NotificationsYou must be signed in to change notification settings
Fork26.3k
Star96k

[CUDA][cuBLASLt] Fix scale setting for`allowFP16AccumulationCuBLAStrue` case#153083

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 ourterms of service andprivacy statement. We’ll occasionally send you account related emails.

Already on GitHub?Sign in to your account

Jump to bottom

Closed

eqy wants to merge3 commits intopytorch:mainfromeqy:fixscalefp16

Closed

[CUDA][cuBLASLt] Fix scale setting for`allowFP16AccumulationCuBLAStrue` case#153083

eqy wants to merge3 commits intopytorch:mainfromeqy:fixscalefp16

Conversation

Copy link

Collaborator

eqy commentedMay 7, 2025•
edited by pytorch-botbot
Loading

Also add some missing@onlyCUDA / support check decorators intest_matmul_cuda.py
Should helpresolve#151890

cc@ptrblck @msaroufim @jerryzh168 @csarofeen @xwang233

check in

8ce0ed5

eqy requested a review fromsyed-ahmed as acode owner

May 7, 2025 19:06

eqy added module: cuda

Related to torch.cuda, and CUDA support in general

module: cublas

Problem related to cublas support

open source module: half

Related to float16 half-precision floats

topic: not user facingtopic category labels

May 7, 2025

Copy link

pytorch-botbot commentedMay 7, 2025•
edited
Loading

🔗 Helpful Links

🧪 See artifacts and rendered test results athud.pytorch.org/pr/153083

📄 PreviewPython docs built from this PR
📄 PreviewC++ docs built from this PR
❓ Need help or want to give feedback on the CI? Visit thebot commands wiki or ouroffice hours

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

✅ You can merge normally! (1 Unrelated Failure)

As of commit167c4a6 with merge base590965f ():

BROKEN TRUNK - The following job failed but were present on the merge base:

👉Rebase onto the `viable/strict` branch to avoid these failures

pull / linux-focal-py3_9-clang9-xla / build (gh) (trunk failure)
ninja: build stopped: subcommand failed

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

janeyx99 reviewed

May 7, 2025

View reviewed changes

test/test_matmul_cuda.py OutdatedShow resolvedHide resolved

janeyx99 reviewed

May 7, 2025

View reviewed changes

aten/src/ATen/cuda/CUDABlas.cpp

		cudaDeviceProp* prop =at::cuda::getCurrentDeviceProperties();
		if (prop->major >=7 &&at::globalContext().allowFP16AccumulationCuBLAS()) {
		computeType = CUBLAS_COMPUTE_16F;
		scaleType = CUDA_R_16F;

Copy link

Contributor

janeyx99May 7, 2025

Choose a reason for hiding this comment

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

👀

janeyx99 approved these changes

May 7, 2025

View reviewed changes

Copy link

Contributor

janeyx99 left a comment

Choose a reason for hiding this comment

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

looks ok

janeyx99 added release notes: cuda

release notes category

triagedThis issue has been looked at a team member, and triaged and prioritized into an appropriate module and removed topic: not user facingtopic category labels

May 7, 2025

address comments

5f97fcd

Copy link

CollaboratorAuthor

eqy commentedMay 8, 2025

@pytorchmergebot merge

pytorch-botbot added the ciflow/trunkTrigger trunk jobs on your pull request label

May 8, 2025

pytorchmergebot added the merging label

May 8, 2025

Copy link

Collaborator

pytorchmergebot commentedMay 8, 2025

Merge started

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

Learn more about merging in thewiki.

Questions? Feedback? Please reach out to thePyTorch DevX Team

Advanced Debugging

Check the merge workflow status
here

Copy link

Collaborator

pytorchmergebot commentedMay 8, 2025

Merge failed

Reason: Commandgit -C /home/runner/work/pytorch/pytorch rebase origin/main returned non-zero exit code 1

Rebasing (1/1)Auto-merging test/test_matmul_cuda.pyCONFLICT (content): Merge conflict in test/test_matmul_cuda.pyerror: could not apply c5d1ed74a17... [CUDA][cuBLASLt] Fix scale setting for `allowFP16AccumulationCuBLAS` `true` case (#153083)hint: Resolve all conflicts manually, mark them as resolved withhint: "git add/rm <conflicted_files>", then run "git rebase --continue".hint: You can instead skip this commit: run "git rebase --skip".hint: To abort and get back to the state before "git rebase", run "git rebase --abort".hint: Disable this message with "git config set advice.mergeConflict false"Could not apply c5d1ed74a17... [CUDA][cuBLASLt] Fix scale setting for `allowFP16AccumulationCuBLAS` `true` case (#153083)

Details for Dev Infra team

Raised byworkflow job