1; RUN: opt < %s  -loop-vectorize -force-vector-width=4 -force-vector-interleave=2 -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@a = common global [2048 x i32] zeroinitializer, align 16
7
8; This is the loop.
9;  for (i=0; i<n; i++){
10;    a[i] += i;
11;  }
12;CHECK-LABEL: @inc(
13;CHECK: load <4 x i32>
14;CHECK: load <4 x i32>
15;CHECK: add nsw <4 x i32>
16;CHECK: add nsw <4 x i32>
17;CHECK: store <4 x i32>
18;CHECK: store <4 x i32>
19;CHECK: ret void
20define void @inc(i32 %n) nounwind uwtable noinline ssp {
21  %1 = icmp sgt i32 %n, 0
22  br i1 %1, label %.lr.ph, label %._crit_edge
23
24.lr.ph:                                           ; preds = %0, %.lr.ph
25  %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
26  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
27  %3 = load i32* %2, align 4
28  %4 = trunc i64 %indvars.iv to i32
29  %5 = add nsw i32 %3, %4
30  store i32 %5, i32* %2, align 4
31  %indvars.iv.next = add i64 %indvars.iv, 1
32  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
33  %exitcond = icmp eq i32 %lftr.wideiv, %n
34  br i1 %exitcond, label %._crit_edge, label %.lr.ph
35
36._crit_edge:                                      ; preds = %.lr.ph, %0
37  ret void
38}
39
40