================
@@ -746,9 +758,14 @@ void CGOpenMPRuntimeGPU::emitKernelInit(const
OMPExecutableDirective &D,
CodeGenFunction &CGF,
EntryFunctionState &EST, bool IsSPMD) {
llvm::OpenMPIRBuilder::TargetKernelDefaultAttrs Attrs;
- Attrs.ExecFlags =
- IsSPMD ? llvm::omp::OMPTgtExecModeFlags::OMP_TGT_EXEC_MODE_SPMD
- : llvm::omp::OMPTgtExecModeFlags::OMP_TGT_EXEC_MODE_GENERIC;
+ if (IsSPMD && canPromoteToNoLoop(CGM.getLangOpts(), D))
+ Attrs.ExecFlags =
----------------
ro-i wrote:
You set the exec flag, which optimizes the grid size selection, but you didn't
actually change the codegen, no? Do you plan this as part of another PR?
https://github.com/llvm/llvm-project/pull/205325
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
