[clang] [llvm] [AMDGPU] Emit a waitcnt instruction after each memory instruction (PR #79236)

[Jay Foad via cfe-commits]

jayfoad wrote:

Can you add at least one test for a VMEM (flat or scratch or global or buffer or image) atomic without return? That should use vscnt on GFX10.

Apart from that the SIInsertWaitcnts.cpp and tests look good to me. I have not reviewed the clang parts but it looks like @Pierre-vh approved them previously?

https://github.com/llvm/llvm-project/pull/79236