: 026331c4d9b5 ("cfg80211/mac80211: allow registering for and sending action frames") Cc: stable@vger.kernel.org Signed-off-by: Dmitry Ivanov <dima@ubnt.com> [rewrite commit message] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-04-06 14:23:18 +00:00
if (state != NETLINK_URELEASE || notify->protocol != NETLINK_GENERIC)
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
return NOTIFY_DONE;
rcu_read_lock();
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
list_for_each_entry_rcu(rdev, &cfg80211_rdev_list, list) {
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
struct cfg80211_sched_scan_request *sched_scan_req;
nl80211: fix sched scan netlink socket owner destruction A single netlink socket might own multiple interfaces *and* a scheduled scan request (which might belong to another interface), so when it goes away both may need to be destroyed. Remove the schedule_scan_stop indirection to fix this - it's only needed for interface destruction because of the way this works right now, with a single work taking care of all interfaces. Cc: stable@vger.kernel.org Fixes: 93a1e86ce10e4 ("nl80211: Stop scheduled scan if netlink client disappears") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-05 09:57:14 +00:00
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
list_for_each_entry_rcu(sched_scan_req,
&rdev->sched_scan_req_list,
list) {
if (sched_scan_req->owner_nlportid == notify->portid) {
sched_scan_req->nl_owner_dead = true;
nl80211: fix sched scan netlink socket owner destruction A single netlink socket might own multiple interfaces *and* a scheduled scan request (which might belong to another interface), so when it goes away both may need to be destroyed. Remove the schedule_scan_stop indirection to fix this - it's only needed for interface destruction because of the way this works right now, with a single work taking care of all interfaces. Cc: stable@vger.kernel.org Fixes: 93a1e86ce10e4 ("nl80211: Stop scheduled scan if netlink client disappears") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-05 09:57:14 +00:00
schedule_work(&rdev->sched_scan_stop_wk);
nl80211: allow multiple active scheduled scan requests This patch implements the idea to have multiple scheduled scan requests running concurrently. It mainly illustrates how to deal with the incoming request from user-space in terms of backward compatibility. In order to use multiple scheduled scans user-space needs to provide a flag attribute NL80211_ATTR_SCHED_SCAN_MULTI to indicate support. If not the request is treated as a legacy scan. Drivers currently supporting scheduled scan are now indicating they support a single scheduled scan request. This obsoletes WIPHY_FLAG_SUPPORTS_SCHED_SCAN. Reviewed-by: Hante Meuleman <hante.meuleman@broadcom.com> Reviewed-by: Pieter-Paul Giesberts <pieter-paul.giesberts@broadcom.com> Reviewed-by: Franky Lin <franky.lin@broadcom.com> Signed-off-by: Arend van Spriel <arend.vanspriel@broadcom.com> [clean up netlink destroy path to avoid allocations, code cleanups] Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-21 12:05:00 +00:00
}
nl80211: fix sched scan netlink socket owner destruction A single netlink socket might own multiple interfaces *and* a scheduled scan request (which might belong to another interface), so when it goes away both may need to be destroyed. Remove the schedule_scan_stop indirection to fix this - it's only needed for interface destruction because of the way this works right now, with a single work taking care of all interfaces. Cc: stable@vger.kernel.org Fixes: 93a1e86ce10e4 ("nl80211: Stop scheduled scan if netlink client disappears") Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-05 09:57:14 +00:00
}
cfg80211: allow userspace to take ownership of interfaces When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-03-24 16:57:27 +00:00
cfg80211: make wdev_list accessible to drivers There's no harm in having drivers read the list, since they can use RCU protection or RTNL locking; allow this to not require each and every driver to also implement its own bookkeeping. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2016-05-03 13:52:04 +00:00
list_for_each_entry_rcu(wdev, &rdev->wiphy.wdev_list, list) {
netlink: Rename pid to portid to avoid confusion It is a frequent mistake to confuse the netlink port identifier with a process identifier. Try to reduce this confusion by renaming fields that hold port identifiers portid instead of pid. I have carefully avoided changing the structures exported to userspace to avoid changing the userspace API. I have successfully built an allyesconfig kernel with this change. Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com> Acked-by: Stephen Hemminger <shemminger@vyatta.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2012-09-07 20:12:54 +00:00
cfg80211_mlme_unregister_socket(wdev, notify->portid);
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
cfg80211: simplify netlink socket owner interface deletion There's no need to allocate a portid structure and then, for each of those, walk the interfaces - we can just add a flag to each interface and walk those directly. Due to padding in the struct, we can even do it without any memory cost, and it even simplifies the code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 05:43:41 +00:00
if (wdev->owner_nlportid == notify->portid) {
wdev->nl_owner_dead = true;
schedule_work(&rdev->destroy_work);
} else if (wdev->conn_owner_nlportid == notify->portid) {
cfg80211: NL80211_ATTR_SOCKET_OWNER support for CMD_CONNECT Disconnect or deauthenticate when the owning socket is closed if this flag is supplied to CMD_CONNECT or CMD_ASSOCIATE. This may be used to ensure userspace daemon doesn't leave an unmanaged connection behind. In some situations it would be possible to account for that, to some degree, in the deamon restart code or in the up/down scripts without the use of this attribute. But there will be systems where the daemon can go away for varying periods without a warning due to local resource management. Signed-off-by: Andrew Zaborowski <andrew.zaborowski@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-06 21:33:43 +00:00
schedule_work(&wdev->disconnect_wk);
cfg80211: simplify netlink socket owner interface deletion There's no need to allocate a portid structure and then, for each of those, walk the interfaces - we can just add a flag to each interface and walk those directly. Due to padding in the struct, we can even do it without any memory cost, and it even simplifies the code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-04-26 05:43:41 +00:00
}
cfg80211: allow userspace to take ownership of interfaces When dynamically creating interfaces from userspace, e.g. for P2P usage, such interfaces are usually owned by the process that created them, i.e. wpa_supplicant. Should wpa_supplicant crash, such interfaces will often cease operating properly and cause problems on restarting the process. To avoid this problem, introduce an ownership concept for interfaces. If an interface is owned by a netlink socket, then it will be destroyed if the netlink socket is closed for any reason, including if the process it belongs to crashed. This gives us a race-free way to get rid of any such interfaces. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-03-24 16:57:27 +00:00
}
cfg80211: allow registering more than one beacon listener The commit: commit 5e760230e42cf759bd923457ca2753aacf2e656e Author: Johannes Berg <johannes.berg@intel.com> Date: Fri Nov 4 11:18:17 2011 +0100 cfg80211: allow registering to beacons allowed only a single process to register for beacon events per wiphy. This breaks cases where a user may want two or more VIFs on a wiphy and run a seperate hostapd process on each vif. This patch allows multiple beacon listeners, fixing the regression. Signed-off-by: Ben Greear <greearb@candelatech.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2012-10-26 21:49:25 +00:00
spin_lock_bh(&rdev->beacon_registrations_lock);
list_for_each_entry_safe(reg, tmp, &rdev->beacon_registrations,
list) {
if (reg->nlportid == notify->portid) {
list_del(&reg->list);
kfree(reg);
break;
}
}
spin_unlock_bh(&rdev->beacon_registrations_lock);
cfg80211: allow registering to beacons Add the ability to register to received beacon frames to allow implementing OLBC logic in userspace. The registration is per wiphy since there's no point in receiving the same frame multiple times. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-11-04 10:18:17 +00:00
}
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
rcu_read_unlock();
cfg80211: Add API to change the indoor regulatory setting Previously, the indoor setting configuration assumed that as long as a station interface is connected, the indoor environment setting does not change. However, this assumption is problematic as: - It is possible that a station interface is connected to a mobile AP, e.g., softAP or a P2P GO, where it is possible that both the station and the mobile AP move out of the indoor environment making the indoor setting invalid. In such a case, user space has no way to invalidate the setting. - A station interface disconnection does not necessarily imply that the device is no longer operating in an indoor environment, e.g., it is possible that the station interface is roaming but is still stays indoor. To handle the above, extend the indoor configuration API to allow user space to indicate a change of indoor settings, and allow it to indicate weather it controls the indoor setting, such that: 1. If the user space process explicitly indicates that it is going to control the indoor setting, do not clear the indoor setting internally, unless the socket is released. The user space process should use the NL80211_ATTR_SOCKET_OWNER attribute in the command to state that it is going to control the indoor setting. 2. Reset the indoor setting when restoring the regulatory settings in case it is not owned by a user space process. Based on the above, a user space tool that continuously monitors the indoor settings, i.e., tracking power setting, location etc., can indicate environment changes to the regulatory core. It should be noted that currently user space is the only provided mechanism used to hint to the regulatory core over the indoor/outdoor environment -- while the country IEs do have an environment setting this has been completely ignored by the regulatory core by design for a while now since country IEs typically can contain bogus data. Acked-by: Luis R. Rodriguez <mcgrof@suse.com> Signed-off-by: ArikX Nemtsov <arik@wizery.com> Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2015-03-04 05:32:06 +00:00
/*
* It is possible that the user space process that is controlling the
* indoor setting disappeared, so notify the regulatory core.
*/
regulatory_netlink_notify(notify->portid);
cfg80211: change return value of notifier function Return NOTIFY_DONE if we don't care this time's notification, return NOTIFY_OK if we successfully handled this time's notification. That's the formal way to do it. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:04 +00:00
return NOTIFY_OK;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
}
static struct notifier_block nl80211_netlink_notifier = {
.notifier_call = nl80211_netlink_notify,
};
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
void cfg80211_ft_event(struct net_device *netdev,
struct cfg80211_ft_event_params *ft_event)
{
struct wiphy *wiphy = netdev->ieee80211_ptr->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
struct sk_buff *msg;
void *hdr;
trace_cfg80211_ft_event(wiphy, netdev, ft_event);
if (!ft_event->target_ap)
return;
cfg80211: size various nl80211 messages correctly Ilan reported that sometimes nl80211 messages weren't working if the frames being transported got very large, which was really a problem for userspace-to-kernel messages, but prompted me to look at the code. Upon review, I found various places where variable-length data is transported in an nl80211 message but the message isn't allocated taking that into account. This shouldn't cause any problems since the frames aren't really that long, apart in one place where two (possibly very long frames) might not fit. Fix all the places (that I found) that get variable length data from the driver and put it into a message to take the length of the variable data into account. The 100 there is just a safe constant for the remaining message overhead (it's usually around 50 for most messages.) Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2017-01-09 10:10:42 +00:00
msg = nlmsg_new(100 + ft_event->ric_ies_len, GFP_KERNEL);
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_FT_EVENT);
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (!hdr)
goto out;
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, netdev->ifindex) ||
nla_put(msg, NL80211_ATTR_MAC, ETH_ALEN, ft_event->target_ap))
goto out;
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
if (ft_event->ies &&
nla_put(msg, NL80211_ATTR_IE, ft_event->ies_len, ft_event->ies))
goto out;
if (ft_event->ric_ies &&
nla_put(msg, NL80211_ATTR_IE_RIC, ft_event->ric_ies_len,
ft_event->ric_ies))
goto out;
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
nl80211: remove bogus genlmsg_end() error checking genlmsg_end() can't return an error since it returns the skb length so remove checks treating the return value as an error code. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-06-04 10:46:03 +00:00
genlmsg_end(msg, hdr);
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
genetlink: pass family to functions using groups This doesn't really change anything, but prepares for the next patch that will change the APIs to pass the group ID within the family, rather than the global group ID. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:38 +00:00
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(&rdev->wiphy), msg, 0,
genetlink: make multicast groups const, prevent abuse Register generic netlink multicast groups as an array with the family and give them contiguous group IDs. Then instead of passing the global group ID to the various functions that send messages, pass the ID relative to the family - for most families that's just 0 because the only have one group. This avoids the list_head and ID in each group, adding a new field for the mcast group ID offset to the family. At the same time, this allows us to prevent abusing groups again like the quota and dropmon code did, since we can now check that a family only uses a group it owns. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2013-11-19 14:19:39 +00:00
NL80211_MCGRP_MLME, GFP_KERNEL);
nl80211: check nla_put_* return values Coverity pointed out that in a few functions we don't check the return value of the nla_put_*() calls. Most of these are fairly harmless because the input isn't very dynamic and controlled by the kernel, but the pattern is simply wrong, so fix this. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-10-25 09:05:22 +00:00
return;
out:
nlmsg_free(msg);
cfg80211: Extend support for IEEE 802.11r Fast BSS Transition Add NL80211_CMD_UPDATE_FT_IES to support update of FT IEs to the WLAN driver and NL80211_CMD_FT_EVENT to send FT events from the WLAN driver. This will carry the target AP's MAC address along with the relevant Information Elements. This event is used to report received FT IEs (MDIE, FTIE, RSN IE, TIE, RICIE). These changes allow FT to be supported with drivers that use an internal SME instead of user space option (like FT implementation in wpa_supplicant with mac80211-based drivers). Signed-off-by: Jouni Malinen <jouni@qca.qualcomm.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-02-27 15:14:27 +00:00
}
EXPORT_SYMBOL(cfg80211_ft_event);
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
void cfg80211_crit_proto_stopped(struct wireless_dev *wdev, gfp_t gfp)
{
struct cfg80211_registered_device *rdev;
struct sk_buff *msg;
void *hdr;
u32 nlportid;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
rdev = wiphy_to_rdev(wdev->wiphy);
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
if (!rdev->crit_proto_nlportid)
return;
nlportid = rdev->crit_proto_nlportid;
rdev->crit_proto_nlportid = 0;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, gfp);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_CRIT_PROTOCOL_STOP);
if (!hdr)
goto nla_put_failure;
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
cfg80211: introduce critical protocol indication from user-space Some protocols need a more reliable connection to complete successful in reasonable time. This patch adds a user-space API to indicate the wireless driver that a critical protocol is about to commence and when it is done, using nl80211 primitives NL80211_CMD_CRIT_PROTOCOL_START and NL80211_CRIT_PROTOCOL_STOP. There can be only on critical protocol session started per registered cfg80211 device. The driver can support this by implementing the cfg80211 callbacks .crit_proto_start() and .crit_proto_stop(). Examples of protocols that can benefit from this are DHCP, EAPOL, APIPA. Exactly how the link can/should be made more reliable is up to the driver. Things to consider are avoid scanning, no multi-channel operations, and alter coexistence schemes. Reviewed-by: Pieter-Paul Giesberts <pieterpg@broadcom.com> Reviewed-by: Franky (Zhenhui) Lin <frankyl@broadcom.com> Signed-off-by: Arend van Spriel <arend@broadcom.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2013-04-18 13:49:00 +00:00
goto nla_put_failure;
genlmsg_end(msg, hdr);
genlmsg_unicast(wiphy_net(&rdev->wiphy), msg, nlportid);
return;
nla_put_failure:
if (hdr)
genlmsg_cancel(msg, hdr);
nlmsg_free(msg);
}
EXPORT_SYMBOL(cfg80211_crit_proto_stopped);
nl80211: send event when AP operation is stopped There are a few cases, e.g. suspend, where an AP interface is stopped by the kernel rather than by userspace request, most commonly when suspending. To let userspace know about this, send the NL80211_CMD_STOP_AP command as an event every time an AP interface is stopped. This also happens when userspace did in fact request the AP stop, but that's not a problem. For full-MAC drivers this may need to be extended to also cover cases where the device stopped the AP operation for some reason, this a bit more complicated because then all cfg80211 state also needs to be reset; such API is not part of this patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 13:06:29 +00:00
void nl80211_send_ap_stopped(struct wireless_dev *wdev)
{
struct wiphy *wiphy = wdev->wiphy;
cfg80211: change wiphy_to_dev function name Name wiphy_to_rdev is more accurate to describe what the function does, i.e., return a pointer pointing to struct cfg80211_registered_device. Signed-off-by: Zhao, Gang <gamerh2o@gmail.com> Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-04-21 04:53:03 +00:00
struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
nl80211: send event when AP operation is stopped There are a few cases, e.g. suspend, where an AP interface is stopped by the kernel rather than by userspace request, most commonly when suspending. To let userspace know about this, send the NL80211_CMD_STOP_AP command as an event every time an AP interface is stopped. This also happens when userspace did in fact request the AP stop, but that's not a problem. For full-MAC drivers this may need to be extended to also cover cases where the device stopped the AP operation for some reason, this a bit more complicated because then all cfg80211 state also needs to be reset; such API is not part of this patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 13:06:29 +00:00
struct sk_buff *msg;
void *hdr;
msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
if (!msg)
return;
hdr = nl80211hdr_put(msg, 0, 0, 0, NL80211_CMD_STOP_AP);
if (!hdr)
goto out;
if (nla_put_u32(msg, NL80211_ATTR_WIPHY, rdev->wiphy_idx) ||
nla_put_u32(msg, NL80211_ATTR_IFINDEX, wdev->netdev->ifindex) ||
wireless: use nla_put_u64_64bit() Signed-off-by: Nicolas Dichtel <nicolas.dichtel@6wind.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-04-25 08:25:22 +00:00
nla_put_u64_64bit(msg, NL80211_ATTR_WDEV, wdev_id(wdev),
NL80211_ATTR_PAD))
nl80211: send event when AP operation is stopped There are a few cases, e.g. suspend, where an AP interface is stopped by the kernel rather than by userspace request, most commonly when suspending. To let userspace know about this, send the NL80211_CMD_STOP_AP command as an event every time an AP interface is stopped. This also happens when userspace did in fact request the AP stop, but that's not a problem. For full-MAC drivers this may need to be extended to also cover cases where the device stopped the AP operation for some reason, this a bit more complicated because then all cfg80211 state also needs to be reset; such API is not part of this patch. Signed-off-by: Johannes Berg <johannes.berg@intel.com>
2014-01-24 13:06:29 +00:00
goto out;
genlmsg_end(msg, hdr);
genlmsg_multicast_netns(&nl80211_fam, wiphy_net(wiphy), msg, 0,
NL80211_MCGRP_MLME, GFP_KERNEL);
return;
out:
nlmsg_free(msg);
}
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
/* initialisation/exit functions */
genetlink: mark families as __ro_after_init Now genl_register_family() is the only thing (other than the users themselves, perhaps, but I didn't find any doing that) writing to the family struct. In all families that I found, genl_register_family() is only called from __init functions (some indirectly, in which case I've add __init annotations to clarifly things), so all can actually be marked __ro_after_init. This protects the data structure from accidental corruption. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:05 +00:00
int __init nl80211_init(void)
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
{
wireless: Use genl_register_family_with_ops() Use genl_register_family_with_ops() instead of a copy. Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl> Signed-off-by: David S. Miller <davem@davemloft.net>
2009-05-21 10:34:06 +00:00
int err;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
genetlink: statically initialize families Instead of providing macros/inline functions to initialize the families, make all users initialize them statically and get rid of the macros. This reduces the kernel code size by about 1.6k on x86-64 (with allyesconfig). Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2016-10-24 12:40:03 +00:00
err = genl_register_family(&nl80211_fam);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
if (err)
return err;
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
err = netlink_register_notifier(&nl80211_netlink_notifier);
if (err)
goto err_out;
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
return 0;
err_out:
genl_unregister_family(&nl80211_fam);
return err;
}
void nl80211_exit(void)
{
cfg80211/mac80211: allow registering for and sending action frames This implements a new command to register for action frames that userspace wants to handle instead of the in-kernel rejection. It is then responsible for rejecting ones that it decided not to handle. There is no unregistration, but the socket can be closed for that. Frames that are not registered for will not be forwarded to userspace and will be rejected by the kernel, the cfg80211 API helps implementing that. Additionally, this patch adds a new command that allows doing action frame transmission from userspace. It can be used either to exchange action frames on the current operational channel (e.g., with the AP with which we are currently associated) or to exchange off-channel Public Action frames with the remain-on-channel command. Signed-off-by: Jouni Malinen <jouni.malinen@atheros.com> Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com>
2010-02-15 10:53:10 +00:00
netlink_unregister_notifier(&nl80211_netlink_notifier);
[NL80211]: add netlink interface to cfg80211 Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Signed-off-by: John W. Linville <linville@tuxdriver.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2007-09-20 17:09:35 +00:00
genl_unregister_family(&nl80211_fam);
}
Reference in New Issue Copy Permalink