[NVPTX] CUDA provides a memcpy and memset

[jdoerfert]

Summary

https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#dynamic-global-memory-allocation-and-operations

[tra]

It would be good to add a test.

Both NVCC and clang currently lower memcpy to an explicit loop. I'm not sure what effect (if any) allowing memcpy/memset libcall would have on performance. We may want to benchmark it before landing.

[jdoerfert]

In D98607#2626634, @tra wrote:

It would be good to add a test.

Both NVCC and clang currently lower memcpy to an explicit loop. I'm not sure what effect (if any) allowing memcpy/memset libcall would have on performance. We may want to benchmark it before landing.

I doubt I have the proper setup to do such benchmarking. I care about malloc/free, this was just a follow up because the same CUDA documentation paragraph says they are available.
I'm fine with dropping this for now.

[tra]

I'd incorporate the changes into your free/malloc patch, but leave them commented out with a TODO explaining that they are available, but disabled until we can prove that they are beneficial.

[jdoerfert]

"merged" into D98606

llvm/lib/Analysis/TargetLibraryInfo.cpp

static void initialize(TargetLibraryInfoImpl &TLI, const Triple &T,
   //
   // FIXME: Having no standard library prevents e.g. many fastmath
   // optimizations, so this situation should be fixed.
   if (T.isNVPTX()) {
     TLI.disableAllFunctions();
     TLI.setAvailable(LibFunc_nvvm_reflect);
     TLI.setAvailable(llvm::LibFunc_malloc);
     TLI.setAvailable(llvm::LibFunc_free);
+    TLI.setAvailable(llvm::LibFunc_memcpy);
+    TLI.setAvailable(llvm::LibFunc_memset);
   } else {
     TLI.setUnavailable(LibFunc_nvvm_reflect);
   }
 
   // These vec_malloc/free routines are only available on AIX.
   if (!T.isOSAIX()) {
     TLI.setUnavailable(LibFunc_vec_calloc);
     TLI.setUnavailable(LibFunc_vec_malloc);