; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -basicaa -slp-vectorizer -dce -S -mtriple=i386-apple-macosx10.8.0 -mcpu=corei7-avx | FileCheck %s

target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32-S128"
target triple = "i386-apple-macosx10.8.0"

define double @foo(double* nocapture %D) {
; CHECK-LABEL: @foo(
; CHECK-NEXT:    br label [[TMP1:%.*]]
; CHECK:         [[I_02:%.*]] = phi i32 [ 0, [[TMP0:%.*]] ], [ [[TMP12:%.*]], [[TMP1]] ]
; CHECK-NEXT:    [[SUM_01:%.*]] = phi double [ 0.000000e+00, [[TMP0]] ], [ [[TMP11:%.*]], [[TMP1]] ]
; CHECK-NEXT:    [[TMP2:%.*]] = shl nsw i32 [[I_02]], 1
; CHECK-NEXT:    [[TMP3:%.*]] = getelementptr inbounds double, double* [[D:%.*]], i32 [[TMP2]]
; CHECK-NEXT:    [[TMP4:%.*]] = bitcast double* [[TMP3]] to <2 x double>*
; CHECK-NEXT:    [[TMP5:%.*]] = load <2 x double>, <2 x double>* [[TMP4]], align 4
; CHECK-NEXT:    [[TMP6:%.*]] = fmul <2 x double> [[TMP5]], [[TMP5]]
; CHECK-NEXT:    [[TMP7:%.*]] = fmul <2 x double> [[TMP6]], [[TMP6]]
; CHECK-NEXT:    [[TMP8:%.*]] = extractelement <2 x double> [[TMP7]], i32 0
; CHECK-NEXT:    [[TMP9:%.*]] = extractelement <2 x double> [[TMP7]], i32 1
; CHECK-NEXT:    [[TMP10:%.*]] = fadd double [[TMP8]], [[TMP9]]
; CHECK-NEXT:    [[TMP11]] = fadd double [[SUM_01]], [[TMP10]]
; CHECK-NEXT:    [[TMP12]] = add nsw i32 [[I_02]], 1
; CHECK-NEXT:    [[EXITCOND:%.*]] = icmp eq i32 [[TMP12]], 100
; CHECK-NEXT:    br i1 [[EXITCOND]], label [[TMP13:%.*]], label [[TMP1]]
; CHECK:         ret double [[TMP11]]
;
  br label %1

; <label>:1                                       ; preds = %1, %0
  %i.02 = phi i32 [ 0, %0 ], [ %10, %1 ]
  %sum.01 = phi double [ 0.000000e+00, %0 ], [ %9, %1 ]
  %2 = shl nsw i32 %i.02, 1
  %3 = getelementptr inbounds double, double* %D, i32 %2
  %4 = load double, double* %3, align 4
  %A4 = fmul double %4, %4
  %A42 = fmul double %A4, %A4
  %5 = or i32 %2, 1
  %6 = getelementptr inbounds double, double* %D, i32 %5
  %7 = load double, double* %6, align 4
  %A7 = fmul double %7, %7
  %A72 = fmul double %A7, %A7
  %8 = fadd double %A42, %A72
  %9 = fadd double %sum.01, %8
  %10 = add nsw i32 %i.02, 1
  %exitcond = icmp eq i32 %10, 100
  br i1 %exitcond, label %11, label %1

; <label>:11                                      ; preds = %1
  ret double %9
}