Skip to content

UCT/GDA: Move peermem check outside the loop. #11001

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Nov 14, 2025
+20 −19
Merged

UCT/GDA: Move peermem check outside the loop. #11001

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
39 changes: 20 additions & 19 deletions src/uct/ib/mlx5/gdaki/gdaki.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -615,6 +615,25 @@ uct_gdaki_query_tl_devices(uct_md_h tl_md,
ucs_sys_dev_distance_t dist;
int i, num_gpus;

/*
* Save the result of peermem driver check in a global flag to avoid
* printing diag message for each MD.
*/
if (peermem_loaded == -1) {
peermem_loaded = !!(md->super.reg_mem_types &
UCS_BIT(UCS_MEMORY_TYPE_CUDA));
if (peermem_loaded == 0) {
ucs_diag("GDAKI not supported, please load "
"Nvidia peermem driver by running "
"\"modprobe nvidia_peermem\"");
}
}

if (peermem_loaded == 0) {
status = UCS_ERR_NO_DEVICE;
goto out;
}

status = UCT_CUDADRV_FUNC_LOG_ERR(cuDeviceGetCount(&num_gpus));
if (status != UCS_OK) {
return status;
Expand Down Expand Up @@ -652,25 +671,6 @@ uct_gdaki_query_tl_devices(uct_md_h tl_md,
goto err;
}

/*
* Save the result of peermem driver check in a global flag to avoid
* printing diag message for each GPU and MD.
*/
if (peermem_loaded == -1) {
peermem_loaded = !!(md->super.reg_mem_types &
UCS_BIT(UCS_MEMORY_TYPE_CUDA));
if (peermem_loaded == 0) {
ucs_diag("GDAKI not supported, please load "
"Nvidia peermem driver by running "
"\"modprobe nvidia_peermem\"");
}
}

if (peermem_loaded == 0) {
status = UCS_ERR_NO_DEVICE;
goto err;
}

uct_cuda_base_get_sys_dev(device, &dev);
status = ucs_topo_get_distance(dev, md->super.dev.sys_dev, &dist);
if (status != UCS_OK) {
Expand All @@ -697,6 +697,7 @@ uct_gdaki_query_tl_devices(uct_md_h tl_md,

err:
ucs_free(tl_devices);
out:
return status;
}

Expand Down
Loading