[cuda] Include GPU binary into host object file and generate init/deinit code.

- added -fcuda-include-gpubinary option to incorporate results of
  device-side compilation into host-side one.
- generate code to register GPU binaries and associated kernels
  with CUDA runtime and clean-up on exit.
- added test case for init/deinit code generation.

Differential Revision: http://reviews.llvm.org/D9507

llvm-svn: 236765

clang/lib/CodeGen/CodeGenModule.cpp

@@ -350,6 +350,13 @@ void CodeGenModule::Release() {
   if (ObjCRuntime)
     if (llvm::Function *ObjCInitFunction = ObjCRuntime->ModuleInitFunction())
       AddGlobalCtor(ObjCInitFunction);
+  if (Context.getLangOpts().CUDA && !Context.getLangOpts().CUDAIsDevice &&
+      CUDARuntime) {
+    if (llvm::Function *CudaCtorFunction = CUDARuntime->makeModuleCtorFunction())
+      AddGlobalCtor(CudaCtorFunction);
+    if (llvm::Function *CudaDtorFunction = CUDARuntime->makeModuleDtorFunction())
+      AddGlobalDtor(CudaDtorFunction);
+  }
   if (PGOReader && PGOStats.hasDiagnostics())
     PGOStats.reportDiagnostics(getDiags(), getCodeGenOpts().MainFileName);
   EmitCtorList(GlobalCtors, "llvm.global_ctors");
@@ -3678,4 +3685,3 @@ void CodeGenModule::EmitOMPThreadPrivateDecl(const OMPThreadPrivateDecl *D) {
       CXXGlobalInits.push_back(InitFunction);
   }
 }
-