[mlir][bufferization] Allow returning allocs in empty tensor elimination

[matthias-springer]

This flag was renamed in 6a91dfe and accidentally removed from the pass.

[llvmbot]

@llvm/pr-subscribers-mlir-bufferization

Changes

This flag was renamed in 6a91dfe and accidentally removed from the pass. -- Full diff: https://github.com//pull/66198.diff

1 Files Affected:

• (modified) mlir/lib/Dialect/Bufferization/Transforms/EmptyTensorElimination.cpp (+1)

diff --git a/mlir/lib/Dialect/Bufferization/Transforms/EmptyTensorElimination.cpp b/mlir/lib/Dialect/Bufferization/Transforms/EmptyTensorElimination.cpp
index 4dbf22593b911ae..d2e3f512ff30c66 100644
--- a/mlir/lib/Dialect/Bufferization/Transforms/EmptyTensorElimination.cpp
+++ b/mlir/lib/Dialect/Bufferization/Transforms/EmptyTensorElimination.cpp
@@ -268,6 +268,7 @@ struct EmptyTensorElimination
 void EmptyTensorElimination::runOnOperation() {
   Operation *op = getOperation();
   OneShotBufferizationOptions options;
+  options.allowReturnAllocsFromLoops = true;
   OneShotAnalysisState state(op, options);
   if (failed(analyzeOp(op, state))) {
     signalPassFailure();