[CUDA] Improve CUDA version detection and diagnostics.

Always use cuda.h to detect CUDA version. It's a more universal approach
compared to version.txt which is no longer present in recent CUDA versions.

Split the 'unknown CUDA version' warning in two:

* when detected CUDA version is partially supported by clang. It's expected to
work in general, at the feature parity with the latest supported CUDA
version. and may be missing support for the new features/instructions/GPU
variants. Clang will issue a warning.

* when detected version is new. Recent CUDA versions have been working with
clang reasonably well, and will likely to work similarly to the partially
supported ones above. Or it may not work at all. Clang will issue a warning and
proceed as if the latest known CUDA version was detected.

Differential Revision: https://reviews.llvm.org/D108247

1 files changed, 4 insertions, 2 deletions

diff --git a/clang/lib/Basic/Cuda.cpp b/clang/lib/Basic/Cuda.cpp
index 2e34da7..f5ee1fb 100644
--- a/clang/lib/Basic/Cuda.cpp
+++ b/clang/lib/Basic/Cuda.cpp
@@ -40,6 +40,8 @@ const char *CudaVersionToString(CudaVersion V) {
     return "11.3";
   case CudaVersion::CUDA_114:
     return "11.4";
+  case CudaVersion::NEW:
+    return "";
   }
   llvm_unreachable("invalid enum");
 }
@@ -192,7 +194,7 @@ CudaVersion MinVersionForCudaArch(CudaArch A) {
 CudaVersion MaxVersionForCudaArch(CudaArch A) {
   // AMD GPUs do not depend on CUDA versions.
   if (IsAMDGpuArch(A))
-    return CudaVersion::LATEST;
+    return CudaVersion::NEW;
 
   switch (A) {
   case CudaArch::UNKNOWN:
@@ -203,7 +205,7 @@ CudaVersion MaxVersionForCudaArch(CudaArch A) {
   case CudaArch::SM_30:
     return CudaVersion::CUDA_110;
   default:
-    return CudaVersion::LATEST;
+    return CudaVersion::NEW;
   }
 }