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
|
/*
* ====================================================
* Copyright (C) 1998, 2002, 2008 by Red Hat Inc. All rights reserved.
*
* Permission to use, copy, modify, and distribute this
* software is freely granted, provided that this notice
* is preserved.
* ====================================================
*/
#include "i386mach.h"
.global SYM (strchr)
SOTYPE_FUNCTION(strchr)
SYM (strchr):
#ifdef __iamcu__
xorl ecx,ecx
movb dl,cl
/* loop while (*s && *s++ != c) */
leal -1(eax),eax
L15:
incl eax
movb (eax),dl
testb dl,dl
je L14
cmpb cl,dl
jne L15
L14:
/* if (*s == c) return address otherwise return NULL */
cmpb cl,(eax)
je L19
xorl eax,eax
L19:
ret
#else
pushl ebp
movl esp,ebp
pushl edi
pushl ebx
xorl ebx,ebx
movl 8(ebp),edi
addb 12(ebp),bl
#ifndef __OPTIMIZE_SIZE__
/* Special case strchr(p,0). */
je L25
/* Do byte-wise checks until string is aligned. */
test $3,edi
je L5
movl edi,eax
movb (eax),cl
testb cl,cl
je L14
cmpb bl,cl
je L19
incl edi
test $3,edi
je L5
movl edi,eax
movb (eax),cl
testb cl,cl
je L14
cmpb bl,cl
je L19
incl edi
test $3,edi
je L5
movl edi,eax
movb (eax),cl
testb cl,cl
je L14
cmpb bl,cl
je L19
incl edi
/* create 4 byte mask which is just the desired byte repeated 4 times */
L5:
movl ebx,ecx
sall $8,ebx
subl $4,edi
orl ecx,ebx
movl ebx,edx
sall $16,ebx
orl edx,ebx
/* loop performing 4 byte mask checking for 0 byte or desired byte */
.p2align 4,,7
L10:
addl $4,edi
movl (edi),ecx
leal -16843009(ecx),edx
movl ecx,eax
notl eax
andl eax,edx
testl $-2139062144,edx
jne L9
xorl ebx,ecx
leal -16843009(ecx),edx
notl ecx
andl ecx,edx
testl $-2139062144,edx
je L10
#endif /* not __OPTIMIZE_SIZE__ */
/* loop while (*s && *s++ != c) */
L9:
leal -1(edi),eax
.p2align 4,,7
L15:
incl eax
movb (eax),dl
testb dl,dl
je L14
cmpb bl,dl
jne L15
L14:
/* if (*s == c) return address otherwise return NULL */
cmpb bl,(eax)
je L19
xorl eax,eax
L19:
leal -8(ebp),esp
popl ebx
popl edi
leave
ret
#ifndef __OPTIMIZE_SIZE__
/* Special case strchr(p,0). */
#if 0
/* Hideous performance on modern machines. */
L25:
cld
movl $-1,ecx
xor eax,eax
repnz
scasb
leal -1(edi),eax
jmp L19
#endif
L25:
/* Do byte-wise checks until string is aligned. */
test $3,edi
je L26
movl edi,eax
movb (eax),cl
testb cl,cl
je L19
incl edi
test $3,edi
je L26
movl edi,eax
movb (eax),cl
testb cl,cl
je L19
incl edi
test $3,edi
je L26
movl edi,eax
movb (eax),cl
testb cl,cl
je L19
incl edi
L26:
subl $4,edi
/* loop performing 4 byte mask checking for desired 0 byte */
.p2align 4,,7
L27:
addl $4,edi
movl (edi),ecx
leal -16843009(ecx),edx
movl ecx,eax
notl eax
andl eax,edx
testl $-2139062144,edx
je L27
jmp L9
#endif /* !__OPTIMIZE_SIZE__ */
#endif /* __iamcu__ */
|