[Mlir-commits] [mlir] [mlir][vector-to-gpu]: Extend MMA Lowerings (PR #176785)

[Hsiangkai Wang]

================
@@ -130,7 +147,13 @@ static std::optional<int64_t> getStaticallyKnownRowStride(ShapedType type) {
   if (failed(memrefType.getStridesAndOffset(strides, offset)) ||
       strides.back() != 1)
     return std::nullopt;
-  int64_t stride = strides[strides.size() - 2];
+
+  int stridePostion = strides.size() - 2;
+  if (!permutationMap.isPermutation()) {
+    if (auto outerResult = dyn_cast<AffineDimExpr>(permutationMap.getResult(0)))
----------------
Hsiangkai wrote:

Why do you check 0-th mapping here? Is it possible the 0-th mapping is not AffineDimExpr?

https://github.com/llvm/llvm-project/pull/176785