leandrof/linux
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
Files
49402b0b7ff706eba22cb67fe7e2dff9f04ce255
linux/drivers/net/ethernet/netronome/nfp/flower/main.h

286 lines
8.8 KiB
C
Raw Normal View History

nfp: replace long license headers with SPDX Replace the repeated license text with SDPX identifiers. While at it bump the Copyright dates for files we touched this year. Signed-off-by: Edwin Peer <edwin.peer@netronome.com> Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Nic Viljoen <nick.viljoen@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-11 08:57:42 -07:00
/* SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) */
/* Copyright (C) 2017-2018 Netronome Systems, Inc. */
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
#ifndef __NFP_FLOWER_H__
#define __NFP_FLOWER_H__ 1
nfp: flower: compile Geneve match fields Compile Geneve match fields for offloading to the NFP. The addition of Geneve overflows the 8 bit key_layer field, so apply extended metadata to the match cmsg allowing up to 32 more key_layer fields. Rather than adding new Geneve blocks, move the vxlan code to generic ipv4 udp tunnel structs and use these for both vxlan and Geneve. Matches are only supported when specifically mentioning well known port 6081. Geneve tunnel options are not yet included in the match. Only offload Geneve if the fw supports it - include check for this. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 17:58:28 +01:00
#include "cmsg.h"
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
#include <linux/circ_buf.h>
#include <linux/hashtable.h>
nfp: flower: use rhashtable for flow caching Make use of relativistic hash tables for tracking flows instead of fixed sized hash tables. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:34 -07:00
#include <linux/rhashtable.h>
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
#include <linux/time64.h>
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
#include <linux/types.h>
net: sched: get rid of struct tc_to_netdev Get rid of struct tc_to_netdev which is now just unnecessary container and rather pass per-type structures down to drivers directly. Along with that, consolidate the naming of per-type structure variables in cls_*. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-07 10:15:32 +02:00
#include <net/pkt_cls.h>
nfp: flower: implement tcp flag match offload Implement tcp flag match offloading. Current tcp flag match support include FIN, SYN, RST, PSH and URG flags, other flags are unsupported. The PSH and URG flags are only set in the hardware fast path when used in combination with the SYN, RST and PSH flags. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-15 20:19:09 -08:00
#include <net/tcp.h>
nfp: process control messages in workqueue in flower app Processing of control messages is not time-critical and future processing of some messages will require taking the RTNL which is not possible in a BH handler. It seems simplest to move all control message processing to a workqueue. Signed-off-by: Simon Horman <simon.horman@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-16 09:37:43 +02:00
#include <linux/workqueue.h>
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
#include <linux/idr.h>
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
nfp: flower: compute link aggregation action If the egress device of an offloaded rule is a LAG port, then encode the output port to the NFP with a LAG identifier and the offloaded group ID. A prelag action is also offloaded which must be the first action of the series (although may appear after other pre-actions - e.g. tunnels). This causes the FW to check that it has the necessary information to output to the requested LAG port. If it does not, the packet is sent to the kernel before any other actions are applied to it. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:55 -07:00
struct nfp_fl_pre_lag;
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
struct net_device;
struct nfp_app;
nfp: flower: support multiple memory units for filter offloads Adds support for multiple memory units which are used for filter offloads. Each filter is assigned a stats id, the MSBs of the id are used to determine which memory unit the filter should be offloaded to. The number of available memory units that could be used for filter offload is obtained from HW. A simple round robin technique is used to allocate and distribute the ids across memory units. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:52 -08:00
#define NFP_FL_STAT_ID_MU_NUM GENMASK(31, 22)
#define NFP_FL_STAT_ID_STAT GENMASK(21, 0)
nfp: flower: use host context count provided by firmware Read the host context count symbols provided by firmware and use it to determine the number of allocated stats ids. Previously it won't be possible to offload more than 2^17 filter even if FW was able to do so. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:36 -07:00
#define NFP_FL_STATS_ELEM_RS FIELD_SIZEOF(struct nfp_fl_stats_id, \
init_unalloc)
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
#define NFP_FLOWER_MASK_ENTRY_RS 256
#define NFP_FLOWER_MASK_ELEMENT_RS 1
#define NFP_FLOWER_MASK_HASH_BITS 10
nfp: fix flower offload metadata flag usage Hardware has no notion of new or last mask id, instead it makes use of the message type (i.e. add flow or del flow) in combination with a single bit in metadata flags to determine when to add or delete a mask id. Previously we made use of the new or last flags to indicate that a new mask should be allocated or deallocated, respectively. This incorrect behaviour is fixed by making use single bit in metadata flags to indicate mask allocation or deallocation. Fixes: 43f84b72c50d ("nfp: add metadata to each flow offload") Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-11-16 17:06:39 -08:00
#define NFP_FL_META_FLAG_MANAGE_MASK BIT(7)
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
#define NFP_FL_MASK_REUSE_TIME_NS 40000
#define NFP_FL_MASK_ID_LOCATION 1
nfp: compile flower vxlan tunnel metadata match fields Compile ovs-tc flower vxlan metadata match fields for offloading. Only support offload of tunnel data when the VXLAN port specifically matches well known port 4789. Signed-off-by: John Hurley <john.hurley@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:36 +02:00
#define NFP_FL_VXLAN_PORT 4789
nfp: flower: compile Geneve match fields Compile Geneve match fields for offloading to the NFP. The addition of Geneve overflows the 8 bit key_layer field, so apply extended metadata to the match cmsg allowing up to 32 more key_layer fields. Rather than adding new Geneve blocks, move the vxlan code to generic ipv4 udp tunnel structs and use these for both vxlan and Geneve. Matches are only supported when specifically mentioning well known port 6081. Geneve tunnel options are not yet included in the match. Only offload Geneve if the fw supports it - include check for this. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 17:58:28 +01:00
#define NFP_FL_GENEVE_PORT 6081
/* Extra features bitmap. */
#define NFP_FL_FEATS_GENEVE BIT(0)
nfp: flower: offload phys port MTU change Trigger a port mod message to request an MTU change on the NIC when any physical port representor is assigned a new MTU value. The driver waits 10 msec for an ack that the FW has set the MTU. If no ack is received the request is rejected and an appropriate warning flagged. Rather than maintain an MTU queue per repr, one is maintained per app. Because the MTU ndo is protected by the rtnl lock, there can never be contention here. Portmod messages from the NIC are also protected by rtnl so we first check if the portmod is an ack and, if so, handle outside rtnl and the cmsg work queue. Acks are detected by the marking of a bit in a portmod response. They are then verfied by checking the port number and MTU value expected by the app. If the expected MTU is 0 then no acks are currently expected. Also, ensure that the packet headroom reserved by the flower firmware is considered when accepting an MTU change on any repr. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-28 18:50:07 -07:00
#define NFP_FL_NBI_MTU_SETTING BIT(1)
nfp: flower: add geneve option push action offload Introduce new push geneve option action. This allows offloading filters configured to entunnel geneve with options. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-08-07 17:36:02 +02:00
#define NFP_FL_FEATS_GENEVE_OPT BIT(2)
nfp: flower: fix vlan match by checking both vlan id and vlan pcp Previously we only checked if the vlan id field is present when trying to match a vlan tag. The vlan id and vlan pcp field should be treated independently. Fixes: 5571e8c9f241 ("nfp: extend flower matching capabilities") Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-09-11 06:38:44 -07:00
#define NFP_FL_FEATS_VLAN_PCP BIT(3)
nfp: flower: check for/turn on LAG support in firmware Check if the fw contains the _abi_flower_balance_sync_enable symbol. If it does then write a 1 to this indicating that the driver is willing to receive NIC to kernel LAG related control messages. If the write is successful, update the list of extra features supported by the fw and add a stub to accept LAG cmsgs. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:50 -07:00
#define NFP_FL_FEATS_LAG BIT(31)
nfp: compile flower vxlan tunnel metadata match fields Compile ovs-tc flower vxlan metadata match fields for offloading. Only support offload of tunnel data when the VXLAN port specifically matches well known port 4789. Signed-off-by: John Hurley <john.hurley@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:36 +02:00
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
struct nfp_fl_mask_id {
struct circ_buf mask_id_free_list;
nfp: avoid using getnstimeofday64() getnstimeofday64 is deprecated in favor of the ktime_get() family of functions. The direct replacement would be ktime_get_real_ts64(), but I'm picking the basic ktime_get() instead: - using a ktime_t simplifies the code compared to timespec64 - using monotonic time instead of real time avoids issues caused by a concurrent settimeofday() or during a leap second adjustment. Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-07-11 14:29:53 +02:00
ktime_t *last_used;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
u8 init_unallocated;
};
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
struct nfp_fl_stats_id {
struct circ_buf free_list;
u32 init_unalloc;
u8 repeated_em_count;
};
nfp: flower: tidy tunnel related private data Recent additions to the flower app private data have grouped the variables of a given feature into a struct and added that struct to the main private data struct. In keeping with this, move all tunnel related private data to their own struct. This has no affect on functionality but improves readability and maintenance of the code. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:53 -08:00
/**
* struct nfp_fl_tunnel_offloads - priv data for tunnel offloads
* @mac_index_list: List of unique 8-bit indexes for non NFP netdevs
* @ipv4_off_list: List of IPv4 addresses to offload
* @neigh_off_list: List of neighbour offloads
* @mac_index_lock: Lock for the MAC index list
* @ipv4_off_lock: Lock for the IPv4 address list
* @neigh_off_lock: Lock for the neighbour address list
* @mac_off_ids: IDA to manage id assignment for offloaded MACs
* @neigh_nb: Notifier to monitor neighbour state
*/
struct nfp_fl_tunnel_offloads {
struct list_head mac_index_list;
struct list_head ipv4_off_list;
struct list_head neigh_off_list;
struct mutex mac_index_lock;
struct mutex ipv4_off_lock;
spinlock_t neigh_off_lock;
struct ida mac_off_ids;
struct notifier_block neigh_nb;
};
nfp: flower: offload phys port MTU change Trigger a port mod message to request an MTU change on the NIC when any physical port representor is assigned a new MTU value. The driver waits 10 msec for an ack that the FW has set the MTU. If no ack is received the request is rejected and an appropriate warning flagged. Rather than maintain an MTU queue per repr, one is maintained per app. Because the MTU ndo is protected by the rtnl lock, there can never be contention here. Portmod messages from the NIC are also protected by rtnl so we first check if the portmod is an ack and, if so, handle outside rtnl and the cmsg work queue. Acks are detected by the marking of a bit in a portmod response. They are then verfied by checking the port number and MTU value expected by the app. If the expected MTU is 0 then no acks are currently expected. Also, ensure that the packet headroom reserved by the flower firmware is considered when accepting an MTU change on any repr. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-28 18:50:07 -07:00
/**
* struct nfp_mtu_conf - manage MTU setting
* @portnum: NFP port number of repr with requested MTU change
* @requested_val: MTU value requested for repr
* @ack: Received ack that MTU has been correctly set
* @wait_q: Wait queue for MTU acknowledgements
* @lock: Lock for setting/reading MTU variables
*/
struct nfp_mtu_conf {
u32 portnum;
unsigned int requested_val;
bool ack;
wait_queue_head_t wait_q;
spinlock_t lock;
};
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
/**
* struct nfp_fl_lag - Flower APP priv data for link aggregation
* @work: Work queue for writing configs to the HW
* @lock: Lock to protect lag_group_list
* @group_list: List of all master/slave groups offloaded
* @ida_handle: IDA to handle group ids
* @pkt_num: Incremented for each config packet sent
* @batch_ver: Incremented for each batch of config packets
* @global_inst: Instance allocator for groups
* @rst_cfg: Marker to reset HW LAG config
nfp: flower: implement host cmsg handler for LAG Adds the control message handler to synchronize offloaded group config with that of the kernel. Such messages are sent from fw to driver and feature the following 3 flags: - Data: an attached cmsg could not be processed - store for retransmission - Xon: FW can accept new messages - retransmit any stored cmsgs - Sync: full sync requested so retransmit all kernel LAG group info Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:54 -07:00
* @retrans_skbs: Cmsgs that could not be processed by HW and require
* retransmission
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
*/
struct nfp_fl_lag {
struct delayed_work work;
struct mutex lock;
struct list_head group_list;
struct ida ida_handle;
unsigned int pkt_num;
unsigned int batch_ver;
u8 global_inst;
bool rst_cfg;
nfp: flower: implement host cmsg handler for LAG Adds the control message handler to synchronize offloaded group config with that of the kernel. Such messages are sent from fw to driver and feature the following 3 flags: - Data: an attached cmsg could not be processed - store for retransmission - Xon: FW can accept new messages - retransmit any stored cmsgs - Sync: full sync requested so retransmit all kernel LAG group info Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:54 -07:00
struct sk_buff_head retrans_skbs;
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
};
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
/**
* struct nfp_flower_priv - Flower APP per-vNIC priv data
nfp: separate app vNIC init/clean from alloc/free We currently only have one app callback for vNIC creation and destruction. This is insufficient, because some actions have to be taken before netdev is registered, after it's registered and after it's unregistered. Old callbacks were really corresponding to alloc/free actions. Rename them and add proper init/clean. Apps using representors will be able to use new callbacks to manage lifetime of upper devices. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-02 18:26:00 -07:00
* @app: Back pointer to app
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
* @nn: Pointer to vNIC
* @mask_id_seed: Seed used for mask hash table
* @flower_version: HW version of flower
nfp: flower: read extra feature support from fw Extract the _abi_flower_extra_features symbol from the fw which gives a 64 bit bitmap of new features (on top of the flower base support) that the fw can offload. Store this bitmap in the priv data associated with each app. If the symbol does not exist, set the bitmap to 0. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 17:58:27 +01:00
* @flower_ext_feats: Bitmap of extra features the HW supports
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
* @stats_ids: List of free stats ids
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
* @mask_ids: List of free mask ids
* @mask_table: Hash table used to store masks
nfp: flower: use host context count provided by firmware Read the host context count symbols provided by firmware and use it to determine the number of allocated stats ids. Previously it won't be possible to offload more than 2^17 filter even if FW was able to do so. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:36 -07:00
* @stats_ring_size: Maximum number of allowed stats ids
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
* @flow_table: Hash table used to store flower rules
nfp: flower: use stats array instead of storing stats per flow Make use of an array stats instead of storing stats per flow which would require a hash lookup at critical times. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:35 -07:00
* @stats: Stored stats updates for flower rules
* @stats_lock: Lock for flower rule stats updates
nfp: process control messages in workqueue in flower app Processing of control messages is not time-critical and future processing of some messages will require taking the RTNL which is not possible in a BH handler. It seems simplest to move all control message processing to a workqueue. Signed-off-by: Simon Horman <simon.horman@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-16 09:37:43 +02:00
* @cmsg_work: Workqueue for control messages processing
nfp: flower: split and limit cmsg skb lists Introduce a second skb list for handling control messages and limit the number of allowed messages. Some control messages are considered more crucial than others, resulting in the need for a second skb list. By splitting the list into a separate high and low priority list we can ensure that messages on the high list get added to the head of the list that gets processed, this however has no functional impact. Previously there was no limit on the number of messages allowed on the queue, this could result in the queue growing boundlessly and eventually the host running out of memory. Fixes: b985f870a5f0 ("nfp: process control messages in workqueue in flower app") Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-11 16:47:38 -07:00
* @cmsg_skbs_high: List of higher priority skbs for control message
* processing
* @cmsg_skbs_low: List of lower priority skbs for control message
* processing
nfp: flower: tidy tunnel related private data Recent additions to the flower app private data have grouped the variables of a given feature into a struct and added that struct to the main private data struct. In keeping with this, move all tunnel related private data to their own struct. This has no affect on functionality but improves readability and maintenance of the code. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:53 -08:00
* @tun: Tunnel offload data
nfp: flower: implement the PORT_REIFY message The PORT_REIFY message indicates whether reprs have been created or when they are about to be destroyed. This is necessary so firmware can know which state the driver is in, e.g. the firmware will not send any control messages related to ports when the reprs are destroyed. This prevents nuisance warning messages printed whenever the firmware sends updates for non-existent reprs. Signed-off-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-02 19:19:01 -08:00
* @reify_replies: atomically stores the number of replies received
* from firmware for repr reify
* @reify_wait_queue: wait queue for repr reify response counting
nfp: flower: offload phys port MTU change Trigger a port mod message to request an MTU change on the NIC when any physical port representor is assigned a new MTU value. The driver waits 10 msec for an ack that the FW has set the MTU. If no ack is received the request is rejected and an appropriate warning flagged. Rather than maintain an MTU queue per repr, one is maintained per app. Because the MTU ndo is protected by the rtnl lock, there can never be contention here. Portmod messages from the NIC are also protected by rtnl so we first check if the portmod is an ack and, if so, handle outside rtnl and the cmsg work queue. Acks are detected by the marking of a bit in a portmod response. They are then verfied by checking the port number and MTU value expected by the app. If the expected MTU is 0 then no acks are currently expected. Also, ensure that the packet headroom reserved by the flower firmware is considered when accepting an MTU change on any repr. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-28 18:50:07 -07:00
* @mtu_conf: Configuration of repr MTU value
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
* @nfp_lag: Link aggregation data block
nfp: flower: offload tunnel decap rules via indirect TC blocks Previously, TC block tunnel decap rules were only offloaded when a callback was triggered through registration of the rules egress device. This meant that the driver had no access to the ingress netdev and so could not verify it was the same tunnel type that the rule implied. Register tunnel devices for indirect TC block offloads in NFP, giving access to new rules based on the ingress device rather than egress. Use this to verify the netdev type of VXLAN and Geneve based rules and offload the rules to HW if applicable. Tunnel registration is done via a netdev notifier. On notifier registration, this is triggered for already existing netdevs. This means that NFP can register for offloads from devices that exist before it is loaded (filter rules will be replayed from the TC core). Similarly, on notifier unregister, a call is triggered for each currently active netdev. This allows the driver to unregister any indirect block callbacks that may still be active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-09 21:21:29 -08:00
* @indr_block_cb_priv: List of priv data passed to indirect block cbs
nfp: flower: support multiple memory units for filter offloads Adds support for multiple memory units which are used for filter offloads. Each filter is assigned a stats id, the MSBs of the id are used to determine which memory unit the filter should be offloaded to. The number of available memory units that could be used for filter offload is obtained from HW. A simple round robin technique is used to allocate and distribute the ids across memory units. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:52 -08:00
* @active_mem_unit: Current active memory unit for flower rules
* @total_mem_units: Total number of available memory units for flower rules
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
*/
struct nfp_flower_priv {
nfp: separate app vNIC init/clean from alloc/free We currently only have one app callback for vNIC creation and destruction. This is insufficient, because some actions have to be taken before netdev is registered, after it's registered and after it's unregistered. Old callbacks were really corresponding to alloc/free actions. Rename them and add proper init/clean. Apps using representors will be able to use new callbacks to manage lifetime of upper devices. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-02 18:26:00 -07:00
struct nfp_app *app;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
struct nfp_net *nn;
u32 mask_id_seed;
u64 flower_version;
nfp: flower: read extra feature support from fw Extract the _abi_flower_extra_features symbol from the fw which gives a 64 bit bitmap of new features (on top of the flower base support) that the fw can offload. Store this bitmap in the priv data associated with each app. If the symbol does not exist, set the bitmap to 0. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 17:58:27 +01:00
u64 flower_ext_feats;
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
struct nfp_fl_stats_id stats_ids;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
struct nfp_fl_mask_id mask_ids;
DECLARE_HASHTABLE(mask_table, NFP_FLOWER_MASK_HASH_BITS);
nfp: flower: use host context count provided by firmware Read the host context count symbols provided by firmware and use it to determine the number of allocated stats ids. Previously it won't be possible to offload more than 2^17 filter even if FW was able to do so. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:36 -07:00
u32 stats_ring_size;
nfp: flower: use rhashtable for flow caching Make use of relativistic hash tables for tracking flows instead of fixed sized hash tables. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:34 -07:00
struct rhashtable flow_table;
nfp: flower: use stats array instead of storing stats per flow Make use of an array stats instead of storing stats per flow which would require a hash lookup at critical times. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:35 -07:00
struct nfp_fl_stats *stats;
spinlock_t stats_lock; /* lock stats */
nfp: process control messages in workqueue in flower app Processing of control messages is not time-critical and future processing of some messages will require taking the RTNL which is not possible in a BH handler. It seems simplest to move all control message processing to a workqueue. Signed-off-by: Simon Horman <simon.horman@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-16 09:37:43 +02:00
struct work_struct cmsg_work;
nfp: flower: split and limit cmsg skb lists Introduce a second skb list for handling control messages and limit the number of allowed messages. Some control messages are considered more crucial than others, resulting in the need for a second skb list. By splitting the list into a separate high and low priority list we can ensure that messages on the high list get added to the head of the list that gets processed, this however has no functional impact. Previously there was no limit on the number of messages allowed on the queue, this could result in the queue growing boundlessly and eventually the host running out of memory. Fixes: b985f870a5f0 ("nfp: process control messages in workqueue in flower app") Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-11 16:47:38 -07:00
struct sk_buff_head cmsg_skbs_high;
struct sk_buff_head cmsg_skbs_low;
nfp: flower: tidy tunnel related private data Recent additions to the flower app private data have grouped the variables of a given feature into a struct and added that struct to the main private data struct. In keeping with this, move all tunnel related private data to their own struct. This has no affect on functionality but improves readability and maintenance of the code. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:53 -08:00
struct nfp_fl_tunnel_offloads tun;
nfp: flower: implement the PORT_REIFY message The PORT_REIFY message indicates whether reprs have been created or when they are about to be destroyed. This is necessary so firmware can know which state the driver is in, e.g. the firmware will not send any control messages related to ports when the reprs are destroyed. This prevents nuisance warning messages printed whenever the firmware sends updates for non-existent reprs. Signed-off-by: Dirk van der Merwe <dirk.vandermerwe@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-01-02 19:19:01 -08:00
atomic_t reify_replies;
wait_queue_head_t reify_wait_queue;
nfp: flower: offload phys port MTU change Trigger a port mod message to request an MTU change on the NIC when any physical port representor is assigned a new MTU value. The driver waits 10 msec for an ack that the FW has set the MTU. If no ack is received the request is rejected and an appropriate warning flagged. Rather than maintain an MTU queue per repr, one is maintained per app. Because the MTU ndo is protected by the rtnl lock, there can never be contention here. Portmod messages from the NIC are also protected by rtnl so we first check if the portmod is an ack and, if so, handle outside rtnl and the cmsg work queue. Acks are detected by the marking of a bit in a portmod response. They are then verfied by checking the port number and MTU value expected by the app. If the expected MTU is 0 then no acks are currently expected. Also, ensure that the packet headroom reserved by the flower firmware is considered when accepting an MTU change on any repr. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-03-28 18:50:07 -07:00
struct nfp_mtu_conf mtu_conf;
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
struct nfp_fl_lag nfp_lag;
nfp: flower: offload tunnel decap rules via indirect TC blocks Previously, TC block tunnel decap rules were only offloaded when a callback was triggered through registration of the rules egress device. This meant that the driver had no access to the ingress netdev and so could not verify it was the same tunnel type that the rule implied. Register tunnel devices for indirect TC block offloads in NFP, giving access to new rules based on the ingress device rather than egress. Use this to verify the netdev type of VXLAN and Geneve based rules and offload the rules to HW if applicable. Tunnel registration is done via a netdev notifier. On notifier registration, this is triggered for already existing netdevs. This means that NFP can register for offloads from devices that exist before it is loaded (filter rules will be replayed from the TC core). Similarly, on notifier unregister, a call is triggered for each currently active netdev. This allows the driver to unregister any indirect block callbacks that may still be active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-09 21:21:29 -08:00
struct list_head indr_block_cb_priv;
nfp: flower: support multiple memory units for filter offloads Adds support for multiple memory units which are used for filter offloads. Each filter is assigned a stats id, the MSBs of the id are used to determine which memory unit the filter should be offloaded to. The number of available memory units that could be used for filter offload is obtained from HW. A simple round robin technique is used to allocate and distribute the ids across memory units. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:52 -08:00
unsigned int active_mem_unit;
unsigned int total_mem_units;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
};
nfp: flower: add per repr private data for LAG offload Add a bitmap to each flower repr to track its state if it is enslaved by a bond. This LAG state may be different to the port state - for example, the port may be up but LAG state may be down due to the selection in an active/backup bond. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:51 -07:00
/**
* struct nfp_flower_repr_priv - Flower APP per-repr priv data
* @lag_port_flags: Extended port flags to record lag state of repr
*/
struct nfp_flower_repr_priv {
unsigned long lag_port_flags;
};
nfp: extend flower add flow offload Extends the flower flow add function by calculating which match fields are present in the flower offload structure and allocating the appropriate space to describe these. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:14 +02:00
struct nfp_fl_key_ls {
u32 key_layer_two;
u8 key_layer;
int key_size;
};
struct nfp_fl_rule_metadata {
u8 key_len;
u8 mask_len;
u8 act_len;
u8 flags;
__be32 host_ctx_id;
__be64 host_cookie __packed;
__be64 flow_version __packed;
__be32 shortcut;
};
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
struct nfp_fl_stats {
u64 pkts;
u64 bytes;
u64 used;
};
nfp: extend flower add flow offload Extends the flower flow add function by calculating which match fields are present in the flower offload structure and allocating the appropriate space to describe these. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:14 +02:00
struct nfp_fl_payload {
struct nfp_fl_rule_metadata meta;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
unsigned long tc_flower_cookie;
nfp: flower: use rhashtable for flow caching Make use of relativistic hash tables for tracking flows instead of fixed sized hash tables. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:34 -07:00
struct rhash_head fl_node;
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
struct rcu_head rcu;
nfp: offload vxlan IPv4 endpoints of flower rules Maintain a list of IPv4 addresses used as the tunnel destination IP match fields in currently active flower rules. Offload the entire list of NFP_FL_IPV4_ADDRS_MAX (even if some are unused) when new IPs are added or removed. The NFP should only be aware of tunnel end points that are currently used by rules on the device Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:39 +02:00
__be32 nfp_tun_ipv4_addr;
nfp: flower: support offloading multiple rules with same cookie When multiple netdevs are attached to a tc offload block and register for callbacks, a rule added to the block will be propogated to all netdevs. Previously these were detected as duplicates (based on cookie) and rejected. Modify the rule nfp lookup function to optionally include an ingress netdev and a host context along with the cookie value when searching for a rule. When a new rule is passed to the driver, the netdev the rule is to be attached to is considered when searching for dublicates. When a stats update is received from HW, the host context is used alongside the cookie to map to the correct host rule. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-24 21:17:03 -07:00
struct net_device *ingress_dev;
nfp: extend flower add flow offload Extends the flower flow add function by calculating which match fields are present in the flower offload structure and allocating the appropriate space to describe these. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:14 +02:00
char *unmasked_data;
char *mask_data;
char *action_data;
};
nfp: flower: use rhashtable for flow caching Make use of relativistic hash tables for tracking flows instead of fixed sized hash tables. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-10-08 18:57:34 -07:00
extern const struct rhashtable_params nfp_flower_table_params;
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
struct nfp_fl_stats_frame {
__be32 stats_con_id;
__be32 pkt_count;
__be64 byte_count;
__be64 stats_cookie;
};
nfp: flower: support multiple memory units for filter offloads Adds support for multiple memory units which are used for filter offloads. Each filter is assigned a stats id, the MSBs of the id are used to determine which memory unit the filter should be offloaded to. The number of available memory units that could be used for filter offload is obtained from HW. A simple round robin technique is used to allocate and distribute the ids across memory units. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2019-01-15 19:06:52 -08:00
int nfp_flower_metadata_init(struct nfp_app *app, u64 host_ctx_count,
unsigned int host_ctx_split);
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
void nfp_flower_metadata_cleanup(struct nfp_app *app);
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
int nfp_flower_setup_tc(struct nfp_app *app, struct net_device *netdev,
net: sched: get rid of struct tc_to_netdev Get rid of struct tc_to_netdev which is now just unnecessary container and rather pass per-type structures down to drivers directly. Along with that, consolidate the naming of per-type structure variables in cls_*. Signed-off-by: Jiri Pirko <jiri@mellanox.com> Acked-by: Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-08-07 10:15:32 +02:00
enum tc_setup_type type, void *type_data);
nfp: flower: allow non repr netdev offload Previously the offload functions in NFP assumed that the ingress (or egress) netdev passed to them was an nfp repr. Modify the driver to permit the passing of non repr netdevs as the ingress device for an offload rule candidate. This may include devices such as tunnels. The driver should then base its offload decision on a combination of ingress device and egress port for a rule. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-09 21:21:27 -08:00
int nfp_flower_compile_flow_match(struct nfp_app *app,
struct tc_cls_flower_offload *flow,
nfp: extend flower matching capabilities Extends matching capabilities for flower offloads to include vlan, layer 2, layer 3 and layer 4 type matches. This includes both exact and wildcard matching. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:15 +02:00
struct nfp_fl_key_ls *key_ls,
struct net_device *netdev,
nfp: flower: compile Geneve match fields Compile Geneve match fields for offloading to the NFP. The addition of Geneve overflows the 8 bit key_layer field, so apply extended metadata to the match cmsg allowing up to 32 more key_layer fields. Rather than adding new Geneve blocks, move the vxlan code to generic ipv4 udp tunnel structs and use these for both vxlan and Geneve. Matches are only supported when specifically mentioning well known port 6081. Geneve tunnel options are not yet included in the match. Only offload Geneve if the fw supports it - include check for this. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-12-19 17:58:28 +01:00
struct nfp_fl_payload *nfp_flow,
enum nfp_flower_tun_type tun_type);
nfp: flower: compute link aggregation action If the egress device of an offloaded rule is a LAG port, then encode the output port to the NFP with a LAG identifier and the offloaded group ID. A prelag action is also offloaded which must be the first action of the series (although may appear after other pre-actions - e.g. tunnels). This causes the FW to check that it has the necessary information to output to the requested LAG port. If it does not, the packet is sent to the kernel before any other actions are applied to it. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:55 -07:00
int nfp_flower_compile_action(struct nfp_app *app,
struct tc_cls_flower_offload *flow,
nfp: add basic action capabilities to flower offloads Adds push vlan, pop vlan, output and drop action capabilities to flower offloads. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:16 +02:00
struct net_device *netdev,
struct nfp_fl_payload *nfp_flow);
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
int nfp_compile_flow_metadata(struct nfp_app *app,
struct tc_cls_flower_offload *flow,
nfp: flower: support offloading multiple rules with same cookie When multiple netdevs are attached to a tc offload block and register for callbacks, a rule added to the block will be propogated to all netdevs. Previously these were detected as duplicates (based on cookie) and rejected. Modify the rule nfp lookup function to optionally include an ingress netdev and a host context along with the cookie value when searching for a rule. When a new rule is passed to the driver, the netdev the rule is to be attached to is considered when searching for dublicates. When a stats update is received from HW, the host context is used alongside the cookie to map to the correct host rule. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-24 21:17:03 -07:00
struct nfp_fl_payload *nfp_flow,
struct net_device *netdev);
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
int nfp_modify_flow_metadata(struct nfp_app *app,
struct nfp_fl_payload *nfp_flow);
struct nfp_fl_payload *
nfp: flower: support offloading multiple rules with same cookie When multiple netdevs are attached to a tc offload block and register for callbacks, a rule added to the block will be propogated to all netdevs. Previously these were detected as duplicates (based on cookie) and rejected. Modify the rule nfp lookup function to optionally include an ingress netdev and a host context along with the cookie value when searching for a rule. When a new rule is passed to the driver, the netdev the rule is to be attached to is considered when searching for dublicates. When a stats update is received from HW, the host context is used alongside the cookie to map to the correct host rule. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-04-24 21:17:03 -07:00
nfp_flower_search_fl_table(struct nfp_app *app, unsigned long tc_flower_cookie,
nfp: flower: remove unnecessary code in flow lookup Recent changes to NFP mean that stats updates from fw to driver no longer require a flow lookup and (because egdev offload has been removed) the ingress netdev for a lookup is now always known. Remove obsolete code in a flow lookup that matches on host context and that allows for a netdev to be NULL. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-09 21:21:31 -08:00
struct net_device *netdev);
nfp: add metadata to each flow offload Adds metadata describing the mask id of each flow and keeps track of flows installed in hardware. Previously a flow could not be removed from hardware as there was no way of knowing if that a specific flow was installed. This is solved by storing the offloaded flows in a hash table. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:17 +02:00
struct nfp_fl_payload *
nfp_flower_remove_fl_table(struct nfp_app *app, unsigned long tc_flower_cookie);
nfp: extend flower matching capabilities Extends matching capabilities for flower offloads to include vlan, layer 2, layer 3 and layer 4 type matches. This includes both exact and wildcard matching. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:15 +02:00
nfp: add a stats handler for flower offloads Previously there was no way of updating flow rule stats after they have been offloaded to hardware. This is solved by keeping track of stats received from hardware and providing this to the TC handler on request. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:18 +02:00
void nfp_flower_rx_flow_stats(struct nfp_app *app, struct sk_buff *skb);
nfp: offload flower vxlan endpoint MAC addresses Generate a list of MAC addresses of netdevs that could be used as VXLAN tunnel end points. Give offloaded MACs an index for storage on the NFP in the ranges: 0x100-0x1ff physical port representors 0x200-0x2ff VF port representors 0x300-0x3ff other offloads (e.g. vxlan netdevs, ovs bridges) Assign phys and vf indexes based on unique 8 bit values in the port num. Maintain list of other netdevs to ensure same netdev is not offloaded twice and each gets a unique ID without exhausting the entries. Because the IDs are unique but constant for a netdev, any changes are implemented by overwriting the index on NFP. Signed-off-by: John Hurley <john.hurley@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:38 +02:00
int nfp_tunnel_config_start(struct nfp_app *app);
void nfp_tunnel_config_stop(struct nfp_app *app);
nfp: flower: use the common netdev notifier Use driver's common notifier for LAG and tunnel configuration. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-06 17:07:34 -08:00
int nfp_tunnel_mac_event_handler(struct nfp_app *app,
struct net_device *netdev,
unsigned long event, void *ptr);
nfp: offload vxlan IPv4 endpoints of flower rules Maintain a list of IPv4 addresses used as the tunnel destination IP match fields in currently active flower rules. Offload the entire list of NFP_FL_IPV4_ADDRS_MAX (even if some are unused) when new IPs are added or removed. The NFP should only be aware of tunnel end points that are currently used by rules on the device Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:39 +02:00
void nfp_tunnel_del_ipv4_off(struct nfp_app *app, __be32 ipv4);
void nfp_tunnel_add_ipv4_off(struct nfp_app *app, __be32 ipv4);
nfp: flower vxlan neighbour offload Receive a request when the NFP does not know the next hop for a packet that is to be encapsulated in a VXLAN tunnel. Do a route lookup, determine the next hop entry and update neighbour table on NFP. Monitor the kernel neighbour table for link changes and update NFP with relevant information. Overwrite routes with zero values on the NFP when they expire. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:40 +02:00
void nfp_tunnel_request_route(struct nfp_app *app, struct sk_buff *skb);
nfp: flower vxlan neighbour keep-alive Periodically receive messages containing the destination IPs of tunnels that have recently forwarded traffic. Update the neighbour entries 'used' value for these IPs next hop. This prevents the neighbour entry from expiring on timeout but rather signals an ARP to verify the connection. From an NFP perspective, packets will not fall back mid-flow unless the link is verified to be down. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:41 +02:00
void nfp_tunnel_keep_alive(struct nfp_app *app, struct sk_buff *skb);
nfp: flower: monitor and offload LAG groups Monitor LAG events via the NETDEV_CHANGEUPPER/NETDEV_CHANGELOWERSTATE notifiers to maintain a list of offloadable groups. Sync these groups with HW via a delayed workqueue to prevent excessive re-configuration. When the workqueue is triggered it may generate multiple control messages for different groups. These messages are linked via a batch ID and flags to indicate a new batch and the end of a batch. Update private data in each repr to track their LAG lower state flags. The state of a repr is used to determine the active netdevs that can be offloaded. For example, in active-backup mode, we only offload the netdev currently active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:53 -07:00
void nfp_flower_lag_init(struct nfp_fl_lag *lag);
void nfp_flower_lag_cleanup(struct nfp_fl_lag *lag);
int nfp_flower_lag_reset(struct nfp_fl_lag *lag);
nfp: flower: use the common netdev notifier Use driver's common notifier for LAG and tunnel configuration. Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-06 17:07:34 -08:00
int nfp_flower_lag_netdev_event(struct nfp_flower_priv *priv,
struct net_device *netdev,
unsigned long event, void *ptr);
nfp: flower: implement host cmsg handler for LAG Adds the control message handler to synchronize offloaded group config with that of the kernel. Such messages are sent from fw to driver and feature the following 3 flags: - Data: an attached cmsg could not be processed - store for retransmission - Xon: FW can accept new messages - retransmit any stored cmsgs - Sync: full sync requested so retransmit all kernel LAG group info Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:54 -07:00
bool nfp_flower_lag_unprocessed_msg(struct nfp_app *app, struct sk_buff *skb);
nfp: flower: compute link aggregation action If the egress device of an offloaded rule is a LAG port, then encode the output port to the NFP with a LAG identifier and the offloaded group ID. A prelag action is also offloaded which must be the first action of the series (although may appear after other pre-actions - e.g. tunnels). This causes the FW to check that it has the necessary information to output to the requested LAG port. If it does not, the packet is sent to the kernel before any other actions are applied to it. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-05-23 19:22:55 -07:00
int nfp_flower_lag_populate_pre_action(struct nfp_app *app,
struct net_device *master,
struct nfp_fl_pre_lag *pre_act);
int nfp_flower_lag_get_output_id(struct nfp_app *app,
struct net_device *master);
nfp: flower: offload tunnel decap rules via indirect TC blocks Previously, TC block tunnel decap rules were only offloaded when a callback was triggered through registration of the rules egress device. This meant that the driver had no access to the ingress netdev and so could not verify it was the same tunnel type that the rule implied. Register tunnel devices for indirect TC block offloads in NFP, giving access to new rules based on the ingress device rather than egress. Use this to verify the netdev type of VXLAN and Geneve based rules and offload the rules to HW if applicable. Tunnel registration is done via a netdev notifier. On notifier registration, this is triggered for already existing netdevs. This means that NFP can register for offloads from devices that exist before it is loaded (filter rules will be replayed from the TC core). Similarly, on notifier unregister, a call is triggered for each currently active netdev. This allows the driver to unregister any indirect block callbacks that may still be active. Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2018-11-09 21:21:29 -08:00
int nfp_flower_reg_indir_block_handler(struct nfp_app *app,
struct net_device *netdev,
unsigned long event);
nfp: offload flower vxlan endpoint MAC addresses Generate a list of MAC addresses of netdevs that could be used as VXLAN tunnel end points. Give offloaded MACs an index for storage on the NFP in the ranges: 0x100-0x1ff physical port representors 0x200-0x2ff VF port representors 0x300-0x3ff other offloads (e.g. vxlan netdevs, ovs bridges) Assign phys and vf indexes based on unique 8 bit values in the port num. Maintain list of other netdevs to ensure same netdev is not offloaded twice and each gets a unique ID without exhausting the entries. Because the IDs are unique but constant for a netdev, any changes are implemented by overwriting the index on NFP. Signed-off-by: John Hurley <john.hurley@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-09-25 12:23:38 +02:00
nfp: provide infrastructure for offloading flower based TC filters Adds a flower based TC offload handler for representor devices, this is in addition to the bpf based offload handler. The changes in this patch will be used in a follow-up patch to add tc flower offload to the NFP. The flower app enables tc offloads on representors by default. Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Signed-off-by: Simon Horman <simon.horman@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
2017-06-29 22:08:13 +02:00
#endif
Reference in New Issue Copy Permalink