[Mlir-commits] [llvm] [mlir] [NVVM][NVPTX] Change TMA Tensor reduction ops to use flag for reduction ops (PR #213638)

[Durgadoss R]

================
@@ -16,6 +16,15 @@
 using namespace llvm;
 using namespace nvvm;
 
+void nvvm::printTMAReductionOp(raw_ostream &OS, const Constant *ImmArgVal) {
+  const auto *CI = dyn_cast<ConstantInt>(ImmArgVal);
+  if (!CI || CI->getZExtValue() > static_cast<uint64_t>(TMAReductionOp::XOR))
+    llvm_unreachable(
+        "printTMAReductionOp called with invalid value for immediate argument");
----------------
durga4github wrote:

Since we attach the `Range` attribute, this should be caught by the IR verifier itself? (and hence may not be needed here)

https://github.com/llvm/llvm-project/pull/213638