1 ; RUN: opt -mcpu=kryo -mtriple=aarch64-gnu-linux -passes=loop-data-prefetch -max-prefetch-iters-ahead=1000 -S < %s | FileCheck %s --check-prefix=LARGE_PREFETCH --check-prefix=ALL
2 ; RUN: opt -mcpu=kryo -mtriple=aarch64-gnu-linux -passes=loop-data-prefetch -S < %s | FileCheck %s --check-prefix=NO_LARGE_PREFETCH --check-prefix=ALL
4 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-n32:64-S128"
6 ; ALL-LABEL: @small_stride(
7 define void @small_stride(ptr nocapture %a, ptr nocapture readonly %b) {
12 for.body: ; preds = %for.body, %entry
13 %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
14 %arrayidx = getelementptr inbounds double, ptr %b, i64 %indvars.iv
15 ; ALL-NOT: call void @llvm.prefetch
16 %0 = load double, ptr %arrayidx, align 8
17 %add = fadd double %0, 1.000000e+00
18 %arrayidx2 = getelementptr inbounds double, ptr %a, i64 %indvars.iv
19 store double %add, ptr %arrayidx2, align 8
20 %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
21 %exitcond = icmp eq i64 %indvars.iv.next, 1600
22 br i1 %exitcond, label %for.end, label %for.body
25 for.end: ; preds = %for.body
29 ; ALL-LABEL: @large_stride(
30 define void @large_stride(ptr nocapture %a, ptr nocapture readonly %b) {
35 for.body: ; preds = %for.body, %entry
36 %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
37 %arrayidx = getelementptr inbounds double, ptr %b, i64 %indvars.iv
38 ; LARGE_PREFETCH: call void @llvm.prefetch
39 ; NO_LARGE_PREFETCH-NOT: call void @llvm.prefetch
40 %0 = load double, ptr %arrayidx, align 8
41 %add = fadd double %0, 1.000000e+00
42 %arrayidx2 = getelementptr inbounds double, ptr %a, i64 %indvars.iv
43 store double %add, ptr %arrayidx2, align 8
44 %indvars.iv.next = add nuw nsw i64 %indvars.iv, 150
45 %exitcond = icmp eq i64 %indvars.iv.next, 160000
46 br i1 %exitcond, label %for.end, label %for.body
49 for.end: ; preds = %for.body