[pull request][net 00/15] mlx5 fixes 2023-05-22

[pull request][net 00/15] mlx5 fixes 2023-05-22

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

This series provides bug fixes for the mlx5 driver.
Please pull and let me know if there is any problem.

Thanks,
Saeed.


The following changes since commit 18c40a1cc1d990c51381ef48cd93fdb31d5cd903:

  net/handshake: Fix sock->file allocation (2023-05-22 19:25:24 -0700)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-fixes-2023-05-22

for you to fetch changes up to 1da438c0ae02396dc5018b63237492cb5908608d:

  net/mlx5: Fix indexing of mlx5_irq (2023-05-22 22:38:06 -0700)

----------------------------------------------------------------
mlx5-fixes-2023-05-22

----------------------------------------------------------------
Erez Shitrit (1):
      net/mlx5: DR, Fix crc32 calculation to work on big-endian (BE) CPUs

Paul Blakey (1):
      net/mlx5e: TC, Fix using eswitch mapping in nic mode

Rahul Rameshbabu (1):
      net/mlx5e: Fix SQ wake logic in ptp napi_poll context

Roi Dayan (1):
      net/mlx5: Fix error message when failing to allocate device memory

Shay Drory (8):
      net/mlx5: Collect command failures data only for known commands
      net/mlx5: Handle pairing of E-switch via uplink un/load APIs
      net/mlx5: E-switch, Devcom, sync devcom events and devcom comp register
      net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device
      net/mlx5: Devcom, serialize devcom registration
      net/mlx5: Free irqs only on shutdown callback
      net/mlx5: Fix irq affinity management
      net/mlx5: Fix indexing of mlx5_irq

Vlad Buslov (2):
      net/mlx5e: Use correct encap attribute during invalidation
      net/mlx5e: Fix deadlock in tc route query code

Yevgeny Kliteynik (1):
      net/mlx5: DR, Check force-loopback RC QP capability independently from RoCE

 drivers/net/ethernet/mellanox/mlx5/core/cmd.c      |  3 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c   |  2 +
 .../ethernet/mellanox/mlx5/core/en/tc_tun_encap.c  |  4 +-
 drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h  |  2 +
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c    | 57 ++++++++++++------
 drivers/net/ethernet/mellanox/mlx5/core/en_tx.c    | 19 +++---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h  |  5 ++
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 16 +++--
 .../net/ethernet/mellanox/mlx5/core/lib/devcom.c   | 70 ++++++++++++++++------
 .../net/ethernet/mellanox/mlx5/core/lib/devcom.h   |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/main.c     |  2 +-
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h |  1 +
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  | 40 +++++++++++--
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |  4 +-
 .../ethernet/mellanox/mlx5/core/steering/dr_ste.c  |  3 +-
 include/linux/mlx5/mlx5_ifc.h                      |  4 +-
 17 files changed, 176 insertions(+), 59 deletions(-)

[net 01/15] net/mlx5: Collect command failures data only for known commands

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

DEVX can issue a general command, which is not used by mlx5 driver.
In case such command is failed, mlx5 is trying to collect the failure
data, However, mlx5 doesn't create a storage for this command, since
mlx5 doesn't use it. This lead to array-index-out-of-bounds error.

Fix it by checking whether the command is known before collecting the
failure data.

Fixes: 34f46ae0d4b3 ("net/mlx5: Add command failures data to debugfs")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/cmd.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
index d53de39539a8..d532883b42d7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cmd.c
@@ -1920,9 +1920,10 @@ static void mlx5_cmd_err_trace(struct mlx5_core_dev *dev, u16 opcode, u16 op_mod
 static void cmd_status_log(struct mlx5_core_dev *dev, u16 opcode, u8 status,
 			   u32 syndrome, int err)
 {
+	const char *namep = mlx5_command_str(opcode);
 	struct mlx5_cmd_stats *stats;
 
-	if (!err)
+	if (!err || !(strcmp(namep, "unknown command opcode")))
 		return;
 
 	stats = &dev->cmd.stats[opcode];
-- 
2.40.1

Re: [net 01/15] net/mlx5: Collect command failures data only for known commands

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (main)
by Saeed Mahameed <saeedm@nvidia.com>:

On Mon, 22 May 2023 22:42:28 -0700 you wrote:
> From: Shay Drory <shayd@nvidia.com>
> 
> DEVX can issue a general command, which is not used by mlx5 driver.
> In case such command is failed, mlx5 is trying to collect the failure
> data, However, mlx5 doesn't create a storage for this command, since
> mlx5 doesn't use it. This lead to array-index-out-of-bounds error.
> 
> [...]

Here is the summary with links:
  - [net,01/15] net/mlx5: Collect command failures data only for known commands
    https://git.kernel.org/netdev/net/c/2a0a935fb64e
  - [net,02/15] net/mlx5: Handle pairing of E-switch via uplink un/load APIs
    https://git.kernel.org/netdev/net/c/2be5bd42a5bb
  - [net,03/15] net/mlx5: DR, Fix crc32 calculation to work on big-endian (BE) CPUs
    https://git.kernel.org/netdev/net/c/1e5daf5565b6
  - [net,04/15] net/mlx5: DR, Check force-loopback RC QP capability independently from RoCE
    https://git.kernel.org/netdev/net/c/c7dd225bc224
  - [net,05/15] net/mlx5e: Use correct encap attribute during invalidation
    https://git.kernel.org/netdev/net/c/be071cdb167f
  - [net,06/15] net/mlx5: Fix error message when failing to allocate device memory
    https://git.kernel.org/netdev/net/c/a65735148e03
  - [net,07/15] net/mlx5e: Fix deadlock in tc route query code
    https://git.kernel.org/netdev/net/c/691c041bf208
  - [net,08/15] net/mlx5e: Fix SQ wake logic in ptp napi_poll context
    https://git.kernel.org/netdev/net/c/7aa503801916
  - [net,09/15] net/mlx5e: TC, Fix using eswitch mapping in nic mode
    https://git.kernel.org/netdev/net/c/dfa1e46d6093
  - [net,10/15] net/mlx5: E-switch, Devcom, sync devcom events and devcom comp register
    https://git.kernel.org/netdev/net/c/8c253dfc89ef
  - [net,11/15] net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device
    https://git.kernel.org/netdev/net/c/af87194352ca
  - [net,12/15] net/mlx5: Devcom, serialize devcom registration
    https://git.kernel.org/netdev/net/c/1f893f57a3bf
  - [net,13/15] net/mlx5: Free irqs only on shutdown callback
    https://git.kernel.org/netdev/net/c/9c2d08010963
  - [net,14/15] net/mlx5: Fix irq affinity management
    https://git.kernel.org/netdev/net/c/ef8c063cf88e
  - [net,15/15] net/mlx5: Fix indexing of mlx5_irq
    https://git.kernel.org/netdev/net/c/1da438c0ae02

You are awesome, thank you!
-- 
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html

[net 02/15] net/mlx5: Handle pairing of E-switch via uplink un/load APIs

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

In case user switch a device from switchdev mode to legacy mode, mlx5
first unpair the E-switch and afterwards unload the uplink vport.
From the other hand, in case user remove or reload a device, mlx5
first unload the uplink vport and afterwards unpair the E-switch.

The latter is causing a bug[1], hence, handle pairing of E-switch as
part of uplink un/load APIs.

[1]
In case VF_LAG is used, every tc fdb flow is duplicated to the peer
esw. However, the original esw keeps a pointer to this duplicated
flow, not the peer esw.
e.g.: if user create tc fdb flow over esw0, the flow is duplicated
over esw1, in FW/HW, but in SW, esw0 keeps a pointer to the duplicated
flow.
During module unload while a peer tc fdb flow is still offloaded, in
case the first device to be removed is the peer device (esw1 in the
example above), the peer net-dev is destroyed, and so the mlx5e_priv
is memset to 0.
Afterwards, the peer device is trying to unpair himself from the
original device (esw0 in the example above). Unpair API invoke the
original device to clear peer flow from its eswitch (esw0), but the
peer flow, which is stored over the original eswitch (esw0), is
trying to use the peer mlx5e_priv, which is memset to 0 and result in
bellow kernel-oops.

[  157.964081 ] BUG: unable to handle page fault for address: 000000000002ce60
[  157.964662 ] #PF: supervisor read access in kernel mode
[  157.965123 ] #PF: error_code(0x0000) - not-present page
[  157.965582 ] PGD 0 P4D 0
[  157.965866 ] Oops: 0000 [#1] SMP
[  157.967670 ] RIP: 0010:mlx5e_tc_del_fdb_flow+0x48/0x460 [mlx5_core]
[  157.976164 ] Call Trace:
[  157.976437 ]  <TASK>
[  157.976690 ]  __mlx5e_tc_del_fdb_peer_flow+0xe6/0x100 [mlx5_core]
[  157.977230 ]  mlx5e_tc_clean_fdb_peer_flows+0x67/0x90 [mlx5_core]
[  157.977767 ]  mlx5_esw_offloads_unpair+0x2d/0x1e0 [mlx5_core]
[  157.984653 ]  mlx5_esw_offloads_devcom_event+0xbf/0x130 [mlx5_core]
[  157.985212 ]  mlx5_devcom_send_event+0xa3/0xb0 [mlx5_core]
[  157.985714 ]  esw_offloads_disable+0x5a/0x110 [mlx5_core]
[  157.986209 ]  mlx5_eswitch_disable_locked+0x152/0x170 [mlx5_core]
[  157.986757 ]  mlx5_eswitch_disable+0x51/0x80 [mlx5_core]
[  157.987248 ]  mlx5_unload+0x2a/0xb0 [mlx5_core]
[  157.987678 ]  mlx5_uninit_one+0x5f/0xd0 [mlx5_core]
[  157.988127 ]  remove_one+0x64/0xe0 [mlx5_core]
[  157.988549 ]  pci_device_remove+0x31/0xa0
[  157.988933 ]  device_release_driver_internal+0x18f/0x1f0
[  157.989402 ]  driver_detach+0x3f/0x80
[  157.989754 ]  bus_remove_driver+0x70/0xf0
[  157.990129 ]  pci_unregister_driver+0x34/0x90
[  157.990537 ]  mlx5_cleanup+0xc/0x1c [mlx5_core]
[  157.990972 ]  __x64_sys_delete_module+0x15a/0x250
[  157.991398 ]  ? exit_to_user_mode_prepare+0xea/0x110
[  157.991840 ]  do_syscall_64+0x3d/0x90
[  157.992198 ]  entry_SYSCALL_64_after_hwframe+0x46/0xb0

Fixes: 04de7dda7394 ("net/mlx5e: Infrastructure for duplicated offloading of TC flows")
Fixes: 1418ddd96afd ("net/mlx5e: Duplicate offloaded TC eswitch rules under uplink LAG")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_tc.c            | 4 +++-
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h          | 4 ++++
 drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 7 ++-----
 3 files changed, 9 insertions(+), 6 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 728b82ce4031..65fe40f55d84 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -5301,6 +5301,8 @@ int mlx5e_tc_esw_init(struct mlx5_rep_uplink_priv *uplink_priv)
 		goto err_action_counter;
 	}
 
+	mlx5_esw_offloads_devcom_init(esw);
+
 	return 0;
 
 err_action_counter:
@@ -5329,7 +5331,7 @@ void mlx5e_tc_esw_cleanup(struct mlx5_rep_uplink_priv *uplink_priv)
 	priv = netdev_priv(rpriv->netdev);
 	esw = priv->mdev->priv.eswitch;
 
-	mlx5e_tc_clean_fdb_peer_flows(esw);
+	mlx5_esw_offloads_devcom_cleanup(esw);
 
 	mlx5e_tc_tun_cleanup(uplink_priv->encap);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 1a042c981713..9f007c5438ee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -369,6 +369,8 @@ int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs);
 void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf);
 void mlx5_eswitch_disable_locked(struct mlx5_eswitch *esw);
 void mlx5_eswitch_disable(struct mlx5_eswitch *esw);
+void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw);
+void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw);
 int mlx5_eswitch_set_vport_mac(struct mlx5_eswitch *esw,
 			       u16 vport, const u8 *mac);
 int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw,
@@ -767,6 +769,8 @@ static inline void mlx5_eswitch_cleanup(struct mlx5_eswitch *esw) {}
 static inline int mlx5_eswitch_enable(struct mlx5_eswitch *esw, int num_vfs) { return 0; }
 static inline void mlx5_eswitch_disable_sriov(struct mlx5_eswitch *esw, bool clear_vf) {}
 static inline void mlx5_eswitch_disable(struct mlx5_eswitch *esw) {}
+static inline void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw) {}
+static inline void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw) {}
 static inline bool mlx5_eswitch_is_funcs_handler(struct mlx5_core_dev *dev) { return false; }
 static inline
 int mlx5_eswitch_set_vport_state(struct mlx5_eswitch *esw, u16 vport, int link_state) { return 0; }
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 69215ffb9999..7c34c7cf506f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2779,7 +2779,7 @@ static int mlx5_esw_offloads_devcom_event(int event,
 	return err;
 }
 
-static void esw_offloads_devcom_init(struct mlx5_eswitch *esw)
+void mlx5_esw_offloads_devcom_init(struct mlx5_eswitch *esw)
 {
 	struct mlx5_devcom *devcom = esw->dev->priv.devcom;
 
@@ -2802,7 +2802,7 @@ static void esw_offloads_devcom_init(struct mlx5_eswitch *esw)
 			       ESW_OFFLOADS_DEVCOM_PAIR, esw);
 }
 
-static void esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw)
+void mlx5_esw_offloads_devcom_cleanup(struct mlx5_eswitch *esw)
 {
 	struct mlx5_devcom *devcom = esw->dev->priv.devcom;
 
@@ -3250,8 +3250,6 @@ int esw_offloads_enable(struct mlx5_eswitch *esw)
 	if (err)
 		goto err_vports;
 
-	esw_offloads_devcom_init(esw);
-
 	return 0;
 
 err_vports:
@@ -3292,7 +3290,6 @@ static int esw_offloads_stop(struct mlx5_eswitch *esw,
 
 void esw_offloads_disable(struct mlx5_eswitch *esw)
 {
-	esw_offloads_devcom_cleanup(esw);
 	mlx5_eswitch_disable_pf_vf_vports(esw);
 	esw_offloads_unload_rep(esw, MLX5_VPORT_UPLINK);
 	esw_set_passing_vport_metadata(esw, false);
-- 
2.40.1

[net 03/15] net/mlx5: DR, Fix crc32 calculation to work on big-endian (BE) CPUs

[Saeed Mahameed]

From: Erez Shitrit <erezsh@nvidia.com>

When calculating crc for hash index we use the function crc32 that
calculates for little-endian (LE) arch.
Then we convert it to network endianness using htonl(), but it's wrong
to do the conversion in BE archs since the crc32 value is already LE.

The solution is to switch the bytes from the crc result for all types
of arc.

Fixes: 40416d8ede65 ("net/mlx5: DR, Replace CRC32 implementation to use kernel lib")
Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Alex Vesker <valex@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
index 9413aaf51251..e94fbb015efa 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c
@@ -15,7 +15,8 @@ static u32 dr_ste_crc32_calc(const void *input_data, size_t length)
 {
 	u32 crc = crc32(0, input_data, length);
 
-	return (__force u32)htonl(crc);
+	return (__force u32)((crc >> 24) & 0xff) | ((crc << 8) & 0xff0000) |
+			    ((crc >> 8) & 0xff00) | ((crc << 24) & 0xff000000);
 }
 
 bool mlx5dr_ste_supp_ttl_cs_recalc(struct mlx5dr_cmd_caps *caps)
-- 
2.40.1

[net 04/15] net/mlx5: DR, Check force-loopback RC QP capability independently from RoCE

[Saeed Mahameed]

From: Yevgeny Kliteynik <kliteyn@nvidia.com>

SW Steering uses RC QP for writing STEs to ICM. This writingis done in LB
(loopback), and FL (force-loopback) QP is preferred for performance. FL is
available when RoCE is enabled or disabled based on RoCE caps.
This patch adds reading of FL capability from HCA caps in addition to the
existing reading from RoCE caps, thus fixing the case where we didn't
have loopback enabled when RoCE was disabled.

Fixes: 7304d603a57a ("net/mlx5: DR, Add support for force-loopback QP")
Signed-off-by: Itamar Gozlan <igozlan@nvidia.com>
Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c | 4 +++-
 include/linux/mlx5/mlx5_ifc.h                             | 4 +++-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
index 3835ba3f4dda..1aa525e509f1 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c
@@ -117,6 +117,8 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
 	caps->gvmi		= MLX5_CAP_GEN(mdev, vhca_id);
 	caps->flex_protocols	= MLX5_CAP_GEN(mdev, flex_parser_protocols);
 	caps->sw_format_ver	= MLX5_CAP_GEN(mdev, steering_format_version);
+	caps->roce_caps.fl_rc_qp_when_roce_disabled =
+		MLX5_CAP_GEN(mdev, fl_rc_qp_when_roce_disabled);
 
 	if (MLX5_CAP_GEN(mdev, roce)) {
 		err = dr_cmd_query_nic_vport_roce_en(mdev, 0, &roce_en);
@@ -124,7 +126,7 @@ int mlx5dr_cmd_query_device(struct mlx5_core_dev *mdev,
 			return err;
 
 		caps->roce_caps.roce_en = roce_en;
-		caps->roce_caps.fl_rc_qp_when_roce_disabled =
+		caps->roce_caps.fl_rc_qp_when_roce_disabled |=
 			MLX5_CAP_ROCE(mdev, fl_rc_qp_when_roce_disabled);
 		caps->roce_caps.fl_rc_qp_when_roce_enabled =
 			MLX5_CAP_ROCE(mdev, fl_rc_qp_when_roce_enabled);
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index dc5e2cb302a5..b89778d0d326 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1705,7 +1705,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         rc[0x1];
 
 	u8         uar_4k[0x1];
-	u8         reserved_at_241[0x9];
+	u8         reserved_at_241[0x7];
+	u8         fl_rc_qp_when_roce_disabled[0x1];
+	u8         regexp_params[0x1];
 	u8         uar_sz[0x6];
 	u8         port_selection_cap[0x1];
 	u8         reserved_at_248[0x1];
-- 
2.40.1

[net 05/15] net/mlx5e: Use correct encap attribute during invalidation

[Saeed Mahameed]

From: Vlad Buslov <vladbu@nvidia.com>

With introduction of post action infrastructure most of the users of encap
attribute had been modified in order to obtain the correct attribute by
calling mlx5e_tc_get_encap_attr() helper instead of assuming encap action
is always on default attribute. However, the cited commit didn't modify
mlx5e_invalidate_encap() which prevents it from destroying correct modify
header action which leads to a warning [0]. Fix the issue by using correct
attribute.

[0]:

Feb 21 09:47:35 c-237-177-40-045 kernel: WARNING: CPU: 17 PID: 654 at drivers/net/ethernet/mellanox/mlx5/core/en_tc.c:684 mlx5e_tc_attach_mod_hdr+0x1cc/0x230 [mlx5_core]
Feb 21 09:47:35 c-237-177-40-045 kernel: RIP: 0010:mlx5e_tc_attach_mod_hdr+0x1cc/0x230 [mlx5_core]
Feb 21 09:47:35 c-237-177-40-045 kernel: Call Trace:
Feb 21 09:47:35 c-237-177-40-045 kernel:  <TASK>
Feb 21 09:47:35 c-237-177-40-045 kernel:  mlx5e_tc_fib_event_work+0x8e3/0x1f60 [mlx5_core]
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? mlx5e_take_all_encap_flows+0xe0/0xe0 [mlx5_core]
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? lock_downgrade+0x6d0/0x6d0
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? lockdep_hardirqs_on_prepare+0x273/0x3f0
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? lockdep_hardirqs_on_prepare+0x273/0x3f0
Feb 21 09:47:35 c-237-177-40-045 kernel:  process_one_work+0x7c2/0x1310
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? lockdep_hardirqs_on_prepare+0x3f0/0x3f0
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? pwq_dec_nr_in_flight+0x230/0x230
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? rwlock_bug.part.0+0x90/0x90
Feb 21 09:47:35 c-237-177-40-045 kernel:  worker_thread+0x59d/0xec0
Feb 21 09:47:35 c-237-177-40-045 kernel:  ? __kthread_parkme+0xd9/0x1d0

Fixes: 8300f225268b ("net/mlx5e: Create new flow attr for multi table actions")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
index 20c2d2ecaf93..6a052c6cfc15 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun_encap.c
@@ -1369,11 +1369,13 @@ static void mlx5e_invalidate_encap(struct mlx5e_priv *priv,
 	struct mlx5e_tc_flow *flow;
 
 	list_for_each_entry(flow, encap_flows, tmp_list) {
-		struct mlx5_flow_attr *attr = flow->attr;
 		struct mlx5_esw_flow_attr *esw_attr;
+		struct mlx5_flow_attr *attr;
 
 		if (!mlx5e_is_offloaded_flow(flow))
 			continue;
+
+		attr = mlx5e_tc_get_encap_attr(flow);
 		esw_attr = attr->esw_attr;
 
 		if (flow_flag_test(flow, SLOW))
-- 
2.40.1

[net 06/15] net/mlx5: Fix error message when failing to allocate device memory

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

Fix spacing for the error and also the correct error code pointer.

Fixes: c9b9dcb430b3 ("net/mlx5: Move device memory management to mlx5_core")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index 995eb2d5ace0..a7eb65cd0bdd 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1049,7 +1049,7 @@ static int mlx5_init_once(struct mlx5_core_dev *dev)
 
 	dev->dm = mlx5_dm_create(dev);
 	if (IS_ERR(dev->dm))
-		mlx5_core_warn(dev, "Failed to init device memory%d\n", err);
+		mlx5_core_warn(dev, "Failed to init device memory %ld\n", PTR_ERR(dev->dm));
 
 	dev->tracer = mlx5_fw_tracer_create(dev);
 	dev->hv_vhca = mlx5_hv_vhca_create(dev);
-- 
2.40.1

[net 07/15] net/mlx5e: Fix deadlock in tc route query code

[Saeed Mahameed]

From: Vlad Buslov <vladbu@nvidia.com>

Cited commit causes ABBA deadlock[0] when peer flows are created while
holding the devcom rw semaphore. Due to peer flows offload implementation
the lock is taken much higher up the call chain and there is no obvious way
to easily fix the deadlock. Instead, since tc route query code needs the
peer eswitch structure only to perform a lookup in xarray and doesn't
perform any sleeping operations with it, refactor the code for lockless
execution in following ways:

- RCUify the devcom 'data' pointer. When resetting the pointer
synchronously wait for RCU grace period before returning. This is fine
since devcom is currently only used for synchronization of
pairing/unpairing of eswitches which is rare and already expensive as-is.

- Wrap all usages of 'paired' boolean in {READ|WRITE}_ONCE(). The flag has
already been used in some unlocked contexts without proper
annotations (e.g. users of mlx5_devcom_is_paired() function), but it wasn't
an issue since all relevant code paths checked it again after obtaining the
devcom semaphore. Now it is also used by mlx5_devcom_get_peer_data_rcu() as
"best effort" check to return NULL when devcom is being unpaired. Note that
while RCU read lock doesn't prevent the unpaired flag from being changed
concurrently it still guarantees that reader can continue to use 'data'.

- Refactor mlx5e_tc_query_route_vport() function to use new
mlx5_devcom_get_peer_data_rcu() API which fixes the deadlock.

[0]:

[  164.599612] ======================================================
[  164.600142] WARNING: possible circular locking dependency detected
[  164.600667] 6.3.0-rc3+ #1 Not tainted
[  164.601021] ------------------------------------------------------
[  164.601557] handler1/3456 is trying to acquire lock:
[  164.601998] ffff88811f1714b0 (&esw->offloads.encap_tbl_lock){+.+.}-{3:3}, at: mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.603078]
               but task is already holding lock:
[  164.603617] ffff88810137fc98 (&comp->sem){++++}-{3:3}, at: mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core]
[  164.604459]
               which lock already depends on the new lock.

[  164.605190]
               the existing dependency chain (in reverse order) is:
[  164.605848]
               -> #1 (&comp->sem){++++}-{3:3}:
[  164.606380]        down_read+0x39/0x50
[  164.606772]        mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core]
[  164.607336]        mlx5e_tc_query_route_vport+0x86/0xc0 [mlx5_core]
[  164.607914]        mlx5e_tc_tun_route_lookup+0x1a4/0x1d0 [mlx5_core]
[  164.608495]        mlx5e_attach_decap_route+0xc6/0x1e0 [mlx5_core]
[  164.609063]        mlx5e_tc_add_fdb_flow+0x1ea/0x360 [mlx5_core]
[  164.609627]        __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core]
[  164.610175]        mlx5e_configure_flower+0x952/0x1a20 [mlx5_core]
[  164.610741]        tc_setup_cb_add+0xd4/0x200
[  164.611146]        fl_hw_replace_filter+0x14c/0x1f0 [cls_flower]
[  164.611661]        fl_change+0xc95/0x18a0 [cls_flower]
[  164.612116]        tc_new_tfilter+0x3fc/0xd20
[  164.612516]        rtnetlink_rcv_msg+0x418/0x5b0
[  164.612936]        netlink_rcv_skb+0x54/0x100
[  164.613339]        netlink_unicast+0x190/0x250
[  164.613746]        netlink_sendmsg+0x245/0x4a0
[  164.614150]        sock_sendmsg+0x38/0x60
[  164.614522]        ____sys_sendmsg+0x1d0/0x1e0
[  164.614934]        ___sys_sendmsg+0x80/0xc0
[  164.615320]        __sys_sendmsg+0x51/0x90
[  164.615701]        do_syscall_64+0x3d/0x90
[  164.616083]        entry_SYSCALL_64_after_hwframe+0x46/0xb0
[  164.616568]
               -> #0 (&esw->offloads.encap_tbl_lock){+.+.}-{3:3}:
[  164.617210]        __lock_acquire+0x159e/0x26e0
[  164.617638]        lock_acquire+0xc2/0x2a0
[  164.618018]        __mutex_lock+0x92/0xcd0
[  164.618401]        mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.618943]        post_process_attr+0x153/0x2d0 [mlx5_core]
[  164.619471]        mlx5e_tc_add_fdb_flow+0x164/0x360 [mlx5_core]
[  164.620021]        __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core]
[  164.620564]        mlx5e_configure_flower+0xe33/0x1a20 [mlx5_core]
[  164.621125]        tc_setup_cb_add+0xd4/0x200
[  164.621531]        fl_hw_replace_filter+0x14c/0x1f0 [cls_flower]
[  164.622047]        fl_change+0xc95/0x18a0 [cls_flower]
[  164.622500]        tc_new_tfilter+0x3fc/0xd20
[  164.622906]        rtnetlink_rcv_msg+0x418/0x5b0
[  164.623324]        netlink_rcv_skb+0x54/0x100
[  164.623727]        netlink_unicast+0x190/0x250
[  164.624138]        netlink_sendmsg+0x245/0x4a0
[  164.624544]        sock_sendmsg+0x38/0x60
[  164.624919]        ____sys_sendmsg+0x1d0/0x1e0
[  164.625340]        ___sys_sendmsg+0x80/0xc0
[  164.625731]        __sys_sendmsg+0x51/0x90
[  164.626117]        do_syscall_64+0x3d/0x90
[  164.626502]        entry_SYSCALL_64_after_hwframe+0x46/0xb0
[  164.626995]
               other info that might help us debug this:

[  164.627725]  Possible unsafe locking scenario:

[  164.628268]        CPU0                    CPU1
[  164.628683]        ----                    ----
[  164.629098]   lock(&comp->sem);
[  164.629421]                                lock(&esw->offloads.encap_tbl_lock);
[  164.630066]                                lock(&comp->sem);
[  164.630555]   lock(&esw->offloads.encap_tbl_lock);
[  164.630993]
                *** DEADLOCK ***

[  164.631575] 3 locks held by handler1/3456:
[  164.631962]  #0: ffff888124b75130 (&block->cb_lock){++++}-{3:3}, at: tc_setup_cb_add+0x5b/0x200
[  164.632703]  #1: ffff888116e512b8 (&esw->mode_lock){++++}-{3:3}, at: mlx5_esw_hold+0x39/0x50 [mlx5_core]
[  164.633552]  #2: ffff88810137fc98 (&comp->sem){++++}-{3:3}, at: mlx5_devcom_get_peer_data+0x37/0x80 [mlx5_core]
[  164.634435]
               stack backtrace:
[  164.634883] CPU: 17 PID: 3456 Comm: handler1 Not tainted 6.3.0-rc3+ #1
[  164.635431] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[  164.636340] Call Trace:
[  164.636616]  <TASK>
[  164.636863]  dump_stack_lvl+0x47/0x70
[  164.637217]  check_noncircular+0xfe/0x110
[  164.637601]  __lock_acquire+0x159e/0x26e0
[  164.637977]  ? mlx5_cmd_set_fte+0x5b0/0x830 [mlx5_core]
[  164.638472]  lock_acquire+0xc2/0x2a0
[  164.638828]  ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.639339]  ? lock_is_held_type+0x98/0x110
[  164.639728]  __mutex_lock+0x92/0xcd0
[  164.640074]  ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.640576]  ? __lock_acquire+0x382/0x26e0
[  164.640958]  ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.641468]  ? mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.641965]  mlx5e_attach_encap+0xd8/0x8b0 [mlx5_core]
[  164.642454]  ? lock_release+0xbf/0x240
[  164.642819]  post_process_attr+0x153/0x2d0 [mlx5_core]
[  164.643318]  mlx5e_tc_add_fdb_flow+0x164/0x360 [mlx5_core]
[  164.643835]  __mlx5e_add_fdb_flow+0x2d2/0x430 [mlx5_core]
[  164.644340]  mlx5e_configure_flower+0xe33/0x1a20 [mlx5_core]
[  164.644862]  ? lock_acquire+0xc2/0x2a0
[  164.645219]  tc_setup_cb_add+0xd4/0x200
[  164.645588]  fl_hw_replace_filter+0x14c/0x1f0 [cls_flower]
[  164.646067]  fl_change+0xc95/0x18a0 [cls_flower]
[  164.646488]  tc_new_tfilter+0x3fc/0xd20
[  164.646861]  ? tc_del_tfilter+0x810/0x810
[  164.647236]  rtnetlink_rcv_msg+0x418/0x5b0
[  164.647621]  ? rtnl_setlink+0x160/0x160
[  164.647982]  netlink_rcv_skb+0x54/0x100
[  164.648348]  netlink_unicast+0x190/0x250
[  164.648722]  netlink_sendmsg+0x245/0x4a0
[  164.649090]  sock_sendmsg+0x38/0x60
[  164.649434]  ____sys_sendmsg+0x1d0/0x1e0
[  164.649804]  ? copy_msghdr_from_user+0x6d/0xa0
[  164.650213]  ___sys_sendmsg+0x80/0xc0
[  164.650563]  ? lock_acquire+0xc2/0x2a0
[  164.650926]  ? lock_acquire+0xc2/0x2a0
[  164.651286]  ? __fget_files+0x5/0x190
[  164.651644]  ? find_held_lock+0x2b/0x80
[  164.652006]  ? __fget_files+0xb9/0x190
[  164.652365]  ? lock_release+0xbf/0x240
[  164.652723]  ? __fget_files+0xd3/0x190
[  164.653079]  __sys_sendmsg+0x51/0x90
[  164.653435]  do_syscall_64+0x3d/0x90
[  164.653784]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[  164.654229] RIP: 0033:0x7f378054f8bd
[  164.654577] Code: 28 89 54 24 1c 48 89 74 24 10 89 7c 24 08 e8 6a c3 f4 ff 8b 54 24 1c 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 33 44 89 c7 48 89 44 24 08 e8 be c3 f4 ff 48
[  164.656041] RSP: 002b:00007f377fa114b0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e
[  164.656701] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007f378054f8bd
[  164.657297] RDX: 0000000000000000 RSI: 00007f377fa11540 RDI: 0000000000000014
[  164.657885] RBP: 00007f377fa12278 R08: 0000000000000000 R09: 000000000000015c
[  164.658472] R10: 00007f377fa123d0 R11: 0000000000000293 R12: 0000560962d99bd0
[  164.665317] R13: 0000000000000000 R14: 0000560962d99bd0 R15: 00007f377fa11540

Fixes: f9d196bd632b ("net/mlx5e: Use correct eswitch for stack devices with lag")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Reviewed-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 19 ++++----
 .../ethernet/mellanox/mlx5/core/lib/devcom.c  | 48 ++++++++++++++-----
 .../ethernet/mellanox/mlx5/core/lib/devcom.h  |  1 +
 3 files changed, 48 insertions(+), 20 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 65fe40f55d84..416ab6b6da97 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -1665,11 +1665,9 @@ bool mlx5e_tc_is_vf_tunnel(struct net_device *out_dev, struct net_device *route_
 int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *route_dev, u16 *vport)
 {
 	struct mlx5e_priv *out_priv, *route_priv;
-	struct mlx5_devcom *devcom = NULL;
 	struct mlx5_core_dev *route_mdev;
 	struct mlx5_eswitch *esw;
 	u16 vhca_id;
-	int err;
 
 	out_priv = netdev_priv(out_dev);
 	esw = out_priv->mdev->priv.eswitch;
@@ -1678,6 +1676,9 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro
 
 	vhca_id = MLX5_CAP_GEN(route_mdev, vhca_id);
 	if (mlx5_lag_is_active(out_priv->mdev)) {
+		struct mlx5_devcom *devcom;
+		int err;
+
 		/* In lag case we may get devices from different eswitch instances.
 		 * If we failed to get vport num, it means, mostly, that we on the wrong
 		 * eswitch.
@@ -1686,16 +1687,16 @@ int mlx5e_tc_query_route_vport(struct net_device *out_dev, struct net_device *ro
 		if (err != -ENOENT)
 			return err;
 
+		rcu_read_lock();
 		devcom = out_priv->mdev->priv.devcom;
-		esw = mlx5_devcom_get_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
-		if (!esw)
-			return -ENODEV;
+		esw = mlx5_devcom_get_peer_data_rcu(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
+		err = esw ? mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport) : -ENODEV;
+		rcu_read_unlock();
+
+		return err;
 	}
 
-	err = mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport);
-	if (devcom)
-		mlx5_devcom_release_peer_data(devcom, MLX5_DEVCOM_ESW_OFFLOADS);
-	return err;
+	return mlx5_eswitch_vhca_id_to_vport(esw, vhca_id, vport);
 }
 
 static int
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
index adefde3ea941..070d55f13419 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
@@ -13,7 +13,7 @@ static LIST_HEAD(devcom_list);
 
 struct mlx5_devcom_component {
 	struct {
-		void *data;
+		void __rcu *data;
 	} device[MLX5_DEVCOM_PORTS_SUPPORTED];
 
 	mlx5_devcom_event_handler_t handler;
@@ -162,7 +162,7 @@ void mlx5_devcom_register_component(struct mlx5_devcom *devcom,
 	comp = &devcom->priv->components[id];
 	down_write(&comp->sem);
 	comp->handler = handler;
-	comp->device[devcom->idx].data = data;
+	rcu_assign_pointer(comp->device[devcom->idx].data, data);
 	up_write(&comp->sem);
 }
 
@@ -176,8 +176,9 @@ void mlx5_devcom_unregister_component(struct mlx5_devcom *devcom,
 
 	comp = &devcom->priv->components[id];
 	down_write(&comp->sem);
-	comp->device[devcom->idx].data = NULL;
+	RCU_INIT_POINTER(comp->device[devcom->idx].data, NULL);
 	up_write(&comp->sem);
+	synchronize_rcu();
 }
 
 int mlx5_devcom_send_event(struct mlx5_devcom *devcom,
@@ -193,12 +194,15 @@ int mlx5_devcom_send_event(struct mlx5_devcom *devcom,
 
 	comp = &devcom->priv->components[id];
 	down_write(&comp->sem);
-	for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++)
-		if (i != devcom->idx && comp->device[i].data) {
-			err = comp->handler(event, comp->device[i].data,
-					    event_data);
+	for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++) {
+		void *data = rcu_dereference_protected(comp->device[i].data,
+						       lockdep_is_held(&comp->sem));
+
+		if (i != devcom->idx && data) {
+			err = comp->handler(event, data, event_data);
 			break;
 		}
+	}
 
 	up_write(&comp->sem);
 	return err;
@@ -213,7 +217,7 @@ void mlx5_devcom_set_paired(struct mlx5_devcom *devcom,
 	comp = &devcom->priv->components[id];
 	WARN_ON(!rwsem_is_locked(&comp->sem));
 
-	comp->paired = paired;
+	WRITE_ONCE(comp->paired, paired);
 }
 
 bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom,
@@ -222,7 +226,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom,
 	if (IS_ERR_OR_NULL(devcom))
 		return false;
 
-	return devcom->priv->components[id].paired;
+	return READ_ONCE(devcom->priv->components[id].paired);
 }
 
 void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom,
@@ -236,7 +240,7 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom,
 
 	comp = &devcom->priv->components[id];
 	down_read(&comp->sem);
-	if (!comp->paired) {
+	if (!READ_ONCE(comp->paired)) {
 		up_read(&comp->sem);
 		return NULL;
 	}
@@ -245,7 +249,29 @@ void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom,
 		if (i != devcom->idx)
 			break;
 
-	return comp->device[i].data;
+	return rcu_dereference_protected(comp->device[i].data, lockdep_is_held(&comp->sem));
+}
+
+void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id)
+{
+	struct mlx5_devcom_component *comp;
+	int i;
+
+	if (IS_ERR_OR_NULL(devcom))
+		return NULL;
+
+	for (i = 0; i < MLX5_DEVCOM_PORTS_SUPPORTED; i++)
+		if (i != devcom->idx)
+			break;
+
+	comp = &devcom->priv->components[id];
+	/* This can change concurrently, however 'data' pointer will remain
+	 * valid for the duration of RCU read section.
+	 */
+	if (!READ_ONCE(comp->paired))
+		return NULL;
+
+	return rcu_dereference(comp->device[i].data);
 }
 
 void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom,
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
index 94313c18bb64..9a496f4722da 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.h
@@ -41,6 +41,7 @@ bool mlx5_devcom_is_paired(struct mlx5_devcom *devcom,
 
 void *mlx5_devcom_get_peer_data(struct mlx5_devcom *devcom,
 				enum mlx5_devcom_components id);
+void *mlx5_devcom_get_peer_data_rcu(struct mlx5_devcom *devcom, enum mlx5_devcom_components id);
 void mlx5_devcom_release_peer_data(struct mlx5_devcom *devcom,
 				   enum mlx5_devcom_components id);
 
-- 
2.40.1

[net 08/15] net/mlx5e: Fix SQ wake logic in ptp napi_poll context

[Saeed Mahameed]

From: Rahul Rameshbabu <rrameshbabu@nvidia.com>

Check in the mlx5e_ptp_poll_ts_cq context if the ptp tx sq should be woken
up. Before change, the ptp tx sq may never wake up if the ptp tx ts skb
fifo is full when mlx5e_poll_tx_cq checks if the queue should be woken up.

Fixes: 1880bc4e4a96 ("net/mlx5e: Add TX port timestamp support")
Signed-off-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en/ptp.c  |  2 ++
 .../net/ethernet/mellanox/mlx5/core/en/txrx.h |  2 ++
 .../net/ethernet/mellanox/mlx5/core/en_tx.c   | 19 ++++++++++++-------
 3 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
index eb5abd0e55d9..3cbebfba582b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/ptp.c
@@ -175,6 +175,8 @@ static bool mlx5e_ptp_poll_ts_cq(struct mlx5e_cq *cq, int budget)
 	/* ensure cq space is freed before enabling more cqes */
 	wmb();
 
+	mlx5e_txqsq_wake(&ptpsq->txqsq);
+
 	return work_done == budget;
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
index 47381e949f1f..879d698b6119 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/txrx.h
@@ -193,6 +193,8 @@ static inline u16 mlx5e_txqsq_get_next_pi(struct mlx5e_txqsq *sq, u16 size)
 	return pi;
 }
 
+void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq);
+
 static inline u16 mlx5e_shampo_get_cqe_header_index(struct mlx5e_rq *rq, struct mlx5_cqe64 *cqe)
 {
 	return be16_to_cpu(cqe->shampo.header_entry_index) & (rq->mpwqe.shampo->hd_per_wq - 1);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
index df5e780e8e6a..c7eb6b238c2b 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tx.c
@@ -762,6 +762,17 @@ static void mlx5e_tx_wi_consume_fifo_skbs(struct mlx5e_txqsq *sq, struct mlx5e_t
 	}
 }
 
+void mlx5e_txqsq_wake(struct mlx5e_txqsq *sq)
+{
+	if (netif_tx_queue_stopped(sq->txq) &&
+	    mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) &&
+	    mlx5e_ptpsq_fifo_has_room(sq) &&
+	    !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) {
+		netif_tx_wake_queue(sq->txq);
+		sq->stats->wake++;
+	}
+}
+
 bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 {
 	struct mlx5e_sq_stats *stats;
@@ -861,13 +872,7 @@ bool mlx5e_poll_tx_cq(struct mlx5e_cq *cq, int napi_budget)
 
 	netdev_tx_completed_queue(sq->txq, npkts, nbytes);
 
-	if (netif_tx_queue_stopped(sq->txq) &&
-	    mlx5e_wqc_has_room_for(&sq->wq, sq->cc, sq->pc, sq->stop_room) &&
-	    mlx5e_ptpsq_fifo_has_room(sq) &&
-	    !test_bit(MLX5E_SQ_STATE_RECOVERING, &sq->state)) {
-		netif_tx_wake_queue(sq->txq);
-		stats->wake++;
-	}
+	mlx5e_txqsq_wake(sq);
 
 	return (i == MLX5E_TX_CQ_POLL_BUDGET);
 }
-- 
2.40.1

[net 09/15] net/mlx5e: TC, Fix using eswitch mapping in nic mode

[Saeed Mahameed]

From: Paul Blakey <paulb@nvidia.com>

Cited patch is using the eswitch object mapping pool while
in nic mode where it isn't initialized. This results in the
trace below [0].

Fix that by using either nic or eswitch object mapping pool
depending if eswitch is enabled or not.

[0]:
[  826.446057] ==================================================================
[  826.446729] BUG: KASAN: slab-use-after-free in mlx5_add_flow_rules+0x30/0x490 [mlx5_core]
[  826.447515] Read of size 8 at addr ffff888194485830 by task tc/6233

[  826.448243] CPU: 16 PID: 6233 Comm: tc Tainted: G        W          6.3.0-rc6+ #1
[  826.448890] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
[  826.449785] Call Trace:
[  826.450052]  <TASK>
[  826.450302]  dump_stack_lvl+0x33/0x50
[  826.450650]  print_report+0xc2/0x610
[  826.450998]  ? __virt_addr_valid+0xb1/0x130
[  826.451385]  ? mlx5_add_flow_rules+0x30/0x490 [mlx5_core]
[  826.451935]  kasan_report+0xae/0xe0
[  826.452276]  ? mlx5_add_flow_rules+0x30/0x490 [mlx5_core]
[  826.452829]  mlx5_add_flow_rules+0x30/0x490 [mlx5_core]
[  826.453368]  ? __kmalloc_node+0x5a/0x120
[  826.453733]  esw_add_restore_rule+0x20f/0x270 [mlx5_core]
[  826.454288]  ? mlx5_eswitch_add_send_to_vport_meta_rule+0x260/0x260 [mlx5_core]
[  826.455011]  ? mutex_unlock+0x80/0xd0
[  826.455361]  ? __mutex_unlock_slowpath.constprop.0+0x210/0x210
[  826.455862]  ? mapping_add+0x2cb/0x440 [mlx5_core]
[  826.456425]  mlx5e_tc_action_miss_mapping_get+0x139/0x180 [mlx5_core]
[  826.457058]  ? mlx5e_tc_update_skb_nic+0xb0/0xb0 [mlx5_core]
[  826.457636]  ? __kasan_kmalloc+0x77/0x90
[  826.458000]  ? __kmalloc+0x57/0x120
[  826.458336]  mlx5_tc_ct_flow_offload+0x325/0xe40 [mlx5_core]
[  826.458916]  ? ct_kernel_enter.constprop.0+0x48/0xa0
[  826.459360]  ? mlx5_tc_ct_parse_action+0xf0/0xf0 [mlx5_core]
[  826.459933]  ? mlx5e_mod_hdr_attach+0x491/0x520 [mlx5_core]
[  826.460507]  ? mlx5e_mod_hdr_get+0x12/0x20 [mlx5_core]
[  826.461046]  ? mlx5e_tc_attach_mod_hdr+0x154/0x170 [mlx5_core]
[  826.461635]  mlx5e_configure_flower+0x969/0x2110 [mlx5_core]
[  826.462217]  ? _raw_spin_lock_bh+0x85/0xe0
[  826.462597]  ? __mlx5e_add_fdb_flow+0x750/0x750 [mlx5_core]
[  826.463163]  ? kasan_save_stack+0x2e/0x40
[  826.463534]  ? down_read+0x115/0x1b0
[  826.463878]  ? down_write_killable+0x110/0x110
[  826.464288]  ? tc_setup_action.part.0+0x9f/0x3b0
[  826.464701]  ? mlx5e_is_uplink_rep+0x4c/0x90 [mlx5_core]
[  826.465253]  ? mlx5e_tc_reoffload_flows_work+0x130/0x130 [mlx5_core]
[  826.465878]  tc_setup_cb_add+0x112/0x250
[  826.466247]  fl_hw_replace_filter+0x230/0x310 [cls_flower]
[  826.466724]  ? fl_hw_destroy_filter+0x1a0/0x1a0 [cls_flower]
[  826.467212]  fl_change+0x14e1/0x2030 [cls_flower]
[  826.467636]  ? sock_def_readable+0x89/0x120
[  826.468019]  ? fl_tmplt_create+0x2d0/0x2d0 [cls_flower]
[  826.468509]  ? kasan_unpoison+0x23/0x50
[  826.468873]  ? get_random_u16+0x180/0x180
[  826.469244]  ? __radix_tree_lookup+0x2b/0x130
[  826.469640]  ? fl_get+0x7b/0x140 [cls_flower]
[  826.470042]  ? fl_mask_put+0x200/0x200 [cls_flower]
[  826.470478]  ? __mutex_unlock_slowpath.constprop.0+0x210/0x210
[  826.470973]  ? fl_tmplt_create+0x2d0/0x2d0 [cls_flower]
[  826.471427]  tc_new_tfilter+0x644/0x1050
[  826.471795]  ? tc_get_tfilter+0x860/0x860
[  826.472170]  ? __thaw_task+0x130/0x130
[  826.472525]  ? arch_stack_walk+0x98/0xf0
[  826.472892]  ? cap_capable+0x9f/0xd0
[  826.473235]  ? security_capable+0x47/0x60
[  826.473608]  rtnetlink_rcv_msg+0x1d5/0x550
[  826.473985]  ? rtnl_calcit.isra.0+0x1f0/0x1f0
[  826.474383]  ? __stack_depot_save+0x35/0x4c0
[  826.474779]  ? kasan_save_stack+0x2e/0x40
[  826.475149]  ? kasan_save_stack+0x1e/0x40
[  826.475518]  ? __kasan_record_aux_stack+0x9f/0xb0
[  826.475939]  ? task_work_add+0x77/0x1c0
[  826.476305]  netlink_rcv_skb+0xe0/0x210
[  826.476661]  ? rtnl_calcit.isra.0+0x1f0/0x1f0
[  826.477057]  ? netlink_ack+0x7c0/0x7c0
[  826.477412]  ? rhashtable_jhash2+0xef/0x150
[  826.477796]  ? _copy_from_iter+0x105/0x770
[  826.484386]  netlink_unicast+0x346/0x490
[  826.484755]  ? netlink_attachskb+0x400/0x400
[  826.485145]  ? kernel_text_address+0xc2/0xd0
[  826.485535]  netlink_sendmsg+0x3b0/0x6c0
[  826.485902]  ? kernel_text_address+0xc2/0xd0
[  826.486296]  ? netlink_unicast+0x490/0x490
[  826.486671]  ? iovec_from_user.part.0+0x7a/0x1a0
[  826.487083]  ? netlink_unicast+0x490/0x490
[  826.487461]  sock_sendmsg+0x73/0xc0
[  826.487803]  ____sys_sendmsg+0x364/0x380
[  826.488186]  ? import_iovec+0x7/0x10
[  826.488531]  ? kernel_sendmsg+0x30/0x30
[  826.488893]  ? __copy_msghdr+0x180/0x180
[  826.489258]  ? kasan_save_stack+0x2e/0x40
[  826.489629]  ? kasan_save_stack+0x1e/0x40
[  826.490002]  ? __kasan_record_aux_stack+0x9f/0xb0
[  826.490424]  ? __call_rcu_common.constprop.0+0x46/0x580
[  826.490876]  ___sys_sendmsg+0xdf/0x140
[  826.491231]  ? copy_msghdr_from_user+0x110/0x110
[  826.491649]  ? fget_raw+0x120/0x120
[  826.491988]  ? ___sys_recvmsg+0xd9/0x130
[  826.492355]  ? folio_batch_add_and_move+0x80/0xa0
[  826.492776]  ? _raw_spin_lock+0x7a/0xd0
[  826.493137]  ? _raw_spin_lock+0x7a/0xd0
[  826.493500]  ? _raw_read_lock_irq+0x30/0x30
[  826.493880]  ? kasan_set_track+0x21/0x30
[  826.494249]  ? kasan_save_free_info+0x2a/0x40
[  826.494650]  ? do_sys_openat2+0xff/0x270
[  826.495016]  ? __fget_light+0x1b5/0x200
[  826.495377]  ? __virt_addr_valid+0xb1/0x130
[  826.495763]  __sys_sendmsg+0xb2/0x130
[  826.496118]  ? __sys_sendmsg_sock+0x20/0x20
[  826.496501]  ? __x64_sys_rseq+0x2e0/0x2e0
[  826.496874]  ? do_user_addr_fault+0x276/0x820
[  826.497273]  ? fpregs_assert_state_consistent+0x52/0x60
[  826.497727]  ? exit_to_user_mode_prepare+0x30/0x120
[  826.498158]  do_syscall_64+0x3d/0x90
[  826.498502]  entry_SYSCALL_64_after_hwframe+0x46/0xb0
[  826.498949] RIP: 0033:0x7f9b67f4f887
[  826.499294] Code: 0a 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b9 0f 1f 00 f3 0f 1e fa 64 8b 04 25 18 00 00 00 85 c0 75 10 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 51 c3 48 83 ec 28 89 54 24 1c 48 89 74 24 10
[  826.500742] RSP: 002b:00007fff5d1a5498 EFLAGS: 00000246 ORIG_RAX: 000000000000002e
[  826.501395] RAX: ffffffffffffffda RBX: 0000000064413ce6 RCX: 00007f9b67f4f887
[  826.501975] RDX: 0000000000000000 RSI: 00007fff5d1a5500 RDI: 0000000000000003
[  826.502556] RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000001
[  826.503135] R10: 00007f9b67e08708 R11: 0000000000000246 R12: 0000000000000001
[  826.503714] R13: 0000000000000001 R14: 00007fff5d1a9800 R15: 0000000000485400
[  826.504304]  </TASK>

[  826.504753] Allocated by task 3764:
[  826.505090]  kasan_save_stack+0x1e/0x40
[  826.505453]  kasan_set_track+0x21/0x30
[  826.505810]  __kasan_kmalloc+0x77/0x90
[  826.506164]  __mlx5_create_flow_table+0x16d/0xbb0 [mlx5_core]
[  826.506742]  esw_offloads_enable+0x60d/0xfb0 [mlx5_core]
[  826.507292]  mlx5_eswitch_enable_locked+0x4d3/0x680 [mlx5_core]
[  826.507885]  mlx5_devlink_eswitch_mode_set+0x2a3/0x580 [mlx5_core]
[  826.508513]  devlink_nl_cmd_eswitch_set_doit+0xdf/0x1f0
[  826.508969]  genl_family_rcv_msg_doit.isra.0+0x146/0x1c0
[  826.509427]  genl_rcv_msg+0x28d/0x3e0
[  826.509772]  netlink_rcv_skb+0xe0/0x210
[  826.510133]  genl_rcv+0x24/0x40
[  826.510448]  netlink_unicast+0x346/0x490
[  826.510810]  netlink_sendmsg+0x3b0/0x6c0
[  826.511179]  sock_sendmsg+0x73/0xc0
[  826.511519]  __sys_sendto+0x18d/0x220
[  826.511867]  __x64_sys_sendto+0x72/0x80
[  826.512232]  do_syscall_64+0x3d/0x90
[  826.512576]  entry_SYSCALL_64_after_hwframe+0x46/0xb0

[  826.513220] Freed by task 5674:
[  826.513535]  kasan_save_stack+0x1e/0x40
[  826.513893]  kasan_set_track+0x21/0x30
[  826.514245]  kasan_save_free_info+0x2a/0x40
[  826.514629]  ____kasan_slab_free+0x11a/0x1b0
[  826.515021]  __kmem_cache_free+0x14d/0x280
[  826.515399]  tree_put_node+0x109/0x1c0 [mlx5_core]
[  826.515907]  mlx5_destroy_flow_table+0x119/0x630 [mlx5_core]
[  826.516481]  esw_offloads_steering_cleanup+0xe7/0x150 [mlx5_core]
[  826.517084]  esw_offloads_disable+0xe0/0x160 [mlx5_core]
[  826.517632]  mlx5_eswitch_disable_locked+0x26c/0x290 [mlx5_core]
[  826.518225]  mlx5_devlink_eswitch_mode_set+0x128/0x580 [mlx5_core]
[  826.518834]  devlink_nl_cmd_eswitch_set_doit+0xdf/0x1f0
[  826.519286]  genl_family_rcv_msg_doit.isra.0+0x146/0x1c0
[  826.519748]  genl_rcv_msg+0x28d/0x3e0
[  826.520101]  netlink_rcv_skb+0xe0/0x210
[  826.520458]  genl_rcv+0x24/0x40
[  826.520771]  netlink_unicast+0x346/0x490
[  826.521137]  netlink_sendmsg+0x3b0/0x6c0
[  826.521505]  sock_sendmsg+0x73/0xc0
[  826.521842]  __sys_sendto+0x18d/0x220
[  826.522191]  __x64_sys_sendto+0x72/0x80
[  826.522554]  do_syscall_64+0x3d/0x90
[  826.522894]  entry_SYSCALL_64_after_hwframe+0x46/0xb0

[  826.523540] Last potentially related work creation:
[  826.523969]  kasan_save_stack+0x1e/0x40
[  826.524331]  __kasan_record_aux_stack+0x9f/0xb0
[  826.524739]  insert_work+0x30/0x130
[  826.525078]  __queue_work+0x34b/0x690
[  826.525426]  queue_work_on+0x48/0x50
[  826.525766]  __rhashtable_remove_fast_one+0x4af/0x4d0 [mlx5_core]
[  826.526365]  del_sw_flow_group+0x1b5/0x270 [mlx5_core]
[  826.526898]  tree_put_node+0x109/0x1c0 [mlx5_core]
[  826.527407]  esw_offloads_steering_cleanup+0xd3/0x150 [mlx5_core]
[  826.528009]  esw_offloads_disable+0xe0/0x160 [mlx5_core]
[  826.528616]  mlx5_eswitch_disable_locked+0x26c/0x290 [mlx5_core]
[  826.529218]  mlx5_devlink_eswitch_mode_set+0x128/0x580 [mlx5_core]
[  826.529823]  devlink_nl_cmd_eswitch_set_doit+0xdf/0x1f0
[  826.530276]  genl_family_rcv_msg_doit.isra.0+0x146/0x1c0
[  826.530733]  genl_rcv_msg+0x28d/0x3e0
[  826.531079]  netlink_rcv_skb+0xe0/0x210
[  826.531439]  genl_rcv+0x24/0x40
[  826.531755]  netlink_unicast+0x346/0x490
[  826.532123]  netlink_sendmsg+0x3b0/0x6c0
[  826.532487]  sock_sendmsg+0x73/0xc0
[  826.532825]  __sys_sendto+0x18d/0x220
[  826.533175]  __x64_sys_sendto+0x72/0x80
[  826.533533]  do_syscall_64+0x3d/0x90
[  826.533877]  entry_SYSCALL_64_after_hwframe+0x46/0xb0

[  826.534521] The buggy address belongs to the object at ffff888194485800
                which belongs to the cache kmalloc-512 of size 512
[  826.535506] The buggy address is located 48 bytes inside of
                freed 512-byte region [ffff888194485800, ffff888194485a00)

[  826.536666] The buggy address belongs to the physical page:
[  826.537138] page:00000000d75841dd refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x194480
[  826.537915] head:00000000d75841dd order:3 entire_mapcount:0 nr_pages_mapped:0 pincount:0
[  826.538595] flags: 0x200000000010200(slab|head|node=0|zone=2)
[  826.539089] raw: 0200000000010200 ffff888100042c80 ffffea0004523800 dead000000000002
[  826.539755] raw: 0000000000000000 0000000000200020 00000001ffffffff 0000000000000000
[  826.540417] page dumped because: kasan: bad access detected

[  826.541095] Memory state around the buggy address:
[  826.541519]  ffff888194485700: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[  826.542149]  ffff888194485780: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[  826.542773] >ffff888194485800: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[  826.543400]                                      ^
[  826.543822]  ffff888194485880: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[  826.544452]  ffff888194485900: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
[  826.545079] ==================================================================

Fixes: 6702782845a5 ("net/mlx5e: TC, Set CT miss to the specific ct action instance")
Signed-off-by: Paul Blakey <paulb@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_tc.c   | 34 +++++++++++++++----
 1 file changed, 27 insertions(+), 7 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
index 416ab6b6da97..e95414ef1f04 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_tc.c
@@ -5646,22 +5646,43 @@ bool mlx5e_tc_update_skb_nic(struct mlx5_cqe64 *cqe, struct sk_buff *skb)
 				   0, NULL);
 }
 
+static struct mapping_ctx *
+mlx5e_get_priv_obj_mapping(struct mlx5e_priv *priv)
+{
+	struct mlx5e_tc_table *tc;
+	struct mlx5_eswitch *esw;
+	struct mapping_ctx *ctx;
+
+	if (is_mdev_switchdev_mode(priv->mdev)) {
+		esw = priv->mdev->priv.eswitch;
+		ctx = esw->offloads.reg_c0_obj_pool;
+	} else {
+		tc = mlx5e_fs_get_tc(priv->fs);
+		ctx = tc->mapping;
+	}
+
+	return ctx;
+}
+
 int mlx5e_tc_action_miss_mapping_get(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr,
 				     u64 act_miss_cookie, u32 *act_miss_mapping)
 {
-	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
 	struct mlx5_mapped_obj mapped_obj = {};
+	struct mlx5_eswitch *esw;
 	struct mapping_ctx *ctx;
 	int err;
 
-	ctx = esw->offloads.reg_c0_obj_pool;
-
+	ctx = mlx5e_get_priv_obj_mapping(priv);
 	mapped_obj.type = MLX5_MAPPED_OBJ_ACT_MISS;
 	mapped_obj.act_miss_cookie = act_miss_cookie;
 	err = mapping_add(ctx, &mapped_obj, act_miss_mapping);
 	if (err)
 		return err;
 
+	if (!is_mdev_switchdev_mode(priv->mdev))
+		return 0;
+
+	esw = priv->mdev->priv.eswitch;
 	attr->act_id_restore_rule = esw_add_restore_rule(esw, *act_miss_mapping);
 	if (IS_ERR(attr->act_id_restore_rule))
 		goto err_rule;
@@ -5676,10 +5697,9 @@ int mlx5e_tc_action_miss_mapping_get(struct mlx5e_priv *priv, struct mlx5_flow_a
 void mlx5e_tc_action_miss_mapping_put(struct mlx5e_priv *priv, struct mlx5_flow_attr *attr,
 				      u32 act_miss_mapping)
 {
-	struct mlx5_eswitch *esw = priv->mdev->priv.eswitch;
-	struct mapping_ctx *ctx;
+	struct mapping_ctx *ctx = mlx5e_get_priv_obj_mapping(priv);
 
-	ctx = esw->offloads.reg_c0_obj_pool;
-	mlx5_del_flow_rules(attr->act_id_restore_rule);
+	if (is_mdev_switchdev_mode(priv->mdev))
+		mlx5_del_flow_rules(attr->act_id_restore_rule);
 	mapping_remove(ctx, act_miss_mapping);
 }
-- 
2.40.1

[net 10/15] net/mlx5: E-switch, Devcom, sync devcom events and devcom comp register

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

devcom events are sent to all registered component. Following the
cited patch, it is possible for two components, e.g.: two eswitches,
to send devcom events, while both components are registered. This
means eswitch layer will do double un/pairing, which is double
allocation and free of resources, even though only one un/pairing is
needed. flow example:

	cpu0					cpu1
	----					----

 mlx5_devlink_eswitch_mode_set(dev0)
  esw_offloads_devcom_init()
   mlx5_devcom_register_component(esw0)
                                         mlx5_devlink_eswitch_mode_set(dev1)
                                          esw_offloads_devcom_init()
                                           mlx5_devcom_register_component(esw1)
                                           mlx5_devcom_send_event()
   mlx5_devcom_send_event()

Hence, check whether the eswitches are already un/paired before
free/allocation of resources.

Fixes: 09b278462f16 ("net: devlink: enable parallel ops on netlink interface")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eswitch.h        | 1 +
 .../net/ethernet/mellanox/mlx5/core/eswitch_offloads.c   | 9 ++++++++-
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
index 9f007c5438ee..add6cfa432a5 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch.h
@@ -342,6 +342,7 @@ struct mlx5_eswitch {
 		u32             large_group_num;
 	}  params;
 	struct blocking_notifier_head n_head;
+	bool paired[MLX5_MAX_PORTS];
 };
 
 void esw_offloads_disable(struct mlx5_eswitch *esw);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 7c34c7cf506f..8d19c20d3447 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -2742,6 +2742,9 @@ static int mlx5_esw_offloads_devcom_event(int event,
 		    mlx5_eswitch_vport_match_metadata_enabled(peer_esw))
 			break;
 
+		if (esw->paired[mlx5_get_dev_index(peer_esw->dev)])
+			break;
+
 		err = mlx5_esw_offloads_set_ns_peer(esw, peer_esw, true);
 		if (err)
 			goto err_out;
@@ -2753,14 +2756,18 @@ static int mlx5_esw_offloads_devcom_event(int event,
 		if (err)
 			goto err_pair;
 
+		esw->paired[mlx5_get_dev_index(peer_esw->dev)] = true;
+		peer_esw->paired[mlx5_get_dev_index(esw->dev)] = true;
 		mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, true);
 		break;
 
 	case ESW_OFFLOADS_DEVCOM_UNPAIR:
-		if (!mlx5_devcom_is_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS))
+		if (!esw->paired[mlx5_get_dev_index(peer_esw->dev)])
 			break;
 
 		mlx5_devcom_set_paired(devcom, MLX5_DEVCOM_ESW_OFFLOADS, false);
+		esw->paired[mlx5_get_dev_index(peer_esw->dev)] = false;
+		peer_esw->paired[mlx5_get_dev_index(esw->dev)] = false;
 		mlx5_esw_offloads_unpair(peer_esw);
 		mlx5_esw_offloads_unpair(esw);
 		mlx5_esw_offloads_set_ns_peer(esw, peer_esw, false);
-- 
2.40.1

[net 11/15] net/mlx5: Devcom, fix error flow in mlx5_devcom_register_device

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

In case devcom allocation is failed, mlx5 is always freeing the priv.
However, this priv might have been allocated by a different thread,
and freeing it might lead to use-after-free bugs.
Fix it by freeing the priv only in case it was allocated by the
running thread.

Fixes: fadd59fc50d0 ("net/mlx5: Introduce inter-device communication mechanism")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
index 070d55f13419..8f978491dd32 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
@@ -112,7 +112,8 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev)
 	priv->devs[idx] = dev;
 	devcom = mlx5_devcom_alloc(priv, idx);
 	if (!devcom) {
-		kfree(priv);
+		if (new_priv)
+			kfree(priv);
 		return ERR_PTR(-ENOMEM);
 	}
 
-- 
2.40.1

[net 12/15] net/mlx5: Devcom, serialize devcom registration

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

From one hand, mlx5 driver is allowing to probe PFs in parallel.
From the other hand, devcom, which is a share resource between PFs, is
registered without any lock. This might resulted in memory problems.

Hence, use the global mlx5_dev_list_lock in order to serialize devcom
registration.

Fixes: fadd59fc50d0 ("net/mlx5: Introduce inter-device communication mechanism")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/lib/devcom.c  | 19 ++++++++++++++-----
 1 file changed, 14 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
index 8f978491dd32..b7d779d08d83 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c
@@ -3,6 +3,7 @@
 
 #include <linux/mlx5/vport.h>
 #include "lib/devcom.h"
+#include "mlx5_core.h"
 
 static LIST_HEAD(devcom_list);
 
@@ -77,6 +78,7 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev)
 	if (MLX5_CAP_GEN(dev, num_lag_ports) != MLX5_DEVCOM_PORTS_SUPPORTED)
 		return NULL;
 
+	mlx5_dev_list_lock();
 	sguid0 = mlx5_query_nic_system_image_guid(dev);
 	list_for_each_entry(iter, &devcom_list, list) {
 		struct mlx5_core_dev *tmp_dev = NULL;
@@ -102,8 +104,10 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev)
 
 	if (!priv) {
 		priv = mlx5_devcom_list_alloc();
-		if (!priv)
-			return ERR_PTR(-ENOMEM);
+		if (!priv) {
+			devcom = ERR_PTR(-ENOMEM);
+			goto out;
+		}
 
 		idx = 0;
 		new_priv = true;
@@ -114,12 +118,14 @@ struct mlx5_devcom *mlx5_devcom_register_device(struct mlx5_core_dev *dev)
 	if (!devcom) {
 		if (new_priv)
 			kfree(priv);
-		return ERR_PTR(-ENOMEM);
+		devcom = ERR_PTR(-ENOMEM);
+		goto out;
 	}
 
 	if (new_priv)
 		list_add(&priv->list, &devcom_list);
-
+out:
+	mlx5_dev_list_unlock();
 	return devcom;
 }
 
@@ -132,6 +138,7 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom)
 	if (IS_ERR_OR_NULL(devcom))
 		return;
 
+	mlx5_dev_list_lock();
 	priv = devcom->priv;
 	priv->devs[devcom->idx] = NULL;
 
@@ -142,10 +149,12 @@ void mlx5_devcom_unregister_device(struct mlx5_devcom *devcom)
 			break;
 
 	if (i != MLX5_DEVCOM_PORTS_SUPPORTED)
-		return;
+		goto out;
 
 	list_del(&priv->list);
 	kfree(priv);
+out:
+	mlx5_dev_list_unlock();
 }
 
 void mlx5_devcom_register_component(struct mlx5_devcom *devcom,
-- 
2.40.1

[net 13/15] net/mlx5: Free irqs only on shutdown callback

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Whenever a shutdown is invoked, free irqs only and keep mlx5_irq
synthetic wrapper intact in order to avoid use-after-free on
system shutdown.

for example:
==================================================================
BUG: KASAN: use-after-free in _find_first_bit+0x66/0x80
Read of size 8 at addr ffff88823fc0d318 by task kworker/u192:0/13608

CPU: 25 PID: 13608 Comm: kworker/u192:0 Tainted: G    B   W  O  6.1.21-cloudflare-kasan-2023.3.21 #1
Hardware name: GIGABYTE R162-R2-GEN0/MZ12-HD2-CD, BIOS R14 05/03/2021
Workqueue: mlx5e mlx5e_tx_timeout_work [mlx5_core]
Call Trace:
  <TASK>
  dump_stack_lvl+0x34/0x48
  print_report+0x170/0x473
  ? _find_first_bit+0x66/0x80
  kasan_report+0xad/0x130
  ? _find_first_bit+0x66/0x80
  _find_first_bit+0x66/0x80
  mlx5e_open_channels+0x3c5/0x3a10 [mlx5_core]
  ? console_unlock+0x2fa/0x430
  ? _raw_spin_lock_irqsave+0x8d/0xf0
  ? _raw_spin_unlock_irqrestore+0x42/0x80
  ? preempt_count_add+0x7d/0x150
  ? __wake_up_klogd.part.0+0x7d/0xc0
  ? vprintk_emit+0xfe/0x2c0
  ? mlx5e_trigger_napi_sched+0x40/0x40 [mlx5_core]
  ? dev_attr_show.cold+0x35/0x35
  ? devlink_health_do_dump.part.0+0x174/0x340
  ? devlink_health_report+0x504/0x810
  ? mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  ? process_one_work+0x680/0x1050
  mlx5e_safe_switch_params+0x156/0x220 [mlx5_core]
  ? mlx5e_switch_priv_channels+0x310/0x310 [mlx5_core]
  ? mlx5_eq_poll_irq_disabled+0xb6/0x100 [mlx5_core]
  mlx5e_tx_reporter_timeout_recover+0x123/0x240 [mlx5_core]
  ? __mutex_unlock_slowpath.constprop.0+0x2b0/0x2b0
  devlink_health_reporter_recover+0xa6/0x1f0
  devlink_health_report+0x2f7/0x810
  ? vsnprintf+0x854/0x15e0
  mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_reporter_tx_err_cqe+0x1a0/0x1a0 [mlx5_core]
  ? mlx5e_tx_reporter_timeout_dump+0x50/0x50 [mlx5_core]
  ? mlx5e_tx_reporter_dump_sq+0x260/0x260 [mlx5_core]
  ? newidle_balance+0x9b7/0xe30
  ? psi_group_change+0x6a7/0xb80
  ? mutex_lock+0x96/0xf0
  ? __mutex_lock_slowpath+0x10/0x10
  mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  process_one_work+0x680/0x1050
  worker_thread+0x5a0/0xeb0
  ? process_one_work+0x1050/0x1050
  kthread+0x2a2/0x340
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x22/0x30
  </TASK>

Freed by task 1:
  kasan_save_stack+0x23/0x50
  kasan_set_track+0x21/0x30
  kasan_save_free_info+0x2a/0x40
  ____kasan_slab_free+0x169/0x1d0
  slab_free_freelist_hook+0xd2/0x190
  __kmem_cache_free+0x1a1/0x2f0
  irq_pool_free+0x138/0x200 [mlx5_core]
  mlx5_irq_table_destroy+0xf6/0x170 [mlx5_core]
  mlx5_core_eq_free_irqs+0x74/0xf0 [mlx5_core]
  shutdown+0x194/0x1aa [mlx5_core]
  pci_device_shutdown+0x75/0x120
  device_shutdown+0x35c/0x620
  kernel_restart+0x60/0xa0
  __do_sys_reboot+0x1cb/0x2c0
  do_syscall_64+0x3b/0x90
  entry_SYSCALL_64_after_hwframe+0x4b/0xb5

The buggy address belongs to the object at ffff88823fc0d300
  which belongs to the cache kmalloc-192 of size 192
The buggy address is located 24 bytes inside of
  192-byte region [ffff88823fc0d300, ffff88823fc0d3c0)

The buggy address belongs to the physical page:
page:0000000010139587 refcount:1 mapcount:0 mapping:0000000000000000
index:0x0 pfn:0x23fc0c
head:0000000010139587 order:1 compound_mapcount:0 compound_pincount:0
flags: 0x2ffff800010200(slab|head|node=0|zone=2|lastcpupid=0x1ffff)
raw: 002ffff800010200 0000000000000000 dead000000000122 ffff88810004ca00
raw: 0000000000000000 0000000000200020 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
  ffff88823fc0d200: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
  ffff88823fc0d280: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
 >ffff88823fc0d300: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
                             ^
  ffff88823fc0d380: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
  ffff88823fc0d400: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
==================================================================
general protection fault, probably for non-canonical address
0xdffffc005c40d7ac: 0000 [#1] PREEMPT SMP KASAN NOPTI
KASAN: probably user-memory-access in range [0x00000002e206bd60-0x00000002e206bd67]
CPU: 25 PID: 13608 Comm: kworker/u192:0 Tainted: G    B   W  O  6.1.21-cloudflare-kasan-2023.3.21 #1
Hardware name: GIGABYTE R162-R2-GEN0/MZ12-HD2-CD, BIOS R14 05/03/2021
Workqueue: mlx5e mlx5e_tx_timeout_work [mlx5_core]
RIP: 0010:__alloc_pages+0x141/0x5c0
Call Trace:
  <TASK>
  ? sysvec_apic_timer_interrupt+0xa0/0xc0
  ? asm_sysvec_apic_timer_interrupt+0x16/0x20
  ? __alloc_pages_slowpath.constprop.0+0x1ec0/0x1ec0
  ? _raw_spin_unlock_irqrestore+0x3d/0x80
  __kmalloc_large_node+0x80/0x120
  ? kvmalloc_node+0x4e/0x170
  __kmalloc_node+0xd4/0x150
  kvmalloc_node+0x4e/0x170
  mlx5e_open_channels+0x631/0x3a10 [mlx5_core]
  ? console_unlock+0x2fa/0x430
  ? _raw_spin_lock_irqsave+0x8d/0xf0
  ? _raw_spin_unlock_irqrestore+0x42/0x80
  ? preempt_count_add+0x7d/0x150
  ? __wake_up_klogd.part.0+0x7d/0xc0
  ? vprintk_emit+0xfe/0x2c0
  ? mlx5e_trigger_napi_sched+0x40/0x40 [mlx5_core]
  ? dev_attr_show.cold+0x35/0x35
  ? devlink_health_do_dump.part.0+0x174/0x340
  ? devlink_health_report+0x504/0x810
  ? mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  ? process_one_work+0x680/0x1050
  mlx5e_safe_switch_params+0x156/0x220 [mlx5_core]
  ? mlx5e_switch_priv_channels+0x310/0x310 [mlx5_core]
  ? mlx5_eq_poll_irq_disabled+0xb6/0x100 [mlx5_core]
  mlx5e_tx_reporter_timeout_recover+0x123/0x240 [mlx5_core]
  ? __mutex_unlock_slowpath.constprop.0+0x2b0/0x2b0
  devlink_health_reporter_recover+0xa6/0x1f0
  devlink_health_report+0x2f7/0x810
  ? vsnprintf+0x854/0x15e0
  mlx5e_reporter_tx_timeout+0x29d/0x3a0 [mlx5_core]
  ? mlx5e_reporter_tx_err_cqe+0x1a0/0x1a0 [mlx5_core]
  ? mlx5e_tx_reporter_timeout_dump+0x50/0x50 [mlx5_core]
  ? mlx5e_tx_reporter_dump_sq+0x260/0x260 [mlx5_core]
  ? newidle_balance+0x9b7/0xe30
  ? psi_group_change+0x6a7/0xb80
  ? mutex_lock+0x96/0xf0
  ? __mutex_lock_slowpath+0x10/0x10
  mlx5e_tx_timeout_work+0x17c/0x230 [mlx5_core]
  process_one_work+0x680/0x1050
  worker_thread+0x5a0/0xeb0
  ? process_one_work+0x1050/0x1050
  kthread+0x2a2/0x340
  ? kthread_complete_and_exit+0x20/0x20
  ret_from_fork+0x22/0x30
  </TASK>
---[ end trace 0000000000000000  ]---
RIP: 0010:__alloc_pages+0x141/0x5c0
Code: e0 39 a3 96 89 e9 b8 22 01 32 01 83 e1 0f 48 89 fa 01 c9 48 c1 ea
03 d3 f8 83 e0 03 89 44 24 6c 48 b8 00 00 00 00 00 fc ff df <80> 3c 02
00 0f 85 fc 03 00 00 89 e8 4a 8b 14 f5 e0 39 a3 96 4c 89
RSP: 0018:ffff888251f0f438 EFLAGS: 00010202
RAX: dffffc0000000000 RBX: 1ffff1104a3e1e8b RCX: 0000000000000000
RDX: 000000005c40d7ac RSI: 0000000000000003 RDI: 00000002e206bd60
RBP: 0000000000052dc0 R08: ffff8882b0044218 R09: ffff8882b0045e8a
R10: fffffbfff300fefc R11: ffff888167af4000 R12: 0000000000000003
R13: 0000000000000000 R14: 00000000696c7070 R15: ffff8882373f4380
FS:  0000000000000000(0000) GS:ffff88bf2be80000(0000)
knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005641d031eee8 CR3: 0000002e7ca14000 CR4: 0000000000350ee0
Kernel panic - not syncing: Fatal exception
Kernel Offset: 0x11000000 from 0xffffffff81000000 (relocation range:
0xffffffff80000000-0xffffffffbfffffff)
---[ end Kernel panic - not syncing: Fatal exception  ]---]

Reported-by: Frederick Lawler <fred@cloudflare.com>
Link: https://lore.kernel.org/netdev/be5b9271-7507-19c5-ded1-fa78f1980e69@cloudflare.com
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/eq.c  |  2 +-
 .../ethernet/mellanox/mlx5/core/mlx5_irq.h    |  1 +
 .../net/ethernet/mellanox/mlx5/core/pci_irq.c | 29 +++++++++++++++++++
 3 files changed, 31 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 1c35d721a31d..fe698c79616c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -1104,7 +1104,7 @@ void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev)
 	struct mlx5_eq_table *table = dev->priv.eq_table;
 
 	mutex_lock(&table->lock); /* sync with create/destroy_async_eq */
-	mlx5_irq_table_destroy(dev);
+	mlx5_irq_table_free_irqs(dev);
 	mutex_unlock(&table->lock);
 }
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
index efd0c299c5c7..aa403a5ea34e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_irq.h
@@ -15,6 +15,7 @@ int mlx5_irq_table_init(struct mlx5_core_dev *dev);
 void mlx5_irq_table_cleanup(struct mlx5_core_dev *dev);
 int mlx5_irq_table_create(struct mlx5_core_dev *dev);
 void mlx5_irq_table_destroy(struct mlx5_core_dev *dev);
+void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev);
 int mlx5_irq_table_get_num_comp(struct mlx5_irq_table *table);
 int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table);
 struct mlx5_irq_table *mlx5_irq_table_get(struct mlx5_core_dev *dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 2245d3b2f393..ac1304c2d205 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -691,6 +691,24 @@ static void irq_pools_destroy(struct mlx5_irq_table *table)
 	irq_pool_free(table->pcif_pool);
 }
 
+static void mlx5_irq_pool_free_irqs(struct mlx5_irq_pool *pool)
+{
+	struct mlx5_irq *irq;
+	unsigned long index;
+
+	xa_for_each(&pool->irqs, index, irq)
+		free_irq(irq->map.virq, &irq->nh);
+}
+
+static void mlx5_irq_pools_free_irqs(struct mlx5_irq_table *table)
+{
+	if (table->sf_ctrl_pool) {
+		mlx5_irq_pool_free_irqs(table->sf_comp_pool);
+		mlx5_irq_pool_free_irqs(table->sf_ctrl_pool);
+	}
+	mlx5_irq_pool_free_irqs(table->pcif_pool);
+}
+
 /* irq_table API */
 
 int mlx5_irq_table_init(struct mlx5_core_dev *dev)
@@ -774,6 +792,17 @@ void mlx5_irq_table_destroy(struct mlx5_core_dev *dev)
 	pci_free_irq_vectors(dev->pdev);
 }
 
+void mlx5_irq_table_free_irqs(struct mlx5_core_dev *dev)
+{
+	struct mlx5_irq_table *table = dev->priv.irq_table;
+
+	if (mlx5_core_is_sf(dev))
+		return;
+
+	mlx5_irq_pools_free_irqs(table);
+	pci_free_irq_vectors(dev->pdev);
+}
+
 int mlx5_irq_table_get_sfs_vec(struct mlx5_irq_table *table)
 {
 	if (table->sf_comp_pool)
-- 
2.40.1

[net 14/15] net/mlx5: Fix irq affinity management

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

The cited patch deny the user of changing the affinity of mlx5 irqs,
which break backward compatibility.
Hence, allow the user to change the affinity of mlx5 irqs.

Fixes: bbac70c74183 ("net/mlx5: Use newer affinity descriptor")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index ac1304c2d205..86b528aae6d4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -567,7 +567,7 @@ int mlx5_irqs_request_vectors(struct mlx5_core_dev *dev, u16 *cpus, int nirqs,
 	struct mlx5_irq *irq;
 	int i;
 
-	af_desc.is_managed = 1;
+	af_desc.is_managed = false;
 	for (i = 0; i < nirqs; i++) {
 		cpumask_set_cpu(cpus[i], &af_desc.mask);
 		irq = mlx5_irq_request(dev, i + 1, &af_desc, rmap);
-- 
2.40.1

[net 15/15] net/mlx5: Fix indexing of mlx5_irq

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

After the cited patch, mlx5_irq xarray index can be different then
mlx5_irq MSIX table index.
Fix it by storing both mlx5_irq xarray index and MSIX table index.

Fixes: 3354822cde5a ("net/mlx5: Use dynamic msix vectors allocation")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Eli Cohen <elic@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
index 86b528aae6d4..db5687d9fec9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -32,6 +32,7 @@ struct mlx5_irq {
 	struct mlx5_irq_pool *pool;
 	int refcount;
 	struct msi_map map;
+	u32 pool_index;
 };
 
 struct mlx5_irq_table {
@@ -132,7 +133,7 @@ static void irq_release(struct mlx5_irq *irq)
 	struct cpu_rmap *rmap;
 #endif
 
-	xa_erase(&pool->irqs, irq->map.index);
+	xa_erase(&pool->irqs, irq->pool_index);
 	/* free_irq requires that affinity_hint and rmap will be cleared before
 	 * calling it. To satisfy this requirement, we call
 	 * irq_cpu_rmap_remove() to remove the notifier
@@ -276,11 +277,11 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
 	}
 	irq->pool = pool;
 	irq->refcount = 1;
-	irq->map.index = i;
-	err = xa_err(xa_store(&pool->irqs, irq->map.index, irq, GFP_KERNEL));
+	irq->pool_index = i;
+	err = xa_err(xa_store(&pool->irqs, irq->pool_index, irq, GFP_KERNEL));
 	if (err) {
 		mlx5_core_err(dev, "Failed to alloc xa entry for irq(%u). err = %d\n",
-			      irq->map.index, err);
+			      irq->pool_index, err);
 		goto err_xa;
 	}
 	return irq;
-- 
2.40.1