[PATCH mlx5-next 0/2] mlx5-next updates 2026-10-05

[PATCH mlx5-next 0/2] mlx5-next updates 2026-10-05

[Tariq Toukan]

Hi,

This series contains mlx5 shared updates as preparation for upcoming
features.

Regards,
Tariq

Or Har-Toov (1):
  net/mlx5: Add cap_tx_speed and effective_tx_speed support to IFC

Shay Drory (1):
  net/mlx5: IFC, Add capability bits for virtual LAG

 include/linux/mlx5/mlx5_ifc.h | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)


base-commit: 00290b9ff59ef47b990d26a6b1a8b2ea6dedf5e9
-- 
2.44.0

[PATCH mlx5-next 1/2] net/mlx5: Add cap_tx_speed and effective_tx_speed support to IFC

[Tariq Toukan]

From: Or Har-Toov <ohartoov@nvidia.com>

Introduce the cap_tx_speed and effective_tx_speed fields to the query
and modify_vport_state structures.

Add the esw_vport_state_cap_tx_speed and
esw_vport_state_effective_tx_speed capability bits, indicating the
firmware supports modifying the cap_tx_speed and effective_tx_speed
fields, via the MODIFY_VPORT_STATE command.

Signed-off-by: Or Har-Toov <ohartoov@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Nimrod Oren <noren@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 include/linux/mlx5/mlx5_ifc.h | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index a712463e89b4..989408764e1b 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -1070,7 +1070,9 @@ struct mlx5_ifc_e_switch_cap_bits {
 	u8         root_ft_on_other_esw[0x1];
 	u8         reserved_at_a[0x1];
 	u8         esw_vport_state_max_tx_speed[0x1];
-	u8         reserved_at_c[0xd];
+	u8         esw_vport_state_cap_tx_speed[0x1];
+	u8         esw_vport_state_effective_tx_speed[0x1];
+	u8         reserved_at_e[0xb];
 	u8         esw_functions_changed[0x1];
 	u8         reserved_at_1a[0x1];
 	u8         ecpf_vport_exists[0x1];
@@ -5492,6 +5494,9 @@ struct mlx5_ifc_query_vport_state_out_bits {
 	u8         reserved_at_70[0x8];
 	u8         admin_state[0x4];
 	u8         state[0x4];
+
+	u8         cap_tx_speed[0x10];
+	u8         effective_tx_speed[0x10];
 };
 
 struct mlx5_ifc_array1024_auto_bits {
@@ -7840,6 +7845,9 @@ struct mlx5_ifc_modify_vport_state_in_bits {
 	u8         reserved_at_74[0x4];
 	u8         admin_state[0x4];
 	u8         reserved_at_7c[0x4];
+
+	u8         cap_tx_speed[0x10];
+	u8         effective_tx_speed[0x10];
 };
 
 struct mlx5_ifc_modify_tis_out_bits {
-- 
2.44.0

[PATCH mlx5-next 2/2] net/mlx5: IFC, Add capability bits for virtual LAG

[Tariq Toukan]

From: Shay Drory <shayd@nvidia.com>

Virtual LAG can also be built over virtual uplinks rather than physical
ports.
The support for this virtual LAG is indicated by multi_uplink_eswitch
and port_select_eswitch caps bit.

Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Reviewed-by: Moshe Shemesh <moshe@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
---
 include/linux/mlx5/mlx5_ifc.h | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 989408764e1b..f1ee35aa884a 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -929,7 +929,7 @@ struct mlx5_ifc_flow_table_nic_cap_bits {
 struct mlx5_ifc_port_selection_cap_bits {
 	u8         reserved_at_0[0x10];
 	u8         port_select_flow_table[0x1];
-	u8         reserved_at_11[0x1];
+	u8         port_select_eswitch[0x1];
 	u8         port_select_flow_table_bypass[0x1];
 	u8         reserved_at_13[0xd];
 
@@ -2010,7 +2010,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         reserved_at_3b0[0x2];
 	u8         qp_latency_sensitive_disable[0x1];
 	u8         log_max_stride_sz_sq[0x5];
-	u8         reserved_at_3b8[0x3];
+	u8         reserved_at_3b8[0x1];
+	u8         multi_uplink_eswitch[0x1];
+	u8         reserved_at_3ba[0x1];
 	u8         log_min_stride_sz_sq[0x5];
 
 	u8         hairpin[0x1];
-- 
2.44.0

Re: [PATCH mlx5-next 0/2] mlx5-next updates 2026-10-05

[Przemek Kitszel]

On 10/5/26 10:40 AM, Tariq Toukan wrote:
> Hi,
> 
> This series contains mlx5 shared updates as preparation for upcoming
> features.
> 
> Regards,
> Tariq
> 
> Or Har-Toov (1):
>    net/mlx5: Add cap_tx_speed and effective_tx_speed support to IFC
> 
> Shay Drory (1):
>    net/mlx5: IFC, Add capability bits for virtual LAG
> 
>   include/linux/mlx5/mlx5_ifc.h | 16 +++++++++++++---
>   1 file changed, 13 insertions(+), 3 deletions(-)
> 
> 
> base-commit: 00290b9ff59ef47b990d26a6b1a8b2ea6dedf5e9

my understanding is that you should at least link to the
actual code using the new stuff

Re: [PATCH mlx5-next 0/2] mlx5-next updates 2026-10-05

[Jakub Kicinski]

On Mon, 5 Oct 2026 15:29:15 +0200 Przemek Kitszel wrote:
> On 10/5/26 10:40 AM, Tariq Toukan wrote:
> > Hi,
> > 
> > This series contains mlx5 shared updates as preparation for upcoming
> > features.
> > 
> > Regards,
> > Tariq
> > 
> > Or Har-Toov (1):
> >    net/mlx5: Add cap_tx_speed and effective_tx_speed support to IFC
> > 
> > Shay Drory (1):
> >    net/mlx5: IFC, Add capability bits for virtual LAG
> > 
> >   include/linux/mlx5/mlx5_ifc.h | 16 +++++++++++++---
> >   1 file changed, 13 insertions(+), 3 deletions(-)
> > 
> > 
> > base-commit: 00290b9ff59ef47b990d26a6b1a8b2ea6dedf5e9  
> 
> my understanding is that you should at least link to the
> actual code using the new stuff

I guess, tho this is basically a FW API header update.
It's not like we inspect the FW API update patches to make sure all of
the fields / requests are actually used later..
I did not spot anything "funny" here.