[OpenMP][AMDGPU] Do not attach -fcuda-is-device (#99002)

Summary:
-fcuda-is-device flag is not used for OpenMP offloading for AMD GPUs and
it does not need to be added as clang cc1 option for OpenMP code.

This PR has the same functionality as
#96909 but it doesn't introduce
regression for virtual function support.

Test Plan: 

Reviewers: 

Subscribers: 

Tasks: 

Tags: 


Differential Revision: https://phabricator.intern.facebook.com/D60251022

clang/lib/CodeGen/CodeGenModule.h

@@ -1010,7 +1010,7 @@ class CodeGenModule : public CodeGenTypeCache {
   bool shouldEmitRTTI(bool ForEH = false) {
     return (ForEH || getLangOpts().RTTI) && !getLangOpts().CUDAIsDevice &&
            !(getLangOpts().OpenMP && getLangOpts().OpenMPIsTargetDevice &&
-             getTriple().isNVPTX());
+             (getTriple().isNVPTX() || getTriple().isAMDGPU()));
   }
 
   /// Get the address of the RTTI descriptor for the given type.

clang/lib/Driver/ToolChains/AMDGPUOpenMP.cpp

@@ -47,8 +47,6 @@ void AMDGPUOpenMPToolChain::addClangTargetOptions(
   assert(DeviceOffloadingKind == Action::OFK_OpenMP &&
          "Only OpenMP offloading kinds are supported.");
 
-  CC1Args.push_back("-fcuda-is-device");
-
   if (DriverArgs.hasArg(options::OPT_nogpulib))
     return;
 

clang/test/Driver/amdgpu-openmp-toolchain.c

@@ -7,7 +7,7 @@
 
 // verify the tools invocations
 // CHECK: "-cc1" "-triple" "x86_64-unknown-linux-gnu"{{.*}}"-emit-llvm-bc"{{.*}}"-x" "c"
-// CHECK: "-cc1" "-triple" "amdgcn-amd-amdhsa" "-aux-triple" "x86_64-unknown-linux-gnu"{{.*}}"-fcuda-is-device"{{.*}}"-target-cpu" "gfx906"
+// CHECK: "-cc1" "-triple" "amdgcn-amd-amdhsa" "-aux-triple" "x86_64-unknown-linux-gnu"{{.*}}"-target-cpu" "gfx906"
 // CHECK: "-cc1" "-triple" "x86_64-unknown-linux-gnu"{{.*}}"-emit-obj"
 // CHECK: clang-linker-wrapper{{.*}} "-o" "a.out"