1; RUN: opt < %s  -loop-vectorize -force-vector-interleave=1 -force-vector-width=4 -dce -instcombine -S | FileCheck %s
2
3target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
4target triple = "x86_64-apple-macosx10.8.0"
5
6@b = common global [2048 x i32] zeroinitializer, align 16
7@c = common global [2048 x i32] zeroinitializer, align 16
8@a = common global [2048 x i32] zeroinitializer, align 16
9
10;CHECK-LABEL: @example1(
11;CHECK: shl i32
12;CHECK: zext i32
13;CHECK: load <4 x i32>
14;CHECK: add nsw <4 x i32>
15;CHECK: store <4 x i32>
16;CHECK: ret void
17define void @example1(i32 %n) nounwind uwtable ssp {
18  %n4 = shl i32 %n, 2
19  br label %1
20
21; <label>:1                                       ; preds = %1, %0
22  %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
23  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
24  %3 = load i32* %2, align 4
25  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
26  %5 = load i32* %4, align 4
27  %6 = add nsw i32 %5, %3
28  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
29  store i32 %6, i32* %7, align 4
30  %indvars.iv.next = add i64 %indvars.iv, 1
31  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
32  %exitcond = icmp eq i32 %lftr.wideiv, %n4
33  br i1 %exitcond, label %8, label %1
34
35; <label>:8                                       ; preds = %1
36  ret void
37}
38
39