[llvm-branch-commits] [llvm] [AMDGPU] always emit a soft wait even if it is trivially ~0 (PR #147257)

[Pierre van Houtryve via llvm-branch-commits]

================
@@ -669,6 +679,7 @@ define amdgpu_kernel void @global_volatile_store_1(
 ; GFX12-WGP-NEXT:    s_wait_kmcnt 0x0
 ; GFX12-WGP-NEXT:    s_wait_storecnt 0x0
 ; GFX12-WGP-NEXT:    global_store_b32 v0, v1, s[0:1] scope:SCOPE_SYS
+; GFX12-WGP-NEXT:    s_wait_loadcnt 0x3f
----------------
Pierre-vh wrote:

The waitcnts aren't optimized out at O0 because we want to see them in memory legalizer tests, however we're mostly interested in the waitcnt zero, not the waitcnt ~0
We could still optimize out the ~0 ones, I don't think there is a downside to that

https://github.com/llvm/llvm-project/pull/147257