Skip to content

Commit

Permalink
RDMA/mlx5: Add capability for max sge to get optimized performance
Browse files Browse the repository at this point in the history
Allows the IB device to provide a value of maximum scatter gather entries
per RDMA READ.

In certain cases it may be preferable for a device to perform UMR memory
registration rather than have many scatter entries in a single RDMA READ.
This provides a significant performance increase in devices capable of
using different memory registration schemes based on the number of scatter
gather entries. This general capability allows each device vendor to fine
tune when it is better to use memory registration.

Link: https://lore.kernel.org/r/20191007135933.12483-4-leon@kernel.org
Signed-off-by: Yamin Friedman <yaminf@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
  • Loading branch information
Yamin Friedman authored and Jason Gunthorpe committed Oct 22, 2019
1 parent 00bd143 commit 3660905
Showing 1 changed file with 2 additions and 0 deletions.
2 changes: 2 additions & 0 deletions drivers/infiniband/hw/mlx5/main.c
Original file line number Diff line number Diff line change
Expand Up @@ -1011,6 +1011,8 @@ static int mlx5_ib_query_device(struct ib_device *ibdev,
1 << MLX5_CAP_GEN(mdev, log_max_klm_list_size);
props->max_pi_fast_reg_page_list_len =
props->max_fast_reg_page_list_len / 2;
props->max_sgl_rd =
MLX5_CAP_GEN(mdev, max_sgl_for_optimized_performance);
get_atomic_caps_qp(dev, props);
props->masked_atomic_cap = IB_ATOMIC_NONE;
props->max_mcast_grp = 1 << MLX5_CAP_GEN(mdev, log_max_mcg);
Expand Down

0 comments on commit 3660905

Please sign in to comment.