[pull request][net-next 00/15] mlx5 updates 2023-09-19

[pull request][net-next 00/15] mlx5 updates 2023-09-19

[Saeed Mahameed]

From: Saeed Mahameed <saeedm@nvidia.com>

This series adds misc updates to mlx5.

For more information please see tag log below.

Please pull and let me know if there is any problem.

Thanks,
Saeed.


The following changes since commit 5bed8d585aa1db1651745173a66e32df82a5cb05:

  Merge branch 'add-wed-support-for-mt7988-chipset' (2023-09-19 18:28:00 +0200)

are available in the Git repository at:

  git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git tags/mlx5-updates-2023-09-19

for you to fetch changes up to e738e355045237ee8802cb2b31a8ed6f4b7ac534:

  net/mlx5: Enable 4 ports multiport E-switch (2023-09-19 23:33:09 -0700)

----------------------------------------------------------------
mlx5-updates-2023-09-19

Misc updates for mlx5 driver

1) From Erez, Add support for multicast forwarding to multi destination
   in bridge offloads with software steering mode (SMFS).

2) From Jianbo, Utilize the maximum aggregated link speed for police
   action rate.

3) From Moshe, Add a health error syndrome for pci data poisoned

4) From Shay, Enable 4 ports multiport E-switch

5) From Jiri, Trivial SF code cleanup

----------------------------------------------------------------
Erez Shitrit (3):
      net/mlx5: Bridge, Enable mcast in smfs steering mode
      net/mlx5: DR, Add check for multi destination FTE
      net/mlx5: DR, Handle multi destination action in the right order

Jianbo Liu (2):
      net/mlx5e: Consider aggregated port speed during rate configuration
      net/mlx5e: Check police action rate for matchall filter

Jiri Pirko (8):
      net/mlx5: Call mlx5_sf_id_erase() once in mlx5_sf_dealloc()
      net/mlx5: Use devlink port pointer to get the pointer of container SF struct
      net/mlx5: Convert SF port_indices xarray to function_ids xarray
      net/mlx5: Move state lock taking into mlx5_sf_dealloc()
      net/mlx5: Rename mlx5_sf_deactivate_all() to mlx5_sf_del_all()
      net/mlx5: Push common deletion code into mlx5_sf_del()
      net/mlx5: Remove SF table reference counting
      net/mlx5: Remove redundant max_sfs check and field from struct mlx5_sf_dev_table

Moshe Shemesh (1):
      net/mlx5: Add a health error syndrome for pci data poisoned

Shay Drory (1):
      net/mlx5: Enable 4 ports multiport E-switch

 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c |  11 +-
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c  |  96 +++++++--
 drivers/net/ethernet/mellanox/mlx5/core/health.c   |   2 +
 .../net/ethernet/mellanox/mlx5/core/lag/mpesw.c    |  18 +-
 .../net/ethernet/mellanox/mlx5/core/sf/dev/dev.c   |  15 --
 .../net/ethernet/mellanox/mlx5/core/sf/devlink.c   | 222 ++++++---------------
 .../mellanox/mlx5/core/steering/dr_action.c        |  35 +++-
 .../mellanox/mlx5/core/steering/dr_types.h         |   1 +
 .../ethernet/mellanox/mlx5/core/steering/fs_dr.c   |   9 +-
 include/linux/mlx5/fs.h                            |   1 +
 include/linux/mlx5/mlx5_ifc.h                      |   1 +
 11 files changed, 203 insertions(+), 208 deletions(-)

[net-next 01/15] net/mlx5: Call mlx5_sf_id_erase() once in mlx5_sf_dealloc()

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

Before every call of mlx5_sf_dealloc(), there is a call to
mlx5_sf_id_erase(). So move it to the beginning of mlx5_sf_dealloc().
Also remove redundant mlx5_sf_id_erase() call from mlx5_sf_free()
as it is called only from mlx5_sf_dealloc().

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 964a5b1876f3..519033d70e05 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -112,7 +112,6 @@ mlx5_sf_alloc(struct mlx5_sf_table *table, struct mlx5_eswitch *esw,
 
 static void mlx5_sf_free(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	mlx5_sf_id_erase(table, sf);
 	mlx5_sf_hw_table_sf_free(table->dev, sf->controller, sf->id);
 	trace_mlx5_sf_free(table->dev, sf->port_index, sf->controller, sf->hw_fn_id);
 	kfree(sf);
@@ -362,6 +361,8 @@ int mlx5_devlink_sf_port_new(struct devlink *devlink,
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
+	mlx5_sf_id_erase(table, sf);
+
 	if (sf->hw_state == MLX5_VHCA_STATE_ALLOCATED) {
 		mlx5_sf_free(table, sf);
 	} else if (mlx5_sf_is_active(sf)) {
@@ -402,7 +403,6 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 	}
 
 	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-	mlx5_sf_id_erase(table, sf);
 
 	mutex_lock(&table->sf_state_lock);
 	mlx5_sf_dealloc(table, sf);
@@ -474,7 +474,6 @@ static void mlx5_sf_deactivate_all(struct mlx5_sf_table *table)
 	 */
 	xa_for_each(&table->port_indices, index, sf) {
 		mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-		mlx5_sf_id_erase(table, sf);
 		mlx5_sf_dealloc(table, sf);
 	}
 }
-- 
2.41.0

[net-next 02/15] net/mlx5: Use devlink port pointer to get the pointer of container SF struct

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

Benefit from the fact that struct devlink_port is eventually embedded
in struct mlx5_sf and use container_of() macro to get it instead of the
xarray lookup in every devlink port op.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/sf/devlink.c  | 44 +++++--------------
 1 file changed, 12 insertions(+), 32 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 519033d70e05..b4a373d2ba15 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -20,6 +20,13 @@ struct mlx5_sf {
 	u16 hw_state;
 };
 
+static void *mlx5_sf_by_dl_port(struct devlink_port *dl_port)
+{
+	struct mlx5_devlink_port *mlx5_dl_port = mlx5_devlink_port_get(dl_port);
+
+	return container_of(mlx5_dl_port, struct mlx5_sf, dl_port);
+}
+
 struct mlx5_sf_table {
 	struct mlx5_core_dev *dev; /* To refer from notifier context. */
 	struct xarray port_indices; /* port index based lookup. */
@@ -31,12 +38,6 @@ struct mlx5_sf_table {
 	struct notifier_block mdev_nb;
 };
 
-static struct mlx5_sf *
-mlx5_sf_lookup_by_index(struct mlx5_sf_table *table, unsigned int port_index)
-{
-	return xa_load(&table->port_indices, port_index);
-}
-
 static struct mlx5_sf *
 mlx5_sf_lookup_by_function_id(struct mlx5_sf_table *table, unsigned int fn_id)
 {
@@ -172,26 +173,19 @@ int mlx5_devlink_sf_port_fn_state_get(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
-	int err = 0;
 
 	table = mlx5_sf_table_try_get(dev);
 	if (!table)
 		return -EOPNOTSUPP;
 
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -EOPNOTSUPP;
-		goto sf_err;
-	}
 	mutex_lock(&table->sf_state_lock);
 	*state = mlx5_sf_to_devlink_state(sf->hw_state);
 	*opstate = mlx5_sf_to_devlink_opstate(sf->hw_state);
 	mutex_unlock(&table->sf_state_lock);
-sf_err:
 	mlx5_sf_table_put(table);
-	return err;
+	return 0;
 }
 
 static int mlx5_sf_activate(struct mlx5_core_dev *dev, struct mlx5_sf *sf,
@@ -257,8 +251,8 @@ int mlx5_devlink_sf_port_fn_state_set(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
 	int err;
 
 	table = mlx5_sf_table_try_get(dev);
@@ -267,14 +261,7 @@ int mlx5_devlink_sf_port_fn_state_set(struct devlink_port *dl_port,
 				   "Port state set is only supported in eswitch switchdev mode or SF ports are disabled.");
 		return -EOPNOTSUPP;
 	}
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -ENODEV;
-		goto out;
-	}
-
 	err = mlx5_sf_state_set(dev, table, sf, state, extack);
-out:
 	mlx5_sf_table_put(table);
 	return err;
 }
@@ -385,10 +372,9 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 			     struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
 	struct mlx5_eswitch *esw = dev->priv.eswitch;
 	struct mlx5_sf_table *table;
-	struct mlx5_sf *sf;
-	int err = 0;
 
 	table = mlx5_sf_table_try_get(dev);
 	if (!table) {
@@ -396,20 +382,14 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 				   "Port del is only supported in eswitch switchdev mode or SF ports are disabled.");
 		return -EOPNOTSUPP;
 	}
-	sf = mlx5_sf_lookup_by_index(table, dl_port->index);
-	if (!sf) {
-		err = -ENODEV;
-		goto sf_err;
-	}
 
 	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
 
 	mutex_lock(&table->sf_state_lock);
 	mlx5_sf_dealloc(table, sf);
 	mutex_unlock(&table->sf_state_lock);
-sf_err:
 	mlx5_sf_table_put(table);
-	return err;
+	return 0;
 }
 
 static bool mlx5_sf_state_update_check(const struct mlx5_sf *sf, u8 new_state)
-- 
2.41.0

[net-next 03/15] net/mlx5: Convert SF port_indices xarray to function_ids xarray

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

No need to lookup for sf by a port index. Convert the xarray to have
function id as an index and optimize the remaining function id
based lookup.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/sf/devlink.c  | 29 +++++++------------
 1 file changed, 11 insertions(+), 18 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index b4a373d2ba15..78cdfe595a01 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -29,7 +29,7 @@ static void *mlx5_sf_by_dl_port(struct devlink_port *dl_port)
 
 struct mlx5_sf_table {
 	struct mlx5_core_dev *dev; /* To refer from notifier context. */
-	struct xarray port_indices; /* port index based lookup. */
+	struct xarray function_ids; /* function id based lookup. */
 	refcount_t refcount;
 	struct completion disable_complete;
 	struct mutex sf_state_lock; /* Serializes sf state among user cmds & vhca event handler. */
@@ -41,24 +41,17 @@ struct mlx5_sf_table {
 static struct mlx5_sf *
 mlx5_sf_lookup_by_function_id(struct mlx5_sf_table *table, unsigned int fn_id)
 {
-	unsigned long index;
-	struct mlx5_sf *sf;
-
-	xa_for_each(&table->port_indices, index, sf) {
-		if (sf->hw_fn_id == fn_id)
-			return sf;
-	}
-	return NULL;
+	return xa_load(&table->function_ids, fn_id);
 }
 
-static int mlx5_sf_id_insert(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+static int mlx5_sf_function_id_insert(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	return xa_insert(&table->port_indices, sf->port_index, sf, GFP_KERNEL);
+	return xa_insert(&table->function_ids, sf->hw_fn_id, sf, GFP_KERNEL);
 }
 
-static void mlx5_sf_id_erase(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+static void mlx5_sf_function_id_erase(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	xa_erase(&table->port_indices, sf->port_index);
+	xa_erase(&table->function_ids, sf->hw_fn_id);
 }
 
 static struct mlx5_sf *
@@ -95,7 +88,7 @@ mlx5_sf_alloc(struct mlx5_sf_table *table, struct mlx5_eswitch *esw,
 	sf->hw_state = MLX5_VHCA_STATE_ALLOCATED;
 	sf->controller = controller;
 
-	err = mlx5_sf_id_insert(table, sf);
+	err = mlx5_sf_function_id_insert(table, sf);
 	if (err)
 		goto insert_err;
 
@@ -348,7 +341,7 @@ int mlx5_devlink_sf_port_new(struct devlink *devlink,
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
-	mlx5_sf_id_erase(table, sf);
+	mlx5_sf_function_id_erase(table, sf);
 
 	if (sf->hw_state == MLX5_VHCA_STATE_ALLOCATED) {
 		mlx5_sf_free(table, sf);
@@ -452,7 +445,7 @@ static void mlx5_sf_deactivate_all(struct mlx5_sf_table *table)
 	/* At this point, no new user commands can start and no vhca event can
 	 * arrive. It is safe to destroy all user created SFs.
 	 */
-	xa_for_each(&table->port_indices, index, sf) {
+	xa_for_each(&table->function_ids, index, sf) {
 		mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
 		mlx5_sf_dealloc(table, sf);
 	}
@@ -540,7 +533,7 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev)
 
 	mutex_init(&table->sf_state_lock);
 	table->dev = dev;
-	xa_init(&table->port_indices);
+	xa_init(&table->function_ids);
 	dev->priv.sf_table = table;
 	refcount_set(&table->refcount, 0);
 	table->esw_nb.notifier_call = mlx5_sf_esw_event;
@@ -579,6 +572,6 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev)
 	mlx5_esw_event_notifier_unregister(dev->priv.eswitch, &table->esw_nb);
 	WARN_ON(refcount_read(&table->refcount));
 	mutex_destroy(&table->sf_state_lock);
-	WARN_ON(!xa_empty(&table->port_indices));
+	WARN_ON(!xa_empty(&table->function_ids));
 	kfree(table);
 }
-- 
2.41.0

[net-next 04/15] net/mlx5: Move state lock taking into mlx5_sf_dealloc()

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

Instead of taking lock and calling mlx5_sf_dealloc(), move the lock
taking into mlx5_sf_dealloc(). The other caller of mlx5_sf_dealloc()
does not need it now, but will need it after a follow-up patch removing
the table reference counting.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 78cdfe595a01..bed3fe8759d2 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -341,6 +341,8 @@ int mlx5_devlink_sf_port_new(struct devlink *devlink,
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 {
+	mutex_lock(&table->sf_state_lock);
+
 	mlx5_sf_function_id_erase(table, sf);
 
 	if (sf->hw_state == MLX5_VHCA_STATE_ALLOCATED) {
@@ -358,6 +360,8 @@ static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 		mlx5_sf_hw_table_sf_deferred_free(table->dev, sf->controller, sf->id);
 		kfree(sf);
 	}
+
+	mutex_unlock(&table->sf_state_lock);
 }
 
 int mlx5_devlink_sf_port_del(struct devlink *devlink,
@@ -377,10 +381,7 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 	}
 
 	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-
-	mutex_lock(&table->sf_state_lock);
 	mlx5_sf_dealloc(table, sf);
-	mutex_unlock(&table->sf_state_lock);
 	mlx5_sf_table_put(table);
 	return 0;
 }
-- 
2.41.0

[net-next 05/15] net/mlx5: Rename mlx5_sf_deactivate_all() to mlx5_sf_del_all()

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

The function does not do deactivation, but it deletes all SFs instead.
Rename accordingly.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index bed3fe8759d2..454185ef04f3 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -437,7 +437,7 @@ static void mlx5_sf_table_enable(struct mlx5_sf_table *table)
 	refcount_set(&table->refcount, 1);
 }
 
-static void mlx5_sf_deactivate_all(struct mlx5_sf_table *table)
+static void mlx5_sf_del_all(struct mlx5_sf_table *table)
 {
 	struct mlx5_eswitch *esw = table->dev->priv.eswitch;
 	unsigned long index;
@@ -463,7 +463,7 @@ static void mlx5_sf_table_disable(struct mlx5_sf_table *table)
 	mlx5_sf_table_put(table);
 	wait_for_completion(&table->disable_complete);
 
-	mlx5_sf_deactivate_all(table);
+	mlx5_sf_del_all(table);
 }
 
 static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, void *data)
-- 
2.41.0

[net-next 06/15] net/mlx5: Push common deletion code into mlx5_sf_del()

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

Don't call the same functions for SF deletion on multiple places.
Instead, introduce a helper mlx5_sf_del() and move the code there.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/sf/devlink.c  | 19 +++++++++++--------
 1 file changed, 11 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index 454185ef04f3..c8a043b2a8e0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -364,13 +364,20 @@ static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 	mutex_unlock(&table->sf_state_lock);
 }
 
+static void mlx5_sf_del(struct mlx5_sf_table *table, struct mlx5_sf *sf)
+{
+	struct mlx5_eswitch *esw = table->dev->priv.eswitch;
+
+	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
+	mlx5_sf_dealloc(table, sf);
+}
+
 int mlx5_devlink_sf_port_del(struct devlink *devlink,
 			     struct devlink_port *dl_port,
 			     struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
 	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
-	struct mlx5_eswitch *esw = dev->priv.eswitch;
 	struct mlx5_sf_table *table;
 
 	table = mlx5_sf_table_try_get(dev);
@@ -380,8 +387,7 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 		return -EOPNOTSUPP;
 	}
 
-	mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-	mlx5_sf_dealloc(table, sf);
+	mlx5_sf_del(table, sf);
 	mlx5_sf_table_put(table);
 	return 0;
 }
@@ -439,17 +445,14 @@ static void mlx5_sf_table_enable(struct mlx5_sf_table *table)
 
 static void mlx5_sf_del_all(struct mlx5_sf_table *table)
 {
-	struct mlx5_eswitch *esw = table->dev->priv.eswitch;
 	unsigned long index;
 	struct mlx5_sf *sf;
 
 	/* At this point, no new user commands can start and no vhca event can
 	 * arrive. It is safe to destroy all user created SFs.
 	 */
-	xa_for_each(&table->function_ids, index, sf) {
-		mlx5_eswitch_unload_sf_vport(esw, sf->hw_fn_id);
-		mlx5_sf_dealloc(table, sf);
-	}
+	xa_for_each(&table->function_ids, index, sf)
+		mlx5_sf_del(table, sf);
 }
 
 static void mlx5_sf_table_disable(struct mlx5_sf_table *table)
-- 
2.41.0

[net-next 07/15] net/mlx5: Remove SF table reference counting

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

Historically, the SF table reference counting was present in order to
protect parallel executions of devlink ops. However, since currently
this is protected with devlink instance lock, the SF table reference
counting is no longer needed. Remove it entirely.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/sf/devlink.c  | 120 ++++--------------
 1 file changed, 23 insertions(+), 97 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
index c8a043b2a8e0..6c11e075cab0 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/devlink.c
@@ -30,8 +30,6 @@ static void *mlx5_sf_by_dl_port(struct devlink_port *dl_port)
 struct mlx5_sf_table {
 	struct mlx5_core_dev *dev; /* To refer from notifier context. */
 	struct xarray function_ids; /* function id based lookup. */
-	refcount_t refcount;
-	struct completion disable_complete;
 	struct mutex sf_state_lock; /* Serializes sf state among user cmds & vhca event handler. */
 	struct notifier_block esw_nb;
 	struct notifier_block vhca_nb;
@@ -111,22 +109,6 @@ static void mlx5_sf_free(struct mlx5_sf_table *table, struct mlx5_sf *sf)
 	kfree(sf);
 }
 
-static struct mlx5_sf_table *mlx5_sf_table_try_get(struct mlx5_core_dev *dev)
-{
-	struct mlx5_sf_table *table = dev->priv.sf_table;
-
-	if (!table)
-		return NULL;
-
-	return refcount_inc_not_zero(&table->refcount) ? table : NULL;
-}
-
-static void mlx5_sf_table_put(struct mlx5_sf_table *table)
-{
-	if (refcount_dec_and_test(&table->refcount))
-		complete(&table->disable_complete);
-}
-
 static enum devlink_port_fn_state mlx5_sf_to_devlink_state(u8 hw_state)
 {
 	switch (hw_state) {
@@ -166,18 +148,13 @@ int mlx5_devlink_sf_port_fn_state_get(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
+	struct mlx5_sf_table *table = dev->priv.sf_table;
 	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
-	struct mlx5_sf_table *table;
-
-	table = mlx5_sf_table_try_get(dev);
-	if (!table)
-		return -EOPNOTSUPP;
 
 	mutex_lock(&table->sf_state_lock);
 	*state = mlx5_sf_to_devlink_state(sf->hw_state);
 	*opstate = mlx5_sf_to_devlink_opstate(sf->hw_state);
 	mutex_unlock(&table->sf_state_lock);
-	mlx5_sf_table_put(table);
 	return 0;
 }
 
@@ -244,19 +221,10 @@ int mlx5_devlink_sf_port_fn_state_set(struct devlink_port *dl_port,
 				      struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(dl_port->devlink);
+	struct mlx5_sf_table *table = dev->priv.sf_table;
 	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
-	struct mlx5_sf_table *table;
-	int err;
 
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "Port state set is only supported in eswitch switchdev mode or SF ports are disabled.");
-		return -EOPNOTSUPP;
-	}
-	err = mlx5_sf_state_set(dev, table, sf, state, extack);
-	mlx5_sf_table_put(table);
-	return err;
+	return mlx5_sf_state_set(dev, table, sf, state, extack);
 }
 
 static int mlx5_sf_add(struct mlx5_core_dev *dev, struct mlx5_sf_table *table,
@@ -315,28 +283,37 @@ mlx5_sf_new_check_attr(struct mlx5_core_dev *dev, const struct devlink_port_new_
 	return 0;
 }
 
+static bool mlx5_sf_table_supported(const struct mlx5_core_dev *dev)
+{
+	return dev->priv.eswitch && MLX5_ESWITCH_MANAGER(dev) &&
+	       mlx5_sf_hw_table_supported(dev);
+}
+
 int mlx5_devlink_sf_port_new(struct devlink *devlink,
 			     const struct devlink_port_new_attrs *new_attr,
 			     struct netlink_ext_ack *extack,
 			     struct devlink_port **dl_port)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
-	struct mlx5_sf_table *table;
+	struct mlx5_sf_table *table = dev->priv.sf_table;
 	int err;
 
 	err = mlx5_sf_new_check_attr(dev, new_attr, extack);
 	if (err)
 		return err;
 
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
+	if (!mlx5_sf_table_supported(dev)) {
+		NL_SET_ERR_MSG_MOD(extack, "SF ports are not supported.");
+		return -EOPNOTSUPP;
+	}
+
+	if (!is_mdev_switchdev_mode(dev)) {
 		NL_SET_ERR_MSG_MOD(extack,
-				   "Port add is only supported in eswitch switchdev mode or SF ports are disabled.");
+				   "SF ports are only supported in eswitch switchdev mode.");
 		return -EOPNOTSUPP;
 	}
-	err = mlx5_sf_add(dev, table, new_attr, extack, dl_port);
-	mlx5_sf_table_put(table);
-	return err;
+
+	return mlx5_sf_add(dev, table, new_attr, extack, dl_port);
 }
 
 static void mlx5_sf_dealloc(struct mlx5_sf_table *table, struct mlx5_sf *sf)
@@ -377,18 +354,10 @@ int mlx5_devlink_sf_port_del(struct devlink *devlink,
 			     struct netlink_ext_ack *extack)
 {
 	struct mlx5_core_dev *dev = devlink_priv(devlink);
+	struct mlx5_sf_table *table = dev->priv.sf_table;
 	struct mlx5_sf *sf = mlx5_sf_by_dl_port(dl_port);
-	struct mlx5_sf_table *table;
-
-	table = mlx5_sf_table_try_get(dev);
-	if (!table) {
-		NL_SET_ERR_MSG_MOD(extack,
-				   "Port del is only supported in eswitch switchdev mode or SF ports are disabled.");
-		return -EOPNOTSUPP;
-	}
 
 	mlx5_sf_del(table, sf);
-	mlx5_sf_table_put(table);
 	return 0;
 }
 
@@ -414,14 +383,10 @@ static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, v
 	bool update = false;
 	struct mlx5_sf *sf;
 
-	table = mlx5_sf_table_try_get(table->dev);
-	if (!table)
-		return 0;
-
 	mutex_lock(&table->sf_state_lock);
 	sf = mlx5_sf_lookup_by_function_id(table, event->function_id);
 	if (!sf)
-		goto sf_err;
+		goto unlock;
 
 	/* When driver is attached or detached to a function, an event
 	 * notifies such state change.
@@ -431,55 +396,28 @@ static int mlx5_sf_vhca_event(struct notifier_block *nb, unsigned long opcode, v
 		sf->hw_state = event->new_vhca_state;
 	trace_mlx5_sf_update_state(table->dev, sf->port_index, sf->controller,
 				   sf->hw_fn_id, sf->hw_state);
-sf_err:
+unlock:
 	mutex_unlock(&table->sf_state_lock);
-	mlx5_sf_table_put(table);
 	return 0;
 }
 
-static void mlx5_sf_table_enable(struct mlx5_sf_table *table)
-{
-	init_completion(&table->disable_complete);
-	refcount_set(&table->refcount, 1);
-}
-
 static void mlx5_sf_del_all(struct mlx5_sf_table *table)
 {
 	unsigned long index;
 	struct mlx5_sf *sf;
 
-	/* At this point, no new user commands can start and no vhca event can
-	 * arrive. It is safe to destroy all user created SFs.
-	 */
 	xa_for_each(&table->function_ids, index, sf)
 		mlx5_sf_del(table, sf);
 }
 
-static void mlx5_sf_table_disable(struct mlx5_sf_table *table)
-{
-	if (!refcount_read(&table->refcount))
-		return;
-
-	/* Balances with refcount_set; drop the reference so that new user cmd cannot start
-	 * and new vhca event handler cannot run.
-	 */
-	mlx5_sf_table_put(table);
-	wait_for_completion(&table->disable_complete);
-
-	mlx5_sf_del_all(table);
-}
-
 static int mlx5_sf_esw_event(struct notifier_block *nb, unsigned long event, void *data)
 {
 	struct mlx5_sf_table *table = container_of(nb, struct mlx5_sf_table, esw_nb);
 	const struct mlx5_esw_event_info *mode = data;
 
 	switch (mode->new_mode) {
-	case MLX5_ESWITCH_OFFLOADS:
-		mlx5_sf_table_enable(table);
-		break;
 	case MLX5_ESWITCH_LEGACY:
-		mlx5_sf_table_disable(table);
+		mlx5_sf_del_all(table);
 		break;
 	default:
 		break;
@@ -498,9 +436,6 @@ static int mlx5_sf_mdev_event(struct notifier_block *nb, unsigned long event, vo
 	if (event != MLX5_DRIVER_EVENT_SF_PEER_DEVLINK)
 		return NOTIFY_DONE;
 
-	table = mlx5_sf_table_try_get(table->dev);
-	if (!table)
-		return NOTIFY_DONE;
 
 	mutex_lock(&table->sf_state_lock);
 	sf = mlx5_sf_lookup_by_function_id(table, event_ctx->fn_id);
@@ -513,16 +448,9 @@ static int mlx5_sf_mdev_event(struct notifier_block *nb, unsigned long event, vo
 	ret = NOTIFY_OK;
 out:
 	mutex_unlock(&table->sf_state_lock);
-	mlx5_sf_table_put(table);
 	return ret;
 }
 
-static bool mlx5_sf_table_supported(const struct mlx5_core_dev *dev)
-{
-	return dev->priv.eswitch && MLX5_ESWITCH_MANAGER(dev) &&
-	       mlx5_sf_hw_table_supported(dev);
-}
-
 int mlx5_sf_table_init(struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_table *table;
@@ -539,7 +467,6 @@ int mlx5_sf_table_init(struct mlx5_core_dev *dev)
 	table->dev = dev;
 	xa_init(&table->function_ids);
 	dev->priv.sf_table = table;
-	refcount_set(&table->refcount, 0);
 	table->esw_nb.notifier_call = mlx5_sf_esw_event;
 	err = mlx5_esw_event_notifier_register(dev->priv.eswitch, &table->esw_nb);
 	if (err)
@@ -574,7 +501,6 @@ void mlx5_sf_table_cleanup(struct mlx5_core_dev *dev)
 	mlx5_blocking_notifier_unregister(dev, &table->mdev_nb);
 	mlx5_vhca_event_notifier_unregister(table->dev, &table->vhca_nb);
 	mlx5_esw_event_notifier_unregister(dev->priv.eswitch, &table->esw_nb);
-	WARN_ON(refcount_read(&table->refcount));
 	mutex_destroy(&table->sf_state_lock);
 	WARN_ON(!xa_empty(&table->function_ids));
 	kfree(table);
-- 
2.41.0

[net-next 08/15] net/mlx5: Remove redundant max_sfs check and field from struct mlx5_sf_dev_table

[Saeed Mahameed]

From: Jiri Pirko <jiri@nvidia.com>

table->max_sfs is initialized in mlx5_sf_dev_table_create() and only
used to check for 0 in mlx5_sf_dev_add(). mlx5_sf_dev_add() is called
either from mlx5_sf_dev_state_change_handler() or
mlx5_sf_dev_add_active_work(). Both ensure max SF count is not 0,
using mlx5_sf_max_functions() helper before calling mlx5_sf_dev_add().

So remove the redundant check and no longer used max_sfs field.

Signed-off-by: Jiri Pirko <jiri@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/sf/dev/dev.c  | 15 ---------------
 1 file changed, 15 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
index 05e148db9889..0f9b280514b8 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/sf/dev/dev.c
@@ -14,7 +14,6 @@
 
 struct mlx5_sf_dev_table {
 	struct xarray devices;
-	unsigned int max_sfs;
 	phys_addr_t base_address;
 	u64 sf_bar_length;
 	struct notifier_block nb;
@@ -110,12 +109,6 @@ static void mlx5_sf_dev_add(struct mlx5_core_dev *dev, u16 sf_index, u16 fn_id,
 	sf_dev->parent_mdev = dev;
 	sf_dev->fn_id = fn_id;
 
-	if (!table->max_sfs) {
-		mlx5_adev_idx_free(id);
-		kfree(sf_dev);
-		err = -EOPNOTSUPP;
-		goto add_err;
-	}
 	sf_dev->bar_base_addr = table->base_address + (sf_index * table->sf_bar_length);
 
 	trace_mlx5_sf_dev_add(dev, sf_dev, id);
@@ -296,7 +289,6 @@ static void mlx5_sf_dev_destroy_active_work(struct mlx5_sf_dev_table *table)
 void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 {
 	struct mlx5_sf_dev_table *table;
-	unsigned int max_sfs;
 	int err;
 
 	if (!mlx5_sf_dev_supported(dev))
@@ -310,13 +302,8 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 
 	table->nb.notifier_call = mlx5_sf_dev_state_change_handler;
 	table->dev = dev;
-	if (MLX5_CAP_GEN(dev, max_num_sf))
-		max_sfs = MLX5_CAP_GEN(dev, max_num_sf);
-	else
-		max_sfs = 1 << MLX5_CAP_GEN(dev, log_max_sf);
 	table->sf_bar_length = 1 << (MLX5_CAP_GEN(dev, log_min_sf_size) + 12);
 	table->base_address = pci_resource_start(dev->pdev, 2);
-	table->max_sfs = max_sfs;
 	xa_init(&table->devices);
 	mutex_init(&table->table_lock);
 	dev->priv.sf_dev_table = table;
@@ -332,7 +319,6 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 	err = mlx5_sf_dev_vhca_arm_all(table);
 	if (err)
 		goto arm_err;
-	mlx5_core_dbg(dev, "SF DEV: max sf devices=%d\n", max_sfs);
 	return;
 
 arm_err:
@@ -340,7 +326,6 @@ void mlx5_sf_dev_table_create(struct mlx5_core_dev *dev)
 add_active_err:
 	mlx5_vhca_event_notifier_unregister(dev, &table->nb);
 vhca_err:
-	table->max_sfs = 0;
 	kfree(table);
 	dev->priv.sf_dev_table = NULL;
 table_err:
-- 
2.41.0

[net-next 09/15] net/mlx5e: Consider aggregated port speed during rate configuration

[Saeed Mahameed]

From: Jianbo Liu <jianbol@nvidia.com>

When LAG is configured, functions (PF,VF,SF) can utilize the maximum
aggregated link speed for transmission. Currently the aggregated link
speed is not considered.

Hence, improve it to use the aggregated link speed by referring to the
physical port's upper bonding device when LAG is configured.

Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../net/ethernet/mellanox/mlx5/core/esw/qos.c | 84 ++++++++++++++++---
 1 file changed, 72 insertions(+), 12 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index 1887a24ee414..f76c8f0562e9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -2,6 +2,7 @@
 /* Copyright (c) 2021, NVIDIA CORPORATION & AFFILIATES. All rights reserved. */
 
 #include "eswitch.h"
+#include "lib/mlx5.h"
 #include "esw/qos.h"
 #include "en/port.h"
 #define CREATE_TRACE_POINTS
@@ -701,6 +702,70 @@ int mlx5_esw_qos_set_vport_rate(struct mlx5_eswitch *esw, struct mlx5_vport *vpo
 	return err;
 }
 
+static u32 mlx5_esw_qos_lag_link_speed_get_locked(struct mlx5_core_dev *mdev)
+{
+	struct ethtool_link_ksettings lksettings;
+	struct net_device *slave, *master;
+	u32 speed = SPEED_UNKNOWN;
+
+	/* Lock ensures a stable reference to master and slave netdevice
+	 * while port speed of master is queried.
+	 */
+	ASSERT_RTNL();
+
+	slave = mlx5_uplink_netdev_get(mdev);
+	if (!slave)
+		goto out;
+
+	master = netdev_master_upper_dev_get(slave);
+	if (master && !__ethtool_get_link_ksettings(master, &lksettings))
+		speed = lksettings.base.speed;
+
+out:
+	return speed;
+}
+
+static int mlx5_esw_qos_max_link_speed_get(struct mlx5_core_dev *mdev, u32 *link_speed_max,
+					   bool hold_rtnl_lock, struct netlink_ext_ack *extack)
+{
+	int err;
+
+	if (!mlx5_lag_is_active(mdev))
+		goto skip_lag;
+
+	if (hold_rtnl_lock)
+		rtnl_lock();
+
+	*link_speed_max = mlx5_esw_qos_lag_link_speed_get_locked(mdev);
+
+	if (hold_rtnl_lock)
+		rtnl_unlock();
+
+	if (*link_speed_max != (u32)SPEED_UNKNOWN)
+		return 0;
+
+skip_lag:
+	err = mlx5_port_max_linkspeed(mdev, link_speed_max);
+	if (err)
+		NL_SET_ERR_MSG_MOD(extack, "Failed to get link maximum speed");
+
+	return err;
+}
+
+static int mlx5_esw_qos_link_speed_verify(struct mlx5_core_dev *mdev,
+					  const char *name, u32 link_speed_max,
+					  u64 value, struct netlink_ext_ack *extack)
+{
+	if (value > link_speed_max) {
+		pr_err("%s rate value %lluMbps exceed link maximum speed %u.\n",
+		       name, value, link_speed_max);
+		NL_SET_ERR_MSG_MOD(extack, "TX rate value exceed link maximum speed");
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
 int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32 rate_mbps)
 {
 	u32 ctx[MLX5_ST_SZ_DW(scheduling_context)] = {};
@@ -744,12 +809,6 @@ static int esw_qos_devlink_rate_to_mbps(struct mlx5_core_dev *mdev, const char *
 	u64 value;
 	int err;
 
-	err = mlx5_port_max_linkspeed(mdev, &link_speed_max);
-	if (err) {
-		NL_SET_ERR_MSG_MOD(extack, "Failed to get link maximum speed");
-		return err;
-	}
-
 	value = div_u64_rem(*rate, MLX5_LINKSPEED_UNIT, &remainder);
 	if (remainder) {
 		pr_err("%s rate value %lluBps not in link speed units of 1Mbps.\n",
@@ -758,12 +817,13 @@ static int esw_qos_devlink_rate_to_mbps(struct mlx5_core_dev *mdev, const char *
 		return -EINVAL;
 	}
 
-	if (value > link_speed_max) {
-		pr_err("%s rate value %lluMbps exceed link maximum speed %u.\n",
-		       name, value, link_speed_max);
-		NL_SET_ERR_MSG_MOD(extack, "TX rate value exceed link maximum speed");
-		return -EINVAL;
-	}
+	err = mlx5_esw_qos_max_link_speed_get(mdev, &link_speed_max, true, extack);
+	if (err)
+		return err;
+
+	err = mlx5_esw_qos_link_speed_verify(mdev, name, link_speed_max, value, extack);
+	if (err)
+		return err;
 
 	*rate = value;
 	return 0;
-- 
2.41.0

[net-next 10/15] net/mlx5e: Check police action rate for matchall filter

[Saeed Mahameed]

From: Jianbo Liu <jianbol@nvidia.com>

As matchall filter uses TSAR (Transmit Scheduling Arbiter) for rate
limit, the rate of police action should not be over the port's max
link speed, or the maximum aggregated speed of both ports if LAG is
configured.

Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Reviewed-by: Parav Pandit <parav@nvidia.com>
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
index f76c8f0562e9..d2ebe56c3977 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/qos.c
@@ -770,6 +770,7 @@ int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32
 {
 	u32 ctx[MLX5_ST_SZ_DW(scheduling_context)] = {};
 	struct mlx5_vport *vport;
+	u32 link_speed_max;
 	u32 bitmask;
 	int err;
 
@@ -777,6 +778,17 @@ int mlx5_esw_qos_modify_vport_rate(struct mlx5_eswitch *esw, u16 vport_num, u32
 	if (IS_ERR(vport))
 		return PTR_ERR(vport);
 
+	if (rate_mbps) {
+		err = mlx5_esw_qos_max_link_speed_get(esw->dev, &link_speed_max, false, NULL);
+		if (err)
+			return err;
+
+		err = mlx5_esw_qos_link_speed_verify(esw->dev, "Police",
+						     link_speed_max, rate_mbps, NULL);
+		if (err)
+			return err;
+	}
+
 	mutex_lock(&esw->state_lock);
 	if (!vport->qos.enabled) {
 		/* Eswitch QoS wasn't enabled yet. Enable it and vport QoS. */
-- 
2.41.0

[net-next 11/15] net/mlx5: Bridge, Enable mcast in smfs steering mode

[Saeed Mahameed]

From: Erez Shitrit <erezsh@nvidia.com>

In order to have mcast offloads the driver needs the following:
It should know if that mcast comes from wire port, in addition the flow
should not be marked as any specific source, that way it will give the
flexibility for the driver not to be depended on the way iterator
implemented in the FW.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Vlad Buslov <vladbu@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../ethernet/mellanox/mlx5/core/esw/bridge_mcast.c    | 11 ++---------
 include/linux/mlx5/fs.h                               |  1 +
 2 files changed, 3 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
index 7a01714b3780..a7ed87e9d842 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/esw/bridge_mcast.c
@@ -78,6 +78,8 @@ mlx5_esw_bridge_mdb_flow_create(u16 esw_owner_vhca_id, struct mlx5_esw_bridge_md
 	xa_for_each(&entry->ports, idx, port) {
 		dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 		dests[i].ft = port->mcast.ft;
+		if (port->vport_num == MLX5_VPORT_UPLINK)
+			dests[i].ft->flags |= MLX5_FLOW_TABLE_UPLINK_VPORT;
 		i++;
 	}
 
@@ -585,10 +587,6 @@ mlx5_esw_bridge_mcast_vlan_flow_create(u16 vlan_proto, struct mlx5_esw_bridge_po
 	if (!rule_spec)
 		return ERR_PTR(-ENOMEM);
 
-	if (MLX5_CAP_ESW_FLOWTABLE(bridge->br_offloads->esw->dev, flow_source) &&
-	    port->vport_num == MLX5_VPORT_UPLINK)
-		rule_spec->flow_context.flow_source =
-			MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
 	rule_spec->match_criteria_enable = MLX5_MATCH_OUTER_HEADERS;
 
 	flow_act.action |= MLX5_FLOW_CONTEXT_ACTION_PACKET_REFORMAT;
@@ -660,11 +658,6 @@ mlx5_esw_bridge_mcast_fwd_flow_create(struct mlx5_esw_bridge_port *port)
 	if (!rule_spec)
 		return ERR_PTR(-ENOMEM);
 
-	if (MLX5_CAP_ESW_FLOWTABLE(bridge->br_offloads->esw->dev, flow_source) &&
-	    port->vport_num == MLX5_VPORT_UPLINK)
-		rule_spec->flow_context.flow_source =
-			MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
-
 	if (MLX5_CAP_ESW(bridge->br_offloads->esw->dev, merged_eswitch)) {
 		dest.vport.flags = MLX5_FLOW_DEST_VPORT_VHCA_ID;
 		dest.vport.vhca_id = port->esw_owner_vhca_id;
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index 1e00c2436377..6f7725238abc 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -67,6 +67,7 @@ enum {
 	MLX5_FLOW_TABLE_TERMINATION = BIT(2),
 	MLX5_FLOW_TABLE_UNMANAGED = BIT(3),
 	MLX5_FLOW_TABLE_OTHER_VPORT = BIT(4),
+	MLX5_FLOW_TABLE_UPLINK_VPORT = BIT(5),
 };
 
 #define LEFTOVERS_RULE_NUM	 2
-- 
2.41.0

[net-next 12/15] net/mlx5: DR, Add check for multi destination FTE

[Saeed Mahameed]

From: Erez Shitrit <erezsh@nvidia.com>

The driver should not allow rule that forward to more than one FT in TX
flow unless there is a specific support from the FW.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c         | 13 +++++++++++++
 1 file changed, 13 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 5b83da08692d..7179542e9164 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -55,6 +55,12 @@ static const char *dr_action_id_to_str(enum mlx5dr_action_type action_id)
 	return action_type_to_str[action_id];
 }
 
+static bool mlx5dr_action_supp_fwd_fdb_multi_ft(struct mlx5_core_dev *dev)
+{
+	return (MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_any_table_limit_regc) ||
+		MLX5_CAP_ESW_FLOWTABLE(dev, fdb_multi_path_any_table));
+}
+
 static const enum dr_action_valid_state
 next_action_state[DR_ACTION_DOMAIN_MAX][DR_ACTION_STATE_MAX][DR_ACTION_TYP_MAX] = {
 	[DR_ACTION_DOMAIN_NIC_INGRESS] = {
@@ -1167,6 +1173,7 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 	struct mlx5dr_action **ref_actions;
 	struct mlx5dr_action *action;
 	bool reformat_req = false;
+	u16 num_dst_ft = 0;
 	u32 num_of_ref = 0;
 	u32 ref_act_cnt;
 	int ret;
@@ -1210,6 +1217,12 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 			break;
 
 		case DR_ACTION_TYP_FT:
+			if (num_dst_ft &&
+			    !mlx5dr_action_supp_fwd_fdb_multi_ft(dmn->mdev)) {
+				mlx5dr_dbg(dmn, "multiple FT destinations not supported\n");
+				goto free_ref_actions;
+			}
+			num_dst_ft++;
 			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
 			if (dest_action->dest_tbl->is_fw_tbl)
 				hw_dests[i].ft_id = dest_action->dest_tbl->fw_tbl.id;
-- 
2.41.0

[net-next 13/15] net/mlx5: DR, Handle multi destination action in the right order

[Saeed Mahameed]

From: Erez Shitrit <erezsh@nvidia.com>

Whenever we have few destinations from Flow-table type we need to put
the one that goes to the wire to be the last one.

We are using FW in order to get iterator, the FW uses RX for the first
destinations and TX for the last destination, if we want the packet to
be directed to the wire it should be done in the TX path and not in the
RX.
The code now checks if the FT is directed to the wire and if so puts it
as the last destination.

Signed-off-by: Erez Shitrit <erezsh@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Yevgeny Kliteynik <kliteyn@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 .../mellanox/mlx5/core/steering/dr_action.c   | 22 +++++++++++++++++--
 .../mellanox/mlx5/core/steering/dr_types.h    |  1 +
 .../mellanox/mlx5/core/steering/fs_dr.c       |  9 +++++++-
 3 files changed, 29 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
index 7179542e9164..6ea88a581804 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_action.c
@@ -1169,13 +1169,16 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 				   bool ignore_flow_level,
 				   u32 flow_source)
 {
+	struct mlx5dr_cmd_flow_destination_hw_info tmp_hw_dest;
 	struct mlx5dr_cmd_flow_destination_hw_info *hw_dests;
 	struct mlx5dr_action **ref_actions;
 	struct mlx5dr_action *action;
 	bool reformat_req = false;
+	bool is_ft_wire = false;
 	u16 num_dst_ft = 0;
 	u32 num_of_ref = 0;
 	u32 ref_act_cnt;
+	u16 last_dest;
 	int ret;
 	int i;
 
@@ -1224,10 +1227,15 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 			}
 			num_dst_ft++;
 			hw_dests[i].type = MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE;
-			if (dest_action->dest_tbl->is_fw_tbl)
+			if (dest_action->dest_tbl->is_fw_tbl) {
 				hw_dests[i].ft_id = dest_action->dest_tbl->fw_tbl.id;
-			else
+			} else {
 				hw_dests[i].ft_id = dest_action->dest_tbl->tbl->table_id;
+				if (dest_action->dest_tbl->is_wire_ft) {
+					is_ft_wire = true;
+					last_dest = i;
+				}
+			}
 			break;
 
 		default:
@@ -1236,6 +1244,16 @@ mlx5dr_action_create_mult_dest_tbl(struct mlx5dr_domain *dmn,
 		}
 	}
 
+	/* In multidest, the FW does the iterator in the RX except of the last
+	 * one that done in the TX.
+	 * So, if one of the ft target is wire, put it at the end of the dest list.
+	 */
+	if (is_ft_wire && num_dst_ft > 1) {
+		tmp_hw_dest = hw_dests[last_dest];
+		hw_dests[last_dest] = hw_dests[num_of_dests - 1];
+		hw_dests[num_of_dests - 1] = tmp_hw_dest;
+	}
+
 	action = dr_action_create_generic(DR_ACTION_TYP_FT);
 	if (!action)
 		goto free_ref_actions;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
index 6c59de3e28f6..55dc7383477c 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h
@@ -1064,6 +1064,7 @@ struct mlx5dr_action_sampler {
 
 struct mlx5dr_action_dest_tbl {
 	u8 is_fw_tbl:1;
+	u8 is_wire_ft:1;
 	union {
 		struct mlx5dr_table *tbl;
 		struct {
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
index 14f6df88b1f9..50c2554c9ccf 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/steering/fs_dr.c
@@ -209,10 +209,17 @@ static struct mlx5dr_action *create_ft_action(struct mlx5dr_domain *domain,
 					      struct mlx5_flow_rule *dst)
 {
 	struct mlx5_flow_table *dest_ft = dst->dest_attr.ft;
+	struct mlx5dr_action *tbl_action;
 
 	if (mlx5dr_is_fw_table(dest_ft))
 		return mlx5dr_action_create_dest_flow_fw_table(domain, dest_ft);
-	return mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+
+	tbl_action = mlx5dr_action_create_dest_table(dest_ft->fs_dr_table.dr_table);
+	if (tbl_action)
+		tbl_action->dest_tbl->is_wire_ft =
+			dest_ft->flags & MLX5_FLOW_TABLE_UPLINK_VPORT ? 1 : 0;
+
+	return tbl_action;
 }
 
 static struct mlx5dr_action *create_range_action(struct mlx5dr_domain *domain,
-- 
2.41.0

[net-next 14/15] net/mlx5: Add a health error syndrome for pci data poisoned

[Saeed Mahameed]

From: Moshe Shemesh <moshe@nvidia.com>

Add new health error syndrome to indicate that pci data poisoned error
has been received while fetching device ICM data.

Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlx5/core/health.c | 2 ++
 include/linux/mlx5/mlx5_ifc.h                    | 1 +
 2 files changed, 3 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/health.c b/drivers/net/ethernet/mellanox/mlx5/core/health.c
index 2fb2598b775e..1c220048ae9a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/health.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/health.c
@@ -365,6 +365,8 @@ static const char *hsynd_str(u8 synd)
 		return "FFSER error";
 	case MLX5_INITIAL_SEG_HEALTH_SYNDROME_HIGH_TEMP_ERR:
 		return "High temperature";
+	case MLX5_INITIAL_SEG_HEALTH_SYNDROME_ICM_PCI_POISONED_ERR:
+		return "ICM fetch PCI data poisoned error";
 	default:
 		return "unrecognized error";
 	}
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index dd8421d021cf..b23d8ff286a1 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -10574,6 +10574,7 @@ enum {
 	MLX5_INITIAL_SEG_HEALTH_SYNDROME_EQ_INV                       = 0xe,
 	MLX5_INITIAL_SEG_HEALTH_SYNDROME_FFSER_ERR                    = 0xf,
 	MLX5_INITIAL_SEG_HEALTH_SYNDROME_HIGH_TEMP_ERR                = 0x10,
+	MLX5_INITIAL_SEG_HEALTH_SYNDROME_ICM_PCI_POISONED_ERR         = 0x12,
 };
 
 struct mlx5_ifc_initial_seg_bits {
-- 
2.41.0

[net-next 15/15] net/mlx5: Enable 4 ports multiport E-switch

[Saeed Mahameed]

From: Shay Drory <shayd@nvidia.com>

enable_mpesw() assumed only 2 ports are available, fix this by removing
that assumption and looping through the existing lag ports to enable multi-port
E-switch for cards with more than 2 ports.

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/lag/mpesw.c    | 18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
index 4bf15391525c..0857eebf4f07 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lag/mpesw.c
@@ -65,12 +65,12 @@ static int mlx5_mpesw_metadata_set(struct mlx5_lag *ldev)
 	return err;
 }
 
-#define MLX5_LAG_MPESW_OFFLOADS_SUPPORTED_PORTS 2
+#define MLX5_LAG_MPESW_OFFLOADS_SUPPORTED_PORTS 4
 static int enable_mpesw(struct mlx5_lag *ldev)
 {
 	struct mlx5_core_dev *dev0 = ldev->pf[MLX5_LAG_P1].dev;
-	struct mlx5_core_dev *dev1 = ldev->pf[MLX5_LAG_P2].dev;
 	int err;
+	int i;
 
 	if (ldev->mode != MLX5_LAG_MODE_NONE)
 		return -EINVAL;
@@ -98,11 +98,11 @@ static int enable_mpesw(struct mlx5_lag *ldev)
 
 	dev0->priv.flags &= ~MLX5_PRIV_FLAGS_DISABLE_IB_ADEV;
 	mlx5_rescan_drivers_locked(dev0);
-	err = mlx5_eswitch_reload_reps(dev0->priv.eswitch);
-	if (!err)
-		err = mlx5_eswitch_reload_reps(dev1->priv.eswitch);
-	if (err)
-		goto err_rescan_drivers;
+	for (i = 0; i < ldev->ports; i++) {
+		err = mlx5_eswitch_reload_reps(ldev->pf[i].dev->priv.eswitch);
+		if (err)
+			goto err_rescan_drivers;
+	}
 
 	return 0;
 
@@ -112,8 +112,8 @@ static int enable_mpesw(struct mlx5_lag *ldev)
 	mlx5_deactivate_lag(ldev);
 err_add_devices:
 	mlx5_lag_add_devices(ldev);
-	mlx5_eswitch_reload_reps(dev0->priv.eswitch);
-	mlx5_eswitch_reload_reps(dev1->priv.eswitch);
+	for (i = 0; i < ldev->ports; i++)
+		mlx5_eswitch_reload_reps(ldev->pf[i].dev->priv.eswitch);
 	mlx5_mpesw_metadata_cleanup(ldev);
 	return err;
 }
-- 
2.41.0

Re: [net-next 01/15] net/mlx5: Call mlx5_sf_id_erase() once in mlx5_sf_dealloc()

[patchwork-bot+netdevbpf]

Hello:

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

On Tue, 19 Sep 2023 23:35:38 -0700 you wrote:
> From: Jiri Pirko <jiri@nvidia.com>
> 
> Before every call of mlx5_sf_dealloc(), there is a call to
> mlx5_sf_id_erase(). So move it to the beginning of mlx5_sf_dealloc().
> Also remove redundant mlx5_sf_id_erase() call from mlx5_sf_free()
> as it is called only from mlx5_sf_dealloc().
> 
> [...]

Here is the summary with links:
  - [net-next,01/15] net/mlx5: Call mlx5_sf_id_erase() once in mlx5_sf_dealloc()
    https://git.kernel.org/netdev/net-next/c/2597ee190b4e
  - [net-next,02/15] net/mlx5: Use devlink port pointer to get the pointer of container SF struct
    https://git.kernel.org/netdev/net-next/c/9caeb1475c3e
  - [net-next,03/15] net/mlx5: Convert SF port_indices xarray to function_ids xarray
    https://git.kernel.org/netdev/net-next/c/2284a4836251
  - [net-next,04/15] net/mlx5: Move state lock taking into mlx5_sf_dealloc()
    https://git.kernel.org/netdev/net-next/c/9497036dfbb8
  - [net-next,05/15] net/mlx5: Rename mlx5_sf_deactivate_all() to mlx5_sf_del_all()
    https://git.kernel.org/netdev/net-next/c/a65362f2be8d
  - [net-next,06/15] net/mlx5: Push common deletion code into mlx5_sf_del()
    https://git.kernel.org/netdev/net-next/c/a3cc822beacc
  - [net-next,07/15] net/mlx5: Remove SF table reference counting
    https://git.kernel.org/netdev/net-next/c/2fe6545ef541
  - [net-next,08/15] net/mlx5: Remove redundant max_sfs check and field from struct mlx5_sf_dev_table
    https://git.kernel.org/netdev/net-next/c/7c35cd836f21
  - [net-next,09/15] net/mlx5e: Consider aggregated port speed during rate configuration
    https://git.kernel.org/netdev/net-next/c/8d88e198dcaf
  - [net-next,10/15] net/mlx5e: Check police action rate for matchall filter
    https://git.kernel.org/netdev/net-next/c/4291ab7112ea
  - [net-next,11/15] net/mlx5: Bridge, Enable mcast in smfs steering mode
    https://git.kernel.org/netdev/net-next/c/653b7eb9d744
  - [net-next,12/15] net/mlx5: DR, Add check for multi destination FTE
    https://git.kernel.org/netdev/net-next/c/f6f46e7173cb
  - [net-next,13/15] net/mlx5: DR, Handle multi destination action in the right order
    https://git.kernel.org/netdev/net-next/c/3b81bcbaee28
  - [net-next,14/15] net/mlx5: Add a health error syndrome for pci data poisoned
    https://git.kernel.org/netdev/net-next/c/e0cc92fd945a
  - [net-next,15/15] net/mlx5: Enable 4 ports multiport E-switch
    https://git.kernel.org/netdev/net-next/c/e738e3550452

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