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
Fix mkldnn_matmul error on AArch64 #110150
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/110150
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 69f60f3 with merge base a51b8df (): UNSTABLE - The following job failed but was likely due to flakiness present on trunk and has been marked as unstable:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
The reason is that when the last dimension of c is 1, c will not be marked as transpose, thus m1 and m2 will not be swapped. We should use If last dimension of c is 1, there is no need to use ACL, the performance of BLAS(OpenBLAS) is good. Another reason is that this mkldnn_matmul was added to support bf16 gemm on AArch64, but here it's a matrix multiply a vector, and it's unable to take advantage of the performance gains from bfmmla(bf16 matrix multiplication instructions). |
hi @jgong5, do I need other reviewers to approve to merge the pr? |
Let me see. |
@pytorchbot merge |
Merge failedReason: Approval needed from one of the following: |
Hi, @nkaretnikov, looks like it needs your approval. |
Thanks, but merge failed. @jgong5 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This isn't ideal but seems okay as a hotfix.
@pytorchbot merge |
Merge startedYour 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 |
Where is this merged @imzhuhl |
Fixes #110149