Skip to content

[Backport branch/3.6.x] [libcu++] Skip fabric status decoding when a shared_memory_barrier wait has no report - #12228

Open
github-actions[bot] wants to merge 1 commit into
branch/3.6.xfrom
backport-12131-to-branch/3.6.x
Open

github-actions[bot] wants to merge 1 commit into
branch/3.6.xfrom
backport-12131-to-branch/3.6.x

Conversation

@github-actions

Copy link
Copy Markdown
Contributor

Description

Backport of #12131 to branch/3.6.x.

…report (#12131)

operation_status::get_error_count() and for_each_error() called the
device-side fabric status decoder even when the completed wait produced
no report. On a GeForce RTX 5070 Ti (sm_120) with driver 617.42
(CUDA 13.4), cudaFabricOpErrorStatusCount() returns
cudaErrorNotSupported, so __verify_fabric_status() trapped although
there was nothing to decode.

Return 0 without calling the decoder when there is no report.

Co-authored-by: SUKUNA-AI <180781973+SUKUNA-AI@users.noreply.github.com>
Co-authored-by: pciolkosz <pciolkosz@nvidia.com>
(cherry picked from commit 4811a1c)
@copy-pr-bot

copy-pr-bot Bot commented Oct 23, 2026

Copy link
Copy Markdown
Contributor

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@github-project-automation github-project-automation Bot moved this to Todo in CCCL Oct 23, 2026
@pciolkosz

Copy link
Copy Markdown
Contributor

/ok to test 5302fcd

@github-actions

Copy link
Copy Markdown
Contributor Author

🥳 CI Workflow Results

🟩 Finished in 2h 31m: Pass: 100%/228 | Total: 2d 12h | Max: 1h 29m | Hits: 97%/650213

See results here.

@github-project-automation github-project-automation Bot moved this from Todo to In Review in CCCL Oct 26, 2026

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Status: In Review

Development

Successfully merging this pull request may close these issues.

3 participants