[clang] [llvm] [AMDGPU][GFX12.5] Reimplement monitor load as an atomic operation (PR #177343)

[Matt Arsenault via cfe-commits]

================
@@ -366,6 +366,41 @@ void CodeGenFunction::AddAMDGPUFenceAddressSpaceMMRA(llvm::Instruction *Inst,
   Inst->setMetadata(LLVMContext::MD_mmra, MMRAMetadata::getMD(Ctx, MMRAs));
 }
 
+static llvm::MetadataAsValue *getSyncscopeIDAsMDString(llvm::LLVMContext &Ctx,
+                                                       clang::SyncScope Scope) {
+  StringRef Name;
+  switch (Scope) {
----------------
arsenm wrote:

I'd prefer to keep this as a simpler SyncScope -> StringRef switch+return function, and leave the MD management for the caller 

https://github.com/llvm/llvm-project/pull/177343