[clang] [llvm] [AMDGPU] Emit a waitcnt instruction after each memory instruction (PR #68932)

Matt Arsenault via llvm-commits llvm-commits at lists.llvm.org
Wed Nov 8 00:17:44 PST 2023


================
@@ -1809,6 +1816,23 @@ bool SIInsertWaitcnts::shouldFlushVmCnt(MachineLoop *ML,
   return HasVMemLoad && UsesVgprLoadedOutside;
 }
 
+bool SIInsertWaitcnts::insertWaitcntAfterMemOp(MachineFunction &MF) {
+  bool Modified = false;
+
+  for (auto &MBB : MF) {
----------------
arsenm wrote:

Should try to integrate with the rest of the logic instead of adding a separate pass over the function 

https://github.com/llvm/llvm-project/pull/68932


More information about the llvm-commits mailing list