Update disabling fast-path for strict-export inside MultiheadAttention #164544
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
[ghstack-poisoned]
tugsbayasgalan
requested review from
albanD,
jbschlosser and
mikaylagawarecki
as code owners
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/164544
Note: Links to docs will display an error until the docs builds have been completed. ⏳ No Failures, 1 PendingAs of commit 802e9de with merge base 2a7c486 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@tugsbayasgalan has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
anijain2305
reviewed
Oct 3, 2025
torch/nn/modules/activation.py
Outdated
| why_not_fast_path = "some Tensor argument has_torch_function" | ||
| elif _is_make_fx_tracing(): | ||
| why_not_fast_path = "we are running make_fx tracing" | ||
| elif torch.compiler.is_exporting(): |
There was a problem hiding this comment.
Maybe you need another elif branch for export.
For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. Differential Revision: [D83810733](https://our.internmc.facebook.com/intern/diff/D83810733) [ghstack-poisoned]
|
@tugsbayasgalan has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
tugsbayasgalan
changed the title
…eadAttention" For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. We still have to keep the old flag because lot of code assumes this exist.... grr Differential Revision: [D83810733](https://our.internmc.facebook.com/intern/diff/D83810733) [ghstack-poisoned]
|
@tugsbayasgalan has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
anijain2305
reviewed
Oct 3, 2025
…eadAttention" For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. We still have to keep the old flag because lot of code assumes this exist.... grr Differential Revision: [D83810733](https://our.internmc.facebook.com/intern/diff/D83810733) [ghstack-poisoned]
|
@tugsbayasgalan has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
anijain2305
approved these changes
Oct 3, 2025
mikaylagawarecki
approved these changes
Oct 3, 2025
|
@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 |
cyyever
pushed a commit
to cyyever/pytorch
that referenced
this pull request
Oct 4, 2025
pytorch#164544) For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. We still have to keep the old flag because lot of code assumes this exist.... grr Differential Revision: [D83810733](https://our.internmc.facebook.com/intern/diff/D83810733) Pull Request resolved: pytorch#164544 Approved by: https://github.com/anijain2305, https://github.com/mikaylagawarecki
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Oct 21, 2025
pytorch#164544) For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. We still have to keep the old flag because lot of code assumes this exist.... grr Differential Revision: [D83810733](https://our.internmc.facebook.com/intern/diff/D83810733) Pull Request resolved: pytorch#164544 Approved by: https://github.com/anijain2305, https://github.com/mikaylagawarecki
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
4 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Stack from ghstack (oldest at bottom):
For some reason, executorch needs the slow path. But the original flag doesn't work for new export because we inline torch modules even before getting into make_fx. We still have to keep the old flag because lot of code assumes this exist.... grr
Differential Revision: D83810733