Fix to keep stride in return_and_correct_aliasing() #117860
Closed
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]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/117860
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (2 Unrelated Failures)As of commit 2ff2303 with merge base 26fbbc3 ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
BROKEN TRUNK - The following job failed but was present on the merge base:👉 Rebase onto the `viable/strict` branch to avoid these failures
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 TODO: investigate why the fix trips the assert mentioned in the issue [ghstack-poisoned]
Fixes #117794 Fix tripped the assert here: https://github.com/pytorch/pytorch/blob/86dedebeafdd7b08d21432cebd7538437d3b7509/torch/utils/_python_dispatch.py#L216 From investigation: I found that functionalization of an in-place op (`mul_` in this test case) results in the strides of `TwoTensor`'s `a` / `b` components being mutated to be contiguous. This is not reflected in the outer tensor, causing the assert to be tripped. To address this, I set the `dispatch_sizes_strides_policy` of `TwoTensor` so that sizes / strides / storage offset are always queried from the underlying components. Now the stride mutation of the inner tensors due to functionalization is properly reflected in the outer tensor. This seems like a broader issue affecting subclasses whose sizes / strides are dependent on those of the inner tensors; for these to work properly with functionalization / PT2 as a whole, they probably want to follow this `dispatch_sizes_strides_policy="sizes"` pattern. [ghstack-poisoned]
Fixes #117794 Fix tripped the assert here: https://github.com/pytorch/pytorch/blob/86dedebeafdd7b08d21432cebd7538437d3b7509/torch/utils/_python_dispatch.py#L216 From investigation: I found that functionalization of an in-place op (`mul_` in this test case) results in the strides of `TwoTensor`'s `a` / `b` components being mutated to be contiguous. This is not reflected in the outer tensor, causing the assert to be tripped. After discussion with Brian, I address this by disallowing input mutations on non-contiguous tensor subclass inputs for now. [ghstack-poisoned]
This was referenced Feb 20, 2024
bdhirsh
reviewed
Feb 20, 2024
Fixes #117794 Fix tripped the assert here: https://github.com/pytorch/pytorch/blob/86dedebeafdd7b08d21432cebd7538437d3b7509/torch/utils/_python_dispatch.py#L216 From investigation: I found that functionalization of an in-place op (`mul_` in this test case) results in the strides of `TwoTensor`'s `a` / `b` components being mutated to be contiguous. This is not reflected in the outer tensor, causing the assert to be tripped. After discussion with Brian, I address this in this PR by disallowing input mutations on non-contiguous tensor subclass inputs for now. [ghstack-poisoned]
|
@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 |
Merge failedReason: 1 mandatory check(s) failed. The first few are: Dig deeper by viewing the failures on hud |
|
@pytorchbot merge -f "ignore pre-existing failure" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
|
@davidberard98 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
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):
Fixes #117794
Fix tripped the assert here:
pytorch/torch/utils/_python_dispatch.py
Line 216 in 86dedeb
From investigation: I found that functionalization of an in-place op (
mul_in this test case) results in the strides ofTwoTensor'sa/bcomponents being mutated to be contiguous. This is not reflected in the outer tensor, causing the assert to be tripped.After discussion with Brian, I address this in this PR by disallowing input mutations on non-contiguous tensor subclass inputs for now.
Differential Revision: D54214617