[clang] [llvm] [AMDGPU] Introduce asyncmark/wait intrinsics (PR #173259)

[Pierre van Houtryve via cfe-commits]

================
@@ -0,0 +1,58 @@
+; RUN: llc -global-isel=0 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn -mcpu=gfx900 < %s | FileCheck %s
+
+define float @raw.buffer.load(<4 x i32> inreg %rsrc, ptr addrspace(3) inreg %lds) {
----------------
Pierre-vh wrote:

autogenerate these for ease of maintenance?

https://github.com/llvm/llvm-project/pull/173259