; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt -S -passes=slp-vectorizer %s -slp-threshold=-10 | FileCheck %s
target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
target triple = "aarch64--linux-gnu"

%structA = type { [2 x float] }

define void @test1(ptr nocapture readonly %J, i32 %xmin, i32 %ymin) {
; CHECK-LABEL: @test1(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[FOR_BODY3_LR_PH:%.*]]
; CHECK:       for.body3.lr.ph:
; CHECK-NEXT:    [[CONV5:%.*]] = sitofp i32 [[YMIN:%.*]] to float
; CHECK-NEXT:    [[CONV:%.*]] = sitofp i32 [[XMIN:%.*]] to float
; CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr [[J:%.*]], align 4
; CHECK-NEXT:    [[SUB:%.*]] = fsub fast float [[CONV]], [[TMP0]]
; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds [[STRUCTA:%.*]], ptr [[J]], i64 0, i32 0, i64 1
; CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[ARRAYIDX9]], align 4
; CHECK-NEXT:    [[SUB10:%.*]] = fsub fast float [[CONV5]], [[TMP1]]
; CHECK-NEXT:    [[MUL11:%.*]] = fmul fast float [[SUB]], [[SUB]]
; CHECK-NEXT:    [[MUL12:%.*]] = fmul fast float [[SUB10]], [[SUB10]]
; CHECK-NEXT:    [[ADD:%.*]] = fadd fast float [[MUL11]], [[MUL12]]
; CHECK-NEXT:    [[CMP:%.*]] = fcmp oeq float [[ADD]], 0.000000e+00
; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY3_LR_PH]], label [[FOR_END27:%.*]]
; CHECK:       for.end27:
; CHECK-NEXT:    ret void
;

entry:
  br label %for.body3.lr.ph

for.body3.lr.ph:
  %conv5 = sitofp i32 %ymin to float
  %conv = sitofp i32 %xmin to float
  %0 = load float, ptr %J, align 4
  %sub = fsub fast float %conv, %0
  %arrayidx9 = getelementptr inbounds %structA, ptr %J, i64 0, i32 0, i64 1
  %1 = load float, ptr %arrayidx9, align 4
  %sub10 = fsub fast float %conv5, %1
  %mul11 = fmul fast float %sub, %sub
  %mul12 = fmul fast float %sub10, %sub10
  %add = fadd fast float %mul11, %mul12
  %cmp = fcmp oeq float %add, 0.000000e+00
  br i1 %cmp, label %for.body3.lr.ph, label %for.end27

for.end27:
  ret void
}

define void @test2(ptr nocapture readonly %J, i32 %xmin, i32 %ymin) {
; CHECK-LABEL: @test2(
; CHECK-NEXT:  entry:
; CHECK-NEXT:    br label [[FOR_BODY3_LR_PH:%.*]]
; CHECK:       for.body3.lr.ph:
; CHECK-NEXT:    [[CONV5:%.*]] = sitofp i32 [[YMIN:%.*]] to float
; CHECK-NEXT:    [[CONV:%.*]] = sitofp i32 [[XMIN:%.*]] to float
; CHECK-NEXT:    [[TMP0:%.*]] = load float, ptr [[J:%.*]], align 4
; CHECK-NEXT:    [[SUB:%.*]] = fsub fast float [[CONV]], [[TMP0]]
; CHECK-NEXT:    [[ARRAYIDX9:%.*]] = getelementptr inbounds [[STRUCTA:%.*]], ptr [[J]], i64 0, i32 0, i64 1
; CHECK-NEXT:    [[TMP1:%.*]] = load float, ptr [[ARRAYIDX9]], align 4
; CHECK-NEXT:    [[SUB10:%.*]] = fsub fast float [[CONV5]], [[TMP1]]
; CHECK-NEXT:    [[MUL11:%.*]] = fmul fast float [[SUB]], [[SUB]]
; CHECK-NEXT:    [[MUL12:%.*]] = fmul fast float [[SUB10]], [[SUB10]]
; CHECK-NEXT:    [[ADD:%.*]] = fadd fast float [[MUL12]], [[MUL11]]
; CHECK-NEXT:    [[CMP:%.*]] = fcmp oeq float [[ADD]], 0.000000e+00
; CHECK-NEXT:    br i1 [[CMP]], label [[FOR_BODY3_LR_PH]], label [[FOR_END27:%.*]]
; CHECK:       for.end27:
; CHECK-NEXT:    ret void
;

entry:
  br label %for.body3.lr.ph

for.body3.lr.ph:
  %conv5 = sitofp i32 %ymin to float
  %conv = sitofp i32 %xmin to float
  %0 = load float, ptr %J, align 4
  %sub = fsub fast float %conv, %0
  %arrayidx9 = getelementptr inbounds %structA, ptr %J, i64 0, i32 0, i64 1
  %1 = load float, ptr %arrayidx9, align 4
  %sub10 = fsub fast float %conv5, %1
  %mul11 = fmul fast float %sub, %sub
  %mul12 = fmul fast float %sub10, %sub10
  %add = fadd fast float %mul12, %mul11         ;;;<---- Operands commuted!!
  %cmp = fcmp oeq float %add, 0.000000e+00
  br i1 %cmp, label %for.body3.lr.ph, label %for.end27

for.end27:
  ret void
}