[clang] cuda clang: Add support for CUDA surfaces (PR #132883)

[Jorge Gorbe Moya via cfe-commits]

================
@@ -0,0 +1,3329 @@
+// RUN: %clang_cc1 -triple nvptx-unknown-unknown -fcuda-is-device -O3 -o - %s -emit-llvm | FileCheck %s
+// RUN: %clang_cc1 -triple nvptx64-unknown-unknown -fcuda-is-device -O3 -o - %s -emit-llvm | FileCheck %s
+#include "../Headers/Inputs/include/cuda.h"
----------------
slackito wrote:

Should "include/cuda.h" be in `clang/test/CodeGen/Inputs`  because it's needed for a CodeGen test?

test/Headers contains (if I'm not mistaken) test cases related to headers, not header files for other tests. And it seems unusual for a test to reach into the input files of a different category of tests.

https://github.com/llvm/llvm-project/pull/132883