[Mlir-commits] [mlir] [MLIR][NVVM] Add prefetch Ops (PR #141737)

[Srinivasa Ravi]

================
@@ -2333,6 +2353,60 @@ def NVVM_CpAsyncBulkTensorSharedCTAToGlobalOp :
   let hasVerifier = 1;
 }
 
+//===----------------------------------------------------------------------===//
+// NVVM Prefetch Ops
----------------
Wolfram70 wrote:

There is also the `prefetch.tensormap` Op below this so does it make sense to have Ops?

https://github.com/llvm/llvm-project/pull/141737