]> git.itanic.dy.fi Git - linux-stable/commit
RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile
authorMaher Sanalla <msanalla@nvidia.com>
Mon, 29 Aug 2022 09:02:27 +0000 (12:02 +0300)
committerLeon Romanovsky <leon@kernel.org>
Mon, 5 Sep 2022 11:49:46 +0000 (14:49 +0300)
commit9ca05b0f27de928be121cccf07735819dc9e1ed3
treee7ac6395ac67162e60eba9e12e44889f08fa2445
parent85eaeb5058f0f04dffb124c97c86b4f18db0b833
RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile

When the RDMA auxiliary driver probes, it sets its profile based on
devlink driverinit value. The latter might not be in sync with FW yet
(In case devlink reload is not performed), thus causing a mismatch
between RDMA driver and FW. This results in the following FW syndrome
when the RDMA driver tries to adjust RoCE state, which fails the probe:

"0xC1F678 | modify_nic_vport_context: roce_en set on a vport that
doesn't support roce"

To prevent this, select the PF profile based on FW RoCE capability
instead of relying on devlink driverinit value.
To provide backward compatibility of the RoCE disable feature, on older
FW's where roce_rw is not set (FW RoCE capability is read-only), keep
the current behavior e.g., rely on devlink driverinit value.

Fixes: fbfa97b4d79f ("net/mlx5: Disable roce at HCA level")
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://lore.kernel.org/r/cb34ce9a1df4a24c135cb804db87f7d2418bd6cc.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
drivers/infiniband/hw/mlx5/main.c
drivers/net/ethernet/mellanox/mlx5/core/main.c
include/linux/mlx5/driver.h