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
|
/***************************************************************************
* __________ __ ___.
* Open \______ \ ____ ____ | | _\_ |__ _______ ___
* Source | _// _ \_/ ___\| |/ /| __ \ / _ \ \/ /
* Jukebox | | ( <_> ) \___| < | \_\ ( <_> > < <
* Firmware |____|_ /\____/ \___ >__|_ \|___ /\____/__/\_ \
* \/ \/ \/ \/ \/
* $Id$
*
* Copyright (C) 2002 by Linus Nielsen Feltzing
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
* KIND, either express or implied.
*
****************************************************************************/
#include "config.h"
#include "cpu.h"
.section .init.text,"ax",@progbits
.global start
start:
mov.l .vbr_k,r1
#ifdef DEBUG
/* If we have built our code to be loaded via the standalone GDB
* stub, we will have out VBR at some other location than 0x9000000.
* We must copy the trap vectors for the GDB stub to our vector table. */
mov.l .orig_vbr_k,r2
/* Move the invalid instruction vector (4) */
mov #4,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the invalid slot vector (6) */
mov #6,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the bus error vector (9) */
mov #9,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the DMA bus error vector (10) */
mov #10,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the NMI vector as well (11) */
mov #11,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the UserBreak vector as well (12) */
mov #12,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the breakpoint trap vector (32) */
mov #32,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the IO trap vector (33) */
mov #33,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the serial Rx interrupt vector (105) */
mov #105,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
/* Move the single step trap vector (127) */
mov #127,r0
shll2 r0
mov.l @(r0,r2),r3
mov.l r3,@(r0,r1)
#endif /* DEBUG */
ldc r1,vbr
mov #0,r0
ldc r0,gbr
/* .iram copy is done first since it is reclaimed for other
* uninitialized sections */
/* copy the .iram section */
mov.l .iramcopy_k,r0
mov.l .iram_k,r1
mov.l .iramend_k,r2
/* Note: We cannot put a PC relative load into the delay slot of a 'bra'
instruction (the offset would be wrong), but there is nothing else to
do before the loop, so the delay slot would be 'nop'. The cmp / bf
sequence is the same length, but more efficient. */
cmp/hi r1,r2
bf .noiramcopy
.iramloop:
mov.l @r0+,r3
mov.l r3,@r1
add #4,r1
cmp/hi r1,r2
bt .iramloop
.noiramcopy:
/* zero out .ibss */
mov.l .iedata_k,r0
mov.l .iend_k,r1
bra .iedatastart
mov #0,r2
.iedataloop: /* backwards is faster and shorter */
mov.l r2,@-r1
.iedatastart:
cmp/hi r0,r1
bt .iedataloop
/* zero out bss */
mov.l .edata_k,r0
mov.l .end_k,r1
bra .edatastart
mov #0,r2
.edataloop: /* backwards is faster and shorter */
mov.l r2,@-r1
.edatastart:
cmp/hi r0,r1
bt .edataloop
/* copy the .data section, for rombased execution */
mov.l .datacopy_k,r0
mov.l .data_k,r1
cmp/eq r0,r1
bt .nodatacopy /* Don't copy if src and dest are equal */
mov.l .dataend_k,r2
cmp/hi r1,r2
bf .nodatacopy
.dataloop:
mov.l @r0+,r3
mov.l r3,@r1
add #4,r1
cmp/hi r1,r2
bt .dataloop
.nodatacopy:
/* Munge the main thread stack */
mov.l .stackbegin_k,r0
mov.l .stackend_k,r1
mov r1,r15
mov.l .deadbeef_k,r2
.mungeloop: /* backwards is faster and shorter */
mov.l r2,@-r1
cmp/hi r0,r1
bt .mungeloop
/* call the mainline */
mov.l .main_k,r0
jsr @r0
nop
.hoo:
bra .hoo
nop
.align 2
.vbr_k:
.long vectors
#ifdef DEBUG
.orig_vbr_k:
.long 0x09000000
#endif
.iedata_k:
.long _iedata
.iend_k:
.long _iend
.iramcopy_k:
.long _iramcopy
.iram_k:
.long _iramstart
.iramend_k:
.long _iramend
.edata_k:
.long _edata
.end_k:
.long _end
.datacopy_k:
.long _datacopy
.data_k:
.long _datastart
.dataend_k:
.long _dataend
.stackbegin_k:
.long _stackbegin
.stackend_k:
.long _stackend
.deadbeef_k:
.long 0xdeadbeef
.main_k:
.long _main
.section .resetvectors
vectors:
.long start
.long _stackend
.long start
.long _stackend
|
