1 ; RUN: opt %loadPolly -polly-detect-unprofitable -basicaa -polly-codegen-isl %vector-opt -dce -S < %s | FileCheck %s
7 ;void simple_vec_const(void) {
10 ; for (i = 0; i < 4; i++)
19 target 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"
20 target triple = "x86_64-unknown-linux-gnu"
22 @A = common global [1024 x float] zeroinitializer, align 16
23 @B = common global [1024 x float] zeroinitializer, align 16
25 define void @simple_vec_const() nounwind {
29 bb2: ; preds = %bb5, %bb
30 %indvar = phi i64 [ %indvar.next, %bb5 ], [ 0, %bb ]
31 %scevgep = getelementptr [1024 x float], [1024 x float]* @B, i64 0, i64 %indvar
32 %scevgep1 = getelementptr [1024 x float], [1024 x float]* @A, i64 0, i64 %indvar
33 %exitcond = icmp ne i64 %indvar, 4
34 br i1 %exitcond, label %bb3, label %bb6
37 %tmp = load float* %scevgep1, align 4
38 %tmp4 = fadd float %tmp, 1.000000e+00
39 store float %tmp4, float* %scevgep, align 4
43 %indvar.next = add i64 %indvar, 1
50 define i32 @main() nounwind {
52 call void @simple_vec_const()
53 %tmp = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 42), align 8
54 %tmp1 = fptosi float %tmp to i32
58 ; CHECK: %tmp_p_vec_full = load <4 x float>* bitcast ([1024 x float]* @A to <4 x float>*), align 8, !alias.scope !0, !noalias !2
59 ; CHECK: %tmp4p_vec = fadd <4 x float> %tmp_p_vec_full, <float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00>
60 ; CHECK: store <4 x float> %tmp4p_vec, <4 x float>* bitcast ([1024 x float]* @B to <4 x float>*), align 8, !alias.scope !3, !noalias !4