Skip to content

Commit e167d94

Browse files
[mlir][Bufferization] Do not have read semantics for destination of tensor.parallel_insert_slice.
Signed-off-by: MaheshRavishankar <[email protected]>
1 parent 4b67c53 commit e167d94

File tree

1 file changed

+1
-2
lines changed

1 file changed

+1
-2
lines changed

mlir/lib/Dialect/Tensor/Transforms/BufferizableOpInterfaceImpl.cpp

Lines changed: 1 addition & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -930,8 +930,7 @@ struct ParallelInsertSliceOpInterface
930930

931931
bool bufferizesToMemoryRead(Operation *op, OpOperand &opOperand,
932932
const AnalysisState &state) const {
933-
return insertSliceOpRequiresRead(cast<tensor::ParallelInsertSliceOp>(op),
934-
opOperand);
933+
return opOperand == cast<ParallelInsertSliceOp>(op).getSourceMutable();
935934
}
936935

937936
bool bufferizesToMemoryWrite(Operation *op, OpOperand &opOperand,

0 commit comments

Comments
 (0)