[llvm] [X86][CodeGen] Return true when MIR is changed after optimizeCompareInstr (PR #91723)

[Shengchen Kan via llvm-commits]

================
@@ -5169,7 +5169,7 @@ bool X86InstrInfo::optimizeCompareInstr(MachineInstr &CmpInstr, Register SrcReg,
     // Fall through to optimize Cmp if Cmp is CMPrr or CMPri.
     if (NewOpcode == X86::CMP64rm || NewOpcode == X86::CMP32rm ||
         NewOpcode == X86::CMP16rm || NewOpcode == X86::CMP8rm)
-      return false;
+      return true;
----------------
KanRobert wrote:

The following is not to check EFLAGS, but for more aggressive optimizations for EFLAGS.  Memory analysis is not safe in MIR pass, so we don't do that for CMPrm.

https://github.com/llvm/llvm-project/pull/91723