[Mlir-commits] [mlir] [MLIR][GPUToLLVMSPV] Relax the width check in gpu.shuffle lowering (PR #183445)

[Adam Siemieniuk]

================
@@ -345,7 +345,7 @@ struct GPUShuffleConversion final : ConvertOpToLLVMPattern<gpu::ShuffleOp> {
   LogicalResult
   matchAndRewrite(gpu::ShuffleOp op, OpAdaptor adaptor,
                   ConversionPatternRewriter &rewriter) const final {
-    if (!hasValidWidth(op))
+    if (getSubgroupSize(op) && !hasValidWidth(op))
----------------
adam-smnk wrote:

nit: I'd move the extra subgroup size check inside `hasValidWidth` for better readability and to make this logical coupling stay inside the validator

https://github.com/llvm/llvm-project/pull/183445