[AOTI] Fix a two-pass kernel missmatch #141041
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
Summary: Fixes #140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim. [ghstack-poisoned]
desertfire
added a commit
that referenced
this pull request
Nov 19, 2024
Summary: Fixes #140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim. ghstack-source-id: e431769 Pull Request resolved: #141041
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/141041
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 FailuresAs of commit cbe5448 with merge base b379a28 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
shunting314
approved these changes
Nov 19, 2024
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
Summary: Fixes #140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames chauhang aakhundov Differential Revision: [D66203298](https://our.internmc.facebook.com/intern/diff/D66203298) [ghstack-poisoned]
desertfire
added a commit
that referenced
this pull request
Nov 20, 2024
Summary: Fixes #140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim. ghstack-source-id: 4fbeb19 Pull Request resolved: #141041
|
@desertfire has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@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 |
pobin6
pushed a commit
to pobin6/pytorch
that referenced
this pull request
Dec 5, 2024
Summary: Fixes pytorch#140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim. Differential Revision: [D66203298](https://our.internmc.facebook.com/intern/diff/D66203298) Pull Request resolved: pytorch#141041 Approved by: https://github.com/shunting314
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):
Summary: Fixes #140766. In AOTI's two-pass codegen, the first pass generates triton_per_fused_add_native_layer_norm_4, and the second pass generates triton_red_fused_add_native_layer_norm_4. While this problem will go away with the incoming one-pass implementation, further debugging reveals there is a mismatch in has_non_contiguous_pw_in_reduction_kernel between the two passes, due to a symbol comparsion problem in stride1_for_last_dim.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @chauhang @aakhundov
Differential Revision: D66203298