[AOTI][reland] Update zero size computation in clone_preserve_strides #139458
Conversation
Summary: Reland #139224. clone_preserve_strides implemented in _inductor/utils.py does not handle multi-dimensional 0-size tensor correctly. [ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/139458
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit dcd3eb4 with merge base c8a648d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@pytorchbot merge (Initiating merge automatically since Phabricator Diff has merged) |
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 |
|
@huydhn , I am seeing a test failure gain, https://github.com/pytorch/pytorch/actions/runs/11630431539/job/32390170943, but I don't understand why there was no test failure at the merge time. Also I can't reproduce the problem locally. I will push a test skip as forward fix, since AOTI's arrayref mode has a low priority now. |
|
Ugh, I know why now. It's a miss in target determination. In your PR, I see it wrongly put Atm, we have a the mechanism to add cc @clee2000 |
Summary: A new test added in #139458 only fails in certain CI instance. Skip for now as the failing test has a low priority. @diff-train-skip-merge (to silent fb bot so that I can land this myself) Pull Request resolved: #139485 Approved by: https://github.com/huydhn, https://github.com/hl475
…pytorch#139458) Summary: Reland pytorch#139224. clone_preserve_strides implemented in _inductor/utils.py does not handle multi-dimensional 0-size tensor correctly. Differential Revision: [D65317451](https://our.internmc.facebook.com/intern/diff/D65317451) Pull Request resolved: pytorch#139458 Approved by: https://github.com/hl475
Summary: A new test added in pytorch#139458 only fails in certain CI instance. Skip for now as the failing test has a low priority. @diff-train-skip-merge (to silent fb bot so that I can land this myself) Pull Request resolved: pytorch#139485 Approved by: https://github.com/huydhn, https://github.com/hl475
5 participants
Stack from ghstack (oldest at bottom):
Summary: Reland #139224. clone_preserve_strides implemented in _inductor/utils.py does not handle multi-dimensional 0-size tensor correctly.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @chauhang @aakhundov
Differential Revision: D65317451