[inductor] Fix ReinterpretView call in TMADescriptor IR #138759
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/138759
Note: Links to docs will display an error until the docs builds have been completed. ✅ You can merge normally! (1 Unrelated Failure)As of commit 8e797e2 with merge base 72ea7ba ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
You landed your PR on 10/14 before I landed mine on 10/17. And, apparently, my base rev was older than 10/14 when I landed. I guess, it's a good habit to rebase to the newest |
|
eh I should probably have warned the chat that this PR had a high chance of land races. |
|
@pytorchbot merge |
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 |
Merge failedReason: 1 jobs have failed, first few of them are: inductor-periodic / cuda12.1-py3.10-gcc9-sm80 / test (inductor_torchbench_smoketest_perf, 1, 1, linux.gcp.a100) Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot merge |
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 |
|
The merge job was canceled or timed out. This most often happen if two merge requests were issued for the same PR, or if merge job was waiting for more than 6 hours for tests to finish. In later case, please do not hesitate to reissue the merge command |
|
@pytorchbot merge -f "unrelated failing and hanging CI jobs" |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
This fixes some leftover typos in CreateTMADescriptorVariable.call_function (and close). Pull Request resolved: #138877 Approved by: https://github.com/davidberard98, https://github.com/zou3519, https://github.com/Skylion007 ghstack dependencies: #138759
This adds host-side Triton TMA support to AOTInductor. Notes: - Two helper functions, `init1DTMADescriptor` and `init2DTMADescriptor` are added to the C++ wrapper codegen on GPU, conditioned on the model having user-defined Triton kernels with host-side TMA (CUDA-specific). - C++ wrapper codegen on GPU emits TMA descriptor initialization via the aforementioned helper functions. - Special handling added for the TMA descriptors (in the Python wrapper codegen) during the compile-time autotuning, as the underlying tensor can't be passed directly to the user-defined Triton kernel. TMA descriptors are generated in-between the source tensor's buffer and the kernel call, like in the full Python wrapper codegen. - This PR concludes the host-side Triton TMA support in PT2. Pull Request resolved: #138878 Approved by: https://github.com/desertfire, https://github.com/chenyang78 ghstack dependencies: #138759, #138877
This adds host-side Triton TMA support to AOTInductor. Notes: - Two helper functions, `init1DTMADescriptor` and `init2DTMADescriptor` are added to the C++ wrapper codegen on GPU, conditioned on the model having user-defined Triton kernels with host-side TMA (CUDA-specific). - C++ wrapper codegen on GPU emits TMA descriptor initialization via the aforementioned helper functions. - Special handling added for the TMA descriptors (in the Python wrapper codegen) during the compile-time autotuning, as the underlying tensor can't be passed directly to the user-defined Triton kernel. TMA descriptors are generated in-between the source tensor's buffer and the kernel call, like in the full Python wrapper codegen. - This PR concludes the host-side Triton TMA support in PT2. Pull Request resolved: pytorch#138878 Approved by: https://github.com/desertfire, https://github.com/chenyang78 ghstack dependencies: pytorch#138759, pytorch#138877
This adds host-side Triton TMA support to AOTInductor. Notes: - Two helper functions, `init1DTMADescriptor` and `init2DTMADescriptor` are added to the C++ wrapper codegen on GPU, conditioned on the model having user-defined Triton kernels with host-side TMA (CUDA-specific). - C++ wrapper codegen on GPU emits TMA descriptor initialization via the aforementioned helper functions. - Special handling added for the TMA descriptors (in the Python wrapper codegen) during the compile-time autotuning, as the underlying tensor can't be passed directly to the user-defined Triton kernel. TMA descriptors are generated in-between the source tensor's buffer and the kernel call, like in the full Python wrapper codegen. - This PR concludes the host-side Triton TMA support in PT2. Pull Request resolved: pytorch#138878 Approved by: https://github.com/desertfire, https://github.com/chenyang78 ghstack dependencies: pytorch#138759, pytorch#138877
4 participants
Stack from ghstack (oldest at bottom):
As a result of #137768,
ReinterpretViewcall in theTMADescriptorhas become invalid. This leads to some TMA tests breaking in
test_triton_kernels.py. In this PR, we fix this.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang