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

Jun Wang via llvm-commits llvm-commits at lists.llvm.org
Thu Nov 9 19:13:10 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) {
----------------
jwanggit86 wrote:

@arsenm Pull req has been updated. Now the code for the new feature has been integrated with the existing pass.

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


More information about the llvm-commits mailing list