[llvm] [AMDGPU] Simplify S_WAIT_XCNT insertion. NFC. (PR #145682)

Jay Foad via llvm-commits llvm-commits at lists.llvm.org
Wed Jun 25 04:59:54 PDT 2025


================
@@ -1006,12 +999,8 @@ void WaitcntBrackets::updateByEvent(const SIInstrInfo *TII,
       }
     }
   } else if (T == X_CNT) {
-    for (const MachineOperand &Op : Inst.all_uses()) {
-      RegInterval Interval = getRegInterval(&Inst, MRI, TRI, Op);
-      for (int RegNo = Interval.first; RegNo < Interval.second; ++RegNo) {
-        setRegScore(RegNo, T, CurrScore);
----------------
jayfoad wrote:

Calling setRegScore in a loop over an interval was silly, because setRegScore just calls setScoreByInterval with an interval of size 1.

https://github.com/llvm/llvm-project/pull/145682


More information about the llvm-commits mailing list