1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
|
; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu -mcpu=pwr7 < %s | FileCheck %s
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-f128:128:128-v128:128:128-n32:64"
target triple = "powerpc64-unknown-linux-gnu"
; Function Attrs: nounwind
define void @foo(ptr noalias nocapture %x, ptr noalias nocapture readonly %y) #0 {
entry:
br label %vector.body
vector.body: ; preds = %vector.body, %entry
; CHECK-LABEL: @foo
; CHECK: lvsl
; CHECK: blr
%index = phi i64 [ 0, %entry ], [ %index.next.15, %vector.body ]
%0 = getelementptr inbounds float, ptr %y, i64 %index
%wide.load = load <4 x float>, ptr %0, align 4
%1 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load)
%2 = getelementptr inbounds float, ptr %x, i64 %index
store <4 x float> %1, ptr %2, align 4
%index.next = add i64 %index, 4
%3 = getelementptr inbounds float, ptr %y, i64 %index.next
%wide.load.1 = load <4 x float>, ptr %3, align 4
%4 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.1)
%5 = getelementptr inbounds float, ptr %x, i64 %index.next
store <4 x float> %4, ptr %5, align 4
%index.next.1 = add i64 %index.next, 4
%6 = getelementptr inbounds float, ptr %y, i64 %index.next.1
%wide.load.2 = load <4 x float>, ptr %6, align 4
%7 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.2)
%8 = getelementptr inbounds float, ptr %x, i64 %index.next.1
store <4 x float> %7, ptr %8, align 4
%index.next.2 = add i64 %index.next.1, 4
%9 = getelementptr inbounds float, ptr %y, i64 %index.next.2
%wide.load.3 = load <4 x float>, ptr %9, align 4
%10 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.3)
%11 = getelementptr inbounds float, ptr %x, i64 %index.next.2
store <4 x float> %10, ptr %11, align 4
%index.next.3 = add i64 %index.next.2, 4
%12 = getelementptr inbounds float, ptr %y, i64 %index.next.3
%wide.load.4 = load <4 x float>, ptr %12, align 4
%13 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.4)
%14 = getelementptr inbounds float, ptr %x, i64 %index.next.3
store <4 x float> %13, ptr %14, align 4
%index.next.4 = add i64 %index.next.3, 4
%15 = getelementptr inbounds float, ptr %y, i64 %index.next.4
%wide.load.5 = load <4 x float>, ptr %15, align 4
%16 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.5)
%17 = getelementptr inbounds float, ptr %x, i64 %index.next.4
store <4 x float> %16, ptr %17, align 4
%index.next.5 = add i64 %index.next.4, 4
%18 = getelementptr inbounds float, ptr %y, i64 %index.next.5
%wide.load.6 = load <4 x float>, ptr %18, align 4
%19 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.6)
%20 = getelementptr inbounds float, ptr %x, i64 %index.next.5
store <4 x float> %19, ptr %20, align 4
%index.next.6 = add i64 %index.next.5, 4
%21 = getelementptr inbounds float, ptr %y, i64 %index.next.6
%wide.load.7 = load <4 x float>, ptr %21, align 4
%22 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.7)
%23 = getelementptr inbounds float, ptr %x, i64 %index.next.6
store <4 x float> %22, ptr %23, align 4
%index.next.7 = add i64 %index.next.6, 4
%24 = getelementptr inbounds float, ptr %y, i64 %index.next.7
%wide.load.8 = load <4 x float>, ptr %24, align 4
%25 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.8)
%26 = getelementptr inbounds float, ptr %x, i64 %index.next.7
store <4 x float> %25, ptr %26, align 4
%index.next.8 = add i64 %index.next.7, 4
%27 = getelementptr inbounds float, ptr %y, i64 %index.next.8
%wide.load.9 = load <4 x float>, ptr %27, align 4
%28 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.9)
%29 = getelementptr inbounds float, ptr %x, i64 %index.next.8
store <4 x float> %28, ptr %29, align 4
%index.next.9 = add i64 %index.next.8, 4
%30 = getelementptr inbounds float, ptr %y, i64 %index.next.9
%wide.load.10 = load <4 x float>, ptr %30, align 4
%31 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.10)
%32 = getelementptr inbounds float, ptr %x, i64 %index.next.9
store <4 x float> %31, ptr %32, align 4
%index.next.10 = add i64 %index.next.9, 4
%33 = getelementptr inbounds float, ptr %y, i64 %index.next.10
%wide.load.11 = load <4 x float>, ptr %33, align 4
%34 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.11)
%35 = getelementptr inbounds float, ptr %x, i64 %index.next.10
store <4 x float> %34, ptr %35, align 4
%index.next.11 = add i64 %index.next.10, 4
%36 = getelementptr inbounds float, ptr %y, i64 %index.next.11
%wide.load.12 = load <4 x float>, ptr %36, align 4
%37 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.12)
%38 = getelementptr inbounds float, ptr %x, i64 %index.next.11
store <4 x float> %37, ptr %38, align 4
%index.next.12 = add i64 %index.next.11, 4
%39 = getelementptr inbounds float, ptr %y, i64 %index.next.12
%wide.load.13 = load <4 x float>, ptr %39, align 4
%40 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.13)
%41 = getelementptr inbounds float, ptr %x, i64 %index.next.12
store <4 x float> %40, ptr %41, align 4
%index.next.13 = add i64 %index.next.12, 4
%42 = getelementptr inbounds float, ptr %y, i64 %index.next.13
%wide.load.14 = load <4 x float>, ptr %42, align 4
%43 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.14)
%44 = getelementptr inbounds float, ptr %x, i64 %index.next.13
store <4 x float> %43, ptr %44, align 4
%index.next.14 = add i64 %index.next.13, 4
%45 = getelementptr inbounds float, ptr %y, i64 %index.next.14
%wide.load.15 = load <4 x float>, ptr %45, align 4
%46 = call <4 x float> @llvm_cos_v4f32(<4 x float> %wide.load.15)
%47 = getelementptr inbounds float, ptr %x, i64 %index.next.14
store <4 x float> %46, ptr %47, align 4
%index.next.15 = add i64 %index.next.14, 4
%48 = icmp eq i64 %index.next.15, 2048
br i1 %48, label %for.end, label %vector.body
for.end: ; preds = %vector.body
ret void
}
; Function Attrs: nounwind readonly
declare <4 x float> @llvm_cos_v4f32(<4 x float>) #1
define <2 x double> @bar(ptr %x) {
entry:
%r = load <2 x double>, ptr %x, align 8
; CHECK-LABEL: @bar
; CHECK-NOT: lvsl
; CHECK: blr
ret <2 x double> %r
}
attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }
|
