[Mlir-commits] [mlir] [mlir][nvgpu] Fix crash in optimize-shared-memory pass with vector element types (PR #179111)

[Guray Ozen]

================
@@ -156,6 +156,11 @@ mlir::nvgpu::optimizeSharedMemoryReadsAndWrites(Operation *parentOp,
   if (memRefType.getRank() == 0)
     return failure();
 
+  // Only support memrefs with scalar element types (i.e., int or float).
+  // Memrefs with vector element types are not supported.
+  if (!memRefType.getElementType().isIntOrFloat())
----------------
grypp wrote:

Not blocking but one can actually support vector types.

https://github.com/llvm/llvm-project/pull/179111