[llvm] [AMDGPU] Optimize out s_barrier_signal/_wait (PR #116993)

[Matt Arsenault via llvm-commits]

================
@@ -0,0 +1,56 @@
+; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 < %s | FileCheck %s
+; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 < %s -global-isel | FileCheck %s
----------------
arsenm wrote:

move -global-isel to the front and use explicit =0 for the dag checks 

https://github.com/llvm/llvm-project/pull/116993