[llvm] [AMDGPU] Mark ASYNCMARK as meta instruction to fix hazard cycle miscounting (PR #189981)

[Sameer Sahasrabuddhe via llvm-commits]

================
@@ -366,10 +366,11 @@ define void @test_pipelined_loop(ptr addrspace(1) %foo, ptr addrspace(3) %lds, p
 ; SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
 ; SDAG-NEXT:    v_readfirstlane_b32 s4, v2
 ; SDAG-NEXT:    s_mov_b32 m0, s4
-; SDAG-NEXT:    v_mov_b32_e32 v5, 0
+; SDAG-NEXT:    s_nop 0
 ; SDAG-NEXT:    global_load_dword v[0:1], off lds
 ; SDAG-NEXT:    ; asyncmark
 ; SDAG-NEXT:    global_load_dword v[0:1], off lds
+; SDAG-NEXT:    v_mov_b32_e32 v5, 0
----------------
ssahasra wrote:

So the `v_mov` got replaced by an `s_nop`, and then it was scheduled later. What does that mean?

https://github.com/llvm/llvm-project/pull/189981