[Triton-MLIR][FRONTEND] [BACKEND] fix atomics (#879)
minor fix to backend and frontend of atomics, we can pass 1 test without mask and the shape aligned with CTA size now Co-authored-by: dongdongl <dongdongl@nvidia.com>
This commit is contained in:
@@ -378,7 +378,7 @@ void populateTritonPatterns(TritonGPUTypeConverter &typeConverter,
|
||||
TritonGenericPattern<triton::AddPtrOp>, TritonReducePattern,
|
||||
TritonExpandDimsPattern, TritonMakeRangePattern, TritonDotPattern,
|
||||
TritonLoadPattern, TritonStorePattern, TritonExtElemwisePattern,
|
||||
TritonPrintfPattern>(typeConverter, context);
|
||||
TritonPrintfPattern, TritonAtomicRMWPattern>(typeConverter, context);
|
||||
}
|
||||
|
||||
//
|
||||
|
Reference in New Issue
Block a user