[clang] [CUDA] make kernel stub ICF-proof (PR #90155)
Reid Kleckner via cfe-commits
cfe-commits at lists.llvm.org
Mon Apr 29 16:12:11 PDT 2024
================
@@ -424,6 +424,34 @@ void CGNVCUDARuntime::emitDeviceStubBodyNew(CodeGenFunction &CGF,
CGM.CreateRuntimeFunction(FTy, LaunchKernelName);
CGF.EmitCall(FI, CGCallee::forDirect(cudaLaunchKernelFn), ReturnValueSlot(),
LaunchKernelArgs);
+
+ // To prevent CUDA device stub functions from being merged by ICF in MSVC
+ // environment, create an unique global variable for each kernel and write to
+ // the variable in the device stub.
+ if (CGM.getContext().getTargetInfo().getCXXABI().isMicrosoft() &&
+ !CGF.getLangOpts().HIP) {
+ llvm::Function *KernelFunction = llvm::cast<llvm::Function>(Kernel);
+ if (KernelFunction->hasComdat()) {
+ std::string KernelName = KernelFunction->getName().str();
----------------
rnk wrote:
Please avoid making an extra copy of the original name, since C++ names can be many kilobytes. Use `(KernelFunction->getName() + ".id").str()` to construct the `std::string`.
https://github.com/llvm/llvm-project/pull/90155
More information about the cfe-commits
mailing list