[InstCombine] Y - ~X --> X + Y + 1 fold (PR42457)

Summary:
I *think* we'd want this new variant, because we obviously
have better handling for `add` as compared to `sub`/`not`.

https://rise4fun.com/Alive/WMn

Fixes [[ https://bugs.llvm.org/show_bug.cgi?id=42457 | PR42457 ]]

Reviewers: spatel, nikic, huihuiz, efriedma

Reviewed By: spatel

Subscribers: RKSimon, llvm-commits

Tags: #llvm

Differential Revision: https://reviews.llvm.org/D63992

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@365011 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Transforms/InstCombine/InstCombineAddSub.cpp

@@ -1580,6 +1580,12 @@ Instruction *InstCombiner::visitSub(BinaryOperator &I) {
   if (match(Op1, m_OneUse(m_Add(m_Value(X), m_One()))))
     return BinaryOperator::CreateAdd(Builder.CreateNot(X), Op0);
 
+  // Y - ~X --> (X + 1) + Y
+  if (match(Op1, m_OneUse(m_Not(m_Value(X))))) {
+    return BinaryOperator::CreateAdd(
+        Builder.CreateAdd(Op0, ConstantInt::get(I.getType(), 1)), X);
+  }
+
   if (Constant *C = dyn_cast<Constant>(Op0)) {
     bool IsNegate = match(C, m_ZeroInt());
     Value *X;

test/Transforms/InstCombine/fold-sub-of-not-to-inc-of-add.ll

@@ -4,7 +4,8 @@
 ; Given:
 ;   sub %y, (xor %x, -1)
 ; Transform it to:
-;   add (add %x, %y), 1
+;   add (add %x, 1), %y
+; We prefer this form because that is what -reassociate would produce.
 
 ;------------------------------------------------------------------------------;
 ; Scalar tests
@@ -12,8 +13,8 @@
 
 define i32 @p0_scalar(i32 %x, i32 %y) {
 ; CHECK-LABEL: @p0_scalar(
-; CHECK-NEXT:    [[T0:%.*]] = xor i32 [[X:%.*]], -1
-; CHECK-NEXT:    [[T1:%.*]] = sub i32 [[Y:%.*]], [[T0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add i32 [[Y:%.*]], 1
+; CHECK-NEXT:    [[T1:%.*]] = add i32 [[TMP1]], [[X:%.*]]
 ; CHECK-NEXT:    ret i32 [[T1]]
 ;
   %t0 = xor i32 %x, -1
@@ -27,8 +28,8 @@ define i32 @p0_scalar(i32 %x, i32 %y) {
 
 define <4 x i32> @p1_vector_splat(<4 x i32> %x, <4 x i32> %y) {
 ; CHECK-LABEL: @p1_vector_splat(
-; CHECK-NEXT:    [[T0:%.*]] = xor <4 x i32> [[X:%.*]], <i32 -1, i32 -1, i32 -1, i32 -1>
-; CHECK-NEXT:    [[T1:%.*]] = sub <4 x i32> [[Y:%.*]], [[T0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add <4 x i32> [[Y:%.*]], <i32 1, i32 1, i32 1, i32 1>
+; CHECK-NEXT:    [[T1:%.*]] = add <4 x i32> [[TMP1]], [[X:%.*]]
 ; CHECK-NEXT:    ret <4 x i32> [[T1]]
 ;
   %t0 = xor <4 x i32> %x, <i32 -1, i32 -1, i32 -1, i32 -1>
@@ -38,8 +39,8 @@ define <4 x i32> @p1_vector_splat(<4 x i32> %x, <4 x i32> %y) {
 
 define <4 x i32> @p2_vector_undef(<4 x i32> %x, <4 x i32> %y) {
 ; CHECK-LABEL: @p2_vector_undef(
-; CHECK-NEXT:    [[T0:%.*]] = xor <4 x i32> [[X:%.*]], <i32 -1, i32 -1, i32 undef, i32 -1>
-; CHECK-NEXT:    [[T1:%.*]] = sub <4 x i32> [[Y:%.*]], [[T0]]
+; CHECK-NEXT:    [[TMP1:%.*]] = add <4 x i32> [[Y:%.*]], <i32 1, i32 1, i32 1, i32 1>
+; CHECK-NEXT:    [[T1:%.*]] = add <4 x i32> [[TMP1]], [[X:%.*]]
 ; CHECK-NEXT:    ret <4 x i32> [[T1]]
 ;
   %t0 = xor <4 x i32> %x, <i32 -1, i32 -1, i32 undef, i32 -1>