2 ; RUN: opt < %s -loop-vectorize -debug -disable-output -force-ordered-reductions=true -hints-allow-reordering=false \
3 ; RUN: -force-vector-width=4 -force-vector-interleave=1 -S 2>&1 | FileCheck %s --check-prefix=CHECK-VF4
4 ; RUN: opt < %s -loop-vectorize -debug -disable-output -force-ordered-reductions=true -hints-allow-reordering=false \
5 ; RUN: -force-vector-width=8 -force-vector-interleave=1 -S 2>&1 | FileCheck %s --check-prefix=CHECK-VF8
7 target triple="aarch64-unknown-linux-gnu"
9 ; CHECK-VF4: Found an estimated cost of 21 for VF 4 For instruction: %add = fadd float %0, %sum.07
10 ; CHECK-VF8: Found an estimated cost of 42 for VF 8 For instruction: %add = fadd float %0, %sum.07
12 define float @fadd_strict32(float* noalias nocapture readonly %a, i64 %n) {
17 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
18 %sum.07 = phi float [ 0.000000e+00, %entry ], [ %add, %for.body ]
19 %arrayidx = getelementptr inbounds float, float* %a, i64 %iv
20 %0 = load float, float* %arrayidx, align 4
21 %add = fadd float %0, %sum.07
22 %iv.next = add nuw nsw i64 %iv, 1
23 %exitcond.not = icmp eq i64 %iv.next, %n
24 br i1 %exitcond.not, label %for.end, label %for.body
31 ; CHECK-VF4: Found an estimated cost of 18 for VF 4 For instruction: %add = fadd double %0, %sum.07
32 ; CHECK-VF8: Found an estimated cost of 36 for VF 8 For instruction: %add = fadd double %0, %sum.07
34 define double @fadd_strict64(double* noalias nocapture readonly %a, i64 %n) {
39 %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
40 %sum.07 = phi double [ 0.000000e+00, %entry ], [ %add, %for.body ]
41 %arrayidx = getelementptr inbounds double, double* %a, i64 %iv
42 %0 = load double, double* %arrayidx, align 4
43 %add = fadd double %0, %sum.07
44 %iv.next = add nuw nsw i64 %iv, 1
45 %exitcond.not = icmp eq i64 %iv.next, %n
46 br i1 %exitcond.not, label %for.end, label %for.body