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
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
|
/* Function sincosf vectorized in AVX ISA as wrapper to SSE4 ISA version.
Copyright (C) 2014-2016 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library; if not, see
<http://www.gnu.org/licenses/>. */
#include <sysdep.h>
#include "svml_s_wrapper_impl.h"
.text
ENTRY (_ZGVcN8vl4l4_sincosf)
WRAPPER_IMPL_AVX_fFF _ZGVbN4vl4l4_sincosf
END (_ZGVcN8vl4l4_sincosf)
/* AVX ISA version as wrapper to SSE ISA version (for vector
function declared with #pragma omp declare simd notinbranch). */
.macro WRAPPER_IMPL_AVX_fFF_vvv callee
#ifndef __ILP32__
pushq %rbp
movq %rsp, %rbp
andq $-32, %rsp
subq $224, %rsp
vmovups %ymm0, 64(%rsp)
lea (%rsp), %rdi
vmovdqu %xmm1, 96(%rdi)
vmovdqu %xmm2, 112(%rdi)
vmovdqu %xmm3, 128(%rdi)
vmovdqu %xmm4, 144(%rdi)
vmovdqu %xmm5, 160(%rdi)
lea 32(%rsp), %rsi
vmovdqu %xmm6, 144(%rsi)
vmovdqu %xmm7, 160(%rsi)
vzeroupper
call HIDDEN_JUMPTARGET(\callee)
vmovdqu 80(%rsp), %xmm0
lea 16(%rsp), %rdi
lea 48(%rsp), %rsi
call HIDDEN_JUMPTARGET(\callee)
movq 96(%rsp), %rdx
movq 104(%rsp), %rsi
movq 112(%rsp), %r8
movq 120(%rsp), %r10
movl (%rsp), %eax
movl 4(%rsp), %ecx
movl 8(%rsp), %edi
movl 12(%rsp), %r9d
movl %eax, (%rdx)
movl %ecx, (%rsi)
movq 128(%rsp), %rax
movq 136(%rsp), %rcx
movl %edi, (%r8)
movl %r9d, (%r10)
movq 144(%rsp), %rdi
movq 152(%rsp), %r9
movl 16(%rsp), %r11d
movl 20(%rsp), %edx
movl 24(%rsp), %esi
movl 28(%rsp), %r8d
movl %r11d, (%rax)
movl %edx, (%rcx)
movq 160(%rsp), %r11
movq 168(%rsp), %rdx
movl %esi, (%rdi)
movl %r8d, (%r9)
movq 176(%rsp), %rsi
movq 184(%rsp), %r8
movl 32(%rsp), %r10d
movl 36(%rsp), %eax
movl 40(%rsp), %ecx
movl 44(%rsp), %edi
movl %r10d, (%r11)
movl %eax, (%rdx)
movq 192(%rsp), %r10
movq 200(%rsp), %rax
movl %ecx, (%rsi)
movl %edi, (%r8)
movq 16(%rbp), %rcx
movq 24(%rbp), %rdi
movl 48(%rsp), %r9d
movl 52(%rsp), %r11d
movl 56(%rsp), %edx
movl 60(%rsp), %esi
movl %r9d, (%r10)
movl %r11d, (%rax)
movl %edx, (%rcx)
movl %esi, (%rdi)
movq %rbp, %rsp
popq %rbp
ret
#else
leal 8(%rsp), %r10d
.cfi_def_cfa 10, 0
andl $-32, %esp
pushq -8(%r10d)
pushq %rbp
.cfi_escape 0x10,0x6,0x2,0x76,0
movl %esp, %ebp
pushq %r12
leal -80(%rbp), %esi
pushq %r10
.cfi_escape 0xf,0x3,0x76,0x70,0x6
.cfi_escape 0x10,0xc,0x2,0x76,0x78
leal -112(%rbp), %edi
movq %rsi, %r12
pushq %rbx
.cfi_escape 0x10,0x3,0x2,0x76,0x68
movq %rdi, %rbx
subl $184, %esp
vmovaps %xmm1, -128(%ebp)
vmovaps %xmm2, -144(%ebp)
vmovaps %xmm3, -160(%ebp)
vmovaps %xmm4, -176(%ebp)
vmovaps %ymm0, -208(%ebp)
vzeroupper
call HIDDEN_JUMPTARGET(\callee)
leal 16(%r12), %esi
vmovups -192(%ebp), %xmm0
leal 16(%rbx), %edi
call HIDDEN_JUMPTARGET(\callee)
movq -128(%ebp), %rax
vmovss -112(%ebp), %xmm0
vmovdqa -128(%ebp), %xmm7
vmovdqa -144(%ebp), %xmm3
vmovss %xmm0, (%eax)
vmovss -108(%ebp), %xmm0
vpextrd $1, %xmm7, %eax
vmovss %xmm0, (%eax)
movq -120(%ebp), %rax
vmovss -104(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -100(%ebp), %xmm0
vpextrd $3, %xmm7, %eax
vmovdqa -160(%ebp), %xmm7
vmovss %xmm0, (%eax)
movq -144(%ebp), %rax
vmovss -96(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -92(%ebp), %xmm0
vpextrd $1, %xmm3, %eax
vmovss %xmm0, (%eax)
movq -136(%ebp), %rax
vmovss -88(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -84(%ebp), %xmm0
vpextrd $3, %xmm3, %eax
vmovss %xmm0, (%eax)
movq -160(%ebp), %rax
vmovss -80(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -76(%ebp), %xmm0
vpextrd $1, %xmm7, %eax
vmovss %xmm0, (%eax)
movq -152(%ebp), %rax
vmovss -72(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -68(%ebp), %xmm0
vpextrd $3, %xmm7, %eax
vmovss %xmm0, (%eax)
movq -176(%ebp), %rax
vmovss -64(%ebp), %xmm0
vmovdqa -176(%ebp), %xmm3
vmovss %xmm0, (%eax)
vmovss -60(%ebp), %xmm0
vpextrd $1, %xmm3, %eax
vmovss %xmm0, (%eax)
movq -168(%ebp), %rax
vmovss -56(%ebp), %xmm0
vmovss %xmm0, (%eax)
vmovss -52(%ebp), %xmm0
vpextrd $3, %xmm3, %eax
vmovss %xmm0, (%eax)
addl $184, %esp
popq %rbx
popq %r10
.cfi_def_cfa 10, 0
popq %r12
popq %rbp
leal -8(%r10), %esp
.cfi_def_cfa 7, 8
ret
#endif
.endm
ENTRY (_ZGVcN8vvv_sincosf)
WRAPPER_IMPL_AVX_fFF_vvv _ZGVbN4vl4l4_sincosf
END (_ZGVcN8vvv_sincosf)
|