8000 [Inductor] Skip triton templates for mixedmm on SM70- by malfet · Pull Request #118591 · pytorch/pytorch · GitHub
[go: up one dir, main page]
Skip to content

[Inductor] Skip triton templates for mixedmm on SM70- #118591

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

Jump to bottom
Closed
wants to merge 2 commits into from
Closed

[Inductor] Skip triton templates for mixedmm on SM70- #118591

wants to merge 2 commits into from

Conversation

malfet
Copy link
Contributor
@malfet malfet commented Jan 29, 2024 •
edited by pytorch-bot bot
Loading

@malfet
[Inductor] Skip triton templates for mixedmm on V100
97a5619 
As it results in numerical errors, see #117144

Fixes #117144
@malfet malfet requested review from jansel and HDCharles January 29, 2024 23:55
@pytorch-bot PyTorch Bot
Copy link
pytorch-bot bot commented Jan 29, 2024 •
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/118591

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit a2d0f6a with merge base a40be5f (image):
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@malfet
Copy 8000 link
Contributor Author
malfet commented Jan 29, 2024

I wonder if I should move is_v100_gpu to utils (and perhaps rename it to is_volta_gpu?

jansel
jansel requested changes Jan 30, 2024
View reviewed changes
@jansel
Copy link
Contributor
jansel commented Jan 30, 2024

Seems safer to also disable for older GPUs than Volta.

@malfet malfet changed the title [Inductor] Skip triton templates for mixedmm on V100 [Inductor] Skip triton templates for mixedmm on SM70- Jan 30, 2024
@malfet malfet requested a review from jansel January 30, 2024 19:12
@malfet
Copy link
Contributor Author
malfet commented Jan 30, 2024

@jansel renamed to _is_sm7x_or_older_gpu (I wonder if I should just return False for ROCM GPUs)

@github-actions github-actions bot added the oncall: distributed Add this issue/PR to distributed oncall triage queue label Jan 31, 2024
@malfet malfet force-pushed the malfet/fix-mixed-mm-on-v100 branch from 0034c7f to a2d0f6a Compare January 31, 2024 02:55
@jansel
Copy link
Contributor
jansel commented Jan 31, 2024

Not sure what to do about rocm. Maybe someone from AMD can help figure out minimum requirements there.

@malfet
Copy link
Contributor Author
malfet commented Jan 31, 2024

@pytorchbot merge

@pytorch-bot pytorch-bot bot added the ciflow/trunk Trigger trunk jobs on your pull request label Jan 31, 2024
@pytorchmergebot
Copy link
Collaborator

Merge failed

Reason: This PR needs a release notes: label
If your changes are user facing and intended to be a part of release notes, please use a label starting with release notes:.

If not, please add the topic: not user facing label.

To add a label, you can comment to pytorchbot, for example
@pytorchbot label "topic: not user facing"

For more information, see
https://github.com/pytorch/pytorch/wiki/PyTorch-AutoLabel-Bot#why-categorize-for-release-notes-and-how-does-it-work.

Details for Dev Infra team Raised by workflow job

@malfet
Copy link
Contributor Author
malfet commented Jan 31, 2024

@pytorchbot merge

@pytorchmergebot
Copy link
Collaborator

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Advanced Debugging
Check the merge workflow status
here

pytorch-bot bot pushed a commit that referenced this pull request Feb 8, 2024
@malfet @jeffdaily
[Inductor] Skip triton templates for mixedmm on SM70- (#118591)
885bb45 
As it results in numerical errors, see #117144

Fixes #117144

Pull Request resolved: #118591
Approved by: https://github.com/jansel
atalman pushed a commit to atalman/pytorch that referenced this pull request Feb 14, 2024
@malfet @atalman
[Inductor] Skip triton templates for mixedmm on SM70- (pytorch#118591)
2c2f5e0 
As it results in numerical errors, see pytorch#117144

Fixes pytorch#117144

Pull Request resolved: pytorch#118591
Approved by: https://github.com/jansel
This was referenced Feb 14, 2024
Closed
huydhn pushed a commit that referenced this pull request Feb 14, 2024
@atalman @malfet
[Inductor] Skip triton templates for mixedmm on SM70- (#118591) (#119894
eef51a6 
)

As it results in numerical errors, see #117144

Fixes #117144

Pull Request resolved: #118591
Approved by: https://github.com/jansel

Co-authored-by: Nikita Shulga <nshulga@meta.com>
@github-actions github-actions bot deleted the malfet/fix-mixed-mm-on-v100 branch March 2, 2024 01:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@jansel jansel jansel approved these changes

@HDCharles HDCharles Awaiting requested review from HDCharles

Assignees
No one assigned
Labels
ciflow/inductor ciflow/trunk Trigger trunk jobs on your pull request Merged module: inductor oncall: distributed Add this issue/PR to distributed oncall triage queue release notes: inductor topic: bug fixes topic category
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

test_mixed_mm2_cuda fails on V100
3 participants
@malfet @jansel @pytorchmergebot
0