[pull request][net 00/12] mlx5 fixes 2021-08-09

[pull request][net 00/12] mlx5 fixes 2021-08-09

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

Hi Dave, Jakub,

This series introduces fixes to mlx5 driver.
Please pull and let me know if there is any problem.

Thanks,
Saeed.

---
The following changes since commit d09c548dbf3b31cb07bba562e0f452edfa01efe3:

  net: sched: act_mirred: Reset ct info when mirror/redirect skb (2021-08-09 10:58:47 +0100)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-fixes-2021-08-09

for you to fetch changes up to bd37c2888ccaa5ceb9895718f6909b247cc372e0:

  net/mlx5: Fix return value from tracer initialization (2021-08-09 20:57:03 -0700)

----------------------------------------------------------------
mlx5-fixes-2021-08-09

----------------------------------------------------------------
Alex Vesker (1):
      net/mlx5: DR, Add fail on error check on decap

Aya Levin (2):
      net/mlx5: Block switchdev mode while devlink traps are active
      net/mlx5: Fix return value from tracer initialization

Chris Mi (1):
      net/mlx5e: TC, Fix error handling memory leak

Leon Romanovsky (1):
      net/mlx5: Don't skip subfunction cleanup in case of error in module init

Maxim Mikityanskiy (1):
      net/mlx5e: Destroy page pool after XDP SQ to fix use-after-free

Roi Dayan (1):
      net/mlx5e: Avoid creating tunnel headers for local route

Shay Drory (4):
      net/mlx5: Fix order of functions in mlx5_irq_detach_nb()
      net/mlx5: Set all field of mlx5_irq before inserting it to the xarray
      net/mlx5: Destroy pool->mutex
      net/mlx5: Synchronize correct IRQ when destroying CQ

Vlad Buslov (1):
      net/mlx5: Bridge, fix ageing time

 drivers/infiniband/hw/mlx5/cq.c                    |  4 +--
 drivers/infiniband/hw/mlx5/devx.c                  |  3 +-
 drivers/net/ethernet/mellanox/mlx5/core/cq.c       |  1 +
 .../ethernet/mellanox/mlx5/core/diag/fw_tracer.c   | 11 ++++++--
 .../net/ethernet/mellanox/mlx5/core/en/tc_tun.c    |  5 ++++
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c  | 33 ++++++++--------------
 drivers/net/ethernet/mellanox/mlx5/core/eq.c       | 20 ++++++++++---
 .../net/ethernet/mellanox/mlx5/core/esw/bridge.c   |  4 +--
 .../net/ethernet/mellanox/mlx5/core/esw/sample.c   |  1 +
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 14 +++++++--
 .../net/ethernet/mellanox/mlx5/core/fpga/conn.c    |  4 +--
 drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h   |  2 ++
 drivers/net/ethernet/mellanox/mlx5/core/main.c     | 12 +++-----
 .../net/ethernet/mellanox/mlx5/core/mlx5_core.h    |  5 ++++
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c  | 10 +++++--
 .../ethernet/mellanox/mlx5/core/steering/dr_send.c |  4 +--
 .../mellanox/mlx5/core/steering/dr_ste_v0.c        |  2 ++
 drivers/vdpa/mlx5/net/mlx5_vnet.c                  |  3 +-
 include/linux/mlx5/driver.h                        |  3 +-
 19 files changed, 83 insertions(+), 58 deletions(-)

[net 01/12] net/mlx5: Don't skip subfunction cleanup in case of error in module init

[Saeed Mahameed]

From: Leon Romanovsky <leonro@nvidia.com>

Clean SF resources if mlx5 eth failed to initialize.

Fixes: 1958fc2f0712 ("net/mlx5: SF, Add auxiliary device driver")
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/main.c      | 12 ++++--------
 drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h |  5 +++++
 2 files changed, 9 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/main.c b/drivers/net/ethernet/mellanox/mlx5/core/main.c
index eb1b316560a8..c84ad87c99bb 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/main.c
@@ -1784,16 +1784,14 @@ static int __init init(void)
 	if (err)
 		goto err_sf;
 
-#ifdef CONFIG_MLX5_CORE_EN
 	err = mlx5e_init();
-	if (err) {
-		pci_unregister_driver(&mlx5_core_driver);
-		goto err_debug;
-	}
-#endif
+	if (err)
+		goto err_en;
 
 	return 0;
 
+err_en:
+	mlx5_sf_driver_unregister();
 err_sf:
 	pci_unregister_driver(&mlx5_core_driver);
 err_debug:
@@ -1803,9 +1801,7 @@ static int __init init(void)
 
 static void __exit cleanup(void)
 {
-#ifdef CONFIG_MLX5_CORE_EN
 	mlx5e_cleanup();
-#endif
 	mlx5_sf_driver_unregister();
 	pci_unregister_driver(&mlx5_core_driver);
 	mlx5_unregister_debugfs();
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index 343807ac2036..da365b8f0141 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -206,8 +206,13 @@ int mlx5_firmware_flash(struct mlx5_core_dev *dev, const struct firmware *fw,
 int mlx5_fw_version_query(struct mlx5_core_dev *dev,
 			  u32 *running_ver, u32 *stored_ver);
 
+#ifdef CONFIG_MLX5_CORE_EN
 int mlx5e_init(void);
 void mlx5e_cleanup(void);
+#else
+static inline int mlx5e_init(void){ return 0; }
+static inline void mlx5e_cleanup(void){}
+#endif
 
 static inline bool mlx5_sriov_is_enabled(struct mlx5_core_dev *dev)
 {
-- 
2.31.1

Re: [net 01/12] net/mlx5: Don't skip subfunction cleanup in case of error in module init

[patchwork-bot+netdevbpf]

Hello:

This series was applied to netdev/net.git (refs/heads/master):

On Mon,  9 Aug 2021 20:59:12 -0700 you wrote:
> From: Leon Romanovsky <leonro@nvidia.com>
> 
> Clean SF resources if mlx5 eth failed to initialize.
> 
> Fixes: 1958fc2f0712 ("net/mlx5: SF, Add auxiliary device driver")
> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
> Reviewed-by: Parav Pandit <parav@nvidia.com>
> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
> 
> [...]

Here is the summary with links:
  - [net,01/12] net/mlx5: Don't skip subfunction cleanup in case of error in module init
    https://git.kernel.org/netdev/net/c/c633e799641c
  - [net,02/12] net/mlx5: DR, Add fail on error check on decap
    https://git.kernel.org/netdev/net/c/d3875924dae6
  - [net,03/12] net/mlx5e: Avoid creating tunnel headers for local route
    https://git.kernel.org/netdev/net/c/c623c95afa56
  - [net,04/12] net/mlx5: Bridge, fix ageing time
    https://git.kernel.org/netdev/net/c/6d8680da2e98
  - [net,05/12] net/mlx5e: Destroy page pool after XDP SQ to fix use-after-free
    https://git.kernel.org/netdev/net/c/8ba3e4c85825
  - [net,06/12] net/mlx5: Block switchdev mode while devlink traps are active
    https://git.kernel.org/netdev/net/c/c85a6b8feb16
  - [net,07/12] net/mlx5: Fix order of functions in mlx5_irq_detach_nb()
    https://git.kernel.org/netdev/net/c/3c8946e0e284
  - [net,08/12] net/mlx5: Set all field of mlx5_irq before inserting it to the xarray
    https://git.kernel.org/netdev/net/c/5957cc557dc5
  - [net,09/12] net/mlx5: Destroy pool->mutex
    https://git.kernel.org/netdev/net/c/ba317e832d45
  - [net,10/12] net/mlx5e: TC, Fix error handling memory leak
    https://git.kernel.org/netdev/net/c/88bbd7b2369a
  - [net,11/12] net/mlx5: Synchronize correct IRQ when destroying CQ
    https://git.kernel.org/netdev/net/c/563476ae0c5e
  - [net,12/12] net/mlx5: Fix return value from tracer initialization
    https://git.kernel.org/netdev/net/c/bd37c2888cca

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

[net 02/12] net/mlx5: DR, Add fail on error check on decap

[Saeed Mahameed]

From: Alex Vesker <valex@nvidia.com>

While processing encapsulated packet on RX, one of the fields that is
checked is the inner packet length. If the length as specified in the header
doesn't match the actual inner packet length, the packet is invalid
and should be dropped. However, such packet caused the NIC to hang.

This patch turns on a 'fail_on_error' HW bit which allows HW to drop
such an invalid packet while processing RX packet and trying to decap it.

Fixes: ad17dc8cf910 ("net/mlx5: DR, Move STEv0 action apply logic")
Signed-off-by: Alex Vesker <valex@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_ste_v0.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
index f1950e4968da..e4dd4eed5aee 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste_v0.c
@@ -352,6 +352,7 @@ static void dr_ste_v0_set_rx_decap(u8 *hw_ste_p)
 {
 	MLX5_SET(ste_rx_steering_mult, hw_ste_p, tunneling_action,
 		 DR_STE_TUNL_ACTION_DECAP);
+	MLX5_SET(ste_rx_steering_mult, hw_ste_p, fail_on_error, 1);
 }
 
 static void dr_ste_v0_set_rx_pop_vlan(u8 *hw_ste_p)
@@ -365,6 +366,7 @@ static void dr_ste_v0_set_rx_decap_l3(u8 *hw_ste_p, bool vlan)
 	MLX5_SET(ste_rx_steering_mult, hw_ste_p, tunneling_action,
 		 DR_STE_TUNL_ACTION_L3_DECAP);
 	MLX5_SET(ste_modify_packet, hw_ste_p, action_description, vlan ? 1 : 0);
+	MLX5_SET(ste_rx_steering_mult, hw_ste_p, fail_on_error, 1);
 }
 
 static void dr_ste_v0_set_rewrite_actions(u8 *hw_ste_p, u16 num_of_actions,
-- 
2.31.1

[net 03/12] net/mlx5e: Avoid creating tunnel headers for local route

[Saeed Mahameed]

From: Roi Dayan <roid@nvidia.com>

It could be local and remote are on the same machine and the route
result will be a local route which will result in creating encap id
with src/dst mac address of 0.

Fixes: a54e20b4fcae ("net/mlx5e: Add basic TC tunnel set action for SRIOV offloads")
Signed-off-by: Roi Dayan <roid@nvidia.com>
Reviewed-by: Maor Dickman <maord@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
index 8f79f04eccd6..1e2d117082d4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_tun.c
@@ -124,6 +124,11 @@ static int mlx5e_route_lookup_ipv4_get(struct mlx5e_priv *priv,
 	if (IS_ERR(rt))
 		return PTR_ERR(rt);
 
+	if (rt->rt_type != RTN_UNICAST) {
+		ret = -ENETUNREACH;
+		goto err_rt_release;
+	}
+
 	if (mlx5_lag_is_multipath(mdev) && rt->rt_gw_family != AF_INET) {
 		ret = -ENETUNREACH;
 		goto err_rt_release;
-- 
2.31.1

[net 04/12] net/mlx5: Bridge, fix ageing time

[Saeed Mahameed]

From: Vlad Buslov <vladbu@nvidia.com>

Ageing time is not converted from clock_t to jiffies which results
incorrect ageing timeout calculation in workqueue update task. Fix it by
applying clock_t_to_jiffies() to provided value.

Fixes: c636a0f0f3f0 ("net/mlx5: Bridge, dynamic entry ageing")
Signed-off-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
index a6e1d4f78268..f3f56f32e435 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge.c
@@ -579,7 +579,7 @@ static struct mlx5_esw_bridge *mlx5_esw_bridge_create(int ifindex,
 	xa_init(&bridge->vports);
 	bridge->ifindex = ifindex;
 	bridge->refcnt = 1;
-	bridge->ageing_time = BR_DEFAULT_AGEING_TIME;
+	bridge->ageing_time = clock_t_to_jiffies(BR_DEFAULT_AGEING_TIME);
 	list_add(&bridge->list, &br_offloads->bridges);
 
 	return bridge;
@@ -1006,7 +1006,7 @@ int mlx5_esw_bridge_ageing_time_set(unsigned long ageing_time, struct mlx5_eswit
 	if (!vport->bridge)
 		return -EINVAL;
 
-	vport->bridge->ageing_time = ageing_time;
+	vport->bridge->ageing_time = clock_t_to_jiffies(ageing_time);
 	return 0;
 }
 
-- 
2.31.1

[net 05/12] net/mlx5e: Destroy page pool after XDP SQ to fix use-after-free

[Saeed Mahameed]

From: Maxim Mikityanskiy <maximmi@nvidia.com>

mlx5e_close_xdpsq does the cleanup: it calls mlx5e_free_xdpsq_descs to
free the outstanding descriptors, which relies on
mlx5e_page_release_dynamic and page_pool_release_page. However,
page_pool_destroy is already called by this point, because
mlx5e_close_rq runs before mlx5e_close_xdpsq.

This commit fixes the use-after-free by swapping mlx5e_close_xdpsq and
mlx5e_close_rq.

The commit cited below started calling page_pool_destroy directly from
the driver. Previously, the page pool was destroyed under a call_rcu
from xdp_rxq_info_unreg_mem_model, which would defer the deallocation
until after the XDPSQ is cleaned up.

Fixes: 1da4bbeffe41 ("net: core: page_pool: add user refcnt and reintroduce page_pool_destroy")
Signed-off-by: Maxim Mikityanskiy <maximmi@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 20 +++++++++----------
 1 file changed, 10 insertions(+), 10 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 37c440837945..fd250f7bcd88 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1891,30 +1891,30 @@ static int mlx5e_open_queues(struct mlx5e_channel *c,
 	if (err)
 		goto err_close_icosq;
 
+	err = mlx5e_open_rxq_rq(c, params, &cparam->rq);
+	if (err)
+		goto err_close_sqs;
+
 	if (c->xdp) {
 		err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, NULL,
 				       &c->rq_xdpsq, false);
 		if (err)
-			goto err_close_sqs;
+			goto err_close_rq;
 	}
 
-	err = mlx5e_open_rxq_rq(c, params, &cparam->rq);
-	if (err)
-		goto err_close_xdp_sq;
-
 	err = mlx5e_open_xdpsq(c, params, &cparam->xdp_sq, NULL, &c->xdpsq, true);
 	if (err)
-		goto err_close_rq;
+		goto err_close_xdp_sq;
 
 	return 0;
 
-err_close_rq:
-	mlx5e_close_rq(&c->rq);
-
 err_close_xdp_sq:
 	if (c->xdp)
 		mlx5e_close_xdpsq(&c->rq_xdpsq);
 
+err_close_rq:
+	mlx5e_close_rq(&c->rq);
+
 err_close_sqs:
 	mlx5e_close_sqs(c);
 
@@ -1949,9 +1949,9 @@ static int mlx5e_open_queues(struct mlx5e_channel *c,
 static void mlx5e_close_queues(struct mlx5e_channel *c)
 {
 	mlx5e_close_xdpsq(&c->xdpsq);
-	mlx5e_close_rq(&c->rq);
 	if (c->xdp)
 		mlx5e_close_xdpsq(&c->rq_xdpsq);
+	mlx5e_close_rq(&c->rq);
 	mlx5e_close_sqs(c);
 	mlx5e_close_icosq(&c->icosq);
 	mlx5e_close_icosq(&c->async_icosq);
-- 
2.31.1

[net 06/12] net/mlx5: Block switchdev mode while devlink traps are active

[Saeed Mahameed]

From: Aya Levin <ayal@nvidia.com>

Since switchdev mode can't support  devlink traps, verify there are
no active devlink traps before moving eswitch to switchdev mode. If
there are active traps, prevent the switchdev mode configuration.

Fixes: eb3862a0525d ("net/mlx5e: Enable traps according to link state")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
index 011e766e4f67..3bb71a186004 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c
@@ -48,6 +48,7 @@
 #include "lib/fs_chains.h"
 #include "en_tc.h"
 #include "en/mapping.h"
+#include "devlink.h"
 
 #define mlx5_esw_for_each_rep(esw, i, rep) \
 	xa_for_each(&((esw)->offloads.vport_reps), i, rep)
@@ -3001,12 +3002,19 @@ int mlx5_devlink_eswitch_mode_set(struct devlink *devlink, u16 mode,
 	if (cur_mlx5_mode == mlx5_mode)
 		goto unlock;
 
-	if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV)
+	if (mode == DEVLINK_ESWITCH_MODE_SWITCHDEV) {
+		if (mlx5_devlink_trap_get_num_active(esw->dev)) {
+			NL_SET_ERR_MSG_MOD(extack,
+					   "Can't change mode while devlink traps are active");
+			err = -EOPNOTSUPP;
+			goto unlock;
+		}
 		err = esw_offloads_start(esw, extack);
-	else if (mode == DEVLINK_ESWITCH_MODE_LEGACY)
+	} else if (mode == DEVLINK_ESWITCH_MODE_LEGACY) {
 		err = esw_offloads_stop(esw, extack);
-	else
+	} else {
 		err = -EINVAL;
+	}
 
 unlock:
 	mlx5_esw_unlock(esw);
-- 
2.31.1

[net 07/12] net/mlx5: Fix order of functions in mlx5_irq_detach_nb()

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Change order of functions in mlx5_irq_detach_nb() so it will be
a mirror of mlx5_irq_attach_nb.

Fixes: 71e084e26414 ("net/mlx5: Allocating a pool of MSI-X vectors for SFs")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 5 ++++-
 1 file changed, 4 insertions(+), 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 b25f764daa08..95e60da33f03 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -251,8 +251,11 @@ int mlx5_irq_attach_nb(struct mlx5_irq *irq, struct notifier_block *nb)
 
 int mlx5_irq_detach_nb(struct mlx5_irq *irq, struct notifier_block *nb)
 {
+	int err = 0;
+
+	err = atomic_notifier_chain_unregister(&irq->nh, nb);
 	irq_put(irq);
-	return atomic_notifier_chain_unregister(&irq->nh, nb);
+	return err;
 }
 
 struct cpumask *mlx5_irq_get_affinity_mask(struct mlx5_irq *irq)
-- 
2.31.1

[net 08/12] net/mlx5: Set all field of mlx5_irq before inserting it to the xarray

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Currently irq->pool is set after the irq is insert to the xarray.
Set irq->pool before the irq is inserted to the xarray.

Fixes: 71e084e26414 ("net/mlx5: Allocating a pool of MSI-X vectors for SFs")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Parav Pandit <parav@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 95e60da33f03..7b923f6b5462 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -214,6 +214,7 @@ static struct mlx5_irq *irq_request(struct mlx5_irq_pool *pool, int i)
 		err = -ENOMEM;
 		goto err_cpumask;
 	}
+	irq->pool = pool;
 	kref_init(&irq->kref);
 	irq->index = i;
 	err = xa_err(xa_store(&pool->irqs, irq->index, irq, GFP_KERNEL));
@@ -222,7 +223,6 @@ static struct mlx5_irq *irq_request(struct mlx5_irq_pool *pool, int i)
 			      irq->index, err);
 		goto err_xa;
 	}
-	irq->pool = pool;
 	return irq;
 err_xa:
 	free_cpumask_var(irq->mask);
-- 
2.31.1

[net 09/12] net/mlx5: Destroy pool->mutex

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

Destroy pool->mutex when we destroy the pool.

Fixes: c36326d38d93 ("net/mlx5: Round-Robin EQs over IRQs")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 3 ++-
 1 file changed, 2 insertions(+), 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 7b923f6b5462..3465b363fc2f 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
@@ -440,6 +440,7 @@ irq_pool_alloc(struct mlx5_core_dev *dev, int start, int size, char *name,
 	if (!pool)
 		return ERR_PTR(-ENOMEM);
 	pool->dev = dev;
+	mutex_init(&pool->lock);
 	xa_init_flags(&pool->irqs, XA_FLAGS_ALLOC);
 	pool->xa_num_irqs.min = start;
 	pool->xa_num_irqs.max = start + size - 1;
@@ -448,7 +449,6 @@ irq_pool_alloc(struct mlx5_core_dev *dev, int start, int size, char *name,
 			 name);
 	pool->min_threshold = min_threshold * MLX5_EQ_REFS_PER_IRQ;
 	pool->max_threshold = max_threshold * MLX5_EQ_REFS_PER_IRQ;
-	mutex_init(&pool->lock);
 	mlx5_core_dbg(dev, "pool->name = %s, pool->size = %d, pool->start = %d",
 		      name, size, start);
 	return pool;
@@ -462,6 +462,7 @@ static void irq_pool_free(struct mlx5_irq_pool *pool)
 	xa_for_each(&pool->irqs, index, irq)
 		irq_release(&irq->kref);
 	xa_destroy(&pool->irqs);
+	mutex_destroy(&pool->lock);
 	kvfree(pool);
 }
 
-- 
2.31.1

[net 10/12] net/mlx5e: TC, Fix error handling memory leak

[Saeed Mahameed]

From: Chris Mi <cmi@nvidia.com>

Free the offload sample action on error.

Fixes: f94d6389f6a8 ("net/mlx5e: TC, Add support to offload sample action")
Signed-off-by: Chris Mi <cmi@nvidia.com>
Reviewed-by: Oz Shlomo <ozsh@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/sample.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/sample.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/sample.c
index 794012c5c476..d3ad78aa9d45 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/sample.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/sample.c
@@ -501,6 +501,7 @@ mlx5_esw_sample_offload(struct mlx5_esw_psample *esw_psample,
 err_offload_rule:
 	mlx5_esw_vporttbl_put(esw, &per_vport_tbl_attr);
 err_default_tbl:
+	kfree(sample_flow);
 	return ERR_PTR(err);
 }
 
-- 
2.31.1

[net 11/12] net/mlx5: Synchronize correct IRQ when destroying CQ

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

The CQ destroy is performed based on the IRQ number that is stored in
cq->irqn. That number wasn't set explicitly during CQ creation and as
expected some of the API users of mlx5_core_create_cq() forgot to update
it.

This caused to wrong synchronization call of the wrong IRQ with a number
0 instead of the real one.

As a fix, set the IRQ number directly in the mlx5_core_create_cq() and
update all users accordingly.

Fixes: 1a86b377aa21 ("vdpa/mlx5: Add VDPA driver for supported mlx5 devices")
Fixes: ef1659ade359 ("IB/mlx5: Add DEVX support for CQ events")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/infiniband/hw/mlx5/cq.c               |  4 +---
 drivers/infiniband/hw/mlx5/devx.c             |  3 +--
 drivers/net/ethernet/mellanox/mlx5/core/cq.c  |  1 +
 .../net/ethernet/mellanox/mlx5/core/en_main.c | 13 ++----------
 drivers/net/ethernet/mellanox/mlx5/core/eq.c  | 20 +++++++++++++++----
 .../ethernet/mellanox/mlx5/core/fpga/conn.c   |  4 +---
 .../net/ethernet/mellanox/mlx5/core/lib/eq.h  |  2 ++
 .../mellanox/mlx5/core/steering/dr_send.c     |  4 +---
 drivers/vdpa/mlx5/net/mlx5_vnet.c             |  3 +--
 include/linux/mlx5/driver.h                   |  3 +--
 10 files changed, 27 insertions(+), 30 deletions(-)

diff --git a/drivers/infiniband/hw/mlx5/cq.c b/drivers/infiniband/hw/mlx5/cq.c
index 7abeb576b3c5..b8e5e371bb19 100644
--- a/drivers/infiniband/hw/mlx5/cq.c
+++ b/drivers/infiniband/hw/mlx5/cq.c
@@ -945,7 +945,6 @@ int mlx5_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 	u32 *cqb = NULL;
 	void *cqc;
 	int cqe_size;
-	unsigned int irqn;
 	int eqn;
 	int err;
 
@@ -984,7 +983,7 @@ int mlx5_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 		INIT_WORK(&cq->notify_work, notify_soft_wc_handler);
 	}
 
-	err = mlx5_vector2eqn(dev->mdev, vector, &eqn, &irqn);
+	err = mlx5_vector2eqn(dev->mdev, vector, &eqn);
 	if (err)
 		goto err_cqb;
 
@@ -1007,7 +1006,6 @@ int mlx5_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr,
 		goto err_cqb;
 
 	mlx5_ib_dbg(dev, "cqn 0x%x\n", cq->mcq.cqn);
-	cq->mcq.irqn = irqn;
 	if (udata)
 		cq->mcq.tasklet_ctx.comp = mlx5_ib_cq_comp;
 	else
diff --git a/drivers/infiniband/hw/mlx5/devx.c b/drivers/infiniband/hw/mlx5/devx.c
index eb9b0a2707f8..c869b2a91a28 100644
--- a/drivers/infiniband/hw/mlx5/devx.c
+++ b/drivers/infiniband/hw/mlx5/devx.c
@@ -975,7 +975,6 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_QUERY_EQN)(
 	struct mlx5_ib_dev *dev;
 	int user_vector;
 	int dev_eqn;
-	unsigned int irqn;
 	int err;
 
 	if (uverbs_copy_from(&user_vector, attrs,
@@ -987,7 +986,7 @@ static int UVERBS_HANDLER(MLX5_IB_METHOD_DEVX_QUERY_EQN)(
 		return PTR_ERR(c);
 	dev = to_mdev(c->ibucontext.device);
 
-	err = mlx5_vector2eqn(dev->mdev, user_vector, &dev_eqn, &irqn);
+	err = mlx5_vector2eqn(dev->mdev, user_vector, &dev_eqn);
 	if (err < 0)
 		return err;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/cq.c b/drivers/net/ethernet/mellanox/mlx5/core/cq.c
index df3e4938ecdd..360e093874d4 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/cq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/cq.c
@@ -134,6 +134,7 @@ int mlx5_core_create_cq(struct mlx5_core_dev *dev, struct mlx5_core_cq *cq,
 			      cq->cqn);
 
 	cq->uar = dev->priv.uar;
+	cq->irqn = eq->core.irqn;
 
 	return 0;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index fd250f7bcd88..24f919ef9b8e 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1535,15 +1535,9 @@ static int mlx5e_alloc_cq_common(struct mlx5e_priv *priv,
 {
 	struct mlx5_core_dev *mdev = priv->mdev;
 	struct mlx5_core_cq *mcq = &cq->mcq;
-	int eqn_not_used;
-	unsigned int irqn;
 	int err;
 	u32 i;
 
-	err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn_not_used, &irqn);
-	if (err)
-		return err;
-
 	err = mlx5_cqwq_create(mdev, &param->wq, param->cqc, &cq->wq,
 			       &cq->wq_ctrl);
 	if (err)
@@ -1557,7 +1551,6 @@ static int mlx5e_alloc_cq_common(struct mlx5e_priv *priv,
 	mcq->vector     = param->eq_ix;
 	mcq->comp       = mlx5e_completion_event;
 	mcq->event      = mlx5e_cq_error_event;
-	mcq->irqn       = irqn;
 
 	for (i = 0; i < mlx5_cqwq_get_size(&cq->wq); i++) {
 		struct mlx5_cqe64 *cqe = mlx5_cqwq_get_wqe(&cq->wq, i);
@@ -1605,11 +1598,10 @@ static int mlx5e_create_cq(struct mlx5e_cq *cq, struct mlx5e_cq_param *param)
 	void *in;
 	void *cqc;
 	int inlen;
-	unsigned int irqn_not_used;
 	int eqn;
 	int err;
 
-	err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn, &irqn_not_used);
+	err = mlx5_vector2eqn(mdev, param->eq_ix, &eqn);
 	if (err)
 		return err;
 
@@ -1983,9 +1975,8 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
 	struct mlx5e_channel *c;
 	unsigned int irq;
 	int err;
-	int eqn;
 
-	err = mlx5_vector2eqn(priv->mdev, ix, &eqn, &irq);
+	err = mlx5_vector2irqn(priv->mdev, ix, &irq);
 	if (err)
 		return err;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/eq.c b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
index 6e074cc457de..605c8ecc3610 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/eq.c
@@ -855,8 +855,8 @@ static int create_comp_eqs(struct mlx5_core_dev *dev)
 	return err;
 }
 
-int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn,
-		    unsigned int *irqn)
+static int vector2eqnirqn(struct mlx5_core_dev *dev, int vector, int *eqn,
+			  unsigned int *irqn)
 {
 	struct mlx5_eq_table *table = dev->priv.eq_table;
 	struct mlx5_eq_comp *eq, *n;
@@ -865,8 +865,10 @@ int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn,
 
 	list_for_each_entry_safe(eq, n, &table->comp_eqs_list, list) {
 		if (i++ == vector) {
-			*eqn = eq->core.eqn;
-			*irqn = eq->core.irqn;
+			if (irqn)
+				*irqn = eq->core.irqn;
+			if (eqn)
+				*eqn = eq->core.eqn;
 			err = 0;
 			break;
 		}
@@ -874,8 +876,18 @@ int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn,
 
 	return err;
 }
+
+int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn)
+{
+	return vector2eqnirqn(dev, vector, eqn, NULL);
+}
 EXPORT_SYMBOL(mlx5_vector2eqn);
 
+int mlx5_vector2irqn(struct mlx5_core_dev *dev, int vector, unsigned int *irqn)
+{
+	return vector2eqnirqn(dev, vector, NULL, irqn);
+}
+
 unsigned int mlx5_comp_vectors_count(struct mlx5_core_dev *dev)
 {
 	return dev->priv.eq_table->num_comp_eqs;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
index bd66ab2af5b5..d5da4ab65766 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/fpga/conn.c
@@ -417,7 +417,6 @@ static int mlx5_fpga_conn_create_cq(struct mlx5_fpga_conn *conn, int cq_size)
 	struct mlx5_wq_param wqp;
 	struct mlx5_cqe64 *cqe;
 	int inlen, err, eqn;
-	unsigned int irqn;
 	void *cqc, *in;
 	__be64 *pas;
 	u32 i;
@@ -446,7 +445,7 @@ static int mlx5_fpga_conn_create_cq(struct mlx5_fpga_conn *conn, int cq_size)
 		goto err_cqwq;
 	}
 
-	err = mlx5_vector2eqn(mdev, smp_processor_id(), &eqn, &irqn);
+	err = mlx5_vector2eqn(mdev, smp_processor_id(), &eqn);
 	if (err) {
 		kvfree(in);
 		goto err_cqwq;
@@ -476,7 +475,6 @@ static int mlx5_fpga_conn_create_cq(struct mlx5_fpga_conn *conn, int cq_size)
 	*conn->cq.mcq.arm_db    = 0;
 	conn->cq.mcq.vector     = 0;
 	conn->cq.mcq.comp       = mlx5_fpga_conn_cq_complete;
-	conn->cq.mcq.irqn       = irqn;
 	conn->cq.mcq.uar        = fdev->conn_res.uar;
 	tasklet_setup(&conn->cq.tasklet, mlx5_fpga_conn_cq_tasklet);
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
index 624cedebb510..d3d628b862f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/eq.h
@@ -104,4 +104,6 @@ void mlx5_core_eq_free_irqs(struct mlx5_core_dev *dev);
 struct cpu_rmap *mlx5_eq_table_get_rmap(struct mlx5_core_dev *dev);
 #endif
 
+int mlx5_vector2irqn(struct mlx5_core_dev *dev, int vector, unsigned int *irqn);
+
 #endif
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
index 12cf323a5943..9df0e73d1c35 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_send.c
@@ -749,7 +749,6 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev,
 	struct mlx5_cqe64 *cqe;
 	struct mlx5dr_cq *cq;
 	int inlen, err, eqn;
-	unsigned int irqn;
 	void *cqc, *in;
 	__be64 *pas;
 	int vector;
@@ -782,7 +781,7 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev,
 		goto err_cqwq;
 
 	vector = raw_smp_processor_id() % mlx5_comp_vectors_count(mdev);
-	err = mlx5_vector2eqn(mdev, vector, &eqn, &irqn);
+	err = mlx5_vector2eqn(mdev, vector, &eqn);
 	if (err) {
 		kvfree(in);
 		goto err_cqwq;
@@ -818,7 +817,6 @@ static struct mlx5dr_cq *dr_create_cq(struct mlx5_core_dev *mdev,
 	*cq->mcq.arm_db = cpu_to_be32(2 << 28);
 
 	cq->mcq.vector = 0;
-	cq->mcq.irqn = irqn;
 	cq->mcq.uar = uar;
 
 	return cq;
diff --git a/drivers/vdpa/mlx5/net/mlx5_vnet.c b/drivers/vdpa/mlx5/net/mlx5_vnet.c
index 2a31467f7ac5..379a19144a25 100644
--- a/drivers/vdpa/mlx5/net/mlx5_vnet.c
+++ b/drivers/vdpa/mlx5/net/mlx5_vnet.c
@@ -526,7 +526,6 @@ static int cq_create(struct mlx5_vdpa_net *ndev, u16 idx, u32 num_ent)
 	void __iomem *uar_page = ndev->mvdev.res.uar->map;
 	u32 out[MLX5_ST_SZ_DW(create_cq_out)];
 	struct mlx5_vdpa_cq *vcq = &mvq->cq;
-	unsigned int irqn;
 	__be64 *pas;
 	int inlen;
 	void *cqc;
@@ -566,7 +565,7 @@ static int cq_create(struct mlx5_vdpa_net *ndev, u16 idx, u32 num_ent)
 	/* Use vector 0 by default. Consider adding code to choose least used
 	 * vector.
 	 */
-	err = mlx5_vector2eqn(mdev, 0, &eqn, &irqn);
+	err = mlx5_vector2eqn(mdev, 0, &eqn);
 	if (err)
 		goto err_vec;
 
diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 1efe37466969..25a8be58d289 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -1044,8 +1044,7 @@ void mlx5_unregister_debugfs(void);
 void mlx5_fill_page_array(struct mlx5_frag_buf *buf, __be64 *pas);
 void mlx5_fill_page_frag_array_perm(struct mlx5_frag_buf *buf, __be64 *pas, u8 perm);
 void mlx5_fill_page_frag_array(struct mlx5_frag_buf *frag_buf, __be64 *pas);
-int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn,
-		    unsigned int *irqn);
+int mlx5_vector2eqn(struct mlx5_core_dev *dev, int vector, int *eqn);
 int mlx5_core_attach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid, u32 qpn);
 int mlx5_core_detach_mcg(struct mlx5_core_dev *dev, union ib_gid *mgid, u32 qpn);
 
-- 
2.31.1

[net 12/12] net/mlx5: Fix return value from tracer initialization

[Saeed Mahameed]

From: Aya Levin <ayal@nvidia.com>

Check return value of mlx5_fw_tracer_start(), set error path and fix
return value of mlx5_fw_tracer_init() accordingly.

Fixes: c71ad41ccb0c ("net/mlx5: FW tracer, events handling")
Signed-off-by: Aya Levin <ayal@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c  | 11 +++++++++--
 1 file changed, 9 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
index 01a1d02dcf15..3f8a98093f8c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c
@@ -1019,12 +1019,19 @@ int mlx5_fw_tracer_init(struct mlx5_fw_tracer *tracer)
 	MLX5_NB_INIT(&tracer->nb, fw_tracer_event, DEVICE_TRACER);
 	mlx5_eq_notifier_register(dev, &tracer->nb);
 
-	mlx5_fw_tracer_start(tracer);
-
+	err = mlx5_fw_tracer_start(tracer);
+	if (err) {
+		mlx5_core_warn(dev, "FWTracer: Failed to start tracer %d\n", err);
+		goto err_notifier_unregister;
+	}
 	return 0;
 
+err_notifier_unregister:
+	mlx5_eq_notifier_unregister(dev, &tracer->nb);
+	mlx5_core_destroy_mkey(dev, &tracer->buff.mkey);
 err_dealloc_pd:
 	mlx5_core_dealloc_pd(dev, tracer->buff.pdn);
+	cancel_work_sync(&tracer->read_fw_strings_work);
 	return err;
 }
 
-- 
2.31.1