[Mlir-commits] [mlir] [mlir][vector] Improve shape_cast lowering (PR #140800)
Andrzej Warzyński
llvmlistbot at llvm.org
Tue May 27 09:50:51 PDT 2025
================
@@ -157,41 +160,54 @@ class ShapeCastOpRewritePattern : public OpRewritePattern<vector::ShapeCastOp> {
LogicalResult matchAndRewrite(vector::ShapeCastOp op,
PatternRewriter &rewriter) const override {
Location loc = op.getLoc();
- auto sourceVectorType = op.getSourceVectorType();
- auto resultVectorType = op.getResultVectorType();
+ VectorType sourceType = op.getSourceVectorType();
+ VectorType resultType = op.getResultVectorType();
- if (sourceVectorType.isScalable() || resultVectorType.isScalable())
+ if (sourceType.isScalable() || resultType.isScalable())
return failure();
- // Special case for n-D / 1-D lowerings with better implementations.
- int64_t srcRank = sourceVectorType.getRank();
- int64_t resRank = resultVectorType.getRank();
- if ((srcRank > 1 && resRank == 1) || (srcRank == 1 && resRank > 1))
+ // Special case for n-D / 1-D lowerings with implementations that use
+ // extract_strided_slice / insert_strided_slice.
----------------
banach-space wrote:
Could we clarify this comment? (the original part is quite confusing). Right now, combined with the code, it reads a bit like:
> This is a special case, lets fail!
😅 I assume that it was meant to be:
> This special case is handled by other, more optimal patterns.
Or something similar :)
https://github.com/llvm/llvm-project/pull/140800
More information about the Mlir-commits
mailing list