License cleanup: add SPDX GPL-2.0 license identifier to files with no license
Many source files in the tree are missing licensing information, which
makes it harder for compliance tools to determine the correct license.
By default all files without license information are under the default
license of the kernel, which is GPL version 2.
Update the files which contain no license information with the 'GPL-2.0'
SPDX license identifier. The SPDX identifier is a legally binding
shorthand, which can be used instead of the full boiler plate text.
This patch is based on work done by Thomas Gleixner and Kate Stewart and
Philippe Ombredanne.
How this work was done:
Patches were generated and checked against linux-4.14-rc6 for a subset of
the use cases:
- file had no licensing information it it.
- file was a */uapi/* one with no licensing information in it,
- file was a */uapi/* one with existing licensing information,
Further patches will be generated in subsequent months to fix up cases
where non-standard license headers were used, and references to license
had to be inferred by heuristics based on keywords.
The analysis to determine which SPDX License Identifier to be applied to
a file was done in a spreadsheet of side by side results from of the
output of two independent scanners (ScanCode & Windriver) producing SPDX
tag:value files created by Philippe Ombredanne. Philippe prepared the
base worksheet, and did an initial spot review of a few 1000 files.
The 4.13 kernel was the starting point of the analysis with 60,537 files
assessed. Kate Stewart did a file by file comparison of the scanner
results in the spreadsheet to determine which SPDX license identifier(s)
to be applied to the file. She confirmed any determination that was not
immediately clear with lawyers working with the Linux Foundation.
Criteria used to select files for SPDX license identifier tagging was:
- Files considered eligible had to be source code files.
- Make and config files were included as candidates if they contained >5
lines of source
- File already had some variant of a license header in it (even if <5
lines).
All documentation files were explicitly excluded.
The following heuristics were used to determine which SPDX license
identifiers to apply.
- when both scanners couldn't find any license traces, file was
considered to have no license information in it, and the top level
COPYING file license applied.
For non */uapi/* files that summary was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 11139
and resulted in the first patch in this series.
If that file was a */uapi/* path one, it was "GPL-2.0 WITH
Linux-syscall-note" otherwise it was "GPL-2.0". Results of that was:
SPDX license identifier # files
---------------------------------------------------|-------
GPL-2.0 WITH Linux-syscall-note 930
and resulted in the second patch in this series.
- if a file had some form of licensing information in it, and was one
of the */uapi/* ones, it was denoted with the Linux-syscall-note if
any GPL family license was found in the file or had no licensing in
it (per prior point). Results summary:
SPDX license identifier # files
---------------------------------------------------|------
GPL-2.0 WITH Linux-syscall-note 270
GPL-2.0+ WITH Linux-syscall-note 169
((GPL-2.0 WITH Linux-syscall-note) OR BSD-2-Clause) 21
((GPL-2.0 WITH Linux-syscall-note) OR BSD-3-Clause) 17
LGPL-2.1+ WITH Linux-syscall-note 15
GPL-1.0+ WITH Linux-syscall-note 14
((GPL-2.0+ WITH Linux-syscall-note) OR BSD-3-Clause) 5
LGPL-2.0+ WITH Linux-syscall-note 4
LGPL-2.1 WITH Linux-syscall-note 3
((GPL-2.0 WITH Linux-syscall-note) OR MIT) 3
((GPL-2.0 WITH Linux-syscall-note) AND MIT) 1
and that resulted in the third patch in this series.
- when the two scanners agreed on the detected license(s), that became
the concluded license(s).
- when there was disagreement between the two scanners (one detected a
license but the other didn't, or they both detected different
licenses) a manual inspection of the file occurred.
- In most cases a manual inspection of the information in the file
resulted in a clear resolution of the license that should apply (and
which scanner probably needed to revisit its heuristics).
- When it was not immediately clear, the license identifier was
confirmed with lawyers working with the Linux Foundation.
- If there was any question as to the appropriate license identifier,
the file was flagged for further research and to be revisited later
in time.
In total, over 70 hours of logged manual review was done on the
spreadsheet to determine the SPDX license identifiers to apply to the
source files by Kate, Philippe, Thomas and, in some cases, confirmation
by lawyers working with the Linux Foundation.
Kate also obtained a third independent scan of the 4.13 code base from
FOSSology, and compared selected files where the other two scanners
disagreed against that SPDX file, to see if there was new insights. The
Windriver scanner is based on an older version of FOSSology in part, so
they are related.
Thomas did random spot checks in about 500 files from the spreadsheets
for the uapi headers and agreed with SPDX license identifier in the
files he inspected. For the non-uapi files Thomas did random spot checks
in about 15000 files.
In initial set of patches against 4.14-rc6, 3 files were found to have
copy/paste license identifier errors, and have been fixed to reflect the
correct identifier.
Additionally Philippe spent 10 hours this week doing a detailed manual
inspection and review of the 12,461 patched files from the initial patch
version early this week with:
- a full scancode scan run, collecting the matched texts, detected
license ids and scores
- reviewing anything where there was a license detected (about 500+
files) to ensure that the applied SPDX license was correct
- reviewing anything where there was no detection but the patch license
was not GPL-2.0 WITH Linux-syscall-note to ensure that the applied
SPDX license was correct
This produced a worksheet with 20 files needing minor correction. This
worksheet was then exported into 3 different .csv files for the
different types of files to be modified.
These .csv files were then reviewed by Greg. Thomas wrote a script to
parse the csv files and add the proper SPDX tag to the file, in the
format that the file expected. This script was further refined by Greg
based on the output to detect more types of files automatically and to
distinguish between header and source .c files (which need different
comment types.) Finally Greg ran the script using the .csv files to
generate the patches.
Reviewed-by: Kate Stewart <kstewart@linuxfoundation.org>
Reviewed-by: Philippe Ombredanne <pombredanne@nexb.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2017-11-01 14:07:57 +00:00
|
|
|
/* SPDX-License-Identifier: GPL-2.0 */
|
2005-11-10 01:25:51 +00:00
|
|
|
#ifndef __NET_NETLINK_H
|
|
|
|
#define __NET_NETLINK_H
|
|
|
|
|
|
|
|
#include <linux/types.h>
|
|
|
|
#include <linux/netlink.h>
|
2006-12-04 04:15:30 +00:00
|
|
|
#include <linux/jiffies.h>
|
2015-03-29 14:59:25 +00:00
|
|
|
#include <linux/in6.h>
|
2005-11-10 01:25:51 +00:00
|
|
|
|
|
|
|
/* ========================================================================
|
|
|
|
* Netlink Messages and Attributes Interface (As Seen On TV)
|
|
|
|
* ------------------------------------------------------------------------
|
|
|
|
* Messages Interface
|
|
|
|
* ------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* Message Format:
|
|
|
|
* <--- nlmsg_total_size(payload) --->
|
|
|
|
* <-- nlmsg_msg_size(payload) ->
|
|
|
|
* +----------+- - -+-------------+- - -+-------- - -
|
|
|
|
* | nlmsghdr | Pad | Payload | Pad | nlmsghdr
|
|
|
|
* +----------+- - -+-------------+- - -+-------- - -
|
|
|
|
* nlmsg_data(nlh)---^ ^
|
|
|
|
* nlmsg_next(nlh)-----------------------+
|
|
|
|
*
|
|
|
|
* Payload Format:
|
|
|
|
* <---------------------- nlmsg_len(nlh) --------------------->
|
|
|
|
* <------ hdrlen ------> <- nlmsg_attrlen(nlh, hdrlen) ->
|
|
|
|
* +----------------------+- - -+--------------------------------+
|
|
|
|
* | Family Header | Pad | Attributes |
|
|
|
|
* +----------------------+- - -+--------------------------------+
|
|
|
|
* nlmsg_attrdata(nlh, hdrlen)---^
|
|
|
|
*
|
|
|
|
* Data Structures:
|
|
|
|
* struct nlmsghdr netlink message header
|
|
|
|
*
|
|
|
|
* Message Construction:
|
|
|
|
* nlmsg_new() create a new netlink message
|
|
|
|
* nlmsg_put() add a netlink message to an skb
|
|
|
|
* nlmsg_put_answer() callback based nlmsg_put()
|
2010-06-21 20:45:42 +00:00
|
|
|
* nlmsg_end() finalize netlink message
|
2006-08-05 06:03:05 +00:00
|
|
|
* nlmsg_get_pos() return current position in message
|
|
|
|
* nlmsg_trim() trim part of message
|
2005-11-10 01:25:51 +00:00
|
|
|
* nlmsg_cancel() cancel message construction
|
2024-04-03 20:22:59 +00:00
|
|
|
* nlmsg_consume() free a netlink message (expected)
|
|
|
|
* nlmsg_free() free a netlink message (drop)
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Message Sending:
|
|
|
|
* nlmsg_multicast() multicast message to several groups
|
|
|
|
* nlmsg_unicast() unicast a message to a single socket
|
2006-08-15 07:31:06 +00:00
|
|
|
* nlmsg_notify() send notification message
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Message Length Calculations:
|
|
|
|
* nlmsg_msg_size(payload) length of message w/o padding
|
|
|
|
* nlmsg_total_size(payload) length of message w/ padding
|
|
|
|
* nlmsg_padlen(payload) length of padding at tail
|
|
|
|
*
|
|
|
|
* Message Payload Access:
|
|
|
|
* nlmsg_data(nlh) head of message payload
|
|
|
|
* nlmsg_len(nlh) length of message payload
|
|
|
|
* nlmsg_attrdata(nlh, hdrlen) head of attributes data
|
|
|
|
* nlmsg_attrlen(nlh, hdrlen) length of attributes data
|
|
|
|
*
|
|
|
|
* Message Parsing:
|
|
|
|
* nlmsg_ok(nlh, remaining) does nlh fit into remaining bytes?
|
|
|
|
* nlmsg_next(nlh, remaining) get next netlink message
|
|
|
|
* nlmsg_parse() parse attributes of a message
|
|
|
|
* nlmsg_find_attr() find an attribute in a message
|
|
|
|
* nlmsg_for_each_msg() loop over all messages
|
|
|
|
* nlmsg_validate() validate netlink message incl. attrs
|
|
|
|
* nlmsg_for_each_attr() loop over all attributes
|
|
|
|
*
|
2006-08-15 07:31:41 +00:00
|
|
|
* Misc:
|
|
|
|
* nlmsg_report() report back to application?
|
|
|
|
*
|
2005-11-10 01:25:51 +00:00
|
|
|
* ------------------------------------------------------------------------
|
|
|
|
* Attributes Interface
|
|
|
|
* ------------------------------------------------------------------------
|
|
|
|
*
|
|
|
|
* Attribute Format:
|
|
|
|
* <------- nla_total_size(payload) ------->
|
|
|
|
* <---- nla_attr_size(payload) ----->
|
|
|
|
* +----------+- - -+- - - - - - - - - +- - -+-------- - -
|
|
|
|
* | Header | Pad | Payload | Pad | Header
|
|
|
|
* +----------+- - -+- - - - - - - - - +- - -+-------- - -
|
|
|
|
* <- nla_len(nla) -> ^
|
|
|
|
* nla_data(nla)----^ |
|
|
|
|
* nla_next(nla)-----------------------------'
|
|
|
|
*
|
|
|
|
* Data Structures:
|
2007-10-11 04:09:48 +00:00
|
|
|
* struct nlattr netlink attribute header
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Attribute Construction:
|
2006-08-05 06:03:05 +00:00
|
|
|
* nla_reserve(skb, type, len) reserve room for an attribute
|
|
|
|
* nla_reserve_nohdr(skb, len) reserve room for an attribute w/o hdr
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_put(skb, type, len, data) add attribute to skb
|
2006-08-05 06:03:05 +00:00
|
|
|
* nla_put_nohdr(skb, len, data) add attribute w/o hdr
|
2008-01-23 06:10:59 +00:00
|
|
|
* nla_append(skb, len, data) append data to skb
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Attribute Construction for Basic Types:
|
|
|
|
* nla_put_u8(skb, type, value) add u8 attribute to skb
|
|
|
|
* nla_put_u16(skb, type, value) add u16 attribute to skb
|
|
|
|
* nla_put_u32(skb, type, value) add u32 attribute to skb
|
2017-07-13 14:46:43 +00:00
|
|
|
* nla_put_u64_64bit(skb, type,
|
|
|
|
* value, padattr) add u64 attribute to skb
|
2012-07-27 06:28:52 +00:00
|
|
|
* nla_put_s8(skb, type, value) add s8 attribute to skb
|
|
|
|
* nla_put_s16(skb, type, value) add s16 attribute to skb
|
|
|
|
* nla_put_s32(skb, type, value) add s32 attribute to skb
|
2016-04-22 15:31:20 +00:00
|
|
|
* nla_put_s64(skb, type, value,
|
|
|
|
* padattr) add s64 attribute to skb
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_put_string(skb, type, str) add string attribute to skb
|
|
|
|
* nla_put_flag(skb, type) add flag attribute to skb
|
2016-04-22 15:31:21 +00:00
|
|
|
* nla_put_msecs(skb, type, jiffies,
|
|
|
|
* padattr) add msecs attribute to skb
|
2015-03-29 14:59:25 +00:00
|
|
|
* nla_put_in_addr(skb, type, addr) add IPv4 address attribute to skb
|
|
|
|
* nla_put_in6_addr(skb, type, addr) add IPv6 address attribute to skb
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Nested Attributes Construction:
|
|
|
|
* nla_nest_start(skb, type) start a nested attribute
|
|
|
|
* nla_nest_end(skb, nla) finalize a nested attribute
|
|
|
|
* nla_nest_cancel(skb, nla) cancel nested attribute construction
|
|
|
|
*
|
|
|
|
* Attribute Length Calculations:
|
|
|
|
* nla_attr_size(payload) length of attribute w/o padding
|
|
|
|
* nla_total_size(payload) length of attribute w/ padding
|
|
|
|
* nla_padlen(payload) length of padding
|
|
|
|
*
|
|
|
|
* Attribute Payload Access:
|
|
|
|
* nla_data(nla) head of attribute payload
|
|
|
|
* nla_len(nla) length of attribute payload
|
|
|
|
*
|
|
|
|
* Attribute Payload Access for Basic Types:
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* nla_get_uint(nla) get payload for a uint attribute
|
|
|
|
* nla_get_sint(nla) get payload for a sint attribute
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_get_u8(nla) get payload for a u8 attribute
|
|
|
|
* nla_get_u16(nla) get payload for a u16 attribute
|
|
|
|
* nla_get_u32(nla) get payload for a u32 attribute
|
|
|
|
* nla_get_u64(nla) get payload for a u64 attribute
|
2012-07-27 06:28:52 +00:00
|
|
|
* nla_get_s8(nla) get payload for a s8 attribute
|
|
|
|
* nla_get_s16(nla) get payload for a s16 attribute
|
|
|
|
* nla_get_s32(nla) get payload for a s32 attribute
|
|
|
|
* nla_get_s64(nla) get payload for a s64 attribute
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_get_flag(nla) return 1 if flag is true
|
|
|
|
* nla_get_msecs(nla) get payload for a msecs attribute
|
|
|
|
*
|
|
|
|
* Attribute Misc:
|
|
|
|
* nla_memcpy(dest, nla, count) copy attribute into memory
|
|
|
|
* nla_memcmp(nla, data, size) compare attribute with memory area
|
2020-11-15 17:08:06 +00:00
|
|
|
* nla_strscpy(dst, nla, size) copy attribute to a sized string
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_strcmp(nla, str) compare attribute with string
|
|
|
|
*
|
|
|
|
* Attribute Parsing:
|
|
|
|
* nla_ok(nla, remaining) does nla fit into remaining bytes?
|
|
|
|
* nla_next(nla, remaining) get next netlink attribute
|
|
|
|
* nla_validate() validate a stream of attributes
|
2006-09-25 22:54:03 +00:00
|
|
|
* nla_validate_nested() validate a stream of nested attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_find() find attribute in stream of attributes
|
2006-08-05 06:03:05 +00:00
|
|
|
* nla_find_nested() find attribute in nested attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_parse() parse and validate stream of attrs
|
2018-09-26 20:19:42 +00:00
|
|
|
* nla_parse_nested() parse nested attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* nla_for_each_attr() loop over all attributes
|
netlink: introduce type-checking attribute iteration
There are, especially with multi-attr arrays, many cases
of needing to iterate all attributes of a specific type
in a netlink message or a nested attribute. Add specific
macros to support that case.
Also convert many instances using this spatch:
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
Although I had to undo one bad change this made, and
I also adjusted some other code for whitespace and to
use direct variable initialization now.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://lore.kernel.org/r/20240328203144.b5a6c895fb80.I1869b44767379f204998ff44dd239803f39c23e0@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-03-28 19:31:45 +00:00
|
|
|
* nla_for_each_attr_type() loop over all attributes with the
|
|
|
|
* given type
|
2006-09-25 22:53:37 +00:00
|
|
|
* nla_for_each_nested() loop over the nested attributes
|
netlink: introduce type-checking attribute iteration
There are, especially with multi-attr arrays, many cases
of needing to iterate all attributes of a specific type
in a netlink message or a nested attribute. Add specific
macros to support that case.
Also convert many instances using this spatch:
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
Although I had to undo one bad change this made, and
I also adjusted some other code for whitespace and to
use direct variable initialization now.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://lore.kernel.org/r/20240328203144.b5a6c895fb80.I1869b44767379f204998ff44dd239803f39c23e0@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-03-28 19:31:45 +00:00
|
|
|
* nla_for_each_nested_type() loop over the nested attributes with
|
|
|
|
* the given type
|
2005-11-10 01:25:51 +00:00
|
|
|
*=========================================================================
|
|
|
|
*/
|
|
|
|
|
|
|
|
/**
|
|
|
|
* Standard attribute types to specify validation policy
|
|
|
|
*/
|
|
|
|
enum {
|
|
|
|
NLA_UNSPEC,
|
|
|
|
NLA_U8,
|
|
|
|
NLA_U16,
|
|
|
|
NLA_U32,
|
|
|
|
NLA_U64,
|
|
|
|
NLA_STRING,
|
|
|
|
NLA_FLAG,
|
|
|
|
NLA_MSECS,
|
|
|
|
NLA_NESTED,
|
2018-09-26 09:15:34 +00:00
|
|
|
NLA_NESTED_ARRAY,
|
2006-08-27 03:11:47 +00:00
|
|
|
NLA_NUL_STRING,
|
2007-03-23 18:37:48 +00:00
|
|
|
NLA_BINARY,
|
2012-07-27 06:28:52 +00:00
|
|
|
NLA_S8,
|
|
|
|
NLA_S16,
|
|
|
|
NLA_S32,
|
|
|
|
NLA_S64,
|
2017-07-30 17:24:49 +00:00
|
|
|
NLA_BITFIELD32,
|
2018-09-17 09:57:28 +00:00
|
|
|
NLA_REJECT,
|
2022-10-31 12:34:07 +00:00
|
|
|
NLA_BE16,
|
|
|
|
NLA_BE32,
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
NLA_SINT,
|
|
|
|
NLA_UINT,
|
2005-11-10 01:25:51 +00:00
|
|
|
__NLA_TYPE_MAX,
|
|
|
|
};
|
|
|
|
|
|
|
|
#define NLA_TYPE_MAX (__NLA_TYPE_MAX - 1)
|
|
|
|
|
2020-04-30 20:13:08 +00:00
|
|
|
struct netlink_range_validation {
|
|
|
|
u64 min, max;
|
|
|
|
};
|
|
|
|
|
|
|
|
struct netlink_range_validation_signed {
|
|
|
|
s64 min, max;
|
|
|
|
};
|
|
|
|
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
enum nla_policy_validation {
|
|
|
|
NLA_VALIDATE_NONE,
|
|
|
|
NLA_VALIDATE_RANGE,
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
NLA_VALIDATE_RANGE_WARN_TOO_LONG,
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
NLA_VALIDATE_MIN,
|
|
|
|
NLA_VALIDATE_MAX,
|
2020-10-05 22:07:38 +00:00
|
|
|
NLA_VALIDATE_MASK,
|
2020-04-30 20:13:08 +00:00
|
|
|
NLA_VALIDATE_RANGE_PTR,
|
2018-09-27 09:28:36 +00:00
|
|
|
NLA_VALIDATE_FUNCTION,
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
};
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* struct nla_policy - attribute validation policy
|
|
|
|
* @type: Type of attribute or NLA_UNSPEC
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* @validation_type: type of attribute validation done in addition to
|
2018-09-27 09:28:36 +00:00
|
|
|
* type-specific validation (e.g. range, function call), see
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* &enum nla_policy_validation
|
2006-08-27 03:11:47 +00:00
|
|
|
* @len: Type specific length of payload
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* Policies are defined as arrays of this struct, the array must be
|
|
|
|
* accessible by attribute type up to the highest identifier to be expected.
|
|
|
|
*
|
2006-08-27 03:11:47 +00:00
|
|
|
* Meaning of `len' field:
|
|
|
|
* NLA_STRING Maximum length of string
|
|
|
|
* NLA_NUL_STRING Maximum length of string (excluding NUL)
|
|
|
|
* NLA_FLAG Unused
|
2007-03-23 18:37:48 +00:00
|
|
|
* NLA_BINARY Maximum length of attribute payload
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
* (but see also below with the validation type)
|
2018-09-26 09:15:34 +00:00
|
|
|
* NLA_NESTED,
|
|
|
|
* NLA_NESTED_ARRAY Length verification is done by checking len of
|
2018-09-26 09:15:33 +00:00
|
|
|
* nested header (or empty); len field is used if
|
2020-04-30 20:13:05 +00:00
|
|
|
* nested_policy is also used, for the max attr
|
2018-09-26 09:15:33 +00:00
|
|
|
* number in the nested policy.
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* NLA_SINT, NLA_UINT,
|
2011-11-03 00:10:05 +00:00
|
|
|
* NLA_U8, NLA_U16,
|
|
|
|
* NLA_U32, NLA_U64,
|
2012-07-27 06:28:52 +00:00
|
|
|
* NLA_S8, NLA_S16,
|
|
|
|
* NLA_S32, NLA_S64,
|
2022-10-31 12:34:07 +00:00
|
|
|
* NLA_BE16, NLA_BE32,
|
2011-11-03 00:10:05 +00:00
|
|
|
* NLA_MSECS Leaving the length field zero will verify the
|
|
|
|
* given type fits, using it verifies minimum length
|
|
|
|
* just like "All other"
|
2018-09-17 09:57:28 +00:00
|
|
|
* NLA_BITFIELD32 Unused
|
|
|
|
* NLA_REJECT Unused
|
2011-11-03 00:10:05 +00:00
|
|
|
* All other Minimum length of attribute payload
|
2006-08-27 03:11:47 +00:00
|
|
|
*
|
2020-04-30 20:13:05 +00:00
|
|
|
* Meaning of validation union:
|
2018-09-17 09:57:28 +00:00
|
|
|
* NLA_BITFIELD32 This is a 32-bit bitmap/bitselector attribute and
|
2020-04-30 20:13:05 +00:00
|
|
|
* `bitfield32_valid' is the u32 value of valid flags
|
|
|
|
* NLA_REJECT This attribute is always rejected and `reject_message'
|
2018-09-17 09:57:28 +00:00
|
|
|
* may point to a string to report as the error instead
|
|
|
|
* of the generic one in extended ACK.
|
2020-04-30 20:13:05 +00:00
|
|
|
* NLA_NESTED `nested_policy' to a nested policy to validate, must
|
|
|
|
* also set `len' to the max attribute number. Use the
|
|
|
|
* provided NLA_POLICY_NESTED() macro.
|
2018-09-26 09:15:33 +00:00
|
|
|
* Note that nla_parse() will validate, but of course not
|
|
|
|
* parse, the nested sub-policies.
|
2020-04-30 20:13:05 +00:00
|
|
|
* NLA_NESTED_ARRAY `nested_policy' points to a nested policy to validate,
|
|
|
|
* must also set `len' to the max attribute number. Use
|
|
|
|
* the provided NLA_POLICY_NESTED_ARRAY() macro.
|
|
|
|
* The difference to NLA_NESTED is the structure:
|
|
|
|
* NLA_NESTED has the nested attributes directly inside
|
|
|
|
* while an array has the nested attributes at another
|
|
|
|
* level down and the attribute types directly in the
|
|
|
|
* nesting don't matter.
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* NLA_UINT,
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* NLA_U8,
|
|
|
|
* NLA_U16,
|
|
|
|
* NLA_U32,
|
|
|
|
* NLA_U64,
|
2022-10-31 12:34:07 +00:00
|
|
|
* NLA_BE16,
|
|
|
|
* NLA_BE32,
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* NLA_SINT,
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* NLA_S8,
|
|
|
|
* NLA_S16,
|
|
|
|
* NLA_S32,
|
2020-04-30 20:13:05 +00:00
|
|
|
* NLA_S64 The `min' and `max' fields are used depending on the
|
|
|
|
* validation_type field, if that is min/max/range then
|
|
|
|
* the min, max or both are used (respectively) to check
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* the value of the integer attribute.
|
|
|
|
* Note that in the interest of code simplicity and
|
|
|
|
* struct size both limits are s16, so you cannot
|
|
|
|
* enforce a range that doesn't fall within the range
|
2023-01-27 07:45:06 +00:00
|
|
|
* of s16 - do that using the NLA_POLICY_FULL_RANGE()
|
|
|
|
* or NLA_POLICY_FULL_RANGE_SIGNED() macros instead.
|
2020-04-30 20:13:05 +00:00
|
|
|
* Use the NLA_POLICY_MIN(), NLA_POLICY_MAX() and
|
|
|
|
* NLA_POLICY_RANGE() macros.
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* NLA_UINT,
|
2020-04-30 20:13:08 +00:00
|
|
|
* NLA_U8,
|
|
|
|
* NLA_U16,
|
|
|
|
* NLA_U32,
|
|
|
|
* NLA_U64 If the validation_type field instead is set to
|
|
|
|
* NLA_VALIDATE_RANGE_PTR, `range' must be a pointer
|
|
|
|
* to a struct netlink_range_validation that indicates
|
|
|
|
* the min/max values.
|
|
|
|
* Use NLA_POLICY_FULL_RANGE().
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
* NLA_SINT,
|
2020-04-30 20:13:08 +00:00
|
|
|
* NLA_S8,
|
|
|
|
* NLA_S16,
|
|
|
|
* NLA_S32,
|
|
|
|
* NLA_S64 If the validation_type field instead is set to
|
|
|
|
* NLA_VALIDATE_RANGE_PTR, `range_signed' must be a
|
|
|
|
* pointer to a struct netlink_range_validation_signed
|
|
|
|
* that indicates the min/max values.
|
|
|
|
* Use NLA_POLICY_FULL_RANGE_SIGNED().
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
*
|
|
|
|
* NLA_BINARY If the validation type is like the ones for integers
|
|
|
|
* above, then the min/max length (not value like for
|
|
|
|
* integers) of the attribute is enforced.
|
|
|
|
*
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
* All other Unused - but note that it's a union
|
2018-09-17 09:57:28 +00:00
|
|
|
*
|
2018-09-27 09:28:36 +00:00
|
|
|
* Meaning of `validate' field, use via NLA_POLICY_VALIDATE_FN:
|
2020-04-30 20:13:05 +00:00
|
|
|
* NLA_BINARY Validation function called for the attribute.
|
2018-09-27 09:28:36 +00:00
|
|
|
* All other Unused - but note that it's a union
|
|
|
|
*
|
2005-11-10 01:25:51 +00:00
|
|
|
* Example:
|
2020-04-30 20:13:05 +00:00
|
|
|
*
|
|
|
|
* static const u32 myvalidflags = 0xff231023;
|
|
|
|
*
|
2010-02-18 08:14:31 +00:00
|
|
|
* static const struct nla_policy my_policy[ATTR_MAX+1] = {
|
2005-11-10 01:25:51 +00:00
|
|
|
* [ATTR_FOO] = { .type = NLA_U16 },
|
2007-03-23 18:37:48 +00:00
|
|
|
* [ATTR_BAR] = { .type = NLA_STRING, .len = BARSIZ },
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
* [ATTR_BAZ] = NLA_POLICY_EXACT_LEN(sizeof(struct mystruct)),
|
2020-04-30 20:13:05 +00:00
|
|
|
* [ATTR_GOO] = NLA_POLICY_BITFIELD32(myvalidflags),
|
2005-11-10 01:25:51 +00:00
|
|
|
* };
|
|
|
|
*/
|
|
|
|
struct nla_policy {
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
u8 type;
|
|
|
|
u8 validation_type;
|
2006-08-27 03:11:47 +00:00
|
|
|
u16 len;
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
union {
|
2022-10-27 21:21:07 +00:00
|
|
|
/**
|
|
|
|
* @strict_start_type: first attribute to validate strictly
|
|
|
|
*
|
|
|
|
* This entry is special, and used for the attribute at index 0
|
netlink: add strict parsing for future attributes
Unfortunately, we cannot add strict parsing for all attributes, as
that would break existing userspace. We currently warn about it, but
that's about all we can do.
For new attributes, however, the story is better: nobody is using
them, so we can reject bad sizes.
Also, for new attributes, we need not accept them when the policy
doesn't declare their usage.
David Ahern and I went back and forth on how to best encode this, and
the best way we found was to have a "boundary type", from which point
on new attributes have all possible validation applied, and NLA_UNSPEC
is rejected.
As we didn't want to add another argument to all functions that get a
netlink policy, the workaround is to encode that boundary in the first
entry of the policy array (which is for type 0 and thus probably not
really valid anyway). I put it into the validation union for the rare
possibility that somebody is actually using attribute 0, which would
continue to work fine unless they tried to use the extended validation,
which isn't likely. We also didn't find any in-tree users with type 0.
The reason for setting the "start strict here" attribute is that we
never really need to start strict from 0, which is invalid anyway (or
in legacy families where that isn't true, it cannot be set to strict),
so we can thus reserve the value 0 for "don't do this check" and don't
have to add the tag to all policies right now.
Thus, policies can now opt in to this validation, which we should do
for all existing policies, at least when adding new attributes.
Note that entirely *new* policies won't need to set it, as the use
of that should be using nla_parse()/nlmsg_parse() etc. which anyway
do fully strict validation now, regardless of this.
So in effect, this patch only covers the "existing command with new
attribute" case.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:30 +00:00
|
|
|
* only, and specifies special data about the policy, namely it
|
|
|
|
* specifies the "boundary type" where strict length validation
|
|
|
|
* starts for any attribute types >= this value, also, strict
|
|
|
|
* nesting validation starts here.
|
|
|
|
*
|
|
|
|
* Additionally, it means that NLA_UNSPEC is actually NLA_REJECT
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
* for any types >= this, so need to use NLA_POLICY_MIN_LEN() to
|
|
|
|
* get the previous pure { .len = xyz } behaviour. The advantage
|
|
|
|
* of this is that types not specified in the policy will be
|
|
|
|
* rejected.
|
netlink: add strict parsing for future attributes
Unfortunately, we cannot add strict parsing for all attributes, as
that would break existing userspace. We currently warn about it, but
that's about all we can do.
For new attributes, however, the story is better: nobody is using
them, so we can reject bad sizes.
Also, for new attributes, we need not accept them when the policy
doesn't declare their usage.
David Ahern and I went back and forth on how to best encode this, and
the best way we found was to have a "boundary type", from which point
on new attributes have all possible validation applied, and NLA_UNSPEC
is rejected.
As we didn't want to add another argument to all functions that get a
netlink policy, the workaround is to encode that boundary in the first
entry of the policy array (which is for type 0 and thus probably not
really valid anyway). I put it into the validation union for the rare
possibility that somebody is actually using attribute 0, which would
continue to work fine unless they tried to use the extended validation,
which isn't likely. We also didn't find any in-tree users with type 0.
The reason for setting the "start strict here" attribute is that we
never really need to start strict from 0, which is invalid anyway (or
in legacy families where that isn't true, it cannot be set to strict),
so we can thus reserve the value 0 for "don't do this check" and don't
have to add the tag to all policies right now.
Thus, policies can now opt in to this validation, which we should do
for all existing policies, at least when adding new attributes.
Note that entirely *new* policies won't need to set it, as the use
of that should be using nla_parse()/nlmsg_parse() etc. which anyway
do fully strict validation now, regardless of this.
So in effect, this patch only covers the "existing command with new
attribute" case.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:30 +00:00
|
|
|
*
|
|
|
|
* For completely new families it should be set to 1 so that the
|
|
|
|
* validation is enforced for all attributes. For existing ones
|
|
|
|
* it should be set at least when new attributes are added to
|
|
|
|
* the enum used by the policy, and be set to the new value that
|
|
|
|
* was added to enforce strict validation from thereon.
|
|
|
|
*/
|
|
|
|
u16 strict_start_type;
|
2022-10-27 21:21:07 +00:00
|
|
|
|
|
|
|
/* private: use NLA_POLICY_*() to set */
|
|
|
|
const u32 bitfield32_valid;
|
|
|
|
const u32 mask;
|
|
|
|
const char *reject_message;
|
|
|
|
const struct nla_policy *nested_policy;
|
2023-10-25 16:22:04 +00:00
|
|
|
const struct netlink_range_validation *range;
|
|
|
|
const struct netlink_range_validation_signed *range_signed;
|
2022-10-27 21:21:07 +00:00
|
|
|
struct {
|
|
|
|
s16 min, max;
|
|
|
|
};
|
|
|
|
int (*validate)(const struct nlattr *attr,
|
|
|
|
struct netlink_ext_ack *extack);
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
};
|
2005-11-10 01:25:51 +00:00
|
|
|
};
|
|
|
|
|
2018-09-17 09:57:29 +00:00
|
|
|
#define NLA_POLICY_ETH_ADDR NLA_POLICY_EXACT_LEN(ETH_ALEN)
|
|
|
|
#define NLA_POLICY_ETH_ADDR_COMPAT NLA_POLICY_EXACT_LEN_WARN(ETH_ALEN)
|
|
|
|
|
2019-01-25 09:08:28 +00:00
|
|
|
#define _NLA_POLICY_NESTED(maxattr, policy) \
|
2020-04-30 20:13:05 +00:00
|
|
|
{ .type = NLA_NESTED, .nested_policy = policy, .len = maxattr }
|
2019-01-25 09:08:28 +00:00
|
|
|
#define _NLA_POLICY_NESTED_ARRAY(maxattr, policy) \
|
2020-04-30 20:13:05 +00:00
|
|
|
{ .type = NLA_NESTED_ARRAY, .nested_policy = policy, .len = maxattr }
|
2019-01-25 09:08:28 +00:00
|
|
|
#define NLA_POLICY_NESTED(policy) \
|
|
|
|
_NLA_POLICY_NESTED(ARRAY_SIZE(policy) - 1, policy)
|
|
|
|
#define NLA_POLICY_NESTED_ARRAY(policy) \
|
|
|
|
_NLA_POLICY_NESTED_ARRAY(ARRAY_SIZE(policy) - 1, policy)
|
2020-04-30 20:13:05 +00:00
|
|
|
#define NLA_POLICY_BITFIELD32(valid) \
|
|
|
|
{ .type = NLA_BITFIELD32, .bitfield32_valid = valid }
|
2018-09-26 09:15:33 +00:00
|
|
|
|
2023-07-18 07:52:29 +00:00
|
|
|
#define __NLA_IS_UINT_TYPE(tp) \
|
|
|
|
(tp == NLA_U8 || tp == NLA_U16 || tp == NLA_U32 || \
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
tp == NLA_U64 || tp == NLA_UINT || \
|
|
|
|
tp == NLA_BE16 || tp == NLA_BE32)
|
2020-10-05 22:07:37 +00:00
|
|
|
#define __NLA_IS_SINT_TYPE(tp) \
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
(tp == NLA_S8 || tp == NLA_S16 || tp == NLA_S32 || tp == NLA_S64 || \
|
|
|
|
tp == NLA_SINT)
|
2020-10-05 22:07:37 +00:00
|
|
|
|
2018-10-12 10:53:00 +00:00
|
|
|
#define __NLA_ENSURE(condition) BUILD_BUG_ON_ZERO(!(condition))
|
2020-10-05 22:07:38 +00:00
|
|
|
#define NLA_ENSURE_UINT_TYPE(tp) \
|
|
|
|
(__NLA_ENSURE(__NLA_IS_UINT_TYPE(tp)) + tp)
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
#define NLA_ENSURE_UINT_OR_BINARY_TYPE(tp) \
|
2020-10-05 22:07:37 +00:00
|
|
|
(__NLA_ENSURE(__NLA_IS_UINT_TYPE(tp) || \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
tp == NLA_MSECS || \
|
|
|
|
tp == NLA_BINARY) + tp)
|
2020-04-30 20:13:08 +00:00
|
|
|
#define NLA_ENSURE_SINT_TYPE(tp) \
|
2020-10-05 22:07:37 +00:00
|
|
|
(__NLA_ENSURE(__NLA_IS_SINT_TYPE(tp)) + tp)
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
#define NLA_ENSURE_INT_OR_BINARY_TYPE(tp) \
|
2020-10-05 22:07:37 +00:00
|
|
|
(__NLA_ENSURE(__NLA_IS_UINT_TYPE(tp) || \
|
|
|
|
__NLA_IS_SINT_TYPE(tp) || \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
tp == NLA_MSECS || \
|
|
|
|
tp == NLA_BINARY) + tp)
|
2018-09-27 09:28:36 +00:00
|
|
|
#define NLA_ENSURE_NO_VALIDATION_PTR(tp) \
|
|
|
|
(__NLA_ENSURE(tp != NLA_BITFIELD32 && \
|
|
|
|
tp != NLA_REJECT && \
|
|
|
|
tp != NLA_NESTED && \
|
|
|
|
tp != NLA_NESTED_ARRAY) + tp)
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
|
|
|
|
#define NLA_POLICY_RANGE(tp, _min, _max) { \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp), \
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
.validation_type = NLA_VALIDATE_RANGE, \
|
|
|
|
.min = _min, \
|
|
|
|
.max = _max \
|
|
|
|
}
|
|
|
|
|
2020-04-30 20:13:08 +00:00
|
|
|
#define NLA_POLICY_FULL_RANGE(tp, _range) { \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
.type = NLA_ENSURE_UINT_OR_BINARY_TYPE(tp), \
|
2020-04-30 20:13:08 +00:00
|
|
|
.validation_type = NLA_VALIDATE_RANGE_PTR, \
|
|
|
|
.range = _range, \
|
|
|
|
}
|
|
|
|
|
|
|
|
#define NLA_POLICY_FULL_RANGE_SIGNED(tp, _range) { \
|
|
|
|
.type = NLA_ENSURE_SINT_TYPE(tp), \
|
|
|
|
.validation_type = NLA_VALIDATE_RANGE_PTR, \
|
|
|
|
.range_signed = _range, \
|
|
|
|
}
|
|
|
|
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
#define NLA_POLICY_MIN(tp, _min) { \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp), \
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
.validation_type = NLA_VALIDATE_MIN, \
|
|
|
|
.min = _min, \
|
|
|
|
}
|
|
|
|
|
|
|
|
#define NLA_POLICY_MAX(tp, _max) { \
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp), \
|
netlink: add attribute range validation to policy
Without further bloating the policy structs, we can overload
the `validation_data' pointer with a struct of s16 min, max
and use those to validate ranges in NLA_{U,S}{8,16,32,64}
attributes.
It may sound strange to validate NLA_U32 with a s16 max, but
in many cases NLA_U32 is used for enums etc. since there's no
size benefit in using a smaller attribute width anyway, due
to netlink attribute alignment; in cases like that it's still
useful, particularly when the attribute really transports an
enum value.
Doing so lets us remove quite a bit of validation code, if we
can be sure that these attributes aren't used by userspace in
places where they're ignored today.
To achieve all this, split the 'type' field and introduce a
new 'validation_type' field which indicates what further
validation (beyond the validation prescribed by the type of
the attribute) is done. This currently allows for no further
validation (the default), as well as min, max and range checks.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-27 09:28:35 +00:00
|
|
|
.validation_type = NLA_VALIDATE_MAX, \
|
|
|
|
.max = _max, \
|
|
|
|
}
|
|
|
|
|
2020-10-05 22:07:38 +00:00
|
|
|
#define NLA_POLICY_MASK(tp, _mask) { \
|
|
|
|
.type = NLA_ENSURE_UINT_TYPE(tp), \
|
|
|
|
.validation_type = NLA_VALIDATE_MASK, \
|
|
|
|
.mask = _mask, \
|
|
|
|
}
|
|
|
|
|
2018-09-27 09:28:36 +00:00
|
|
|
#define NLA_POLICY_VALIDATE_FN(tp, fn, ...) { \
|
|
|
|
.type = NLA_ENSURE_NO_VALIDATION_PTR(tp), \
|
|
|
|
.validation_type = NLA_VALIDATE_FUNCTION, \
|
|
|
|
.validate = fn, \
|
|
|
|
.len = __VA_ARGS__ + 0, \
|
|
|
|
}
|
|
|
|
|
netlink: make NLA_BINARY validation more flexible
Add range validation for NLA_BINARY, allowing validation of any
combination of combination minimum or maximum lengths, using the
existing NLA_POLICY_RANGE()/NLA_POLICY_FULL_RANGE() macros, just
like for integers where the value is checked.
Also make NLA_POLICY_EXACT_LEN(), NLA_POLICY_EXACT_LEN_WARN()
and NLA_POLICY_MIN_LEN() special cases of this, removing the old
types NLA_EXACT_LEN and NLA_MIN_LEN.
This allows us to save some code where both minimum and maximum
lengths are requires, currently the policy only allows maximum
(NLA_BINARY), minimum (NLA_MIN_LEN) or exact (NLA_EXACT_LEN), so
a range of lengths cannot be accepted and must be checked by the
code that consumes the attributes later.
Also, this allows advertising the correct ranges in the policy
export to userspace. Here, NLA_MIN_LEN and NLA_EXACT_LEN already
were special cases of NLA_BINARY with min and min/max length
respectively.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-08-18 08:17:33 +00:00
|
|
|
#define NLA_POLICY_EXACT_LEN(_len) NLA_POLICY_RANGE(NLA_BINARY, _len, _len)
|
|
|
|
#define NLA_POLICY_EXACT_LEN_WARN(_len) { \
|
|
|
|
.type = NLA_BINARY, \
|
|
|
|
.validation_type = NLA_VALIDATE_RANGE_WARN_TOO_LONG, \
|
|
|
|
.min = _len, \
|
|
|
|
.max = _len \
|
|
|
|
}
|
|
|
|
#define NLA_POLICY_MIN_LEN(_len) NLA_POLICY_MIN(NLA_BINARY, _len)
|
|
|
|
|
2006-08-18 01:14:52 +00:00
|
|
|
/**
|
|
|
|
* struct nl_info - netlink source information
|
|
|
|
* @nlh: Netlink message header of original request
|
2019-06-18 15:12:43 +00:00
|
|
|
* @nl_net: Network namespace
|
2012-09-07 20:12:54 +00:00
|
|
|
* @portid: Netlink PORTID of requesting application
|
2019-06-18 15:12:43 +00:00
|
|
|
* @skip_notify: Skip netlink notifications to user space
|
2019-06-18 15:12:44 +00:00
|
|
|
* @skip_notify_kernel: Skip selected in-kernel notifications
|
2006-08-18 01:14:52 +00:00
|
|
|
*/
|
|
|
|
struct nl_info {
|
|
|
|
struct nlmsghdr *nlh;
|
2008-01-10 11:26:13 +00:00
|
|
|
struct net *nl_net;
|
2012-09-07 20:12:54 +00:00
|
|
|
u32 portid;
|
2019-06-18 15:12:44 +00:00
|
|
|
u8 skip_notify:1,
|
|
|
|
skip_notify_kernel:1;
|
2006-08-18 01:14:52 +00:00
|
|
|
};
|
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
/**
|
|
|
|
* enum netlink_validation - netlink message/attribute validation levels
|
|
|
|
* @NL_VALIDATE_LIBERAL: Old-style "be liberal" validation, not caring about
|
|
|
|
* extra data at the end of the message, attributes being longer than
|
|
|
|
* they should be, or unknown attributes being present.
|
|
|
|
* @NL_VALIDATE_TRAILING: Reject junk data encountered after attribute parsing.
|
|
|
|
* @NL_VALIDATE_MAXTYPE: Reject attributes > max type; Together with _TRAILING
|
|
|
|
* this is equivalent to the old nla_parse_strict()/nlmsg_parse_strict().
|
|
|
|
* @NL_VALIDATE_UNSPEC: Reject attributes with NLA_UNSPEC in the policy.
|
|
|
|
* This can safely be set by the kernel when the given policy has no
|
|
|
|
* NLA_UNSPEC anymore, and can thus be used to ensure policy entries
|
|
|
|
* are enforced going forward.
|
|
|
|
* @NL_VALIDATE_STRICT_ATTRS: strict attribute policy parsing (e.g.
|
|
|
|
* U8, U16, U32 must have exact size, etc.)
|
2019-05-02 14:15:10 +00:00
|
|
|
* @NL_VALIDATE_NESTED: Check that NLA_F_NESTED is set for NLA_NESTED(_ARRAY)
|
|
|
|
* and unset for other policies.
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
*/
|
|
|
|
enum netlink_validation {
|
|
|
|
NL_VALIDATE_LIBERAL = 0,
|
|
|
|
NL_VALIDATE_TRAILING = BIT(0),
|
|
|
|
NL_VALIDATE_MAXTYPE = BIT(1),
|
|
|
|
NL_VALIDATE_UNSPEC = BIT(2),
|
|
|
|
NL_VALIDATE_STRICT_ATTRS = BIT(3),
|
2019-05-02 14:15:10 +00:00
|
|
|
NL_VALIDATE_NESTED = BIT(4),
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
};
|
|
|
|
|
|
|
|
#define NL_VALIDATE_DEPRECATED_STRICT (NL_VALIDATE_TRAILING |\
|
|
|
|
NL_VALIDATE_MAXTYPE)
|
|
|
|
#define NL_VALIDATE_STRICT (NL_VALIDATE_TRAILING |\
|
|
|
|
NL_VALIDATE_MAXTYPE |\
|
|
|
|
NL_VALIDATE_UNSPEC |\
|
2019-05-02 14:15:10 +00:00
|
|
|
NL_VALIDATE_STRICT_ATTRS |\
|
|
|
|
NL_VALIDATE_NESTED)
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
|
2013-09-21 17:22:50 +00:00
|
|
|
int netlink_rcv_skb(struct sk_buff *skb,
|
2017-04-12 12:34:04 +00:00
|
|
|
int (*cb)(struct sk_buff *, struct nlmsghdr *,
|
|
|
|
struct netlink_ext_ack *));
|
2013-09-21 17:22:50 +00:00
|
|
|
int nlmsg_notify(struct sock *sk, struct sk_buff *skb, u32 portid,
|
|
|
|
unsigned int group, int report, gfp_t flags);
|
2005-11-10 01:25:53 +00:00
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
int __nla_validate(const struct nlattr *head, int len, int maxtype,
|
|
|
|
const struct nla_policy *policy, unsigned int validate,
|
|
|
|
struct netlink_ext_ack *extack);
|
|
|
|
int __nla_parse(struct nlattr **tb, int maxtype, const struct nlattr *head,
|
|
|
|
int len, const struct nla_policy *policy, unsigned int validate,
|
|
|
|
struct netlink_ext_ack *extack);
|
2013-09-21 17:22:50 +00:00
|
|
|
int nla_policy_len(const struct nla_policy *, int);
|
|
|
|
struct nlattr *nla_find(const struct nlattr *head, int len, int attrtype);
|
2020-11-15 17:08:06 +00:00
|
|
|
ssize_t nla_strscpy(char *dst, const struct nlattr *nla, size_t dstsize);
|
2017-07-27 14:56:40 +00:00
|
|
|
char *nla_strdup(const struct nlattr *nla, gfp_t flags);
|
2013-09-21 17:22:50 +00:00
|
|
|
int nla_memcpy(void *dest, const struct nlattr *src, int count);
|
|
|
|
int nla_memcmp(const struct nlattr *nla, const void *data, size_t size);
|
|
|
|
int nla_strcmp(const struct nlattr *nla, const char *str);
|
|
|
|
struct nlattr *__nla_reserve(struct sk_buff *skb, int attrtype, int attrlen);
|
2016-04-21 16:58:24 +00:00
|
|
|
struct nlattr *__nla_reserve_64bit(struct sk_buff *skb, int attrtype,
|
|
|
|
int attrlen, int padattr);
|
2013-09-21 17:22:50 +00:00
|
|
|
void *__nla_reserve_nohdr(struct sk_buff *skb, int attrlen);
|
|
|
|
struct nlattr *nla_reserve(struct sk_buff *skb, int attrtype, int attrlen);
|
2016-04-21 16:58:24 +00:00
|
|
|
struct nlattr *nla_reserve_64bit(struct sk_buff *skb, int attrtype,
|
|
|
|
int attrlen, int padattr);
|
2013-09-21 17:22:50 +00:00
|
|
|
void *nla_reserve_nohdr(struct sk_buff *skb, int attrlen);
|
|
|
|
void __nla_put(struct sk_buff *skb, int attrtype, int attrlen,
|
|
|
|
const void *data);
|
2016-04-21 16:58:24 +00:00
|
|
|
void __nla_put_64bit(struct sk_buff *skb, int attrtype, int attrlen,
|
|
|
|
const void *data, int padattr);
|
2013-09-21 17:22:50 +00:00
|
|
|
void __nla_put_nohdr(struct sk_buff *skb, int attrlen, const void *data);
|
|
|
|
int nla_put(struct sk_buff *skb, int attrtype, int attrlen, const void *data);
|
2016-04-21 16:58:24 +00:00
|
|
|
int nla_put_64bit(struct sk_buff *skb, int attrtype, int attrlen,
|
|
|
|
const void *data, int padattr);
|
2013-09-21 17:22:50 +00:00
|
|
|
int nla_put_nohdr(struct sk_buff *skb, int attrlen, const void *data);
|
|
|
|
int nla_append(struct sk_buff *skb, int attrlen, const void *data);
|
2005-11-10 01:25:51 +00:00
|
|
|
|
|
|
|
/**************************************************************************
|
|
|
|
* Netlink Messages
|
|
|
|
**************************************************************************/
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_msg_size - length of netlink message not including padding
|
|
|
|
* @payload: length of message payload
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_msg_size(int payload)
|
|
|
|
{
|
|
|
|
return NLMSG_HDRLEN + payload;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_total_size - length of netlink message including padding
|
|
|
|
* @payload: length of message payload
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_total_size(int payload)
|
|
|
|
{
|
|
|
|
return NLMSG_ALIGN(nlmsg_msg_size(payload));
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_padlen - length of padding at the message's tail
|
|
|
|
* @payload: length of message payload
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_padlen(int payload)
|
|
|
|
{
|
|
|
|
return nlmsg_total_size(payload) - nlmsg_msg_size(payload);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_data - head of message payload
|
2011-05-10 08:16:21 +00:00
|
|
|
* @nlh: netlink message header
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
|
|
|
static inline void *nlmsg_data(const struct nlmsghdr *nlh)
|
|
|
|
{
|
|
|
|
return (unsigned char *) nlh + NLMSG_HDRLEN;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_len - length of message payload
|
|
|
|
* @nlh: netlink message header
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_len(const struct nlmsghdr *nlh)
|
|
|
|
{
|
|
|
|
return nlh->nlmsg_len - NLMSG_HDRLEN;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_attrdata - head of attributes data
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
*/
|
|
|
|
static inline struct nlattr *nlmsg_attrdata(const struct nlmsghdr *nlh,
|
|
|
|
int hdrlen)
|
|
|
|
{
|
|
|
|
unsigned char *data = nlmsg_data(nlh);
|
|
|
|
return (struct nlattr *) (data + NLMSG_ALIGN(hdrlen));
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_attrlen - length of attributes data
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_attrlen(const struct nlmsghdr *nlh, int hdrlen)
|
|
|
|
{
|
|
|
|
return nlmsg_len(nlh) - NLMSG_ALIGN(hdrlen);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_ok - check if the netlink message fits into the remaining bytes
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @remaining: number of bytes remaining in message stream
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_ok(const struct nlmsghdr *nlh, int remaining)
|
|
|
|
{
|
2008-12-26 01:21:17 +00:00
|
|
|
return (remaining >= (int) sizeof(struct nlmsghdr) &&
|
2005-11-10 01:25:51 +00:00
|
|
|
nlh->nlmsg_len >= sizeof(struct nlmsghdr) &&
|
|
|
|
nlh->nlmsg_len <= remaining);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_next - next netlink message in message stream
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @remaining: number of bytes remaining in message stream
|
|
|
|
*
|
|
|
|
* Returns the next netlink message in the message stream and
|
|
|
|
* decrements remaining by the size of the current message.
|
|
|
|
*/
|
2010-11-16 17:52:32 +00:00
|
|
|
static inline struct nlmsghdr *
|
|
|
|
nlmsg_next(const struct nlmsghdr *nlh, int *remaining)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
int totlen = NLMSG_ALIGN(nlh->nlmsg_len);
|
|
|
|
|
|
|
|
*remaining -= totlen;
|
|
|
|
|
|
|
|
return (struct nlmsghdr *) ((unsigned char *) nlh + totlen);
|
|
|
|
}
|
|
|
|
|
2019-04-26 12:07:29 +00:00
|
|
|
/**
|
|
|
|
* nla_parse - Parse a stream of attributes into a tb buffer
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK pointer
|
|
|
|
*
|
|
|
|
* Parses a stream of attributes and stores a pointer to each attribute in
|
|
|
|
* the tb array accessible via the attribute type. Attributes with a type
|
|
|
|
* exceeding maxtype will be rejected, policy must be specified, attributes
|
|
|
|
* will be validated in the strictest way possible.
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_parse(struct nlattr **tb, int maxtype,
|
|
|
|
const struct nlattr *head, int len,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_parse(tb, maxtype, head, len, policy,
|
|
|
|
NL_VALIDATE_STRICT, extack);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* nla_parse_deprecated - Parse a stream of attributes into a tb buffer
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK pointer
|
|
|
|
*
|
|
|
|
* Parses a stream of attributes and stores a pointer to each attribute in
|
|
|
|
* the tb array accessible via the attribute type. Attributes with a type
|
|
|
|
* exceeding maxtype will be ignored and attributes from the policy are not
|
|
|
|
* always strictly validated (only for new attributes).
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_parse_deprecated(struct nlattr **tb, int maxtype,
|
|
|
|
const struct nlattr *head, int len,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_parse(tb, maxtype, head, len, policy,
|
|
|
|
NL_VALIDATE_LIBERAL, extack);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_parse_deprecated_strict - Parse a stream of attributes into a tb buffer
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK pointer
|
|
|
|
*
|
|
|
|
* Parses a stream of attributes and stores a pointer to each attribute in
|
|
|
|
* the tb array accessible via the attribute type. Attributes with a type
|
|
|
|
* exceeding maxtype will be rejected as well as trailing data, but the
|
|
|
|
* policy is not completely strictly validated (only for new attributes).
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_parse_deprecated_strict(struct nlattr **tb, int maxtype,
|
|
|
|
const struct nlattr *head,
|
|
|
|
int len,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_parse(tb, maxtype, head, len, policy,
|
|
|
|
NL_VALIDATE_DEPRECATED_STRICT, extack);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* __nlmsg_parse - parse attributes of a netlink message
|
2005-11-10 01:25:51 +00:00
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @policy: validation policy
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* @validate: validation strictness
|
2017-04-12 12:34:07 +00:00
|
|
|
* @extack: extended ACK report struct
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
|
|
|
* See nla_parse()
|
|
|
|
*/
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
static inline int __nlmsg_parse(const struct nlmsghdr *nlh, int hdrlen,
|
|
|
|
struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
unsigned int validate,
|
|
|
|
struct netlink_ext_ack *extack)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2018-10-08 03:16:23 +00:00
|
|
|
if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen)) {
|
|
|
|
NL_SET_ERR_MSG(extack, "Invalid header length");
|
2005-11-10 01:25:51 +00:00
|
|
|
return -EINVAL;
|
2018-10-08 03:16:23 +00:00
|
|
|
}
|
2005-11-10 01:25:51 +00:00
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
return __nla_parse(tb, maxtype, nlmsg_attrdata(nlh, hdrlen),
|
|
|
|
nlmsg_attrlen(nlh, hdrlen), policy, validate,
|
|
|
|
extack);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2019-04-26 12:07:29 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_parse - parse attributes of a netlink message
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
2022-08-24 01:36:21 +00:00
|
|
|
* @policy: validation policy
|
2019-04-26 12:07:29 +00:00
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* See nla_parse()
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_parse(const struct nlmsghdr *nlh, int hdrlen,
|
|
|
|
struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
2019-08-12 20:07:07 +00:00
|
|
|
return __nlmsg_parse(nlh, hdrlen, tb, maxtype, policy,
|
|
|
|
NL_VALIDATE_STRICT, extack);
|
2019-04-26 12:07:29 +00:00
|
|
|
}
|
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_parse_deprecated - parse attributes of a netlink message
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
2022-08-24 01:36:21 +00:00
|
|
|
* @policy: validation policy
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* See nla_parse_deprecated()
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_parse_deprecated(const struct nlmsghdr *nlh, int hdrlen,
|
|
|
|
struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
2018-10-08 03:16:25 +00:00
|
|
|
{
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
return __nlmsg_parse(nlh, hdrlen, tb, maxtype, policy,
|
|
|
|
NL_VALIDATE_LIBERAL, extack);
|
|
|
|
}
|
2018-10-08 03:16:25 +00:00
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_parse_deprecated_strict - parse attributes of a netlink message
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of family specific header
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
2022-08-24 01:36:21 +00:00
|
|
|
* @policy: validation policy
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* See nla_parse_deprecated_strict()
|
|
|
|
*/
|
|
|
|
static inline int
|
|
|
|
nlmsg_parse_deprecated_strict(const struct nlmsghdr *nlh, int hdrlen,
|
|
|
|
struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nlmsg_parse(nlh, hdrlen, tb, maxtype, policy,
|
|
|
|
NL_VALIDATE_DEPRECATED_STRICT, extack);
|
2018-10-08 03:16:25 +00:00
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_find_attr - find a specific attribute in a netlink message
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of familiy specific header
|
|
|
|
* @attrtype: type of attribute to look for
|
|
|
|
*
|
|
|
|
* Returns the first attribute which matches the specified type.
|
|
|
|
*/
|
2010-11-03 16:35:40 +00:00
|
|
|
static inline struct nlattr *nlmsg_find_attr(const struct nlmsghdr *nlh,
|
2005-11-10 01:25:51 +00:00
|
|
|
int hdrlen, int attrtype)
|
|
|
|
{
|
|
|
|
return nla_find(nlmsg_attrdata(nlh, hdrlen),
|
|
|
|
nlmsg_attrlen(nlh, hdrlen), attrtype);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* nla_validate_deprecated - Validate a stream of attributes
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* Validates all attributes in the specified attribute stream against the
|
|
|
|
* specified policy. Validation is done in liberal mode.
|
|
|
|
* See documenation of struct nla_policy for more details.
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_validate_deprecated(const struct nlattr *head, int len,
|
|
|
|
int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_validate(head, len, maxtype, policy, NL_VALIDATE_LIBERAL,
|
|
|
|
extack);
|
|
|
|
}
|
|
|
|
|
2019-04-26 12:07:29 +00:00
|
|
|
/**
|
|
|
|
* nla_validate - Validate a stream of attributes
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* Validates all attributes in the specified attribute stream against the
|
|
|
|
* specified policy. Validation is done in strict mode.
|
|
|
|
* See documenation of struct nla_policy for more details.
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_validate(const struct nlattr *head, int len, int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_validate(head, len, maxtype, policy, NL_VALIDATE_STRICT,
|
|
|
|
extack);
|
|
|
|
}
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_validate_deprecated - validate a netlink message including attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* @nlh: netlinket message header
|
|
|
|
* @hdrlen: length of familiy specific header
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @policy: validation policy
|
2017-04-12 12:34:07 +00:00
|
|
|
* @extack: extended ACK report struct
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
static inline int nlmsg_validate_deprecated(const struct nlmsghdr *nlh,
|
|
|
|
int hdrlen, int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
if (nlh->nlmsg_len < nlmsg_msg_size(hdrlen))
|
|
|
|
return -EINVAL;
|
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
return __nla_validate(nlmsg_attrdata(nlh, hdrlen),
|
|
|
|
nlmsg_attrlen(nlh, hdrlen), maxtype,
|
|
|
|
policy, NL_VALIDATE_LIBERAL, extack);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
|
|
|
|
|
2006-08-15 07:31:41 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_report - need to report back to application?
|
|
|
|
* @nlh: netlink message header
|
|
|
|
*
|
|
|
|
* Returns 1 if a report back to the application is requested.
|
|
|
|
*/
|
2009-08-25 14:07:40 +00:00
|
|
|
static inline int nlmsg_report(const struct nlmsghdr *nlh)
|
2006-08-15 07:31:41 +00:00
|
|
|
{
|
2021-07-27 03:41:41 +00:00
|
|
|
return nlh ? !!(nlh->nlmsg_flags & NLM_F_ECHO) : 0;
|
2006-08-15 07:31:41 +00:00
|
|
|
}
|
|
|
|
|
2022-10-28 08:42:21 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_seq - return the seq number of netlink message
|
|
|
|
* @nlh: netlink message header
|
|
|
|
*
|
|
|
|
* Returns 0 if netlink message is NULL
|
|
|
|
*/
|
|
|
|
static inline u32 nlmsg_seq(const struct nlmsghdr *nlh)
|
|
|
|
{
|
|
|
|
return nlh ? nlh->nlmsg_seq : 0;
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_for_each_attr - iterate over a stream of attributes
|
|
|
|
* @pos: loop counter, set to current attribute
|
|
|
|
* @nlh: netlink message header
|
|
|
|
* @hdrlen: length of familiy specific header
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nlmsg_for_each_attr(pos, nlh, hdrlen, rem) \
|
|
|
|
nla_for_each_attr(pos, nlmsg_attrdata(nlh, hdrlen), \
|
|
|
|
nlmsg_attrlen(nlh, hdrlen), rem)
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_put - Add a new netlink message to an skb
|
|
|
|
* @skb: socket buffer to store message in
|
2014-10-16 13:47:51 +00:00
|
|
|
* @portid: netlink PORTID of requesting application
|
2005-11-10 01:25:51 +00:00
|
|
|
* @seq: sequence number of message
|
|
|
|
* @type: message type
|
|
|
|
* @payload: length of message payload
|
|
|
|
* @flags: message flags
|
|
|
|
*
|
|
|
|
* Returns NULL if the tailroom of the skb is insufficient to store
|
|
|
|
* the message header and payload.
|
|
|
|
*/
|
2012-09-07 20:12:54 +00:00
|
|
|
static inline struct nlmsghdr *nlmsg_put(struct sk_buff *skb, u32 portid, u32 seq,
|
2005-11-10 01:25:51 +00:00
|
|
|
int type, int payload, int flags)
|
|
|
|
{
|
|
|
|
if (unlikely(skb_tailroom(skb) < nlmsg_total_size(payload)))
|
|
|
|
return NULL;
|
|
|
|
|
2012-09-07 20:12:54 +00:00
|
|
|
return __nlmsg_put(skb, portid, seq, type, payload, flags);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2022-10-27 21:25:53 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_append - Add more data to a nlmsg in a skb
|
|
|
|
* @skb: socket buffer to store message in
|
|
|
|
* @size: length of message payload
|
|
|
|
*
|
|
|
|
* Append data to an existing nlmsg, used when constructing a message
|
|
|
|
* with multiple fixed-format headers (which is rare).
|
|
|
|
* Returns NULL if the tailroom of the skb is insufficient to store
|
|
|
|
* the extra payload.
|
|
|
|
*/
|
|
|
|
static inline void *nlmsg_append(struct sk_buff *skb, u32 size)
|
|
|
|
{
|
|
|
|
if (unlikely(skb_tailroom(skb) < NLMSG_ALIGN(size)))
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
if (NLMSG_ALIGN(size) - size)
|
|
|
|
memset(skb_tail_pointer(skb) + size, 0,
|
|
|
|
NLMSG_ALIGN(size) - size);
|
|
|
|
return __skb_put(skb, NLMSG_ALIGN(size));
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_put_answer - Add a new callback based netlink message to an skb
|
|
|
|
* @skb: socket buffer to store message in
|
|
|
|
* @cb: netlink callback
|
|
|
|
* @type: message type
|
|
|
|
* @payload: length of message payload
|
|
|
|
* @flags: message flags
|
|
|
|
*
|
|
|
|
* Returns NULL if the tailroom of the skb is insufficient to store
|
|
|
|
* the message header and payload.
|
|
|
|
*/
|
|
|
|
static inline struct nlmsghdr *nlmsg_put_answer(struct sk_buff *skb,
|
|
|
|
struct netlink_callback *cb,
|
|
|
|
int type, int payload,
|
|
|
|
int flags)
|
|
|
|
{
|
2012-09-07 20:12:54 +00:00
|
|
|
return nlmsg_put(skb, NETLINK_CB(cb->skb).portid, cb->nlh->nlmsg_seq,
|
2005-11-10 01:25:51 +00:00
|
|
|
type, payload, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_new - Allocate a new netlink message
|
2006-11-10 22:10:15 +00:00
|
|
|
* @payload: size of the message payload
|
2006-08-05 06:03:05 +00:00
|
|
|
* @flags: the type of memory to allocate.
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
2006-11-10 22:10:15 +00:00
|
|
|
* Use NLMSG_DEFAULT_SIZE if the size of the payload isn't known
|
|
|
|
* and a good default is needed.
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2006-11-10 22:10:15 +00:00
|
|
|
static inline struct sk_buff *nlmsg_new(size_t payload, gfp_t flags)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2006-11-10 22:10:15 +00:00
|
|
|
return alloc_skb(nlmsg_total_size(payload), flags);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2023-11-15 12:01:08 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_new_large - Allocate a new netlink message with non-contiguous
|
|
|
|
* physical memory
|
|
|
|
* @payload: size of the message payload
|
|
|
|
*
|
|
|
|
* The allocated skb is unable to have frag page for shinfo->frags*,
|
|
|
|
* as the NULL setting for skb->head in netlink_skb_destructor() will
|
|
|
|
* bypass most of the handling in skb_release_data()
|
|
|
|
*/
|
|
|
|
static inline struct sk_buff *nlmsg_new_large(size_t payload)
|
|
|
|
{
|
|
|
|
return netlink_alloc_large_skb(nlmsg_total_size(payload), 0);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_end - Finalize a netlink message
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
* @nlh: netlink message header
|
|
|
|
*
|
|
|
|
* Corrects the netlink message header to include the appeneded
|
|
|
|
* attributes. Only necessary if attributes have been added to
|
|
|
|
* the message.
|
|
|
|
*/
|
2015-01-16 21:09:00 +00:00
|
|
|
static inline void nlmsg_end(struct sk_buff *skb, struct nlmsghdr *nlh)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2007-04-20 03:29:13 +00:00
|
|
|
nlh->nlmsg_len = skb_tail_pointer(skb) - (unsigned char *)nlh;
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2006-08-05 06:03:05 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_get_pos - return current position in netlink message
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
*
|
|
|
|
* Returns a pointer to the current tail of the message.
|
|
|
|
*/
|
|
|
|
static inline void *nlmsg_get_pos(struct sk_buff *skb)
|
|
|
|
{
|
2007-04-20 03:29:13 +00:00
|
|
|
return skb_tail_pointer(skb);
|
2006-08-05 06:03:05 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_trim - Trim message to a mark
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
* @mark: mark to trim to
|
|
|
|
*
|
2008-06-03 23:36:54 +00:00
|
|
|
* Trims the message to the provided mark.
|
2006-08-05 06:03:05 +00:00
|
|
|
*/
|
2008-06-03 23:36:54 +00:00
|
|
|
static inline void nlmsg_trim(struct sk_buff *skb, const void *mark)
|
2006-08-05 06:03:05 +00:00
|
|
|
{
|
2015-01-06 00:04:21 +00:00
|
|
|
if (mark) {
|
|
|
|
WARN_ON((unsigned char *) mark < skb->data);
|
2006-08-05 06:03:05 +00:00
|
|
|
skb_trim(skb, (unsigned char *) mark - skb->data);
|
2015-01-06 00:04:21 +00:00
|
|
|
}
|
2006-08-05 06:03:05 +00:00
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_cancel - Cancel construction of a netlink message
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
* @nlh: netlink message header
|
|
|
|
*
|
|
|
|
* Removes the complete netlink message including all
|
2008-06-03 23:36:54 +00:00
|
|
|
* attributes from the socket buffer again.
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2008-06-03 23:36:54 +00:00
|
|
|
static inline void nlmsg_cancel(struct sk_buff *skb, struct nlmsghdr *nlh)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2008-06-03 23:36:54 +00:00
|
|
|
nlmsg_trim(skb, nlh);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2024-04-03 20:22:59 +00:00
|
|
|
* nlmsg_free - drop a netlink message
|
2005-11-10 01:25:51 +00:00
|
|
|
* @skb: socket buffer of netlink message
|
|
|
|
*/
|
|
|
|
static inline void nlmsg_free(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
kfree_skb(skb);
|
|
|
|
}
|
|
|
|
|
2024-04-03 20:22:59 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_consume - free a netlink message
|
|
|
|
* @skb: socket buffer of netlink message
|
|
|
|
*/
|
|
|
|
static inline void nlmsg_consume(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
consume_skb(skb);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
2023-12-16 12:29:59 +00:00
|
|
|
* nlmsg_multicast_filtered - multicast a netlink message with filter function
|
2005-11-10 01:25:51 +00:00
|
|
|
* @sk: netlink socket to spread messages to
|
|
|
|
* @skb: netlink message as socket buffer
|
2012-09-07 20:12:54 +00:00
|
|
|
* @portid: own netlink portid to avoid sending to yourself
|
2005-11-10 01:25:51 +00:00
|
|
|
* @group: multicast group id
|
2006-08-15 07:31:06 +00:00
|
|
|
* @flags: allocation flags
|
2023-12-16 12:29:59 +00:00
|
|
|
* @filter: filter function
|
|
|
|
* @filter_data: filter function private data
|
|
|
|
*
|
|
|
|
* Return: 0 on success, negative error code for failure.
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2023-12-16 12:29:59 +00:00
|
|
|
static inline int nlmsg_multicast_filtered(struct sock *sk, struct sk_buff *skb,
|
|
|
|
u32 portid, unsigned int group,
|
|
|
|
gfp_t flags,
|
|
|
|
netlink_filter_fn filter,
|
|
|
|
void *filter_data)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
|
|
|
NETLINK_CB(skb).dst_group = group;
|
|
|
|
|
2023-12-16 12:29:59 +00:00
|
|
|
err = netlink_broadcast_filtered(sk, skb, portid, group, flags,
|
|
|
|
filter, filter_data);
|
2005-11-10 01:25:51 +00:00
|
|
|
if (err > 0)
|
|
|
|
err = 0;
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
2023-12-16 12:29:59 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_multicast - multicast a netlink message
|
|
|
|
* @sk: netlink socket to spread messages to
|
|
|
|
* @skb: netlink message as socket buffer
|
|
|
|
* @portid: own netlink portid to avoid sending to yourself
|
|
|
|
* @group: multicast group id
|
|
|
|
* @flags: allocation flags
|
|
|
|
*/
|
|
|
|
static inline int nlmsg_multicast(struct sock *sk, struct sk_buff *skb,
|
|
|
|
u32 portid, unsigned int group, gfp_t flags)
|
|
|
|
{
|
|
|
|
return nlmsg_multicast_filtered(sk, skb, portid, group, flags,
|
|
|
|
NULL, NULL);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nlmsg_unicast - unicast a netlink message
|
|
|
|
* @sk: netlink socket to spread message to
|
|
|
|
* @skb: netlink message as socket buffer
|
2012-09-07 20:12:54 +00:00
|
|
|
* @portid: netlink portid of the destination socket
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2012-09-07 20:12:54 +00:00
|
|
|
static inline int nlmsg_unicast(struct sock *sk, struct sk_buff *skb, u32 portid)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
int err;
|
|
|
|
|
2012-09-07 20:12:54 +00:00
|
|
|
err = netlink_unicast(sk, skb, portid, MSG_DONTWAIT);
|
2005-11-10 01:25:51 +00:00
|
|
|
if (err > 0)
|
|
|
|
err = 0;
|
|
|
|
|
|
|
|
return err;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nlmsg_for_each_msg - iterate over a stream of messages
|
|
|
|
* @pos: loop counter, set to current message
|
|
|
|
* @head: head of message stream
|
|
|
|
* @len: length of message stream
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nlmsg_for_each_msg(pos, head, len, rem) \
|
|
|
|
for (pos = head, rem = len; \
|
|
|
|
nlmsg_ok(pos, rem); \
|
|
|
|
pos = nlmsg_next(pos, &(rem)))
|
|
|
|
|
netlink: advertise incomplete dumps
Consider the following situation:
* a dump that would show 8 entries, four in the first
round, and four in the second
* between the first and second rounds, 6 entries are
removed
* now the second round will not show any entry, and
even if there is a sequence/generation counter the
application will not know
To solve this problem, add a new flag NLM_F_DUMP_INTR
to the netlink header that indicates the dump wasn't
consistent, this flag can also be set on the MSG_DONE
message that terminates the dump, and as such above
situation can be detected.
To achieve this, add a sequence counter to the netlink
callback struct. Of course, netlink code still needs
to use this new functionality. The correct way to do
that is to always set cb->seq when a dumpit callback
is invoked and call nl_dump_check_consistent() for
each new message. The core code will also call this
function for the final MSG_DONE message.
To make it usable with generic netlink, a new function
genlmsg_nlhdr() is needed to obtain the netlink header
from the genetlink user header.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: John W. Linville <linville@tuxdriver.com>
2011-06-20 11:40:46 +00:00
|
|
|
/**
|
|
|
|
* nl_dump_check_consistent - check if sequence is consistent and advertise if not
|
|
|
|
* @cb: netlink callback structure that stores the sequence number
|
|
|
|
* @nlh: netlink message header to write the flag to
|
|
|
|
*
|
|
|
|
* This function checks if the sequence (generation) number changed during dump
|
|
|
|
* and if it did, advertises it in the netlink message header.
|
|
|
|
*
|
|
|
|
* The correct way to use it is to set cb->seq to the generation counter when
|
|
|
|
* all locks for dumping have been acquired, and then call this function for
|
|
|
|
* each message that is generated.
|
|
|
|
*
|
|
|
|
* Note that due to initialisation concerns, 0 is an invalid sequence number
|
|
|
|
* and must not be used by code that uses this functionality.
|
|
|
|
*/
|
|
|
|
static inline void
|
|
|
|
nl_dump_check_consistent(struct netlink_callback *cb,
|
|
|
|
struct nlmsghdr *nlh)
|
|
|
|
{
|
|
|
|
if (cb->prev_seq && cb->seq != cb->prev_seq)
|
|
|
|
nlh->nlmsg_flags |= NLM_F_DUMP_INTR;
|
|
|
|
cb->prev_seq = cb->seq;
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**************************************************************************
|
|
|
|
* Netlink Attributes
|
|
|
|
**************************************************************************/
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_attr_size - length of attribute not including padding
|
|
|
|
* @payload: length of payload
|
|
|
|
*/
|
|
|
|
static inline int nla_attr_size(int payload)
|
|
|
|
{
|
|
|
|
return NLA_HDRLEN + payload;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_total_size - total length of attribute including padding
|
|
|
|
* @payload: length of payload
|
|
|
|
*/
|
|
|
|
static inline int nla_total_size(int payload)
|
|
|
|
{
|
|
|
|
return NLA_ALIGN(nla_attr_size(payload));
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_padlen - length of padding at the tail of attribute
|
|
|
|
* @payload: length of payload
|
|
|
|
*/
|
|
|
|
static inline int nla_padlen(int payload)
|
|
|
|
{
|
|
|
|
return nla_total_size(payload) - nla_attr_size(payload);
|
|
|
|
}
|
|
|
|
|
2007-09-12 12:44:36 +00:00
|
|
|
/**
|
|
|
|
* nla_type - attribute type
|
|
|
|
* @nla: netlink attribute
|
|
|
|
*/
|
|
|
|
static inline int nla_type(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return nla->nla_type & NLA_TYPE_MASK;
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_data - head of payload
|
|
|
|
* @nla: netlink attribute
|
|
|
|
*/
|
|
|
|
static inline void *nla_data(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return (char *) nla + NLA_HDRLEN;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_len - length of payload
|
|
|
|
* @nla: netlink attribute
|
|
|
|
*/
|
2023-12-06 20:59:07 +00:00
|
|
|
static inline u16 nla_len(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
return nla->nla_len - NLA_HDRLEN;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_ok - check if the netlink attribute fits into the remaining bytes
|
|
|
|
* @nla: netlink attribute
|
|
|
|
* @remaining: number of bytes remaining in attribute stream
|
|
|
|
*/
|
|
|
|
static inline int nla_ok(const struct nlattr *nla, int remaining)
|
|
|
|
{
|
2016-12-13 19:30:15 +00:00
|
|
|
return remaining >= (int) sizeof(*nla) &&
|
|
|
|
nla->nla_len >= sizeof(*nla) &&
|
2005-11-10 01:25:51 +00:00
|
|
|
nla->nla_len <= remaining;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2007-10-11 04:09:48 +00:00
|
|
|
* nla_next - next netlink attribute in attribute stream
|
2005-11-10 01:25:51 +00:00
|
|
|
* @nla: netlink attribute
|
|
|
|
* @remaining: number of bytes remaining in attribute stream
|
|
|
|
*
|
|
|
|
* Returns the next netlink attribute in the attribute stream and
|
|
|
|
* decrements remaining by the size of the current attribute.
|
|
|
|
*/
|
|
|
|
static inline struct nlattr *nla_next(const struct nlattr *nla, int *remaining)
|
|
|
|
{
|
2016-11-19 00:54:35 +00:00
|
|
|
unsigned int totlen = NLA_ALIGN(nla->nla_len);
|
2005-11-10 01:25:51 +00:00
|
|
|
|
|
|
|
*remaining -= totlen;
|
|
|
|
return (struct nlattr *) ((char *) nla + totlen);
|
|
|
|
}
|
|
|
|
|
2006-08-05 06:03:05 +00:00
|
|
|
/**
|
|
|
|
* nla_find_nested - find attribute in a set of nested attributes
|
|
|
|
* @nla: attribute containing the nested attributes
|
|
|
|
* @attrtype: type of attribute to look for
|
|
|
|
*
|
|
|
|
* Returns the first attribute which matches the specified type.
|
|
|
|
*/
|
2010-11-16 17:52:32 +00:00
|
|
|
static inline struct nlattr *
|
|
|
|
nla_find_nested(const struct nlattr *nla, int attrtype)
|
2006-08-05 06:03:05 +00:00
|
|
|
{
|
|
|
|
return nla_find(nla_data(nla), nla_len(nla), attrtype);
|
|
|
|
}
|
|
|
|
|
2019-04-26 12:07:29 +00:00
|
|
|
/**
|
|
|
|
* nla_parse_nested - parse nested attributes
|
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @nla: attribute containing the nested attributes
|
|
|
|
* @policy: validation policy
|
|
|
|
* @extack: extended ACK report struct
|
|
|
|
*
|
|
|
|
* See nla_parse()
|
|
|
|
*/
|
|
|
|
static inline int nla_parse_nested(struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nlattr *nla,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
2019-05-02 14:15:10 +00:00
|
|
|
if (!(nla->nla_type & NLA_F_NESTED)) {
|
|
|
|
NL_SET_ERR_MSG_ATTR(extack, nla, "NLA_F_NESTED is missing");
|
|
|
|
return -EINVAL;
|
|
|
|
}
|
|
|
|
|
2019-04-26 12:07:29 +00:00
|
|
|
return __nla_parse(tb, maxtype, nla_data(nla), nla_len(nla), policy,
|
|
|
|
NL_VALIDATE_STRICT, extack);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* nla_parse_nested_deprecated - parse nested attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* @tb: destination array with maxtype+1 elements
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @nla: attribute containing the nested attributes
|
|
|
|
* @policy: validation policy
|
2017-04-12 12:34:07 +00:00
|
|
|
* @extack: extended ACK report struct
|
2005-11-10 01:25:51 +00:00
|
|
|
*
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* See nla_parse_deprecated()
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
static inline int nla_parse_nested_deprecated(struct nlattr *tb[], int maxtype,
|
|
|
|
const struct nlattr *nla,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
return __nla_parse(tb, maxtype, nla_data(nla), nla_len(nla), policy,
|
|
|
|
NL_VALIDATE_LIBERAL, extack);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
2007-06-25 20:49:35 +00:00
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
2007-10-11 04:09:48 +00:00
|
|
|
* nla_put_u8 - Add a u8 netlink attribute to a socket buffer
|
2005-11-10 01:25:51 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_u8(struct sk_buff *skb, int attrtype, u8 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
/* temporary variables to work around GCC PR81715 with asan-stack=1 */
|
|
|
|
u8 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(u8), &tmp);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_put_u16 - Add a u16 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_u16(struct sk_buff *skb, int attrtype, u16 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
u16 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(u16), &tmp);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2012-03-30 03:18:53 +00:00
|
|
|
/**
|
|
|
|
* nla_put_be16 - Add a __be16 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_be16(struct sk_buff *skb, int attrtype, __be16 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be16 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(__be16), &tmp);
|
2012-03-30 03:18:53 +00:00
|
|
|
}
|
|
|
|
|
2012-04-01 23:11:31 +00:00
|
|
|
/**
|
|
|
|
* nla_put_net16 - Add 16-bit network byte order netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_net16(struct sk_buff *skb, int attrtype, __be16 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be16 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_be16(skb, attrtype | NLA_F_NET_BYTEORDER, tmp);
|
2012-04-01 23:11:31 +00:00
|
|
|
}
|
|
|
|
|
2012-04-02 00:14:27 +00:00
|
|
|
/**
|
|
|
|
* nla_put_le16 - Add a __le16 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_le16(struct sk_buff *skb, int attrtype, __le16 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__le16 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(__le16), &tmp);
|
2012-04-02 00:14:27 +00:00
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_put_u32 - Add a u32 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_u32(struct sk_buff *skb, int attrtype, u32 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
u32 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(u32), &tmp);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
/**
|
|
|
|
* nla_put_uint - Add a variable-size unsigned int to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_uint(struct sk_buff *skb, int attrtype, u64 value)
|
|
|
|
{
|
|
|
|
u64 tmp64 = value;
|
|
|
|
u32 tmp32 = value;
|
|
|
|
|
|
|
|
if (tmp64 == tmp32)
|
|
|
|
return nla_put_u32(skb, attrtype, tmp32);
|
|
|
|
return nla_put(skb, attrtype, sizeof(u64), &tmp64);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
2012-03-30 03:18:53 +00:00
|
|
|
* nla_put_be32 - Add a __be32 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_be32(struct sk_buff *skb, int attrtype, __be32 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be32 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(__be32), &tmp);
|
2012-03-30 03:18:53 +00:00
|
|
|
}
|
|
|
|
|
2012-04-01 23:11:31 +00:00
|
|
|
/**
|
|
|
|
* nla_put_net32 - Add 32-bit network byte order netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_net32(struct sk_buff *skb, int attrtype, __be32 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be32 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_be32(skb, attrtype | NLA_F_NET_BYTEORDER, tmp);
|
2012-04-01 23:11:31 +00:00
|
|
|
}
|
|
|
|
|
2012-04-02 00:14:27 +00:00
|
|
|
/**
|
|
|
|
* nla_put_le32 - Add a __le32 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_le32(struct sk_buff *skb, int attrtype, __le32 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__le32 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(__le32), &tmp);
|
2012-04-02 00:14:27 +00:00
|
|
|
}
|
|
|
|
|
2016-04-22 15:31:22 +00:00
|
|
|
/**
|
|
|
|
* nla_put_u64_64bit - Add a u64 netlink attribute to a skb and align it
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
* @padattr: attribute type for the padding
|
|
|
|
*/
|
|
|
|
static inline int nla_put_u64_64bit(struct sk_buff *skb, int attrtype,
|
|
|
|
u64 value, int padattr)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
u64 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_64bit(skb, attrtype, sizeof(u64), &tmp, padattr);
|
2016-04-22 15:31:22 +00:00
|
|
|
}
|
|
|
|
|
2012-03-30 03:18:53 +00:00
|
|
|
/**
|
2016-04-22 15:31:18 +00:00
|
|
|
* nla_put_be64 - Add a __be64 netlink attribute to a socket buffer and align it
|
2012-03-30 03:18:53 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
2016-04-22 15:31:18 +00:00
|
|
|
* @padattr: attribute type for the padding
|
2012-03-30 03:18:53 +00:00
|
|
|
*/
|
2016-04-22 15:31:18 +00:00
|
|
|
static inline int nla_put_be64(struct sk_buff *skb, int attrtype, __be64 value,
|
|
|
|
int padattr)
|
2012-03-30 03:18:53 +00:00
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be64 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_64bit(skb, attrtype, sizeof(__be64), &tmp, padattr);
|
2012-03-30 03:18:53 +00:00
|
|
|
}
|
|
|
|
|
2012-04-01 23:11:31 +00:00
|
|
|
/**
|
2016-04-22 15:31:19 +00:00
|
|
|
* nla_put_net64 - Add 64-bit network byte order nlattr to a skb and align it
|
2012-04-01 23:11:31 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
2016-04-22 15:31:19 +00:00
|
|
|
* @padattr: attribute type for the padding
|
2012-04-01 23:11:31 +00:00
|
|
|
*/
|
2016-04-22 15:31:19 +00:00
|
|
|
static inline int nla_put_net64(struct sk_buff *skb, int attrtype, __be64 value,
|
|
|
|
int padattr)
|
2012-04-01 23:11:31 +00:00
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be64 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_be64(skb, attrtype | NLA_F_NET_BYTEORDER, tmp,
|
2016-04-22 15:31:19 +00:00
|
|
|
padattr);
|
2012-04-01 23:11:31 +00:00
|
|
|
}
|
|
|
|
|
2012-04-02 00:14:27 +00:00
|
|
|
/**
|
2016-04-22 15:31:17 +00:00
|
|
|
* nla_put_le64 - Add a __le64 netlink attribute to a socket buffer and align it
|
2012-04-02 00:14:27 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
2016-04-22 15:31:17 +00:00
|
|
|
* @padattr: attribute type for the padding
|
2012-04-02 00:14:27 +00:00
|
|
|
*/
|
2016-04-22 15:31:17 +00:00
|
|
|
static inline int nla_put_le64(struct sk_buff *skb, int attrtype, __le64 value,
|
|
|
|
int padattr)
|
2012-04-02 00:14:27 +00:00
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__le64 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_64bit(skb, attrtype, sizeof(__le64), &tmp, padattr);
|
2012-04-02 00:14:27 +00:00
|
|
|
}
|
|
|
|
|
2012-07-27 06:28:52 +00:00
|
|
|
/**
|
|
|
|
* nla_put_s8 - Add a s8 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_s8(struct sk_buff *skb, int attrtype, s8 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
s8 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(s8), &tmp);
|
2012-07-27 06:28:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_put_s16 - Add a s16 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_s16(struct sk_buff *skb, int attrtype, s16 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
s16 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(s16), &tmp);
|
2012-07-27 06:28:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_put_s32 - Add a s32 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_s32(struct sk_buff *skb, int attrtype, s32 value)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
s32 tmp = value;
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(s32), &tmp);
|
2012-07-27 06:28:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2016-04-22 15:31:20 +00:00
|
|
|
* nla_put_s64 - Add a s64 netlink attribute to a socket buffer and align it
|
2012-07-27 06:28:52 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
2016-04-22 15:31:20 +00:00
|
|
|
* @padattr: attribute type for the padding
|
2012-07-27 06:28:52 +00:00
|
|
|
*/
|
2016-04-22 15:31:20 +00:00
|
|
|
static inline int nla_put_s64(struct sk_buff *skb, int attrtype, s64 value,
|
|
|
|
int padattr)
|
2012-07-27 06:28:52 +00:00
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
s64 tmp = value;
|
|
|
|
|
|
|
|
return nla_put_64bit(skb, attrtype, sizeof(s64), &tmp, padattr);
|
2012-07-27 06:28:52 +00:00
|
|
|
}
|
|
|
|
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
/**
|
|
|
|
* nla_put_sint - Add a variable-size signed int to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: numeric value
|
|
|
|
*/
|
|
|
|
static inline int nla_put_sint(struct sk_buff *skb, int attrtype, s64 value)
|
|
|
|
{
|
|
|
|
s64 tmp64 = value;
|
|
|
|
s32 tmp32 = value;
|
|
|
|
|
|
|
|
if (tmp64 == tmp32)
|
|
|
|
return nla_put_s32(skb, attrtype, tmp32);
|
|
|
|
return nla_put(skb, attrtype, sizeof(s64), &tmp64);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_put_string - Add a string netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @str: NUL terminated string
|
|
|
|
*/
|
|
|
|
static inline int nla_put_string(struct sk_buff *skb, int attrtype,
|
|
|
|
const char *str)
|
|
|
|
{
|
|
|
|
return nla_put(skb, attrtype, strlen(str) + 1, str);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_put_flag - Add a flag netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
*/
|
|
|
|
static inline int nla_put_flag(struct sk_buff *skb, int attrtype)
|
|
|
|
{
|
|
|
|
return nla_put(skb, attrtype, 0, NULL);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
2016-04-22 15:31:21 +00:00
|
|
|
* nla_put_msecs - Add a msecs netlink attribute to a skb and align it
|
2005-11-10 01:25:51 +00:00
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
2014-07-25 09:18:17 +00:00
|
|
|
* @njiffies: number of jiffies to convert to msecs
|
2016-04-22 15:31:21 +00:00
|
|
|
* @padattr: attribute type for the padding
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
|
|
|
static inline int nla_put_msecs(struct sk_buff *skb, int attrtype,
|
2016-04-22 15:31:21 +00:00
|
|
|
unsigned long njiffies, int padattr)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2014-07-25 09:18:17 +00:00
|
|
|
u64 tmp = jiffies_to_msecs(njiffies);
|
2016-04-22 15:31:21 +00:00
|
|
|
|
|
|
|
return nla_put_64bit(skb, attrtype, sizeof(u64), &tmp, padattr);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2015-03-29 14:59:25 +00:00
|
|
|
/**
|
|
|
|
* nla_put_in_addr - Add an IPv4 address netlink attribute to a socket
|
|
|
|
* buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @addr: IPv4 address
|
|
|
|
*/
|
|
|
|
static inline int nla_put_in_addr(struct sk_buff *skb, int attrtype,
|
|
|
|
__be32 addr)
|
|
|
|
{
|
2017-09-22 21:29:19 +00:00
|
|
|
__be32 tmp = addr;
|
|
|
|
|
|
|
|
return nla_put_be32(skb, attrtype, tmp);
|
2015-03-29 14:59:25 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_put_in6_addr - Add an IPv6 address netlink attribute to a socket
|
|
|
|
* buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @addr: IPv6 address
|
|
|
|
*/
|
|
|
|
static inline int nla_put_in6_addr(struct sk_buff *skb, int attrtype,
|
|
|
|
const struct in6_addr *addr)
|
|
|
|
{
|
|
|
|
return nla_put(skb, attrtype, sizeof(*addr), addr);
|
|
|
|
}
|
|
|
|
|
2020-03-28 15:37:42 +00:00
|
|
|
/**
|
|
|
|
* nla_put_bitfield32 - Add a bitfield32 netlink attribute to a socket buffer
|
|
|
|
* @skb: socket buffer to add attribute to
|
|
|
|
* @attrtype: attribute type
|
|
|
|
* @value: value carrying bits
|
|
|
|
* @selector: selector of valid bits
|
|
|
|
*/
|
|
|
|
static inline int nla_put_bitfield32(struct sk_buff *skb, int attrtype,
|
|
|
|
__u32 value, __u32 selector)
|
|
|
|
{
|
|
|
|
struct nla_bitfield32 tmp = { value, selector, };
|
|
|
|
|
|
|
|
return nla_put(skb, attrtype, sizeof(tmp), &tmp);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_get_u32 - return payload of u32 attribute
|
|
|
|
* @nla: u32 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline u32 nla_get_u32(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
return *(u32 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2006-09-27 05:14:41 +00:00
|
|
|
/**
|
|
|
|
* nla_get_be32 - return payload of __be32 attribute
|
|
|
|
* @nla: __be32 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline __be32 nla_get_be32(const struct nlattr *nla)
|
2006-09-27 05:14:41 +00:00
|
|
|
{
|
|
|
|
return *(__be32 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2015-09-28 07:00:23 +00:00
|
|
|
/**
|
|
|
|
* nla_get_le32 - return payload of __le32 attribute
|
|
|
|
* @nla: __le32 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline __le32 nla_get_le32(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(__le32 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_get_u16 - return payload of u16 attribute
|
|
|
|
* @nla: u16 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline u16 nla_get_u16(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
return *(u16 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2007-12-18 06:29:26 +00:00
|
|
|
/**
|
|
|
|
* nla_get_be16 - return payload of __be16 attribute
|
|
|
|
* @nla: __be16 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline __be16 nla_get_be16(const struct nlattr *nla)
|
2007-12-18 06:29:26 +00:00
|
|
|
{
|
|
|
|
return *(__be16 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2006-11-25 01:14:51 +00:00
|
|
|
/**
|
|
|
|
* nla_get_le16 - return payload of __le16 attribute
|
|
|
|
* @nla: __le16 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline __le16 nla_get_le16(const struct nlattr *nla)
|
2006-11-25 01:14:51 +00:00
|
|
|
{
|
|
|
|
return *(__le16 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_get_u8 - return payload of u8 attribute
|
|
|
|
* @nla: u8 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline u8 nla_get_u8(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
return *(u8 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_u64 - return payload of u64 attribute
|
|
|
|
* @nla: u64 netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline u64 nla_get_u64(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
u64 tmp;
|
|
|
|
|
|
|
|
nla_memcpy(&tmp, nla, sizeof(tmp));
|
|
|
|
|
|
|
|
return tmp;
|
|
|
|
}
|
|
|
|
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
/**
|
|
|
|
* nla_get_uint - return payload of uint attribute
|
|
|
|
* @nla: uint netlink attribute
|
|
|
|
*/
|
|
|
|
static inline u64 nla_get_uint(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
if (nla_len(nla) == sizeof(u32))
|
|
|
|
return nla_get_u32(nla);
|
|
|
|
return nla_get_u64(nla);
|
|
|
|
}
|
|
|
|
|
2009-05-27 15:50:35 +00:00
|
|
|
/**
|
|
|
|
* nla_get_be64 - return payload of __be64 attribute
|
|
|
|
* @nla: __be64 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline __be64 nla_get_be64(const struct nlattr *nla)
|
|
|
|
{
|
2010-03-16 13:30:44 +00:00
|
|
|
__be64 tmp;
|
|
|
|
|
|
|
|
nla_memcpy(&tmp, nla, sizeof(tmp));
|
|
|
|
|
2012-07-27 06:28:52 +00:00
|
|
|
return tmp;
|
|
|
|
}
|
|
|
|
|
2015-09-28 07:00:23 +00:00
|
|
|
/**
|
|
|
|
* nla_get_le64 - return payload of __le64 attribute
|
|
|
|
* @nla: __le64 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline __le64 nla_get_le64(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(__le64 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
2012-07-27 06:28:52 +00:00
|
|
|
/**
|
|
|
|
* nla_get_s32 - return payload of s32 attribute
|
|
|
|
* @nla: s32 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline s32 nla_get_s32(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(s32 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_s16 - return payload of s16 attribute
|
|
|
|
* @nla: s16 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline s16 nla_get_s16(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(s16 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_s8 - return payload of s8 attribute
|
|
|
|
* @nla: s8 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline s8 nla_get_s8(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(s8 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_s64 - return payload of s64 attribute
|
|
|
|
* @nla: s64 netlink attribute
|
|
|
|
*/
|
|
|
|
static inline s64 nla_get_s64(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
s64 tmp;
|
|
|
|
|
|
|
|
nla_memcpy(&tmp, nla, sizeof(tmp));
|
|
|
|
|
2010-03-16 13:30:44 +00:00
|
|
|
return tmp;
|
2009-05-27 15:50:35 +00:00
|
|
|
}
|
|
|
|
|
netlink: add variable-length / auto integers
We currently push everyone to use padding to align 64b values
in netlink. Un-padded nla_put_u64() doesn't even exist any more.
The story behind this possibly start with this thread:
https://lore.kernel.org/netdev/20121204.130914.1457976839967676240.davem@davemloft.net/
where DaveM was concerned about the alignment of a structure
containing 64b stats. If user space tries to access such struct
directly:
struct some_stats *stats = nla_data(attr);
printf("A: %llu", stats->a);
lack of alignment may become problematic for some architectures.
These days we most often put every single member in a separate
attribute, meaning that the code above would use a helper like
nla_get_u64(), which can deal with alignment internally.
Even for arches which don't have good unaligned access - access
aligned to 4B should be pretty efficient.
Kernel and well known libraries deal with unaligned input already.
Padded 64b is quite space-inefficient (64b + pad means at worst 16B
per attr vs 32b which takes 8B). It is also more typing:
if (nla_put_u64_pad(rsp, NETDEV_A_SOMETHING_SOMETHING,
value, NETDEV_A_SOMETHING_PAD))
Create a new attribute type which will use 32 bits at netlink
level if value is small enough (probably most of the time?),
and (4B-aligned) 64 bits otherwise. Kernel API is just:
if (nla_put_uint(rsp, NETDEV_A_SOMETHING_SOMETHING, value))
Calling this new type "just" sint / uint with no specific size
will hopefully also make people more comfortable with using it.
Currently telling people "don't use u8, you may need the bits,
and netlink will round up to 4B, anyway" is the #1 comment
we give to newcomers.
In terms of netlink layout it looks like this:
0 4 8 12 16
32b: [nlattr][ u32 ]
64b: [ pad ][nlattr][ u64 ]
uint(32) [nlattr][ u32 ]
uint(64) [nlattr][ u64 ]
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Nicolas Dichtel <nicolas.dichtel@6wind.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2023-10-18 21:39:20 +00:00
|
|
|
/**
|
|
|
|
* nla_get_sint - return payload of uint attribute
|
|
|
|
* @nla: uint netlink attribute
|
|
|
|
*/
|
|
|
|
static inline s64 nla_get_sint(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
if (nla_len(nla) == sizeof(s32))
|
|
|
|
return nla_get_s32(nla);
|
|
|
|
return nla_get_s64(nla);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_get_flag - return payload of flag attribute
|
|
|
|
* @nla: flag netlink attribute
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline int nla_get_flag(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
return !!nla;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_msecs - return payload of msecs attribute
|
|
|
|
* @nla: msecs netlink attribute
|
|
|
|
*
|
|
|
|
* Returns the number of milliseconds in jiffies.
|
|
|
|
*/
|
2008-10-28 18:59:11 +00:00
|
|
|
static inline unsigned long nla_get_msecs(const struct nlattr *nla)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
|
|
|
u64 msecs = nla_get_u64(nla);
|
|
|
|
|
|
|
|
return msecs_to_jiffies((unsigned long) msecs);
|
|
|
|
}
|
|
|
|
|
2015-03-29 14:59:26 +00:00
|
|
|
/**
|
|
|
|
* nla_get_in_addr - return payload of IPv4 address attribute
|
|
|
|
* @nla: IPv4 address netlink attribute
|
|
|
|
*/
|
|
|
|
static inline __be32 nla_get_in_addr(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
return *(__be32 *) nla_data(nla);
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_get_in6_addr - return payload of IPv6 address attribute
|
|
|
|
* @nla: IPv6 address netlink attribute
|
|
|
|
*/
|
|
|
|
static inline struct in6_addr nla_get_in6_addr(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
struct in6_addr tmp;
|
|
|
|
|
|
|
|
nla_memcpy(&tmp, nla, sizeof(tmp));
|
|
|
|
return tmp;
|
|
|
|
}
|
|
|
|
|
2017-07-30 17:24:49 +00:00
|
|
|
/**
|
|
|
|
* nla_get_bitfield32 - return payload of 32 bitfield attribute
|
|
|
|
* @nla: nla_bitfield32 attribute
|
|
|
|
*/
|
|
|
|
static inline struct nla_bitfield32 nla_get_bitfield32(const struct nlattr *nla)
|
|
|
|
{
|
|
|
|
struct nla_bitfield32 tmp;
|
|
|
|
|
|
|
|
nla_memcpy(&tmp, nla, sizeof(tmp));
|
|
|
|
return tmp;
|
|
|
|
}
|
|
|
|
|
2016-10-26 08:53:16 +00:00
|
|
|
/**
|
|
|
|
* nla_memdup - duplicate attribute memory (kmemdup)
|
|
|
|
* @src: netlink attribute to duplicate from
|
|
|
|
* @gfp: GFP mask
|
|
|
|
*/
|
2024-04-15 02:07:31 +00:00
|
|
|
static inline void *nla_memdup_noprof(const struct nlattr *src, gfp_t gfp)
|
2016-10-26 08:53:16 +00:00
|
|
|
{
|
2024-04-15 02:07:31 +00:00
|
|
|
return kmemdup_noprof(nla_data(src), nla_len(src), gfp);
|
2016-10-26 08:53:16 +00:00
|
|
|
}
|
2024-04-15 02:07:31 +00:00
|
|
|
#define nla_memdup(...) alloc_hooks(nla_memdup_noprof(__VA_ARGS__))
|
2016-10-26 08:53:16 +00:00
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
2019-04-26 09:13:06 +00:00
|
|
|
* nla_nest_start_noflag - Start a new level of nested attributes
|
2005-11-10 01:25:51 +00:00
|
|
|
* @skb: socket buffer to add attributes to
|
|
|
|
* @attrtype: attribute type of container
|
|
|
|
*
|
2019-04-26 09:13:06 +00:00
|
|
|
* This function exists for backward compatibility to use in APIs which never
|
|
|
|
* marked their nest attributes with NLA_F_NESTED flag. New APIs should use
|
|
|
|
* nla_nest_start() which sets the flag.
|
|
|
|
*
|
|
|
|
* Returns the container attribute or NULL on error
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2019-04-26 09:13:06 +00:00
|
|
|
static inline struct nlattr *nla_nest_start_noflag(struct sk_buff *skb,
|
|
|
|
int attrtype)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2007-04-20 03:29:13 +00:00
|
|
|
struct nlattr *start = (struct nlattr *)skb_tail_pointer(skb);
|
2005-11-10 01:25:51 +00:00
|
|
|
|
|
|
|
if (nla_put(skb, attrtype, 0, NULL) < 0)
|
|
|
|
return NULL;
|
|
|
|
|
|
|
|
return start;
|
|
|
|
}
|
|
|
|
|
2019-04-26 09:13:06 +00:00
|
|
|
/**
|
|
|
|
* nla_nest_start - Start a new level of nested attributes, with NLA_F_NESTED
|
|
|
|
* @skb: socket buffer to add attributes to
|
|
|
|
* @attrtype: attribute type of container
|
|
|
|
*
|
|
|
|
* Unlike nla_nest_start_noflag(), mark the nest attribute with NLA_F_NESTED
|
|
|
|
* flag. This is the preferred function to use in new code.
|
|
|
|
*
|
|
|
|
* Returns the container attribute or NULL on error
|
|
|
|
*/
|
|
|
|
static inline struct nlattr *nla_nest_start(struct sk_buff *skb, int attrtype)
|
|
|
|
{
|
|
|
|
return nla_nest_start_noflag(skb, attrtype | NLA_F_NESTED);
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_nest_end - Finalize nesting of attributes
|
2007-10-11 04:09:48 +00:00
|
|
|
* @skb: socket buffer the attributes are stored in
|
2005-11-10 01:25:51 +00:00
|
|
|
* @start: container attribute
|
|
|
|
*
|
|
|
|
* Corrects the container attribute header to include the all
|
|
|
|
* appeneded attributes.
|
|
|
|
*
|
|
|
|
* Returns the total data length of the skb.
|
|
|
|
*/
|
|
|
|
static inline int nla_nest_end(struct sk_buff *skb, struct nlattr *start)
|
|
|
|
{
|
2007-04-20 03:29:13 +00:00
|
|
|
start->nla_len = skb_tail_pointer(skb) - (unsigned char *)start;
|
2005-11-10 01:25:51 +00:00
|
|
|
return skb->len;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_nest_cancel - Cancel nesting of attributes
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
* @start: container attribute
|
|
|
|
*
|
|
|
|
* Removes the container attribute and including all nested
|
2008-06-03 23:36:54 +00:00
|
|
|
* attributes. Returns -EMSGSIZE
|
2005-11-10 01:25:51 +00:00
|
|
|
*/
|
2008-06-03 23:36:54 +00:00
|
|
|
static inline void nla_nest_cancel(struct sk_buff *skb, struct nlattr *start)
|
2005-11-10 01:25:51 +00:00
|
|
|
{
|
2008-06-03 23:36:54 +00:00
|
|
|
nlmsg_trim(skb, start);
|
2005-11-10 01:25:51 +00:00
|
|
|
}
|
|
|
|
|
2006-09-25 22:54:03 +00:00
|
|
|
/**
|
2019-12-11 09:58:19 +00:00
|
|
|
* __nla_validate_nested - Validate a stream of nested attributes
|
2006-09-25 22:54:03 +00:00
|
|
|
* @start: container attribute
|
|
|
|
* @maxtype: maximum attribute type to be expected
|
|
|
|
* @policy: validation policy
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
* @validate: validation strictness
|
2017-04-12 12:34:07 +00:00
|
|
|
* @extack: extended ACK report struct
|
2006-09-25 22:54:03 +00:00
|
|
|
*
|
|
|
|
* Validates all attributes in the nested attribute stream against the
|
|
|
|
* specified policy. Attributes with a type exceeding maxtype will be
|
|
|
|
* ignored. See documenation of struct nla_policy for more details.
|
|
|
|
*
|
|
|
|
* Returns 0 on success or a negative error code.
|
|
|
|
*/
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
static inline int __nla_validate_nested(const struct nlattr *start, int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
unsigned int validate,
|
|
|
|
struct netlink_ext_ack *extack)
|
|
|
|
{
|
|
|
|
return __nla_validate(nla_data(start), nla_len(start), maxtype, policy,
|
|
|
|
validate, extack);
|
|
|
|
}
|
|
|
|
|
2019-05-28 08:56:03 +00:00
|
|
|
static inline int
|
2019-12-11 09:58:19 +00:00
|
|
|
nla_validate_nested(const struct nlattr *start, int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
2019-05-28 08:56:03 +00:00
|
|
|
{
|
|
|
|
return __nla_validate_nested(start, maxtype, policy,
|
|
|
|
NL_VALIDATE_STRICT, extack);
|
|
|
|
}
|
|
|
|
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
static inline int
|
|
|
|
nla_validate_nested_deprecated(const struct nlattr *start, int maxtype,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
struct netlink_ext_ack *extack)
|
2006-09-25 22:54:03 +00:00
|
|
|
{
|
netlink: make validation more configurable for future strictness
We currently have two levels of strict validation:
1) liberal (default)
- undefined (type >= max) & NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
- garbage at end of message accepted
2) strict (opt-in)
- NLA_UNSPEC attributes accepted
- attribute length >= expected accepted
Split out parsing strictness into four different options:
* TRAILING - check that there's no trailing data after parsing
attributes (in message or nested)
* MAXTYPE - reject attrs > max known type
* UNSPEC - reject attributes with NLA_UNSPEC policy entries
* STRICT_ATTRS - strictly validate attribute size
The default for future things should be *everything*.
The current *_strict() is a combination of TRAILING and MAXTYPE,
and is renamed to _deprecated_strict().
The current regular parsing has none of this, and is renamed to
*_parse_deprecated().
Additionally it allows us to selectively set one of the new flags
even on old policies. Notably, the UNSPEC flag could be useful in
this case, since it can be arranged (by filling in the policy) to
not be an incompatible userspace ABI change, but would then going
forward prevent forgetting attribute entries. Similar can apply
to the POLICY flag.
We end up with the following renames:
* nla_parse -> nla_parse_deprecated
* nla_parse_strict -> nla_parse_deprecated_strict
* nlmsg_parse -> nlmsg_parse_deprecated
* nlmsg_parse_strict -> nlmsg_parse_deprecated_strict
* nla_parse_nested -> nla_parse_nested_deprecated
* nla_validate_nested -> nla_validate_nested_deprecated
Using spatch, of course:
@@
expression TB, MAX, HEAD, LEN, POL, EXT;
@@
-nla_parse(TB, MAX, HEAD, LEN, POL, EXT)
+nla_parse_deprecated(TB, MAX, HEAD, LEN, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression NLH, HDRLEN, TB, MAX, POL, EXT;
@@
-nlmsg_parse_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
+nlmsg_parse_deprecated_strict(NLH, HDRLEN, TB, MAX, POL, EXT)
@@
expression TB, MAX, NLA, POL, EXT;
@@
-nla_parse_nested(TB, MAX, NLA, POL, EXT)
+nla_parse_nested_deprecated(TB, MAX, NLA, POL, EXT)
@@
expression START, MAX, POL, EXT;
@@
-nla_validate_nested(START, MAX, POL, EXT)
+nla_validate_nested_deprecated(START, MAX, POL, EXT)
@@
expression NLH, HDRLEN, MAX, POL, EXT;
@@
-nlmsg_validate(NLH, HDRLEN, MAX, POL, EXT)
+nlmsg_validate_deprecated(NLH, HDRLEN, MAX, POL, EXT)
For this patch, don't actually add the strict, non-renamed versions
yet so that it breaks compile if I get it wrong.
Also, while at it, make nla_validate and nla_parse go down to a
common __nla_validate_parse() function to avoid code duplication.
Ultimately, this allows us to have very strict validation for every
new caller of nla_parse()/nlmsg_parse() etc as re-introduced in the
next patch, while existing things will continue to work as is.
In effect then, this adds fully strict validation for any new command.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2019-04-26 12:07:28 +00:00
|
|
|
return __nla_validate_nested(start, maxtype, policy,
|
|
|
|
NL_VALIDATE_LIBERAL, extack);
|
2006-09-25 22:54:03 +00:00
|
|
|
}
|
|
|
|
|
2016-04-21 16:58:24 +00:00
|
|
|
/**
|
|
|
|
* nla_need_padding_for_64bit - test 64-bit alignment of the next attribute
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
*
|
|
|
|
* Return true if padding is needed to align the next attribute (nla_data()) to
|
|
|
|
* a 64-bit aligned area.
|
|
|
|
*/
|
|
|
|
static inline bool nla_need_padding_for_64bit(struct sk_buff *skb)
|
|
|
|
{
|
|
|
|
#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
|
|
|
|
/* The nlattr header is 4 bytes in size, that's why we test
|
|
|
|
* if the skb->data _is_ aligned. A NOP attribute, plus
|
|
|
|
* nlattr header for next attribute, will make nla_data()
|
|
|
|
* 8-byte aligned.
|
|
|
|
*/
|
|
|
|
if (IS_ALIGNED((unsigned long)skb_tail_pointer(skb), 8))
|
|
|
|
return true;
|
|
|
|
#endif
|
|
|
|
return false;
|
|
|
|
}
|
|
|
|
|
2016-04-19 23:49:29 +00:00
|
|
|
/**
|
|
|
|
* nla_align_64bit - 64-bit align the nla_data() of next attribute
|
|
|
|
* @skb: socket buffer the message is stored in
|
|
|
|
* @padattr: attribute type for the padding
|
|
|
|
*
|
|
|
|
* Conditionally emit a padding netlink attribute in order to make
|
|
|
|
* the next attribute we emit have a 64-bit aligned nla_data() area.
|
|
|
|
* This will only be done in architectures which do not have
|
2016-04-20 14:31:31 +00:00
|
|
|
* CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS defined.
|
2016-04-19 23:49:29 +00:00
|
|
|
*
|
|
|
|
* Returns zero on success or a negative error code.
|
|
|
|
*/
|
|
|
|
static inline int nla_align_64bit(struct sk_buff *skb, int padattr)
|
|
|
|
{
|
2016-04-21 16:58:24 +00:00
|
|
|
if (nla_need_padding_for_64bit(skb) &&
|
2016-04-20 14:31:31 +00:00
|
|
|
!nla_reserve(skb, padattr, 0))
|
|
|
|
return -EMSGSIZE;
|
2016-04-21 16:58:24 +00:00
|
|
|
|
2016-04-19 23:49:29 +00:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/**
|
|
|
|
* nla_total_size_64bit - total length of attribute including padding
|
|
|
|
* @payload: length of payload
|
|
|
|
*/
|
|
|
|
static inline int nla_total_size_64bit(int payload)
|
|
|
|
{
|
|
|
|
return NLA_ALIGN(nla_attr_size(payload))
|
2016-04-20 14:31:31 +00:00
|
|
|
#ifndef CONFIG_HAVE_EFFICIENT_UNALIGNED_ACCESS
|
2016-04-19 23:49:29 +00:00
|
|
|
+ NLA_ALIGN(nla_attr_size(0))
|
|
|
|
#endif
|
|
|
|
;
|
|
|
|
}
|
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
/**
|
|
|
|
* nla_for_each_attr - iterate over a stream of attributes
|
|
|
|
* @pos: loop counter, set to current attribute
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nla_for_each_attr(pos, head, len, rem) \
|
|
|
|
for (pos = head, rem = len; \
|
|
|
|
nla_ok(pos, rem); \
|
|
|
|
pos = nla_next(pos, &(rem)))
|
|
|
|
|
netlink: introduce type-checking attribute iteration
There are, especially with multi-attr arrays, many cases
of needing to iterate all attributes of a specific type
in a netlink message or a nested attribute. Add specific
macros to support that case.
Also convert many instances using this spatch:
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
Although I had to undo one bad change this made, and
I also adjusted some other code for whitespace and to
use direct variable initialization now.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://lore.kernel.org/r/20240328203144.b5a6c895fb80.I1869b44767379f204998ff44dd239803f39c23e0@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-03-28 19:31:45 +00:00
|
|
|
/**
|
|
|
|
* nla_for_each_attr_type - iterate over a stream of attributes
|
|
|
|
* @pos: loop counter, set to current attribute
|
|
|
|
* @type: required attribute type for @pos
|
|
|
|
* @head: head of attribute stream
|
|
|
|
* @len: length of attribute stream
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nla_for_each_attr_type(pos, type, head, len, rem) \
|
|
|
|
nla_for_each_attr(pos, head, len, rem) \
|
|
|
|
if (nla_type(pos) == type)
|
|
|
|
|
2006-08-05 06:03:05 +00:00
|
|
|
/**
|
|
|
|
* nla_for_each_nested - iterate over nested attributes
|
|
|
|
* @pos: loop counter, set to current attribute
|
|
|
|
* @nla: attribute containing the nested attributes
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nla_for_each_nested(pos, nla, rem) \
|
|
|
|
nla_for_each_attr(pos, nla_data(nla), nla_len(nla), rem)
|
|
|
|
|
netlink: introduce type-checking attribute iteration
There are, especially with multi-attr arrays, many cases
of needing to iterate all attributes of a specific type
in a netlink message or a nested attribute. Add specific
macros to support that case.
Also convert many instances using this spatch:
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) == ATTR) {
...
-}
}
@@
iterator nla_for_each_attr;
iterator name nla_for_each_attr_type;
identifier nla;
expression head, len, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_attr(nla, head, len, rem)
+nla_for_each_attr_type(nla, ATTR, head, len, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
@@
identifier nla;
iterator nla_for_each_nested;
iterator name nla_for_each_nested_type;
expression attr, rem;
expression ATTR;
type T;
identifier x;
@@
-nla_for_each_nested(nla, attr, rem)
+nla_for_each_nested_type(nla, ATTR, attr, rem)
{
<... T x; ...>
-if (nla_type(nla) != ATTR) continue;
...
}
Although I had to undo one bad change this made, and
I also adjusted some other code for whitespace and to
use direct variable initialization now.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Link: https://lore.kernel.org/r/20240328203144.b5a6c895fb80.I1869b44767379f204998ff44dd239803f39c23e0@changeid
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2024-03-28 19:31:45 +00:00
|
|
|
/**
|
|
|
|
* nla_for_each_nested_type - iterate over nested attributes
|
|
|
|
* @pos: loop counter, set to current attribute
|
|
|
|
* @type: required attribute type for @pos
|
|
|
|
* @nla: attribute containing the nested attributes
|
|
|
|
* @rem: initialized to len, holds bytes currently remaining in stream
|
|
|
|
*/
|
|
|
|
#define nla_for_each_nested_type(pos, type, nla, rem) \
|
|
|
|
nla_for_each_nested(pos, nla, rem) \
|
|
|
|
if (nla_type(pos) == type)
|
|
|
|
|
2014-10-27 07:12:16 +00:00
|
|
|
/**
|
|
|
|
* nla_is_last - Test if attribute is last in stream
|
|
|
|
* @nla: attribute to test
|
|
|
|
* @rem: bytes remaining in stream
|
|
|
|
*/
|
|
|
|
static inline bool nla_is_last(const struct nlattr *nla, int rem)
|
|
|
|
{
|
|
|
|
return nla->nla_len == rem;
|
|
|
|
}
|
|
|
|
|
2020-04-30 20:13:11 +00:00
|
|
|
void nla_get_range_unsigned(const struct nla_policy *pt,
|
|
|
|
struct netlink_range_validation *range);
|
|
|
|
void nla_get_range_signed(const struct nla_policy *pt,
|
|
|
|
struct netlink_range_validation_signed *range);
|
|
|
|
|
2020-10-02 21:49:55 +00:00
|
|
|
struct netlink_policy_dump_state;
|
|
|
|
|
2020-10-03 08:44:43 +00:00
|
|
|
int netlink_policy_dump_add_policy(struct netlink_policy_dump_state **pstate,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
unsigned int maxtype);
|
|
|
|
int netlink_policy_dump_get_policy_idx(struct netlink_policy_dump_state *state,
|
|
|
|
const struct nla_policy *policy,
|
|
|
|
unsigned int maxtype);
|
2020-10-02 21:49:55 +00:00
|
|
|
bool netlink_policy_dump_loop(struct netlink_policy_dump_state *state);
|
|
|
|
int netlink_policy_dump_write(struct sk_buff *skb,
|
|
|
|
struct netlink_policy_dump_state *state);
|
2020-10-08 10:45:17 +00:00
|
|
|
int netlink_policy_dump_attr_size_estimate(const struct nla_policy *pt);
|
|
|
|
int netlink_policy_dump_write_attr(struct sk_buff *skb,
|
|
|
|
const struct nla_policy *pt,
|
|
|
|
int nestattr);
|
2020-10-02 21:49:55 +00:00
|
|
|
void netlink_policy_dump_free(struct netlink_policy_dump_state *state);
|
netlink: add infrastructure to expose policies to userspace
Add, and use in generic netlink, helpers to dump out a netlink
policy to userspace, including all the range validation data,
nested policies etc.
This lets userspace discover what the kernel understands.
For families/commands other than generic netlink, the helpers
need to be used directly in an appropriate command, or we can
add some infrastructure (a new netlink family) that those can
register their policies with for introspection. I'm not that
familiar with non-generic netlink, so that's left out for now.
The data exposed to userspace also includes min and max length
for binary/string data, I've done that instead of letting the
userspace tools figure out whether min/max is intended based
on the type so that we can extend this later in the kernel, we
might want to just use the range data for example.
Because of this, I opted to not directly expose the NLA_*
values, even if some of them are already exposed via BPF, as
with min/max length we don't need to have different types here
for NLA_BINARY/NLA_MIN_LEN/NLA_EXACT_LEN, we just make them
all NL_ATTR_TYPE_BINARY with min/max length optionally set.
Similarly, we don't really need NLA_MSECS, and perhaps can
remove it in the future - but not if we encode it into the
userspace API now. It gets mapped to NL_ATTR_TYPE_U64 here.
Note that the exposing here corresponds to the strict policy
interpretation, and NLA_UNSPEC items are omitted entirely.
To get those, change them to NLA_MIN_LEN which behaves in
exactly the same way, but is exposed.
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2020-04-30 20:13:12 +00:00
|
|
|
|
2005-11-10 01:25:51 +00:00
|
|
|
#endif
|