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
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
|
/* $NetBSD: cpufunc_asm_xscale.S,v 1.16 2002/08/17 16:36:32 thorpej Exp $ */
/*
* Copyright (c) 2001, 2002 Wasabi Systems, Inc.
* All rights reserved.
*
* Written by Allen Briggs and Jason R. Thorpe for Wasabi Systems, Inc.
*
* 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. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed for the NetBSD Project by
* Wasabi Systems, Inc.
* 4. The name of Wasabi Systems, Inc. may not be used to endorse
* or promote products derived from this software without specific prior
* written permission.
*
* THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``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 WASABI SYSTEMS, INC
* 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.
*/
/*
* Copyright (c) 2001 Matt Thomas.
* Copyright (c) 1997,1998 Mark Brinicombe.
* Copyright (c) 1997 Causality Limited
* All rights reserved.
*
* 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. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This product includes software developed by Causality Limited.
* 4. The name of Causality Limited may not be used to endorse or promote
* products derived from this software without specific prior written
* permission.
*
* THIS SOFTWARE IS PROVIDED BY CAUSALITY LIMITED ``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 CAUSALITY LIMITED 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.
*
* XScale assembly functions for CPU / MMU / TLB specific operations
*/
#include <machine/cpu.h>
#include <machine/asm.h>
/*
* Size of the XScale core D-cache.
*/
#define DCACHE_SIZE 0x00008000
#ifndef CACHE_CLEAN_BLOCK_INTR
.Lblock_userspace_access:
.word _C_LABEL(block_userspace_access)
#endif
/*
* CPWAIT -- Canonical method to wait for CP15 update.
* From: Intel 80200 manual, section 2.3.3.
*
* NOTE: Clobbers the specified temp reg.
*/
#define CPWAIT_BRANCH \
sub pc, pc, #4
#define CPWAIT(tmp) \
mrc p15, 0, tmp, c2, c0, 0 /* arbitrary read of CP15 */ ;\
mov tmp, tmp /* wait for it to complete */ ;\
CPWAIT_BRANCH /* branch to next insn */
#define CPWAIT_AND_RETURN_SHIFTER lsr #32
#define CPWAIT_AND_RETURN(tmp) \
mrc p15, 0, tmp, c2, c0, 0 /* arbitrary read of CP15 */ ;\
/* Wait for it to complete and branch to the return address */ \
sub pc, lr, tmp, CPWAIT_AND_RETURN_SHIFTER
ENTRY(xscale_cpwait)
CPWAIT_AND_RETURN(r0)
/*
* We need a separate cpu_control() entry point, since we have to
* invalidate the Branch Target Buffer in the event the BPRD bit
* changes in the control register.
*/
ENTRY(xscale_control)
mrc p15, 0, r3, c1, c0, 0 /* Read the control register */
bic r2, r3, r0 /* Clear bits */
eor r2, r2, r1 /* XOR bits */
teq r2, r3 /* Only write if there was a change */
mcrne p15, 0, r0, c7, c5, 6 /* Invalidate the BTB */
mcrne p15, 0, r2, c1, c0, 0 /* Write new control register */
mov r0, r3 /* Return old value */
CPWAIT_AND_RETURN(r1)
/*
* Functions to set the MMU Translation Table Base register
*
* We need to clean and flush the cache as it uses virtual
* addresses that are about to change.
*/
ENTRY(xscale_setttb)
#ifdef CACHE_CLEAN_BLOCK_INTR
mrs r3, cpsr_all
orr r1, r3, #(I32_bit | F32_bit)
msr cpsr_all, r1
#else
ldr r3, .Lblock_userspace_access
ldr r2, [r3]
orr r1, r2, #1
str r1, [r3]
#endif
stmfd sp!, {r0-r3, lr}
bl _C_LABEL(xscale_cache_cleanID)
mcr p15, 0, r0, c7, c5, 0 /* invalidate I$ and BTB */
mcr p15, 0, r0, c7, c10, 4 /* drain write and fill buffer */
CPWAIT(r0)
ldmfd sp!, {r0-r3, lr}
/* Write the TTB */
mcr p15, 0, r0, c2, c0, 0
/* If we have updated the TTB we must flush the TLB */
mcr p15, 0, r0, c8, c7, 0 /* invalidate I+D TLB */
/* The cleanID above means we only need to flush the I cache here */
mcr p15, 0, r0, c7, c5, 0 /* invalidate I$ and BTB */
CPWAIT(r0)
#ifdef CACHE_CLEAN_BLOCK_INTR
msr cpsr_all, r3
#else
str r2, [r3]
#endif
mov pc, lr
/*
* TLB functions
*
* Note: We don't need to worry about issuing a CPWAIT after
* TLB operations, because we expect a pmap_update() to follow.
*/
ENTRY(xscale_tlb_flushID_SE)
mcr p15, 0, r0, c8, c6, 1 /* flush D tlb single entry */
mcr p15, 0, r0, c8, c5, 1 /* flush I tlb single entry */
mov pc, lr
/*
* Cache functions
*/
ENTRY(xscale_cache_flushID)
mcr p15, 0, r0, c7, c7, 0 /* flush I+D cache */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_flushI)
mcr p15, 0, r0, c7, c5, 0 /* flush I cache */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_flushD)
mcr p15, 0, r0, c7, c6, 0 /* flush D cache */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_flushI_SE)
mcr p15, 0, r0, c7, c5, 1 /* flush I cache single entry */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_flushD_SE)
/*
* Errata (rev < 2): Must clean-dcache-line to an address
* before invalidate-dcache-line to an address, or dirty
* bits will not be cleared in the dcache array.
*/
mcr p15, 0, r0, c7, c10, 1
mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_cleanD_E)
mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
CPWAIT_AND_RETURN(r0)
/*
* Information for the XScale cache clean/purge functions:
*
* * Virtual address of the memory region to use
* * Size of memory region
*
* Note the virtual address for the Data cache clean operation
* does not need to be backed by physical memory, since no loads
* will actually be performed by the allocate-line operation.
*
* Note that the Mini-Data cache MUST be cleaned by executing
* loads from memory mapped into a region reserved exclusively
* for cleaning of the Mini-Data cache.
*/
.data
.global _C_LABEL(xscale_cache_clean_addr)
_C_LABEL(xscale_cache_clean_addr):
.word 0x00000000
.global _C_LABEL(xscale_cache_clean_size)
_C_LABEL(xscale_cache_clean_size):
.word DCACHE_SIZE
# .global _C_LABEL(xscale_minidata_clean_addr)
#_C_LABEL(xscale_minidata_clean_addr):
# .word 0x00000000
# .global _C_LABEL(xscale_minidata_clean_size)
#_C_LABEL(xscale_minidata_clean_size):
# .word 0x00000800
.text
.Lxscale_cache_clean_addr:
.word _C_LABEL(xscale_cache_clean_addr)
.Lxscale_cache_clean_size:
.word _C_LABEL(xscale_cache_clean_size)
.Lxscale_minidata_clean_addr:
.word _C_LABEL(xscale_minidata_clean_addr)
.Lxscale_minidata_clean_size:
.word _C_LABEL(xscale_minidata_clean_size)
#ifdef CACHE_CLEAN_BLOCK_INTR
#define XSCALE_CACHE_CLEAN_BLOCK \
mrs r3, cpsr_all ; \
orr r0, r3, #(I32_bit | F32_bit) ; \
msr cpsr_all, r0
#define XSCALE_CACHE_CLEAN_UNBLOCK \
msr cpsr_all, r3
#else
#define XSCALE_CACHE_CLEAN_BLOCK \
ldr r3, .Lblock_userspace_access ; \
ldr ip, [r3] ; \
orr r0, ip, #1 ; \
str r0, [r3]
#define XSCALE_CACHE_CLEAN_UNBLOCK \
str ip, [r3]
#endif /* CACHE_CLEAN_BLOCK_INTR */
#define XSCALE_CACHE_CLEAN_PROLOGUE \
XSCALE_CACHE_CLEAN_BLOCK ; \
ldr r2, .Lxscale_cache_clean_addr ; \
ldmia r2, {r0, r1} ; \
/* \
* BUG ALERT! \
* \
* The XScale core has a strange cache eviction bug, which \
* requires us to use 2x the cache size for the cache clean \
* and for that area to be aligned to 2 * cache size. \
* \
* The work-around is to use 2 areas for cache clean, and to \
* alternate between them whenever this is done. No one knows \
* why the work-around works (mmm!). \
*/ \
eor r0, r0, #(DCACHE_SIZE) ; \
str r0, [r2] ; \
add r0, r0, r1
#define XSCALE_CACHE_CLEAN_EPILOGUE \
XSCALE_CACHE_CLEAN_UNBLOCK
ENTRY_NP(xscale_cache_syncI)
ENTRY_NP(xscale_cache_purgeID)
mcr p15, 0, r0, c7, c5, 0 /* flush I cache (D cleaned below) */
ENTRY_NP(xscale_cache_cleanID)
ENTRY_NP(xscale_cache_purgeD)
ENTRY(xscale_cache_cleanD)
XSCALE_CACHE_CLEAN_PROLOGUE
1: subs r0, r0, #32
mcr p15, 0, r0, c7, c2, 5 /* allocate cache line */
subs r1, r1, #32
bne 1b
CPWAIT(r0)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT(r0)
XSCALE_CACHE_CLEAN_EPILOGUE
mov pc, lr
/*
* Clean the mini-data cache.
*
* It's expected that we only use the mini-data cache for
* kernel addresses, so there is no need to purge it on
* context switch, and no need to prevent userspace access
* while we clean it.
*/
ENTRY(xscale_cache_clean_minidata)
ldr r2, .Lxscale_minidata_clean_addr
ldmia r2, {r0, r1}
1: ldr r3, [r0], #32
subs r1, r1, #32
bne 1b
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r1)
ENTRY(xscale_cache_purgeID_E)
mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
CPWAIT(r1)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
mcr p15, 0, r0, c7, c5, 1 /* flush I cache single entry */
mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
CPWAIT_AND_RETURN(r1)
ENTRY(xscale_cache_purgeD_E)
mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
CPWAIT(r1)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
CPWAIT_AND_RETURN(r1)
/*
* Soft functions
*/
/* xscale_cache_syncI is identical to xscale_cache_purgeID */
ENTRY(xscale_cache_cleanID_rng)
ENTRY(xscale_cache_cleanD_rng)
cmp r1, #0x4000
bcs _C_LABEL(xscale_cache_cleanID)
and r2, r0, #0x1f
add r1, r1, r2
bic r0, r0, #0x1f
1: mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
add r0, r0, #32
subs r1, r1, #32
bhi 1b
CPWAIT(r0)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_purgeID_rng)
cmp r1, #0x4000
bcs _C_LABEL(xscale_cache_purgeID)
and r2, r0, #0x1f
add r1, r1, r2
bic r0, r0, #0x1f
1: mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
mcr p15, 0, r0, c7, c5, 1 /* flush I cache single entry */
add r0, r0, #32
subs r1, r1, #32
bhi 1b
CPWAIT(r0)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_purgeD_rng)
cmp r1, #0x4000
bcs _C_LABEL(xscale_cache_purgeD)
and r2, r0, #0x1f
add r1, r1, r2
bic r0, r0, #0x1f
1: mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
add r0, r0, #32
subs r1, r1, #32
bhi 1b
CPWAIT(r0)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_syncI_rng)
cmp r1, #0x4000
bcs _C_LABEL(xscale_cache_syncI)
and r2, r0, #0x1f
add r1, r1, r2
bic r0, r0, #0x1f
1: mcr p15, 0, r0, c7, c10, 1 /* clean D cache entry */
mcr p15, 0, r0, c7, c5, 1 /* flush I cache single entry */
add r0, r0, #32
subs r1, r1, #32
bhi 1b
CPWAIT(r0)
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r0)
ENTRY(xscale_cache_flushD_rng)
and r2, r0, #0x1f
add r1, r1, r2
bic r0, r0, #0x1f
1: mcr p15, 0, r0, c7, c6, 1 /* flush D cache single entry */
add r0, r0, #32
subs r1, r1, #32
bhi 1b
mcr p15, 0, r0, c7, c10, 4 /* drain write buffer */
CPWAIT_AND_RETURN(r0)
/*
* Context switch.
*
* These is the CPU-specific parts of the context switcher cpu_switch()
* These functions actually perform the TTB reload.
*
* NOTE: Special calling convention
* r1, r4-r13 must be preserved
*/
ENTRY(xscale_context_switch)
/*
* CF_CACHE_PURGE_ID will *ALWAYS* be called prior to this.
* Thus the data cache will contain only kernel data and the
* instruction cache will contain only kernel code, and all
* kernel mappings are shared by all processes.
*/
/* Write the TTB */
mcr p15, 0, r0, c2, c0, 0
/* If we have updated the TTB we must flush the TLB */
mcr p15, 0, r0, c8, c7, 0 /* flush the I+D tlb */
CPWAIT_AND_RETURN(r0)
/*
* xscale_cpu_sleep
*
* This is called when there is nothing on any of the run queues.
* We go into IDLE mode so that any IRQ or FIQ will awaken us.
*
* If this is called with anything other than ARM_SLEEP_MODE_IDLE,
* ignore it.
*/
ENTRY(xscale_cpu_sleep)
tst r0, #0x00000000
bne 1f
mov r0, #0x1
mcr p14, 0, r0, c7, c0, 0
1:
mov pc, lr
|