[TIRx][Test] Gate tcgen05.mma collector certification on nvcc >= 13.4 - #20270
Merged
spectrometerHBH merged 1 commit intoSep 4, 2026
Conversation
The collector-qualified block_scale MMA forms are PTX ISA 9.4 and certify at sm_107f, which needs the CUDA 13.4 ptxas. Move that half of test_ptx_tcgen05_mma_block_size_form into its own test guarded by env.has_nvcc_version(13, 4), the same gate the whole-table certification tests already use, so older toolchains skip it instead of failing while the sm_100a block32 certification and the block16 legality check still run.
tqchen
approved these changes
Sep 4, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
test_ptx_tcgen05_mma_block_size_formassembles the collector-qualifiedtcgen05.mma ... block_scaleforms atsm_107f. Those are PTX ISA 9.4 forms (#20266) and need the CUDA 13.4 ptxas; on a CUDA 13.2 host the test fails withUnknown modifier '.collector::b::fill'instead of skipping.This moves that half into
test_ptx_tcgen05_mma_block_size_collector_form, guarded byenv.has_nvcc_version(13, 4), the same gate the whole-table certification tests already use. Thesm_100ablock32 certification and themxf4block16 legality check stay in the original test and keep running on older toolchains.Testing
On a CUDA 13.2 host (B200):
The full
tests/python/tirx/suite on the same host previously had this as its only failure (2749 passed, 321 skipped).