diff options
| author | Yang Bai <baiyang0132@gmail.com> | 2025-08-22 10:46:06 +0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2025-08-21 21:46:06 -0500 |
| commit | f1f194bf10e6ce180bbb199fa219c4d1ec67290f (patch) | |
| tree | bc814c9cef5cc974a989371d8f3b43b7bc56278e /lldb/packages/Python/lldbsuite/test/lldbpexpect.py | |
| parent | 418fb5030149b5fe46d4541c4037883a04c7f3e3 (diff) | |
| download | llvm-f1f194bf10e6ce180bbb199fa219c4d1ec67290f.zip llvm-f1f194bf10e6ce180bbb199fa219c4d1ec67290f.tar.gz llvm-f1f194bf10e6ce180bbb199fa219c4d1ec67290f.tar.bz2 | |
[mlir][vector] fix: unroll vector.from_elements in gpu pipelines (#154774)
### Problem
PR #142944 introduced a new canonicalization pattern which caused
failures in the following GPU-related integration tests:
-
mlir/test/Integration/GPU/CUDA/TensorCore/sm80/transform-mma-sync-matmul-f16-f16-accum.mlir
-
mlir/test/Integration/GPU/CUDA/TensorCore/sm80/transform-mma-sync-matmul-f32.mlir
The issue occurs because the new canonicalization pattern can generate
multi-dimensional `vector.from_elements` operations (rank > 1), but the
GPU lowering pipelines were not equipped to handle these during the
conversion to LLVM.
### Fix
This PR adds `vector::populateVectorFromElementsLoweringPatterns` to the
GPU lowering passes that are integrated in `gpu-lower-to-nvvm-pipeline`:
- `GpuToLLVMConversionPass`: the general GPU-to-LLVM conversion pass.
- `LowerGpuOpsToNVVMOpsPass`: the NVVM-specific lowering pass.
Co-authored-by: Yang Bai <yangb@nvidia.com>
Diffstat (limited to 'lldb/packages/Python/lldbsuite/test/lldbpexpect.py')
0 files changed, 0 insertions, 0 deletions