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
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py UTC_ARGS: --no_x86_scrub_sp --version 5
; RUN: llc < %s -mtriple=i386-unknown-linux-gnu | FileCheck %s
define void @test_sincos_v4f32(<4 x float> %x, ptr noalias %out_sin, ptr noalias %out_cos) {
; CHECK-LABEL: test_sincos_v4f32:
; CHECK: # %bb.0:
; CHECK-NEXT: pushl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: pushl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: subl $52, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 64
; CHECK-NEXT: .cfi_offset %esi, -12
; CHECK-NEXT: .cfi_offset %edi, -8
; CHECK-NEXT: movl 84(%esp), %esi
; CHECK-NEXT: flds 76(%esp)
; CHECK-NEXT: fstps {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
; CHECK-NEXT: flds 64(%esp)
; CHECK-NEXT: fstps {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
; CHECK-NEXT: flds 72(%esp)
; CHECK-NEXT: fstps {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
; CHECK-NEXT: flds 68(%esp)
; CHECK-NEXT: movl 80(%esp), %edi
; CHECK-NEXT: leal 40(%esp), %eax
; CHECK-NEXT: movl %eax, 8(%esp)
; CHECK-NEXT: leal 4(%edi), %eax
; CHECK-NEXT: movl %eax, 4(%esp)
; CHECK-NEXT: fstps (%esp)
; CHECK-NEXT: calll sincosf
; CHECK-NEXT: leal 44(%esp), %eax
; CHECK-NEXT: movl %eax, 8(%esp)
; CHECK-NEXT: leal 8(%edi), %eax
; CHECK-NEXT: movl %eax, 4(%esp)
; CHECK-NEXT: flds {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
; CHECK-NEXT: fstps (%esp)
; CHECK-NEXT: calll sincosf
; CHECK-NEXT: leal 36(%esp), %eax
; CHECK-NEXT: movl %eax, 8(%esp)
; CHECK-NEXT: movl %edi, 4(%esp)
; CHECK-NEXT: flds {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
; CHECK-NEXT: fstps (%esp)
; CHECK-NEXT: calll sincosf
; CHECK-NEXT: leal 48(%esp), %eax
; CHECK-NEXT: movl %eax, 8(%esp)
; CHECK-NEXT: addl $12, %edi
; CHECK-NEXT: movl %edi, 4(%esp)
; CHECK-NEXT: flds {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
; CHECK-NEXT: fstps (%esp)
; CHECK-NEXT: calll sincosf
; CHECK-NEXT: flds 36(%esp)
; CHECK-NEXT: flds 40(%esp)
; CHECK-NEXT: flds 44(%esp)
; CHECK-NEXT: flds 48(%esp)
; CHECK-NEXT: fstps 12(%esi)
; CHECK-NEXT: fstps 8(%esi)
; CHECK-NEXT: fstps 4(%esi)
; CHECK-NEXT: fstps (%esi)
; CHECK-NEXT: addl $52, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: popl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: popl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 4
; CHECK-NEXT: retl
%result = call { <4 x float>, <4 x float> } @llvm.sincos.v4f32(<4 x float> %x)
%result.0 = extractvalue { <4 x float>, <4 x float> } %result, 0
%result.1 = extractvalue { <4 x float>, <4 x float> } %result, 1
store <4 x float> %result.0, ptr %out_sin, align 4
store <4 x float> %result.1, ptr %out_cos, align 4
ret void
}
define void @test_sincos_v2f64(<2 x double> %x, ptr noalias %out_sin, ptr noalias %out_cos) {
; CHECK-LABEL: test_sincos_v2f64:
; CHECK: # %bb.0:
; CHECK-NEXT: pushl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: pushl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: subl $52, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 64
; CHECK-NEXT: .cfi_offset %esi, -12
; CHECK-NEXT: .cfi_offset %edi, -8
; CHECK-NEXT: movl 84(%esp), %esi
; CHECK-NEXT: fldl 72(%esp)
; CHECK-NEXT: fstpl {{[-0-9]+}}(%e{{[sb]}}p) # 8-byte Folded Spill
; CHECK-NEXT: fldl 64(%esp)
; CHECK-NEXT: movl 80(%esp), %edi
; CHECK-NEXT: leal 24(%esp), %eax
; CHECK-NEXT: movl %eax, 12(%esp)
; CHECK-NEXT: movl %edi, 8(%esp)
; CHECK-NEXT: fstpl (%esp)
; CHECK-NEXT: calll sincos
; CHECK-NEXT: leal 32(%esp), %eax
; CHECK-NEXT: movl %eax, 12(%esp)
; CHECK-NEXT: addl $8, %edi
; CHECK-NEXT: movl %edi, 8(%esp)
; CHECK-NEXT: fldl {{[-0-9]+}}(%e{{[sb]}}p) # 8-byte Folded Reload
; CHECK-NEXT: fstpl (%esp)
; CHECK-NEXT: calll sincos
; CHECK-NEXT: fldl 24(%esp)
; CHECK-NEXT: fldl 32(%esp)
; CHECK-NEXT: fstpl 8(%esi)
; CHECK-NEXT: fstpl (%esi)
; CHECK-NEXT: addl $52, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: popl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: popl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 4
; CHECK-NEXT: retl
%result = call { <2 x double>, <2 x double> } @llvm.sincos.v2f64(<2 x double> %x)
%result.0 = extractvalue { <2 x double>, <2 x double> } %result, 0
%result.1 = extractvalue { <2 x double>, <2 x double> } %result, 1
store <2 x double> %result.0, ptr %out_sin, align 8
store <2 x double> %result.1, ptr %out_cos, align 8
ret void
}
declare void @foo(ptr, ptr)
define void @can_fold_with_call_in_chain(float %x, ptr noalias %a, ptr noalias %b) {
; CHECK-LABEL: can_fold_with_call_in_chain:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: pushl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: pushl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: subl $20, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 32
; CHECK-NEXT: .cfi_offset %esi, -12
; CHECK-NEXT: .cfi_offset %edi, -8
; CHECK-NEXT: flds 32(%esp)
; CHECK-NEXT: fstps {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Spill
; CHECK-NEXT: movl 36(%esp), %edi
; CHECK-NEXT: movl 40(%esp), %esi
; CHECK-NEXT: movl %esi, 4(%esp)
; CHECK-NEXT: movl %edi, (%esp)
; CHECK-NEXT: calll foo@PLT
; CHECK-NEXT: leal 16(%esp), %eax
; CHECK-NEXT: movl %eax, 8(%esp)
; CHECK-NEXT: movl %edi, 4(%esp)
; CHECK-NEXT: flds {{[-0-9]+}}(%e{{[sb]}}p) # 4-byte Folded Reload
; CHECK-NEXT: fstps (%esp)
; CHECK-NEXT: calll sincosf
; CHECK-NEXT: flds 16(%esp)
; CHECK-NEXT: fstps (%esi)
; CHECK-NEXT: addl $20, %esp
; CHECK-NEXT: .cfi_def_cfa_offset 12
; CHECK-NEXT: popl %esi
; CHECK-NEXT: .cfi_def_cfa_offset 8
; CHECK-NEXT: popl %edi
; CHECK-NEXT: .cfi_def_cfa_offset 4
; CHECK-NEXT: retl
entry:
%sin = tail call float @llvm.sin.f32(float %x)
%cos = tail call float @llvm.cos.f32(float %x)
call void @foo(ptr %a, ptr %b)
store float %sin, ptr %a, align 4
store float %cos, ptr %b, align 4
ret void
}
|
