llvm/test/Transforms/SLPVectorizer/addsub.ll

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: %if x86-registered-target %{ opt < %s -passes=slp-vectorizer -S -mtriple=x86_64-unknown-linux-gnu | FileCheck %s %}
; RUN: %if aarch64-registered-target %{ opt < %s -passes=slp-vectorizer -S -mtriple=aarch64-unknown-linux-gnu | FileCheck %s %}

@b = common global [4 x i32] zeroinitializer, align 16
@c = common global [4 x i32] zeroinitializer, align 16
@d = common global [4 x i32] zeroinitializer, align 16
@e = common global [4 x i32] zeroinitializer, align 16
@a = common global [4 x i32] zeroinitializer, align 16
@fb = common global [4 x float] zeroinitializer, align 16
@fc = common global [4 x float] zeroinitializer, align 16
@fa = common global [4 x float] zeroinitializer, align 16
@fd = common global [4 x float] zeroinitializer, align 16

; Function Attrs: nounwind uwtable
define void @addsub() #0 {
; CHECK-LABEL: @addsub(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i32>, ptr @b, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @c, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = add nsw <4 x i32> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr @d, align 4
; CHECK-NEXT:    [[TMP4:%.*]] = load <4 x i32>, ptr @e, align 4
; CHECK-NEXT:    [[TMP5:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP4]]
; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP7:%.*]] = sub nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x i32> [[TMP8]], ptr @a, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load i32, ptr @b, align 4
  %1 = load i32, ptr @c, align 4
  %add = add nsw i32 %0, %1
  %2 = load i32, ptr @d, align 4
  %3 = load i32, ptr @e, align 4
  %add1 = add nsw i32 %2, %3
  %add2 = add nsw i32 %add, %add1
  store i32 %add2, ptr @a, align 4
  %4 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 1), align 4
  %5 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 1), align 4
  %add3 = add nsw i32 %4, %5
  %6 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 1), align 4
  %7 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 1), align 4
  %add4 = add nsw i32 %6, %7
  %sub = sub nsw i32 %add3, %add4
  store i32 %sub, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 1), align 4
  %8 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 2), align 4
  %9 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 2), align 4
  %add5 = add nsw i32 %8, %9
  %10 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 2), align 4
  %11 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 2), align 4
  %add6 = add nsw i32 %10, %11
  %add7 = add nsw i32 %add5, %add6
  store i32 %add7, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 2), align 4
  %12 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 3), align 4
  %13 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 3), align 4
  %add8 = add nsw i32 %12, %13
  %14 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 3), align 4
  %15 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 3), align 4
  %add9 = add nsw i32 %14, %15
  %sub10 = sub nsw i32 %add8, %add9
  store i32 %sub10, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 3), align 4
  ret void
}

define void @addsub_freeze() #0 {
; CHECK-LABEL: @addsub_freeze(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i32>, ptr @b, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @c, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = add nsw <4 x i32> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr @d, align 4
; CHECK-NEXT:    [[TMP4:%.*]] = load <4 x i32>, ptr @e, align 4
; CHECK-NEXT:    [[TMP5:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP4]]
; CHECK-NEXT:    [[TMP6:%.*]] = add nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP7:%.*]] = sub nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    [[TMP9:%.*]] = freeze <4 x i32> [[TMP8]]
; CHECK-NEXT:    store <4 x i32> [[TMP9]], ptr @a, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load i32, ptr @b, align 4
  %1 = load i32, ptr @c, align 4
  %add = add nsw i32 %0, %1
  %2 = load i32, ptr @d, align 4
  %3 = load i32, ptr @e, align 4
  %add1 = add nsw i32 %2, %3
  %add2 = add nsw i32 %add, %add1
  %freeze.add2 = freeze i32 %add2
  store i32 %freeze.add2, ptr @a, align 4
  %4 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 1), align 4
  %5 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 1), align 4
  %add3 = add nsw i32 %4, %5
  %6 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 1), align 4
  %7 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 1), align 4
  %add4 = add nsw i32 %6, %7
  %sub = sub nsw i32 %add3, %add4
  %freeze.sub = freeze i32 %sub
  store i32 %freeze.sub, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 1), align 4
  %8 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 2), align 4
  %9 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 2), align 4
  %add5 = add nsw i32 %8, %9
  %10 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 2), align 4
  %11 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 2), align 4
  %add6 = add nsw i32 %10, %11
  %add7 = add nsw i32 %add5, %add6
  %freeze.add7 = freeze i32 %add7
  store i32 %freeze.add7, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 2), align 4
  %12 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 3), align 4
  %13 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 3), align 4
  %add8 = add nsw i32 %12, %13
  %14 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 3), align 4
  %15 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 3), align 4
  %add9 = add nsw i32 %14, %15
  %sub10 = sub nsw i32 %add8, %add9
  %freeze.sub10 = freeze i32 %sub10
  store i32 %freeze.sub10, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 3), align 4
  ret void
}

; Function Attrs: nounwind uwtable
define void @subadd() #0 {
; CHECK-LABEL: @subadd(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x i32>, ptr @b, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x i32>, ptr @c, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = add nsw <4 x i32> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x i32>, ptr @d, align 4
; CHECK-NEXT:    [[TMP4:%.*]] = load <4 x i32>, ptr @e, align 4
; CHECK-NEXT:    [[TMP5:%.*]] = add nsw <4 x i32> [[TMP3]], [[TMP4]]
; CHECK-NEXT:    [[TMP6:%.*]] = sub nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP7:%.*]] = add nsw <4 x i32> [[TMP2]], [[TMP5]]
; CHECK-NEXT:    [[TMP8:%.*]] = shufflevector <4 x i32> [[TMP6]], <4 x i32> [[TMP7]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x i32> [[TMP8]], ptr @a, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load i32, ptr @b, align 4
  %1 = load i32, ptr @c, align 4
  %add = add nsw i32 %0, %1
  %2 = load i32, ptr @d, align 4
  %3 = load i32, ptr @e, align 4
  %add1 = add nsw i32 %2, %3
  %sub = sub nsw i32 %add, %add1
  store i32 %sub, ptr @a, align 4
  %4 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 1), align 4
  %5 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 1), align 4
  %add2 = add nsw i32 %4, %5
  %6 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 1), align 4
  %7 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 1), align 4
  %add3 = add nsw i32 %6, %7
  %add4 = add nsw i32 %add2, %add3
  store i32 %add4, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 1), align 4
  %8 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 2), align 4
  %9 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 2), align 4
  %add5 = add nsw i32 %8, %9
  %10 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 2), align 4
  %11 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 2), align 4
  %add6 = add nsw i32 %10, %11
  %sub7 = sub nsw i32 %add5, %add6
  store i32 %sub7, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 2), align 4
  %12 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @b, i32 0, i64 3), align 4
  %13 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @c, i32 0, i64 3), align 4
  %add8 = add nsw i32 %12, %13
  %14 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @d, i32 0, i64 3), align 4
  %15 = load i32, ptr getelementptr inbounds ([4 x i32], ptr @e, i32 0, i64 3), align 4
  %add9 = add nsw i32 %14, %15
  %add10 = add nsw i32 %add8, %add9
  store i32 %add10, ptr getelementptr inbounds ([4 x i32], ptr @a, i32 0, i64 3), align 4
  ret void
}

; Function Attrs: nounwind uwtable
define void @faddfsub() #0 {
; CHECK-LABEL: @faddfsub(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr @fc, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP4]], ptr @fa, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load float, ptr @fb, align 4
  %1 = load float, ptr @fc, align 4
  %add = fadd float %0, %1
  store float %add, ptr @fa, align 4
  %2 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %3 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %sub = fsub float %2, %3
  store float %sub, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %4 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %5 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %add1 = fadd float %4, %5
  store float %add1, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %6 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  %sub2 = fsub float %6, %7
  store float %sub2, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  ret void
}

; Function Attrs: nounwind uwtable
define void @fsubfadd() #0 {
; CHECK-LABEL: @fsubfadd(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr @fc, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP4]], ptr @fa, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load float, ptr @fb, align 4
  %1 = load float, ptr @fc, align 4
  %sub = fsub float %0, %1
  store float %sub, ptr @fa, align 4
  %2 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %3 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %add = fadd float %2, %3
  store float %add, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %4 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %5 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %sub1 = fsub float %4, %5
  store float %sub1, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %6 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  %add2 = fadd float %6, %7
  store float %add2, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  ret void
}

; Function Attrs: nounwind uwtable
define void @faddfsub_select() #0 {
; CHECK-LABEL: @faddfsub_select(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    [[TMP0:%.*]] = load <4 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr @fc, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = fadd <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP3:%.*]] = fsub <4 x float> [[TMP0]], [[TMP1]]
; CHECK-NEXT:    [[TMP4:%.*]] = shufflevector <4 x float> [[TMP2]], <4 x float> [[TMP3]], <4 x i32> <i32 0, i32 1, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP4]], ptr @fa, align 4
; CHECK-NEXT:    ret void
;
entry:
  %0 = load float, ptr @fb, align 4
  %1 = load float, ptr @fc, align 4
  %add = fadd float %0, %1
  store float %add, ptr @fa, align 4
  %2 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %3 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %add1 = fadd float %2, %3
  store float %add1, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %4 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %5 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %add2 = fadd float %4, %5
  store float %add2, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %6 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  %sub = fsub float %6, %7
  store float %sub, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  ret void
}

; Check vectorization of following code for float data type-
;  fc[0] = fb[0]+fa[0]; //swapped fb and fa
;  fc[1] = fa[1]-fb[1];
;  fc[2] = fa[2]+fb[2];
;  fc[3] = fa[3]-fb[3];

define void @reorder_alt() #0 {
; CHECK-LABEL: @reorder_alt(
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr @fa, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = load <4 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP3:%.*]] = fadd <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[TMP4:%.*]] = fsub <4 x float> [[TMP1]], [[TMP2]]
; CHECK-NEXT:    [[TMP5:%.*]] = shufflevector <4 x float> [[TMP3]], <4 x float> [[TMP4]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP5]], ptr @fc, align 4
; CHECK-NEXT:    ret void
;
  %1 = load float, ptr @fb, align 4
  %2 = load float, ptr @fa, align 4
  %3 = fadd float %1, %2
  store float %3, ptr @fc, align 4
  %4 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %5 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %6 = fsub float %4, %5
  store float %6, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %8 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %9 = fadd float %7, %8
  store float %9, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %10 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  %11 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %12 = fsub float %10, %11
  store float %12, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  ret void
}

; Check vectorization of following code for float data type-
;  fc[0] = fa[0]+(fb[0]-fd[0]);
;  fc[1] = fa[1]-(fb[1]+fd[1]);
;  fc[2] = fa[2]+(fb[2]-fd[2]);
;  fc[3] = fa[3]-(fd[3]+fb[3]); //swapped fd and fb

define void @reorder_alt_subTree() #0 {
; CHECK-LABEL: @reorder_alt_subTree(
; CHECK-NEXT:    [[TMP1:%.*]] = load <4 x float>, ptr @fa, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = load <4 x float>, ptr @fd, align 4
; CHECK-NEXT:    [[TMP3:%.*]] = load <4 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP4:%.*]] = fsub <4 x float> [[TMP3]], [[TMP2]]
; CHECK-NEXT:    [[TMP5:%.*]] = fadd <4 x float> [[TMP3]], [[TMP2]]
; CHECK-NEXT:    [[TMP6:%.*]] = shufflevector <4 x float> [[TMP4]], <4 x float> [[TMP5]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    [[TMP7:%.*]] = fadd <4 x float> [[TMP1]], [[TMP6]]
; CHECK-NEXT:    [[TMP8:%.*]] = fsub <4 x float> [[TMP1]], [[TMP6]]
; CHECK-NEXT:    [[TMP9:%.*]] = shufflevector <4 x float> [[TMP7]], <4 x float> [[TMP8]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP9]], ptr @fc, align 4
; CHECK-NEXT:    ret void
;
  %1 = load float, ptr @fa, align 4
  %2 = load float, ptr @fb, align 4
  %3 = load float, ptr @fd, align 4
  %4 = fsub float %2, %3
  %5 = fadd float %1, %4
  store float %5, ptr @fc, align 4
  %6 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %8 = load float, ptr getelementptr inbounds ([4 x float], ptr @fd, i32 0, i64 1), align 4
  %9 = fadd float %7, %8
  %10 = fsub float %6, %9
  store float %10, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %11 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %12 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %13 = load float, ptr getelementptr inbounds ([4 x float], ptr @fd, i32 0, i64 2), align 4
  %14 = fsub float %12, %13
  %15 = fadd float %11, %14
  store float %15, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %16 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  %17 = load float, ptr getelementptr inbounds ([4 x float], ptr @fd, i32 0, i64 3), align 4
  %18 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %19 = fadd float %17, %18
  %20 = fsub float %16, %19
  store float %20, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  ret void
}

; Check vectorization of following code for double data type-
;  c[0] = (a[0]+b[0])-d[0];
;  c[1] = d[1]+(a[1]+b[1]); //swapped d[1] and (a[1]+b[1])

define void @reorder_alt_rightsubTree(ptr nocapture %c, ptr noalias nocapture readonly %a, ptr noalias nocapture readonly %b, ptr noalias nocapture readonly %d) {
; CHECK-LABEL: @reorder_alt_rightsubTree(
; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x double>, ptr [[D:%.*]], align 8
; CHECK-NEXT:    [[TMP2:%.*]] = load <2 x double>, ptr [[A:%.*]], align 8
; CHECK-NEXT:    [[TMP3:%.*]] = load <2 x double>, ptr [[B:%.*]], align 8
; CHECK-NEXT:    [[TMP4:%.*]] = fadd <2 x double> [[TMP2]], [[TMP3]]
; CHECK-NEXT:    [[TMP5:%.*]] = fsub <2 x double> [[TMP4]], [[TMP1]]
; CHECK-NEXT:    [[TMP6:%.*]] = fadd <2 x double> [[TMP4]], [[TMP1]]
; CHECK-NEXT:    [[TMP7:%.*]] = shufflevector <2 x double> [[TMP5]], <2 x double> [[TMP6]], <2 x i32> <i32 0, i32 3>
; CHECK-NEXT:    store <2 x double> [[TMP7]], ptr [[C:%.*]], align 8
; CHECK-NEXT:    ret void
;
  %1 = load double, ptr %a
  %2 = load double, ptr %b
  %3 = fadd double %1, %2
  %4 = load double, ptr %d
  %5 = fsub double %3, %4
  store double %5, ptr %c
  %6 = getelementptr inbounds double, ptr %d, i64 1
  %7 = load double, ptr %6
  %8 = getelementptr inbounds double, ptr %a, i64 1
  %9 = load double, ptr %8
  %10 = getelementptr inbounds double, ptr %b, i64 1
  %11 = load double, ptr %10
  %12 = fadd double %9, %11
  %13 = fadd double %7, %12
  %14 = getelementptr inbounds double, ptr %c, i64 1
  store double %13, ptr %14
  ret void
}

define void @vec_shuff_reorder() #0 {
; CHECK-LABEL: @vec_shuff_reorder(
; CHECK-NEXT:    [[TMP1:%.*]] = load <2 x float>, ptr @fa, align 4
; CHECK-NEXT:    [[TMP2:%.*]] = load <2 x float>, ptr @fb, align 4
; CHECK-NEXT:    [[TMP3:%.*]] = load <2 x float>, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
; CHECK-NEXT:    [[TMP4:%.*]] = load <2 x float>, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
; CHECK-NEXT:    [[TMP5:%.*]] = call <4 x float> @llvm.vector.insert.v4f32.v2f32(<4 x float> poison, <2 x float> [[TMP1]], i64 0)
; CHECK-NEXT:    [[TMP6:%.*]] = call <4 x float> @llvm.vector.insert.v4f32.v2f32(<4 x float> [[TMP5]], <2 x float> [[TMP3]], i64 2)
; CHECK-NEXT:    [[TMP7:%.*]] = call <4 x float> @llvm.vector.insert.v4f32.v2f32(<4 x float> poison, <2 x float> [[TMP2]], i64 0)
; CHECK-NEXT:    [[TMP8:%.*]] = call <4 x float> @llvm.vector.insert.v4f32.v2f32(<4 x float> [[TMP7]], <2 x float> [[TMP4]], i64 2)
; CHECK-NEXT:    [[TMP9:%.*]] = fadd <4 x float> [[TMP6]], [[TMP8]]
; CHECK-NEXT:    [[TMP10:%.*]] = fsub <4 x float> [[TMP6]], [[TMP8]]
; CHECK-NEXT:    [[TMP11:%.*]] = shufflevector <4 x float> [[TMP9]], <4 x float> [[TMP10]], <4 x i32> <i32 0, i32 5, i32 2, i32 7>
; CHECK-NEXT:    store <4 x float> [[TMP11]], ptr @fc, align 4
; CHECK-NEXT:    ret void
;
  %1 = load float, ptr @fb, align 4
  %2 = load float, ptr @fa, align 4
  %3 = fadd float %1, %2
  store float %3, ptr @fc, align 4
  %4 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 1), align 4
  %5 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 1), align 4
  %6 = fsub float %4, %5
  store float %6, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 1), align 4
  %7 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 2), align 4
  %8 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 2), align 4
  %9 = fadd float %7, %8
  store float %9, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 2), align 4
  %10 = load float, ptr getelementptr inbounds ([4 x float], ptr @fb, i32 0, i64 3), align 4
  %11 = load float, ptr getelementptr inbounds ([4 x float], ptr @fa, i32 0, i64 3), align 4
  %12 = fsub float %10, %11
  store float %12, ptr getelementptr inbounds ([4 x float], ptr @fc, i32 0, i64 3), align 4
  ret void
}


attributes #0 = { nounwind }