[InstCombine] Change order of canonicalization of ADD and AND

Canonicalize ((x + C1) & C2) --> ((x & C2) + C1) for suitable constants
C1 and C2, instead of the other way round. This should allow more
constant ADDs to be matched as part of addressing modes for loads and
stores.

Differential Revision: https://reviews.llvm.org/D130080

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp b/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
index 528c74a..94f21dd 100644 (file)
--- a/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineAddSub.cpp
@@ -969,15 +969,6 @@ Instruction *InstCombinerImpl::foldAddWithConstant(BinaryOperator &Add) {
     }
   }
 
-  // If all bits affected by the add are included in a high-bit-mask, do the
-  // add before the mask op:
-  // (X & 0xFF00) + xx00 --> (X + xx00) & 0xFF00
-  if (match(Op0, m_OneUse(m_And(m_Value(X), m_APInt(C2)))) &&
-      C2->isNegative() && C2->isShiftedMask() && *C == (*C & *C2)) {
-    Value *NewAdd = Builder.CreateAdd(X, ConstantInt::get(Ty, *C));
-    return BinaryOperator::CreateAnd(NewAdd, ConstantInt::get(Ty, *C2));
-  }
-
   return nullptr;
 }
 

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
index 59217d9..321cf65 100644 (file)
--- a/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineAndOrXor.cpp
@@ -1888,6 +1888,14 @@ Instruction *InstCombinerImpl::visitAnd(BinaryOperator &I) {
         Value *NewAnd = Builder.CreateAnd(X, Op1);
         return BinaryOperator::CreateXor(NewAnd, Op1);
       }
+
+      // If all bits affected by the add are included in a high-bit-mask, do the
+      // mask op before the add. Example:
+      // (X + 16) & -4 --> (X & -4) + 16
+      if (Op0->hasOneUse() && C->isNegatedPowerOf2() && *AddC == (*AddC & *C)) {
+        Value *NewAnd = Builder.CreateAnd(X, Op1);
+        return BinaryOperator::CreateAdd(NewAnd, ConstantInt::get(Ty, *AddC));
+      }
     }
 
     // ((C1 OP zext(X)) & C2) -> zext((C1 OP X) & C2) if C2 fits in the

diff --git a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
index f6f8f11..98cf450 100644 (file)
--- a/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineSelect.cpp
@@ -2602,10 +2602,14 @@ foldRoundUpIntegerWithPow2Alignment(SelectInst &SI,
   if (!match(XLowBits, m_And(m_Specific(X), m_APIntAllowUndef(LowBitMaskCst))))
     return nullptr;
 
+  // Match even if the AND and ADD are swapped.
   const APInt *BiasCst, *HighBitMaskCst;
   if (!match(XBiasedHighBits,
              m_And(m_Add(m_Specific(X), m_APIntAllowUndef(BiasCst)),
-                   m_APIntAllowUndef(HighBitMaskCst))))
+                   m_APIntAllowUndef(HighBitMaskCst))) &&
+      !match(XBiasedHighBits,
+             m_Add(m_And(m_Specific(X), m_APIntAllowUndef(HighBitMaskCst)),
+                   m_APIntAllowUndef(BiasCst))))
     return nullptr;
 
   if (!LowBitMaskCst->isMask())

diff --git a/llvm/test/Transforms/InstCombine/add.ll b/llvm/test/Transforms/InstCombine/add.ll
index 86f65e5..0df79e4 100644 (file)
--- a/llvm/test/Transforms/InstCombine/add.ll
+++ b/llvm/test/Transforms/InstCombine/add.ll
@@ -724,12 +724,12 @@ define i8 @test34(i8 %A) {
 }
 
 ; If all bits affected by the add are included
-; in the mask, do the add before the mask op.
+; in the mask, do the mask op before the add.
 
 define i8 @masked_add(i8 %x) {
 ; CHECK-LABEL: @masked_add(
-; CHECK-NEXT:    [[TMP1:%.*]] = add i8 [[X:%.*]], 96
-; CHECK-NEXT:    [[R:%.*]] = and i8 [[TMP1]], -16
+; CHECK-NEXT:    [[AND:%.*]] = and i8 [[X:%.*]], -16
+; CHECK-NEXT:    [[R:%.*]] = add i8 [[AND]], 96
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
   %and = and i8 %x, 240 ; 0xf0
@@ -739,8 +739,8 @@ define i8 @masked_add(i8 %x) {
 
 define <2 x i8> @masked_add_splat(<2 x i8> %x) {
 ; CHECK-LABEL: @masked_add_splat(
-; CHECK-NEXT:    [[TMP1:%.*]] = add <2 x i8> [[X:%.*]], <i8 64, i8 64>
-; CHECK-NEXT:    [[R:%.*]] = and <2 x i8> [[TMP1]], <i8 -64, i8 -64>
+; CHECK-NEXT:    [[AND:%.*]] = and <2 x i8> [[X:%.*]], <i8 -64, i8 -64>
+; CHECK-NEXT:    [[R:%.*]] = add <2 x i8> [[AND]], <i8 64, i8 64>
 ; CHECK-NEXT:    ret <2 x i8> [[R]]
 ;
   %and = and <2 x i8> %x, <i8 192, i8 192> ; 0xc0
@@ -761,14 +761,14 @@ define i8 @not_masked_add(i8 %x) {
 
 define i8 @masked_add_multi_use(i8 %x) {
 ; CHECK-LABEL: @masked_add_multi_use(
-; CHECK-NEXT:    [[TMP:%.*]] = add i8 [[X:%.*]], 96
-; CHECK-NEXT:    [[R:%.*]] = and i8 [[TMP:%.*]], -16
-; CHECK-NEXT:    call void @use(i8 [[X]])
+; CHECK-NEXT:    [[AND:%.*]] = and i8 [[X:%.*]], -16
+; CHECK-NEXT:    [[R:%.*]] = add i8 [[AND]], 96
+; CHECK-NEXT:    call void @use(i8 [[AND]])
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
   %and = and i8 %x, -16 ; 0xf0
   %r = add i8 %and, 96  ; 0x60
-  call void @use(i8 %x) ; extra use
+  call void @use(i8 %and) ; extra use
   ret i8 %r
 }
 

diff --git a/llvm/test/Transforms/InstCombine/integer-round-up-pow2-alignment.ll b/llvm/test/Transforms/InstCombine/integer-round-up-pow2-alignment.ll
index 58dbf14..04f8d2c 100644 (file)
--- a/llvm/test/Transforms/InstCombine/integer-round-up-pow2-alignment.ll
+++ b/llvm/test/Transforms/InstCombine/integer-round-up-pow2-alignment.ll
@@ -247,8 +247,8 @@ define i8 @n9_wrong_x0(i8 %x.0, i8 %x.1) {
 ; CHECK-LABEL: @n9_wrong_x0(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X_0:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X_0]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X_0]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X_1:%.*]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -263,8 +263,8 @@ define i8 @n9_wrong_x1(i8 %x.0, i8 %x.1) {
 ; CHECK-LABEL: @n9_wrong_x1(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X_0:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X_1:%.*]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X_1:%.*]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X_0]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -279,8 +279,8 @@ define i8 @n9_wrong_x2(i8 %x.0, i8 %x.1) {
 ; CHECK-LABEL: @n9_wrong_x2(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X_1:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X_0:%.*]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X_0:%.*]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X_0]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -297,8 +297,8 @@ define i8 @n10_wrong_low_bit_mask(i8 %x) {
 ; CHECK-LABEL: @n10_wrong_low_bit_mask(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X:%.*]], 31
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -333,8 +333,8 @@ define i8 @n12_wrong_bias(i8 %x) {
 ; CHECK-LABEL: @n12_wrong_bias(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X]], 32
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 32
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -369,8 +369,8 @@ define i8 @n14_wrong_comparison_constant(i8 %x) {
 ; CHECK-LABEL: @n14_wrong_comparison_constant(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 1
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -387,8 +387,8 @@ define i8 @n15_wrong_comparison_predicate_and_constant(i8 %x) {
 ; CHECK-LABEL: @n15_wrong_comparison_predicate_and_constant(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X:%.*]], 14
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]
 ;
@@ -405,8 +405,8 @@ define i8 @n16_oneuse(i8 %x) {
 ; CHECK-LABEL: @n16_oneuse(
 ; CHECK-NEXT:    [[X_LOWBITS:%.*]] = and i8 [[X:%.*]], 15
 ; CHECK-NEXT:    [[X_LOWBITS_ARE_ZERO:%.*]] = icmp eq i8 [[X_LOWBITS]], 0
-; CHECK-NEXT:    [[X_BIASED:%.*]] = add i8 [[X]], 16
-; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = and i8 [[X_BIASED]], -16
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X]], -16
+; CHECK-NEXT:    [[X_BIASED_HIGHBITS:%.*]] = add i8 [[TMP1]], 16
 ; CHECK-NEXT:    call void @use.i8(i8 [[X_BIASED_HIGHBITS]])
 ; CHECK-NEXT:    [[X_ROUNDEDUP:%.*]] = select i1 [[X_LOWBITS_ARE_ZERO]], i8 [[X]], i8 [[X_BIASED_HIGHBITS]]
 ; CHECK-NEXT:    ret i8 [[X_ROUNDEDUP]]

diff --git a/llvm/test/Transforms/InstCombine/or.ll b/llvm/test/Transforms/InstCombine/or.ll
index eb7c3a4..7dacb18 100644 (file)
--- a/llvm/test/Transforms/InstCombine/or.ll
+++ b/llvm/test/Transforms/InstCombine/or.ll
@@ -154,10 +154,14 @@ define i32 @test20(i32 %x) {
   ret i32 %z
 }
 
+; TODO: This should combine to t1 + 2.
 define i32 @test21(i32 %t1) {
 ; CHECK-LABEL: @test21(
-; CHECK-NEXT:    [[T1_MASK1:%.*]] = add i32 [[T1:%.*]], 2
-; CHECK-NEXT:    ret i32 [[T1_MASK1]]
+; CHECK-NEXT:    [[TMP1:%.*]] = and i32 [[T1:%.*]], -2
+; CHECK-NEXT:    [[T3:%.*]] = add i32 [[TMP1]], 2
+; CHECK-NEXT:    [[T5:%.*]] = and i32 [[T1]], 1
+; CHECK-NEXT:    [[T6:%.*]] = or i32 [[T5]], [[T3]]
+; CHECK-NEXT:    ret i32 [[T6]]
 ;
   %t1.mask1 = add i32 %t1, 2
   %t3 = and i32 %t1.mask1, -2

diff --git a/llvm/test/Transforms/InstCombine/sub.ll b/llvm/test/Transforms/InstCombine/sub.ll
index f7d99a0..fa243df 100644 (file)
--- a/llvm/test/Transforms/InstCombine/sub.ll
+++ b/llvm/test/Transforms/InstCombine/sub.ll
@@ -1490,8 +1490,8 @@ define i8 @sub_add_sub_reassoc_use2(i8 %w, i8 %x, i8 %y, i8 %z) {
 
 define i8 @sub_mask_lowbits(i8 %x) {
 ; CHECK-LABEL: @sub_mask_lowbits(
-; CHECK-NEXT:    [[A1:%.*]] = add i8 [[X:%.*]], -108
-; CHECK-NEXT:    [[R:%.*]] = and i8 [[A1]], -4
+; CHECK-NEXT:    [[TMP1:%.*]] = and i8 [[X:%.*]], -4
+; CHECK-NEXT:    [[R:%.*]] = add i8 [[TMP1]], -108
 ; CHECK-NEXT:    ret i8 [[R]]
 ;
   %a1 = add i8 %x, 148 ; 0x94