1; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
2; RUN: llc < %s -mtriple=x86_64-- | FileCheck %s
3
4; This tests whether or not we generate vectors large than preferred vector width when
5; lowering memmove.
6
7; Function Attrs: nounwind uwtable
8define weak_odr dso_local void @A(i8* %src, i8* %dst) local_unnamed_addr #0 {
9; CHECK-LABEL: A:
10; CHECK:       # %bb.0: # %entry
11; CHECK-NEXT:    vmovups (%rdi), %xmm0
12; CHECK-NEXT:    vmovups 16(%rdi), %xmm1
13; CHECK-NEXT:    vmovups %xmm1, 16(%rsi)
14; CHECK-NEXT:    vmovups %xmm0, (%rsi)
15; CHECK-NEXT:    retq
16entry:
17  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 1 %dst, i8* align 1 %src, i64 32, i1 false)
18  ret void
19}
20
21; Function Attrs: nounwind uwtable
22define weak_odr dso_local void @B(i8* %src, i8* %dst) local_unnamed_addr #0 {
23; CHECK-LABEL: B:
24; CHECK:       # %bb.0: # %entry
25; CHECK-NEXT:    vmovups (%rdi), %xmm0
26; CHECK-NEXT:    vmovups 16(%rdi), %xmm1
27; CHECK-NEXT:    vmovups 32(%rdi), %xmm2
28; CHECK-NEXT:    vmovups 48(%rdi), %xmm3
29; CHECK-NEXT:    vmovups %xmm3, 48(%rsi)
30; CHECK-NEXT:    vmovups %xmm2, 32(%rsi)
31; CHECK-NEXT:    vmovups %xmm1, 16(%rsi)
32; CHECK-NEXT:    vmovups %xmm0, (%rsi)
33; CHECK-NEXT:    retq
34entry:
35  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 1 %dst, i8* align 1 %src, i64 64, i1 false)
36  ret void
37}
38
39; Function Attrs: nounwind uwtable
40define weak_odr dso_local void @C(i8* %src, i8* %dst) local_unnamed_addr #2 {
41; CHECK-LABEL: C:
42; CHECK:       # %bb.0: # %entry
43; CHECK-NEXT:    vmovups (%rdi), %ymm0
44; CHECK-NEXT:    vmovups %ymm0, (%rsi)
45; CHECK-NEXT:    vzeroupper
46; CHECK-NEXT:    retq
47entry:
48  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 1 %dst, i8* align 1 %src, i64 32, i1 false)
49  ret void
50}
51
52; Function Attrs: nounwind uwtable
53define weak_odr dso_local void @D(i8* %src, i8* %dst) local_unnamed_addr #2 {
54; CHECK-LABEL: D:
55; CHECK:       # %bb.0: # %entry
56; CHECK-NEXT:    vmovups (%rdi), %ymm0
57; CHECK-NEXT:    vmovups 32(%rdi), %ymm1
58; CHECK-NEXT:    vmovups %ymm1, 32(%rsi)
59; CHECK-NEXT:    vmovups %ymm0, (%rsi)
60; CHECK-NEXT:    vzeroupper
61; CHECK-NEXT:    retq
62entry:
63  call void @llvm.memmove.p0i8.p0i8.i64(i8* align 1 %dst, i8* align 1 %src, i64 64, i1 false)
64  ret void
65}
66
67; Function Attrs: argmemonly nounwind
68declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i1 immarg) #1
69
70attributes #0 = { nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "frame-pointer"="none" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "prefer-vector-width"="128" "stack-protector-buffer-size"="8" "target-cpu"="skylake-avx512" "target-features"="+adx,+aes,+avx,+avx2,+avx512bw,+avx512cd,+avx512dq,+avx512f,+avx512vl,+bmi,+bmi2,+clflushopt,+clwb,+cx16,+cx8,+f16c,+fma,+fsgsbase,+fxsr,+invpcid,+lzcnt,+mmx,+movbe,+pclmul,+pku,+popcnt,+prfchw,+rdrnd,+rdseed,+sahf,+sse,+sse2,+sse3,+sse4.1,+sse4.2,+ssse3,+x87,+xsave,+xsavec,+xsaveopt,+xsaves" "unsafe-fp-math"="false" "use-soft-float"="false" }
71attributes #1 = { argmemonly nounwind }
72attributes #2 = { nounwind uwtable "correctly-rounded-divide-sqrt-fp-math"="false" "disable-tail-calls"="false" "less-precise-fpmad"="false" "min-legal-vector-width"="0" "frame-pointer"="none" "no-infs-fp-math"="false" "no-jump-tables"="false" "no-nans-fp-math"="false" "no-signed-zeros-fp-math"="false" "no-trapping-math"="false" "prefer-vector-width"="256" "stack-protector-buffer-size"="8" "target-cpu"="skylake-avx512" "target-features"="+adx,+aes,+avx,+avx2,+avx512bw,+avx512cd,+avx512dq,+avx512f,+avx512vl,+bmi,+bmi2,+clflushopt,+clwb,+cx16,+cx8,+f16c,+fma,+fsgsbase,+fxsr,+invpcid,+lzcnt,+mmx,+movbe,+pclmul,+pku,+popcnt,+prfchw,+rdrnd,+rdseed,+sahf,+sse,+sse2,+sse3,+sse4.1,+sse4.2,+ssse3,+x87,+xsave,+xsavec,+xsaveopt,+xsaves" "unsafe-fp-math"="false" "use-soft-float"="false" }
73
74!0 = !{i32 1, !"wchar_size", i32 4}
75