-
Notifications
You must be signed in to change notification settings - Fork 26.3k
[MPS] Fix copy_cast_mps() on tensors with storage offset #95093
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
Conversation
This should fix the failure with GPT2 when use_cache=True
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/95093
Note: Links to docs will display an error until the docs builds have been completed. ❌ 4 FailuresAs of commit 5167167: NEW FAILURES - The following jobs have failed:
BROKEN TRUNK - The following jobs failed but were present on the merge base d5d5536:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@pytorchbot merge -f "MPS tests are green." |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
- The copy_cast path requires storage_offset to be applied before casting - This should fix some correctness issues in transformer models Fixes pytorch#94980 Pull Request resolved: pytorch#95093 Approved by: https://github.com/kulinseth
- The copy_cast path requires storage_offset to be applied before casting - This should fix some correctness issues in transformer models Fixes pytorch#94980 Pull Request resolved: pytorch#95093 Approved by: https://github.com/kulinseth
* [MPS] Handle broadcasting by expanding src tensor in Copy.mm (#95272) Fixes #ISSUE_NUMBER Pull Request resolved: #95272 Approved by: https://github.com/DenisVieriu97 * [MPS] Fix copy_cast_mps() on tensors with storage offset (#95093) - The copy_cast path requires storage_offset to be applied before casting - This should fix some correctness issues in transformer models Fixes #94980 Pull Request resolved: #95093 Approved by: https://github.com/kulinseth --------- Co-authored-by: Ramin Azarmehr <razarmehr@apple.com>
…orch#95093)" This reverts commit 9511b9f.
* [MPS] Handle broadcasting by expanding src tensor in Copy.mm (pytorch#95272) Fixes #ISSUE_NUMBER Pull Request resolved: pytorch#95272 Approved by: https://github.com/DenisVieriu97 * [MPS] Fix copy_cast_mps() on tensors with storage offset (pytorch#95093) - The copy_cast path requires storage_offset to be applied before casting - This should fix some correctness issues in transformer models Fixes pytorch#94980 Pull Request resolved: pytorch#95093 Approved by: https://github.com/kulinseth --------- Co-authored-by: Ramin Azarmehr <razarmehr@apple.com>
Fixes #94980