[reland] Fixing aliasing behavior for slice in AQT int4wo layout #2176
+89
−21
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
…out (#2174) Summary: slice op is supposed to preserve aliasing (output of slice should alias the input), but this is not true for TensorCoreTiledLayout (used by int4wo), and some others like gemlite Reason is that we do unpacking, pading and prepacking right now, which creates new tensors. We fixes it in this PR by doing slicing on the packed inner Tensor directly, specifically packed_weight and scale_and_zero in TensorCoreTiledLayout. Test Plan: python test/dtypes/test_affine_quantized.py -k test_slice_and_copy_int4wo Reviewers: Subscribers: Tasks: Tags: * simplify code * add check for data_ptr * format * avoid div by zero * format
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/2176
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ⏳ No Failures, 4 PendingAs of commit 1707c6b with merge base a83636d ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
liangel-02
pushed a commit
that referenced
this pull request
Aug 25, 2025
* [reland] Fixing aliasing behavior for slice in AQT TensorCoreTiledLayout (#2174) Summary: slice op is supposed to preserve aliasing (output of slice should alias the input), but this is not true for TensorCoreTiledLayout (used by int4wo), and some others like gemlite Reason is that we do unpacking, pading and prepacking right now, which creates new tensors. We fixes it in this PR by doing slicing on the packed inner Tensor directly, specifically packed_weight and scale_and_zero in TensorCoreTiledLayout. Test Plan: python test/dtypes/test_affine_quantized.py -k test_slice_and_copy_int4wo Reviewers: Subscribers: Tasks: Tags: * simplify code * add check for data_ptr * format * avoid div by zero * format * fix shape
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Summary:
slice op is supposed to preserve aliasing (output of slice should alias the input), but this is not true for TensorCoreTiledLayout (used by int4wo), and some others like gemlite
Reason is that we do unpacking, pading and prepacking right now, which creates new tensors.
We fixes it in this PR by doing slicing on the packed inner Tensor directly, specifically packed_weight and scale_and_zero in TensorCoreTiledLayout.
Test Plan:
python test/dtypes/test_affine_quantized.py -k test_slice_and_copy_int4wo
Reviewers:
Subscribers:
Tasks:
Tags:
simplify code
add check for data_ptr
format
avoid div by zero
format