[clang] [compiler-rt] [libc] [libcxx] [llvm] [AMDGPU] Fix code object version not being set to 'none' (PR #135036)

[Matt Arsenault via cfe-commits]

================
@@ -62,7 +62,7 @@ Value *EmitAMDGPUWorkGroupSize(CodeGenFunction &CGF, unsigned Index) {
 
   auto Cov = CGF.getTarget().getTargetOpts().CodeObjectVersion;
 
-  if (Cov == CodeObjectVersionKind::COV_None) {
+  if (Cov == CodeObjectVersionKind::COV_None && !CGF.getLangOpts().OpenMP) {
----------------
arsenm wrote:

Shouldn't be language dependent? Does this mean the IR module flag?

https://github.com/llvm/llvm-project/pull/135036