[llvm] [AMDGPU] Delete redundant s_or_b32 (PR #165261)

[via llvm-commits]

================
@@ -10628,7 +10628,31 @@ bool SIInstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg,
   if (SrcReg2 && !getFoldableImm(SrcReg2, *MRI, CmpValue))
     return false;
 
-  const auto optimizeCmpSelect = [&CmpInstr, SrcReg, CmpValue, MRI,
+  // SCC is already valid after SCCValid.
+  // SCCRedefine will redefine SCC to the same value already available after
+  // SCCValid. If there are no intervening SCC conflicts delete SCCRedefine and
+  // update kill/dead flags if necessary.
+  const auto optimizeSCC = [this](MachineInstr *SCCValid,
----------------
LU-JOHN wrote:

> Would it be better to move `optimizeSCC` to a method? It's not accessing its surrounding context besides `this`.

optimizeSCC moved to NFC PR https://github.com/llvm/llvm-project/pull/165871.  Discuss method vs. lambda there.

https://github.com/llvm/llvm-project/pull/165261