[X86][APX] Combine xor .., -1 into Cload/Cstore conditions

Remove redundant NOT instruction: https://godbolt.org/z/jM89ejnsh

Author: phoebewang

llvm/lib/Target/X86/X86ISelLowering.cpp

@@ -58071,12 +58071,24 @@ static SDValue combineX86CloadCstore(SDNode *N, SelectionDAG &DAG) {
     Ops[3] = Op1.getOperand(0);
     Ops[4] = Op1.getOperand(1);
   } else if (Op1.getOpcode() == ISD::AND && Sub.getValue(0).use_empty()) {
+    SDValue Src = Op1;
+    SDValue Op10 = Op1.getOperand(0);
+    if (Op10.getOpcode() == ISD::XOR && isAllOnesConstant(Op10.getOperand(1))) {
+      // res, flags2 = sub 0, (and (xor X, -1), Y)
+      // cload/cstore ..., cond_ne, flag2
+      // ->
+      // res, flags2 = sub 0, (and X, Y)
+      // cload/cstore ..., cond_e, flag2
+      Src = DAG.getNode(ISD::AND, DL, Op1.getValueType(), Op10.getOperand(0),
+                        Op1.getOperand(1));
+      Ops[3] = DAG.getTargetConstant(X86::COND_E, DL, MVT::i8);
+    }
     // res, flags2 = sub 0, (and X, Y)
-    // cload/cstore ..., cond_ne, flag2
+    // cload/cstore ..., cc, flag2
     // ->
     // res, flags2 = cmp (and X, Y), 0
-    // cload/cstore ..., cond_ne, flag2
-    Ops[4] = DAG.getNode(X86ISD::CMP, DL, MVT::i32, Op1, Sub.getOperand(0));
+    // cload/cstore ..., cc, flag2
+    Ops[4] = DAG.getNode(X86ISD::CMP, DL, MVT::i32, Src, Sub.getOperand(0));
   } else {
     return SDValue();
   }

llvm/test/CodeGen/X86/apx/cf.ll

@@ -215,3 +215,17 @@ next:
   store <1 x i32> %2, ptr %p, align 4
   ret void
 }
+
+define void @xor_cond(ptr %p, i1 %cond) {
+; CHECK-LABEL: xor_cond:
+; CHECK:       # %bb.0: # %entry
+; CHECK-NEXT:    xorl %eax, %eax
+; CHECK-NEXT:    testb $1, %sil
+; CHECK-NEXT:    cfcmovel %eax, (%rdi)
+; CHECK-NEXT:    retq
+entry:
+  %0 = xor i1 %cond, true
+  %1 = insertelement <1 x i1> zeroinitializer, i1 %0, i64 0
+  call void @llvm.masked.store.v1i32.p0(<1 x i32> zeroinitializer, ptr %p, i32 1, <1 x i1> %1)
+  ret void
+}