[llvm] [NVPTX] Add Volta Atomic SequentiallyConsistent Load and Store Operations (PR #98551)

[Artem Belevich via llvm-commits]

================
@@ -714,21 +714,28 @@ static unsigned int getCodeAddrSpace(MemSDNode *N) {
   return NVPTX::PTXLdStInstCode::GENERIC;
 }
 
-static unsigned int getCodeMemorySemantic(MemSDNode *N,
-                                          const NVPTXSubtarget *Subtarget) {
+struct OperationOrderings {
----------------
Artem-B wrote:

I'd put both the class and the static function below into an anonymous namespace. Or add `NVPTX` prefix to the names. The current names are too generic.


https://github.com/llvm/llvm-project/pull/98551