-
Notifications
You must be signed in to change notification settings - Fork 26.3k
Revert to old behaviour of not padding strides if shape or stride is dynamic #163639
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
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/163639
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 6532526 with merge base bcb893a ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
|
@nandesuka has exported this pull request. If you are a Meta employee, you can view the originating diff in D83053287. |
|
@pytorchbot label "topic: not user facing" |
blaine-rister
left a comment
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.
LGTM. Left a nit about cleaning up some related code.
| isinstance(s, (int, sympy.Integer)) | ||
| for s in itertools.chain(in_strides, size) | ||
| ) | ||
| if not config.pad_dynamic_shapes and is_dynamic: |
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.
nit: is it safe to remove the or config.pad_dynamic_shapes on line 3784? Now that we're returning early, that might be redundant.
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.
let's just leave it for now, can clean up later once we have a root cause
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.
are we padding with out guards if we remove this?
shall we instead make sure we guard properly .
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.
Can you clarify, guard what values with what constraints?
blaine-rister
left a comment
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.
LGTM. Left a nit about possibly removing a redundant check further down in the function.
…dynamic (pytorch#163639) Summary: forward fix for S563814 Test Plan: foward fix for S563814 Reviewed By: blaine-rister Differential Revision: D83053287
845f5f1 to
6532526
Compare
|
@nandesuka has exported this pull request. If you are a Meta employee, you can view the originating diff in D83053287. |
|
@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 |
…dynamic (#163639) Differential Revision: D83053287 Pull Request resolved: #163639 Approved by: https://github.com/blaine-rister
Differential Revision: D83053287
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben