enable torch.compile for mxfp8_cublas recipe#1841
Merged
Merged
Conversation
Contributor
Author
|
Stack from ghstack (oldest at bottom): |
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1841
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
vkuzo
added a commit
that referenced
this pull request
Mar 5, 2025
Summary: This PR enables `MXLinear` with `mxfp8_cublas` recipe to use torch.compile. The current approach is a short term workaround until pytorch/pytorch#148461 is done. Since we can't use e8m0 in torchinductor or triton yet, we create a custom op wrapper around `torch._scaled_mm` which takes `uint8` scales and does the cast to e8m0 inside the wrapper, where torchinductor can't see it. Test Plan: ``` // this now works (although performance is not ideal due to #1788) python benchmarks/float8/profile_lowp_training.py ~/local/tmp/20250305_test --mx_recipe_name mxfp8_cublas // we can also uncomment the hardware check and run the unit test pytest test/prototype/mx_formats -s -k test_linear_compile ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 033d817 ghstack-comment-id: 2701679811 Pull Request resolved: #1841
vkuzo
added a commit
that referenced
this pull request
Mar 5, 2025
Summary: This PR enables `MXLinear` with `mxfp8_cublas` recipe to use torch.compile. The current approach is a short term workaround until pytorch/pytorch#147873 is done. Since we can't use e8m0 in torchinductor or triton yet, we create a custom op wrapper around `torch._scaled_mm` which takes `uint8` scales and does the cast to e8m0 inside the wrapper, where torchinductor can't see it. Test Plan: ``` // this now works (although performance is not ideal due to #1788) python benchmarks/float8/profile_lowp_training.py ~/local/tmp/20250305_test --mx_recipe_name mxfp8_cublas // we can also uncomment the hardware check and run the unit test pytest test/prototype/mx_formats -s -k test_linear_compile ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: f3ebd12 ghstack-comment-id: 2701679811 Pull Request resolved: #1841
vkuzo
added a commit
that referenced
this pull request
Mar 5, 2025
Summary: This PR enables `MXLinear` with `mxfp8_cublas` recipe to use torch.compile. The current approach is a short term workaround until pytorch/pytorch#147873 is done. Since we can't use e8m0 in torchinductor or triton yet, we create a custom op wrapper around `torch._scaled_mm` which takes `uint8` scales and does the cast to e8m0 inside the wrapper, where torchinductor can't see it. Test Plan: ``` // this now works (although performance is not ideal due to #1788) python benchmarks/float8/profile_lowp_training.py ~/local/tmp/20250305_test --mx_recipe_name mxfp8_cublas // we can also uncomment the hardware check and run the unit test pytest test/prototype/mx_formats -s -k test_linear_compile ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: e5687e3 ghstack-comment-id: 2701679811 Pull Request resolved: #1841
danielvegamyhre
approved these changes
Mar 6, 2025
Comment on lines
+298
to
+304
| is_sm_at_least_100(), | ||
| reason="triton does not work yet on CUDA capability 10.0", | ||
| ) | ||
| @pytest.mark.skipif( | ||
| not is_sm_at_least_100(), | ||
| reason="MX gemms require CUDA capability 10.0", | ||
| ) |
Contributor
There was a problem hiding this comment.
Combining skip if is_sm_at_least_100() with skip if not is_sm_at_least_100() will prevent the test from ever running, so I just want to confirm, is this test intentionally being skipped until the new release of pytorch (with triton that supports compute capability 10.0) is part of CI?
Contributor
Author
There was a problem hiding this comment.
yes, that's corrrect. It's skipped in CI because we don't have B200s in CI, and it's skipped locally because it requires building triton from source. I uncomment these tests if I need to run them, for now.
liangel-02
pushed a commit
that referenced
this pull request
Aug 25, 2025
* Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned] * Update [ghstack-poisoned]
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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:
This PR enables
MXLinearwithmxfp8_cublasrecipe to usetorch.compile.
The current approach is a short term workaround until
pytorch/pytorch#148461 is done. Since we can't
use e8m0 in torchinductor or triton yet, we create a custom op wrapper
around
torch._scaled_mmwhich takesuint8scales and does the cast toe8m0 inside the wrapper, where torchinductor can't see it.
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags: