[llvm-branch-commits] [clang] [llvm] [mlir] [RFC][AMDGPU] Add EXECSYNC address space (PR #195613)

[Matt Arsenault via llvm-branch-commits]

================
@@ -3262,10 +3263,19 @@ bool AMDGPULegalizerInfo::legalizeGlobalValue(
   MachineFunction &MF = B.getMF();
   SIMachineFunctionInfo *MFI = MF.getInfo<SIMachineFunctionInfo>();
 
+  if (AS == AMDGPUAS::EXECSYNC) {
+    const GlobalVariable *GVar = dyn_cast<GlobalVariable>(GV);
+    assert(GVar && AMDGPU::isNamedBarrier(*GVar) &&
+           "Unsupported use of EXECSYNC address space!");
----------------
arsenm wrote:

```suggestion
    const GlobalVariable *GVar = cast<GlobalVariable>(GV);
    assert(AMDGPU::isNamedBarrier(*GVar) &&
             "Unsupported use of EXECSYNC address space!");
```

https://github.com/llvm/llvm-project/pull/195613