Fix stride handling in FFT meta registrations#175731
Fix stride handling in FFT meta registrations#175731XAheli wants to merge 3 commits intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/175731
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit feb5132 with merge base a790959 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
This PR needs a
|
|
@pytorchbot label "module: inductor" |
|
@pytorchbot label "module: fft" |
|
@pytorchbot merge |
|
Pull workflow has not been scheduled for the PR yet. It could be because author doesn't have permissions to run those or skip-checks keywords were added to PR/commits, aborting merge. Please get/give approval for the workflows and/or remove skip ci decorators before next merge attempt. If you think this is a mistake, please contact PyTorch Dev Infra. |
|
@pytorchbot merge |
|
Pull workflow has not been scheduled for the PR yet. It could be because author doesn't have permissions to run those or skip-checks keywords were added to PR/commits, aborting merge. Please get/give approval for the workflows and/or remove skip ci decorators before next merge attempt. If you think this is a mistake, please contact PyTorch Dev Infra. |
|
@pytorchbot merge |
|
Pull workflow has not been scheduled for the PR yet. It could be because author doesn't have permissions to run those or skip-checks keywords were added to PR/commits, aborting merge. Please get/give approval for the workflows and/or remove skip ci decorators before next merge attempt. If you think this is a mistake, please contact PyTorch Dev Infra. |
b41647d to
feb5132
Compare
|
@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 |
Fixes pytorch#167636 #### Summary - Fix `meta_fft_r2c` missing CPU/MKL code path - Fix `_sort_dims` no-op also add `reverse=True` to match C++ descending sort - Fix `_exec_fft` signal dim ordering - Add PocketFFT early return paths for `meta_fft_c2c` and `meta_fft_c2r` on CPU without MKL - Move FFT ops from `CHECK_STRIDES_SKIPS` to `CHECK_STRIDES` in `test_meta.py` - Use `torch.backends.mkl.is_available()` per [this](pytorch#167636 (comment)) Pull Request resolved: pytorch#175731 Approved by: https://github.com/isuruf
Fixes #167636
Summary
meta_fft_r2cmissing CPU/MKL code path_sort_dimsno-op also addreverse=Trueto match C++ descending sort_exec_fftsignal dim orderingmeta_fft_c2candmeta_fft_c2ron CPU without MKLCHECK_STRIDES_SKIPStoCHECK_STRIDESintest_meta.pytorch.backends.mkl.is_available()per thiscc @mruberry @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @jataylo @mlazos