[mlir][test][gpu] Migrate CUDA tests to the TargetAttr compilation workflow #65768
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
…rkflow Migrate tests referencing `gpu-to-cubin` to the new compilation workflow using `TargetAttrs`. The `test-lower-to-nvvm` pass pipeline was modified to use the new compilation workflow to simplify the introduction of future tests. The `createLowerGpuOpsToNVVMOpsPass` function was removed, as it didn't allow for passing all options available in the `ConvertGpuOpsToNVVMOp` pass. The LIT configuration was modified to support CUDA tests only when the `ptxas` & `fatbinary` tools are present.
|
Hey @grypp could you help me migrating & validating SM_90 tests? So far I tested this change in SM_70 & SM_80. |
joker-eph
reviewed
Sep 8, 2023
mlir/test/lit.cfg.py
Outdated
| config.available_features.add("host-supports-cuda-runner") | ||
| config.enable_cuda_runner_tests = True | ||
| else: | ||
| config.enable_cuda_runner_tests = False |
There was a problem hiding this comment.
I don't think we should disable these silently. I would rather have this handled in CMake so that the config and the user intent is clear.
For example maybe my bot is misconfigured and I think I am running tests but actually they would just be skipped!
fabianmcg
changed the title
joker-eph
approved these changes
Sep 9, 2023
fabianmcg
added a commit
to fabianmcg/llvm-project
that referenced
this pull request
Sep 9, 2023
…ation workflow (llvm#65768)" This reverts commit d21b672.
fabianmcg
added a commit
that referenced
this pull request
Sep 9, 2023
…tion workflow (#65768) The revert happened due to a build bot failure that threw 'CUDA_ERROR_UNSUPPORTED_PTX_VERSION'. The failure's root cause was a pass using "+ptx76" for compilation and an old CUDA driver on the bot. This commit relands the patch with "+ptx60". Original Gh PR: #65768 Original commit message: Migrate tests referencing `gpu-to-cubin` to the new compilation workflow using `TargetAttrs`. The `test-lower-to-nvvm` pass pipeline was modified to use the new compilation workflow to simplify the introduction of future tests. The `createLowerGpuOpsToNVVMOpsPass` function was removed, as it didn't allow for passing all options available in the `ConvertGpuOpsToNVVMOp` pass.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Migrate tests referencing
gpu-to-cubinto the new compilation workflow usingTargetAttrs. Thetest-lower-to-nvvmpass pipeline was modified to use the new compilation workflow to simplify the introduction of future tests.The
createLowerGpuOpsToNVVMOpsPassfunction was removed, as it didn't allow for passing all options available in theConvertGpuOpsToNVVMOppass.The LIT configuration was modified to support CUDA tests only when the
ptxas&fatbinarytools are present.NOTE: The tests in
test/Integration/GPU/CUDA/sm90were left unmodified as I couldn't verify they work.