Commit 559f418e authored by Jonathan Kim's avatar Jonathan Kim Committed by Alex Deucher

drm/amdkfd: report the numa weight between host and device over xgmi

GPUs connected to CPUs over xGMI are bidirectional so set weight by a
single hop both ways.
Signed-off-by: default avatarJonathan Kim <jonathan.kim@amd.com>
Tested-by: default avatarRamesh Errabolu <ramesh.errabolu@amd.com>
Reviewed-by: default avatarFelix Kuehling <felix.kuehling@amd.com>
Signed-off-by: default avatarAlex Deucher <alexander.deucher@amd.com>
parent deb68983
...@@ -1140,6 +1140,7 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size, ...@@ -1140,6 +1140,7 @@ static int kfd_fill_gpu_direct_io_link_to_cpu(int *avail_size,
*/ */
sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL; sub_type_hdr->flags |= CRAT_IOLINK_FLAGS_BI_DIRECTIONAL;
sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI; sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_XGMI;
sub_type_hdr->num_hops_xgmi = 1;
} else { } else {
sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_PCIEXPRESS; sub_type_hdr->io_interface_type = CRAT_IOLINK_TYPE_PCIEXPRESS;
} }
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment