Skip to content

Conversation

@Sibylau
Copy link
Contributor

@Sibylau Sibylau commented Sep 24, 2025

Stacked PRs:


skip gather_gemv code check for B200 and fb_code

stack-info: PR: #666, branch: Sibylau/stack/4
@meta-cla meta-cla bot added the CLA Signed This label is managed by the Meta Open Source bot. label Sep 24, 2025
@Sibylau Sibylau requested a review from yf225 September 24, 2025 00:39
if is_cuda():
gpu_name = torch.cuda.get_device_name(0)

if is_cuda() and not is_fbcode() and "b200" not in gpu_name.lower():
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Wonder does https://github.com/pytorch/helion/actions/runs/17962225347/job/51087773640 repro on local H100 machine as well? I haven't quite figured out why it fails also on H100... I think worse case we could skip the assertExpectedJournal for gather_gemv for now

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The test_gather_gemv passes on my local H100 and B200. I am not sure why the generated code varies in the CI runner... Let me skip the assertExpectedJournal for gather_gemv.

@yf225 yf225 merged commit c3a59f2 into main Sep 24, 2025
13 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

CLA Signed This label is managed by the Meta Open Source bot.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants