Fix CUDA 13.2 TMA typedef compatibility - #8
Merged
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Fix AdaptiveGEMM build compatibility with CUDA 13.2 by adding a fallback alias for
PFN_cuTensorMapEncodeTiled.CUDA 13.2 exposes the TMA driver entry point typedef as
PFN_cuTensorMapEncodeTiled_v12000, while AdaptiveGEMM currently references the unversionedPFN_cuTensorMapEncodeTiledname. This causes JIT compilation failures in the CUDA 13.2 environment.Change
Add a guarded alias:
PFN_cuTensorMapEncodeTiled -> PFN_cuTensorMapEncodeTiled_v12000Keep the alias behind
#ifndef PFN_cuTensorMapEncodeTiledso existing CUDA versions that still expose the unversioned typedef are unchanged.Validation
Validated in a torch 2.12.1 / CUDA 13.2 environment with XTuner float8 tests after rebuilding AdaptiveGEMM from this patch:
tests/engine/test_moe_train_engine_float8.pyep1 float8 cases: passedThis fixes the CUDA 13.2 typedef compatibility issue without changing runtime kernel logic.