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
|
/*
* CDDL HEADER START
*
* The contents of this file are subject to the terms of the
* Common Development and Distribution License (the "License").
* You may not use this file except in compliance with the License.
*
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
* or http://www.opensolaris.org/os/licensing.
* See the License for the specific language governing permissions
* and limitations under the License.
*
* When distributing Covered Code, include this CDDL HEADER in each
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
* If applicable, add the following below this CDDL HEADER, with the
* fields enclosed by brackets "[]" replaced with your own identifying
* information: Portions Copyright [yyyy] [name of copyright owner]
*
* CDDL HEADER END
*/
/*
* Copyright 2008 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*/
.file "memcmp.s"
/*
* memcmp(s1, s2, len)
*
* Compare n bytes: s1>s2: >0 s1==s2: 0 s1<s2: <0
*
* Fast assembler language version of the following C-program for memcmp
* which represents the `standard' for the C-library.
*
* int
* memcmp(const void *s1, const void *s2, size_t n)
* {
* if (s1 != s2 && n != 0) {
* const char *ps1 = s1;
* const char *ps2 = s2;
* do {
* if (*ps1++ != *ps2++)
* return (ps1[-1] - ps2[-1]);
* } while (--n != 0);
* }
* return (0);
* }
*/
#include <sys/asm_linkage.h>
ANSI_PRAGMA_WEAK(memcmp,function)
ENTRY(memcmp)
cmp %o0, %o1 ! s1 == s2?
be,pn %xcc, .cmpeq
cmp %o2, 17
bleu,a,pn %xcc, .cmpbyt ! for small counts go do bytes
sub %o1, %o0, %o1
andcc %o0, 3, %o3 ! is s1 aligned?
bz,a,pn %icc, .iss2 ! if so go check s2
andcc %o1, 3, %o4 ! is s2 aligned?
cmp %o3, 2
be,pn %icc, .algn2
cmp %o3, 3
.algn1: ldub [%o0], %o4 ! cmp one byte
inc %o0
ldub [%o1], %o5
inc %o1
dec %o2
be,pn %icc, .algn3
cmp %o4, %o5
be,pt %icc, .algn2
nop
b,a .noteq
.algn2: lduh [%o0], %o4
inc 2, %o0
ldub [%o1], %o5
inc 1, %o1
srl %o4, 8, %o3
cmp %o3, %o5
be,a,pt %icc, 1f
ldub [%o1], %o5 ! delay slot, get next byte from s2
b .noteq
mov %o3, %o4 ! delay slot, move *s1 to %o4
1: inc %o1
dec 2, %o2
and %o4, 0xff, %o4
cmp %o4, %o5
.algn3: be,a,pt %icc, .iss2
andcc %o1, 3, %o4 ! delay slot, is s2 aligned?
b,a .noteq
.cmpbyt:b .bytcmp
deccc %o2
1: ldub [%o0 + %o1], %o5 ! byte compare loop
inc %o0
cmp %o4, %o5
be,a,pt %icc, .bytcmp
deccc %o2 ! delay slot, compare count (len)
b,a .noteq
.bytcmp:bgeu,a,pt %xcc, 1b
ldub [%o0], %o4
.cmpeq:
retl ! strings compare equal
clr %o0
.noteq_word: ! words aren't equal. find unequal byte
srl %o4, 24, %o1 ! first byte
srl %o5, 24, %o2
cmp %o1, %o2
bne,pn %icc, 1f
sll %o4, 8, %o4
sll %o5, 8, %o5
srl %o4, 24, %o1
srl %o5, 24, %o2
cmp %o1, %o2
bne,pn %icc, 1f
sll %o4, 8, %o4
sll %o5, 8, %o5
srl %o4, 24, %o1
srl %o5, 24, %o2
cmp %o1, %o2
bne,pn %icc, 1f
sll %o4, 8, %o4
sll %o5, 8, %o5
srl %o4, 24, %o1
srl %o5, 24, %o2
1:
retl
sub %o1, %o2, %o0 ! delay slot
.noteq:
retl ! strings aren't equal
sub %o4, %o5, %o0 ! delay slot, return(*s1 - *s2)
.iss2: andn %o2, 3, %o3 ! count of aligned bytes
and %o2, 3, %o2 ! remaining bytes
bz,pn %icc, .w4cmp ! if s2 word aligned, compare words
cmp %o4, 2
be,pn %icc, .w2cmp ! s2 half aligned
cmp %o4, 1
.w3cmp:
dec 4, %o3 ! avoid reading beyond the last byte
inc 4, %o2
ldub [%o1], %g1 ! read a byte to align for word reads
inc 1, %o1
be,pt %icc, .w1cmp ! aligned to 1 or 3 bytes
sll %g1, 24, %o5
sub %o1, %o0, %o1
2: lduw [%o0 + %o1], %g1
lduw [%o0], %o4
inc 4, %o0
srl %g1, 8, %g5 ! merge with the other half
or %g5, %o5, %o5
cmp %o4, %o5
bne,pt %icc, .noteq_word
deccc 4, %o3
bnz,pt %xcc, 2b
sll %g1, 24, %o5
sub %o1, 1, %o1 ! used 3 bytes of the last word read
b .bytcmp
deccc %o2
.w1cmp:
dec 4, %o3 ! avoid reading beyond the last byte
inc 4, %o2
lduh [%o1], %g1 ! read 3 bytes to word align
inc 2, %o1
sll %g1, 8, %g5
or %o5, %g5, %o5
sub %o1, %o0, %o1
3: lduw [%o0 + %o1], %g1
lduw [%o0], %o4
inc 4, %o0
srl %g1, 24, %g5 ! merge with the other half
or %g5, %o5, %o5
cmp %o4, %o5
bne,pt %icc, .noteq_word
deccc 4, %o3
bnz,pt %xcc, 3b
sll %g1, 8, %o5
sub %o1, 3, %o1 ! used 1 byte of the last word read
b .bytcmp
deccc %o2
.w2cmp:
dec 4, %o3 ! avoid reading beyond the last byte
inc 4, %o2
lduh [%o1], %g1 ! read a halfword to align s2
inc 2, %o1
sll %g1, 16, %o5
sub %o1, %o0, %o1
4: lduw [%o0 + %o1], %g1 ! read a word from s2
lduw [%o0], %o4 ! read a word from s1
inc 4, %o0
srl %g1, 16, %g5 ! merge with the other half
or %g5, %o5, %o5
cmp %o4, %o5
bne,pn %icc, .noteq_word
deccc 4, %o3
bnz,pt %xcc, 4b
sll %g1, 16, %o5
sub %o1, 2, %o1 ! only used half of the last read word
b .bytcmp
deccc %o2
.w4cmp:
sub %o1, %o0, %o1
lduw [%o0 + %o1], %o5
5: lduw [%o0], %o4
inc 4, %o0
cmp %o4, %o5
bne,pt %icc, .noteq_word
deccc 4, %o3
bnz,a,pt %xcc, 5b
lduw [%o0 + %o1], %o5
b .bytcmp ! compare remaining bytes, if any
deccc %o2
SET_SIZE(memcmp)
|
