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
|
/* $OpenBSD: memmove.S,v 1.5 2016/11/14 15:02:54 visa Exp $ */
/*-
* Copyright (c) 1991, 1993
* The Regents of the University of California. All rights reserved.
*
* This code is derived from software contributed to Berkeley by
* Ralph Campbell.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
#include "DEFS.h"
/*
* memcpy(to, from, len)
always copy forward
*
* memmove(to, from, len), bcopy(from, to, len)
* both handle overlap
*/
LEAF(memcpy, 0)
.set noreorder
move v0, a0 # swap from and to
move a0, a1
move a1, v0
PTR_ADDU t0, a0, a2 # t0 = end of s1 region
sltu t1, a1, t0
sltu t2, a0, a1
j forward # do forward copy
slt t2, a2, 12 # check for small copy
ALEAF(memmove)
.set noreorder
move v0, a0 # swap from and to
move a0, a1
move a1, v0
ALEAF(bcopy)
.set noreorder
PTR_ADDU t0, a0, a2 # t0 = end of s1 region
sltu t1, a1, t0
sltu t2, a0, a1
and t1, t1, t2 # t1 = true if from < to < (from+len)
beq t1, zero, forward # non overlapping, do forward copy
slt t2, a2, 12 # check for small copy
ble a2, zero, 2f
PTR_ADDU t1, a1, a2 # t1 = end of to region
1:
lb v1, -1(t0) # copy bytes backwards,
PTR_SUBU t0, t0, 1 # doesnt happen often so do slow way
PTR_SUBU t1, t1, 1
bne t0, a0, 1b
sb v1, 0(t1)
2:
j ra
nop
forward:
#ifdef _STANDALONE
bne t2, zero, smallcpy # do a small bcopy
xor v1, a0, a1 # compare low two bits of addresses
and v1, v1, 3
PTR_SUBU a3, zero, a1 # compute # bytes to word align address
beq v1, zero, aligned # addresses can be word aligned
and a3, a3, 3
beq a3, zero, 1f
PTR_SUBU a2, a2, a3 # subtract from remaining count
LWHI v1, 0(a0) # get next 4 bytes (unaligned)
LWLO v1, 3(a0)
PTR_ADDU a0, a0, a3
SWHI v1, 0(a1) # store 1, 2, or 3 bytes to align a1
PTR_ADDU a1, a1, a3
1:
and v1, a2, 3 # compute number of words left
PTR_SUBU a3, a2, v1
move a2, v1
PTR_ADDU a3, a3, a0 # compute ending address
2:
LWHI v1, 0(a0) # copy words a0 unaligned, a1 aligned
LWLO v1, 3(a0)
PTR_ADDU a0, a0, 4
sw v1, 0(a1)
PTR_ADDU a1, a1, 4
bne a0, a3, 2b
nop # We have to do this mmu-bug.
b smallcpy
nop
aligned:
beq a3, zero, 1f
PTR_SUBU a2, a2, a3 # subtract from remaining count
LWHI v1, 0(a0) # copy 1, 2, or 3 bytes to align
PTR_ADDU a0, a0, a3
SWHI v1, 0(a1)
PTR_ADDU a1, a1, a3
1:
and v1, a2, 3 # compute number of whole words left
PTR_SUBU a3, a2, v1
move a2, v1
PTR_ADDU a3, a3, a0 # compute ending address
2:
lw v1, 0(a0) # copy words
PTR_ADDU a0, a0, 4
sw v1, 0(a1)
bne a0, a3, 2b
PTR_ADDU a1, a1, 4
#else
bne t2, zero, smallcpy # do a small bcopy
xor v1, a0, a1 # compare low three bits of addresses
and v1, v1, 7
PTR_SUBU a3, zero, a1 # compute # bytes to dword align address
beq v1, zero, aligned # addresses can be dword aligned
and a3, a3, 7
beq a3, zero, 1f
PTR_SUBU a2, a2, a3 # subtract from remaining count
LDHI v1, 0(a0) # get next 8 bytes (unaligned)
LDLO v1, 7(a0)
PTR_ADDU a0, a0, a3
SDHI v1, 0(a1) # store 1-7 bytes to align a1
PTR_ADDU a1, a1, a3
1:
and v1, a2, 7 # compute number of dwords left
PTR_SUBU a3, a2, v1
beq a3, zero, smallcpy
move a2, v1
PTR_ADDU a3, a3, a0 # compute ending address
2:
LDHI v1, 0(a0) # copy dwords a0 unaligned, a1 aligned
LDLO v1, 7(a0)
PTR_ADDU a0, a0, 8
sd v1, 0(a1)
PTR_ADDU a1, a1, 8
bne a0, a3, 2b
nop # We have to do this mmu-bug.
b smallcpy
nop
aligned:
beq a3, zero, 1f
PTR_SUBU a2, a2, a3 # subtract from remaining count
LDHI v1, 0(a0) # copy 1-7 bytes to align
PTR_ADDU a0, a0, a3
SDHI v1, 0(a1)
PTR_ADDU a1, a1, a3
1:
and v1, a2, 7 # compute number of whole dwords left
PTR_SUBU a3, a2, v1
beq a3, zero, smallcpy
move a2, v1
PTR_ADDU a3, a3, a0 # compute ending address
2:
ld v1, 0(a0) # copy dwords
PTR_ADDU a0, a0, 8
sd v1, 0(a1)
bne a0, a3, 2b
PTR_ADDU a1, a1, 8
#endif
smallcpy:
ble a2, zero, 2f
PTR_ADDU a3, a2, a0 # compute ending address
1:
lbu v1, 0(a0) # copy bytes
PTR_ADDU a0, a0, 1
sb v1, 0(a1)
bne a0, a3, 1b
PTR_ADDU a1, a1, 1 # MMU BUG ? can not do -1(a1) at 0x80000000!!
2:
j ra
nop
END(memcpy)
|
