diff options
| author | Jianbo Liu <jianbol@nvidia.com> | 2024-06-03 13:26:38 +0300 |
|---|---|---|
| committer | Leon Romanovsky <leon@kernel.org> | 2024-06-16 18:37:50 +0300 |
| commit | 638420115cc4ad6c3a2683bf46a052b505abb202 (patch) | |
| tree | b4299a72adf3813faf0efab971de711a254c601e /drivers/infiniband/hw/mlx5/umr.h | |
| parent | ae6f6dd5fd0f77368ba5888c2c57bd23abfe4f35 (diff) | |
| download | linux-638420115cc4ad6c3a2683bf46a052b505abb202.tar.gz linux-638420115cc4ad6c3a2683bf46a052b505abb202.tar.bz2 linux-638420115cc4ad6c3a2683bf46a052b505abb202.zip | |
IB/mlx5: Create UMR QP just before first reg_mr occurs
UMR QP is not used in some cases, so move QP and its CQ creations from
driver load flow to the time first reg_mr occurs, that is when MR
interfaces are first called.
The initialization of dev->umrc.pd and dev->umrc.lock is still done in
driver load because pd is needed for mlx5_mkey_cache_init and the lock
is reused to protect against the concurrent creation.
When testing 4G bytes memory registration latency with rtool [1] and 8
threads in parallel, there is minor performance degradation (<5% for
the max latency) is seen for the first reg_mr with this change.
Link: https://github.com/paravmellanox/rtool [1]
Signed-off-by: Jianbo Liu <jianbol@nvidia.com>
Link: https://lore.kernel.org/r/55d3c4f8a542fd974d8a4c5816eccfb318a59b38.1717409369.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Diffstat (limited to 'drivers/infiniband/hw/mlx5/umr.h')
| -rw-r--r-- | drivers/infiniband/hw/mlx5/umr.h | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/drivers/infiniband/hw/mlx5/umr.h b/drivers/infiniband/hw/mlx5/umr.h index 3799bb758e49..5f734dc72bef 100644 --- a/drivers/infiniband/hw/mlx5/umr.h +++ b/drivers/infiniband/hw/mlx5/umr.h @@ -16,6 +16,9 @@ int mlx5r_umr_resource_init(struct mlx5_ib_dev *dev); void mlx5r_umr_resource_cleanup(struct mlx5_ib_dev *dev); +int mlx5r_umr_init(struct mlx5_ib_dev *dev); +void mlx5r_umr_cleanup(struct mlx5_ib_dev *dev); + static inline bool mlx5r_umr_can_load_pas(struct mlx5_ib_dev *dev, size_t length) { |