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
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
|
/* Copyright (C) 2011 Free Software Foundation, Inc.
Contributed by Intel Corporation.
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>
/* fast SSE2 version with using pmaxub and 64 byte loop */
.text
ENTRY(memchr)
movd %rsi, %xmm1
mov %rdi, %rcx
punpcklbw %xmm1, %xmm1
test %rdx, %rdx
jz L(return_null)
punpcklbw %xmm1, %xmm1
and $63, %rcx
pshufd $0, %xmm1, %xmm1
cmp $48, %rcx
ja L(crosscache)
movdqu (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches_1)
sub $16, %rdx
jbe L(return_null)
add $16, %rdi
and $15, %rcx
and $-16, %rdi
add %rcx, %rdx
sub $64, %rdx
jbe L(exit_loop)
jmp L(loop_prolog)
.p2align 4
L(crosscache):
and $15, %rcx
and $-16, %rdi
movdqa (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
/* Check if there is a match. */
pmovmskb %xmm0, %eax
/* Remove the leading bytes. */
sar %cl, %eax
test %eax, %eax
je L(unaligned_no_match)
/* Check which byte is a match. */
bsf %eax, %eax
sub %rax, %rdx
jbe L(return_null)
add %rdi, %rax
add %rcx, %rax
ret
.p2align 4
L(unaligned_no_match):
add %rcx, %rdx
sub $16, %rdx
jbe L(return_null)
add $16, %rdi
sub $64, %rdx
jbe L(exit_loop)
.p2align 4
L(loop_prolog):
movdqa (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches)
movdqa 16(%rdi), %xmm2
pcmpeqb %xmm1, %xmm2
pmovmskb %xmm2, %eax
test %eax, %eax
jnz L(matches16)
movdqa 32(%rdi), %xmm3
pcmpeqb %xmm1, %xmm3
pmovmskb %xmm3, %eax
test %eax, %eax
jnz L(matches32)
movdqa 48(%rdi), %xmm4
pcmpeqb %xmm1, %xmm4
add $64, %rdi
pmovmskb %xmm4, %eax
test %eax, %eax
jnz L(matches0)
test $0x3f, %rdi
jz L(align64_loop)
sub $64, %rdx
jbe L(exit_loop)
movdqa (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches)
movdqa 16(%rdi), %xmm2
pcmpeqb %xmm1, %xmm2
pmovmskb %xmm2, %eax
test %eax, %eax
jnz L(matches16)
movdqa 32(%rdi), %xmm3
pcmpeqb %xmm1, %xmm3
pmovmskb %xmm3, %eax
test %eax, %eax
jnz L(matches32)
movdqa 48(%rdi), %xmm3
pcmpeqb %xmm1, %xmm3
pmovmskb %xmm3, %eax
add $64, %rdi
test %eax, %eax
jnz L(matches0)
mov %rdi, %rcx
and $-64, %rdi
and $63, %rcx
add %rcx, %rdx
.p2align 4
L(align64_loop):
sub $64, %rdx
jbe L(exit_loop)
movdqa (%rdi), %xmm0
movdqa 16(%rdi), %xmm2
movdqa 32(%rdi), %xmm3
movdqa 48(%rdi), %xmm4
pcmpeqb %xmm1, %xmm0
pcmpeqb %xmm1, %xmm2
pcmpeqb %xmm1, %xmm3
pcmpeqb %xmm1, %xmm4
pmaxub %xmm0, %xmm3
pmaxub %xmm2, %xmm4
pmaxub %xmm3, %xmm4
pmovmskb %xmm4, %eax
add $64, %rdi
test %eax, %eax
jz L(align64_loop)
sub $64, %rdi
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches)
pmovmskb %xmm2, %eax
test %eax, %eax
jnz L(matches16)
movdqa 32(%rdi), %xmm3
pcmpeqb %xmm1, %xmm3
pcmpeqb 48(%rdi), %xmm1
pmovmskb %xmm3, %eax
test %eax, %eax
jnz L(matches32)
pmovmskb %xmm1, %eax
bsf %eax, %eax
lea 48(%rdi, %rax), %rax
ret
.p2align 4
L(exit_loop):
add $32, %rdx
jle L(exit_loop_32)
movdqa (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches)
movdqa 16(%rdi), %xmm2
pcmpeqb %xmm1, %xmm2
pmovmskb %xmm2, %eax
test %eax, %eax
jnz L(matches16)
movdqa 32(%rdi), %xmm3
pcmpeqb %xmm1, %xmm3
pmovmskb %xmm3, %eax
test %eax, %eax
jnz L(matches32_1)
sub $16, %rdx
jle L(return_null)
pcmpeqb 48(%rdi), %xmm1
pmovmskb %xmm1, %eax
test %eax, %eax
jnz L(matches48_1)
xor %rax, %rax
ret
.p2align 4
L(exit_loop_32):
add $32, %rdx
movdqa (%rdi), %xmm0
pcmpeqb %xmm1, %xmm0
pmovmskb %xmm0, %eax
test %eax, %eax
jnz L(matches_1)
sub $16, %rdx
jbe L(return_null)
pcmpeqb 16(%rdi), %xmm1
pmovmskb %xmm1, %eax
test %eax, %eax
jnz L(matches16_1)
xor %rax, %rax
ret
.p2align 4
L(matches0):
bsf %eax, %eax
lea -16(%rax, %rdi), %rax
ret
.p2align 4
L(matches):
bsf %eax, %eax
add %rdi, %rax
ret
.p2align 4
L(matches16):
bsf %eax, %eax
lea 16(%rax, %rdi), %rax
ret
.p2align 4
L(matches32):
bsf %eax, %eax
lea 32(%rax, %rdi), %rax
ret
.p2align 4
L(matches_1):
bsf %eax, %eax
sub %rax, %rdx
jbe L(return_null)
add %rdi, %rax
ret
.p2align 4
L(matches16_1):
bsf %eax, %eax
sub %rax, %rdx
jbe L(return_null)
lea 16(%rdi, %rax), %rax
ret
.p2align 4
L(matches32_1):
bsf %eax, %eax
sub %rax, %rdx
jbe L(return_null)
lea 32(%rdi, %rax), %rax
ret
.p2align 4
L(matches48_1):
bsf %eax, %eax
sub %rax, %rdx
jbe L(return_null)
lea 48(%rdi, %rax), %rax
ret
.p2align 4
L(return_null):
xor %rax, %rax
ret
END(memchr)
strong_alias (memchr, __memchr)
libc_hidden_builtin_def(memchr)
|
