[libc-commits] [libc] [libc] Partially implement 'rand' for the GPU (PR #66167)

[Matt Arsenault via libc-commits]

================
@@ -21,10 +21,12 @@ namespace gpu {
 constexpr const uint64_t LANE_SIZE = __AMDGCN_WAVEFRONT_SIZE;
 
 /// Type aliases to the address spaces used by the AMDGPU backend.
-template <typename T> using Private = [[clang::opencl_private]] T;
-template <typename T> using Constant = [[clang::opencl_constant]] T;
-template <typename T> using Local = [[clang::opencl_local]] T;
-template <typename T> using Global = [[clang::opencl_global]] T;
+// FIXME: The OpenCL attributes are currently buggy on AMDGPU compared to the
----------------
arsenm wrote:

I don't think the asserts you are running into should be difficult to fix 

https://github.com/llvm/llvm-project/pull/66167