Skip to content
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.

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

UCT/GTEST: Fix cuda_copy md allocations #10004

Merged
merged 1 commit into from
Jul 21, 2024

Conversation

brminich
Copy link
Contributor

What

  • Do not log errors reproted by cuda fabric allocations if FABRIC_ENABLE config is not set to yes
  • Add a test checking that sending cuda memory thru cuda staging buffers is possible

Why ?

Fixes the following errors:

[ RUN      ] rc/test_ucp_am_nbx_rndv_ppln.cuda_buff_cuda_frag/0 <rc_v,cuda_copy,rocm_copy>
[1720611489.711215] [rock06:27460:0]    cuda_copy_md.c:241  UCX  ERROR   cuMemCreate(&alloc_handle->generic_handle, alloc_handle->length, &prop, 0) failed: operation not permitted
[1720611489.726205] [rock06:27460:0]    cuda_copy_md.c:241  UCX  ERROR   cuMemCreate(&alloc_handle->generic_handle, alloc_handle->length, &prop, 0) failed: operation not permitted

yosefe
yosefe previously approved these changes Jul 10, 2024
- Do not log errors reproted by cuda fabric allocations if
  FABRIC_ENABLE config is not set to yes
- Add a test checking that sending cuda memory thru cuda staging
  buffers is possible
@brminich
Copy link
Contributor Author

squashed, as it as previously approved and is relativelly small

@yosefe
Copy link
Contributor

yosefe commented Jul 15, 2024

/azp run UCX PR

Copy link

Azure Pipelines successfully started running 1 pipeline(s).

@yosefe yosefe merged commit 4655c04 into openucx:master Jul 21, 2024
142 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

2 participants