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
|
From 248d9a5b63bba72bfc316b8a48c6163fce5acc22 Mon Sep 17 00:00:00 2001
From: Paulius Zaleckas <paulius.zaleckas@gmail.com>
Date: Thu, 18 Feb 2010 21:53:01 +0200
Subject: [PATCH] ARM: Use cache alignment from asm/cache.h
Make code more optimal for ARM variants with
different cache line size.
Signed-off-by: Paulius Zaleckas <paulius.zaleckas@gmail.com>
---
arch/arm/boot/compressed/head.S | 11 ++++++-----
arch/arm/include/asm/dma-mapping.h | 2 +-
arch/arm/kernel/entry-armv.S | 31 ++++++++++++++++---------------
arch/arm/kernel/entry-common.S | 7 ++++---
arch/arm/kernel/head.S | 3 ++-
arch/arm/kernel/vmlinux.lds.S | 5 +++--
arch/arm/lib/copy_page.S | 2 +-
arch/arm/lib/memchr.S | 3 ++-
arch/arm/lib/memset.S | 3 ++-
arch/arm/lib/memzero.S | 3 ++-
arch/arm/lib/strchr.S | 3 ++-
arch/arm/lib/strncpy_from_user.S | 3 ++-
arch/arm/lib/strnlen_user.S | 3 ++-
arch/arm/lib/strrchr.S | 3 ++-
arch/arm/mm/abort-ev4.S | 3 ++-
arch/arm/mm/abort-nommu.S | 3 ++-
16 files changed, 51 insertions(+), 37 deletions(-)
--- a/arch/arm/boot/compressed/head.S
+++ b/arch/arm/boot/compressed/head.S
@@ -9,6 +9,7 @@
* published by the Free Software Foundation.
*/
#include <linux/linkage.h>
+#include <asm/cache.h>
/*
* Debugging stuff
@@ -349,7 +350,7 @@ params: ldr r0, =0x10000100 @ params_p
* This routine must preserve:
* r4, r5, r6, r7, r8
*/
- .align 5
+ .align L1_CACHE_SHIFT
cache_on: mov r3, #8 @ cache_on function
b call_cache_fn
@@ -537,7 +538,7 @@ __common_mmu_cache_on:
mcr p15, 0, r3, c2, c0, 0 @ load page table pointer
mcr p15, 0, r1, c3, c0, 0 @ load domain access control
b 1f
- .align 5 @ cache line aligned
+ .align L1_CACHE_SHIFT @ cache line aligned
1: mcr p15, 0, r0, c1, c0, 0 @ load control register
mrc p15, 0, r0, c1, c0, 0 @ and read it back to
sub pc, lr, r0, lsr #32 @ properly flush pipeline
@@ -556,7 +557,7 @@ __common_mmu_cache_on:
* r8 = atags pointer
* r9-r12,r14 = corrupted
*/
- .align 5
+ .align L1_CACHE_SHIFT
reloc_start: add r9, r5, r0
sub r9, r9, #128 @ do not copy the stack
debug_reloc_start
@@ -786,7 +787,7 @@ proc_types:
* This routine must preserve:
* r4, r6, r7
*/
- .align 5
+ .align L1_CACHE_SHIFT
cache_off: mov r3, #12 @ cache_off function
b call_cache_fn
@@ -861,7 +862,7 @@ __armv3_mmu_cache_off:
* This routine must preserve:
* r0, r4, r5, r6, r7
*/
- .align 5
+ .align L1_CACHE_SHIFT
cache_clean_flush:
mov r3, #16
b call_cache_fn
--- a/arch/arm/kernel/entry-armv.S
+++ b/arch/arm/kernel/entry-armv.S
@@ -23,6 +23,7 @@
#include <asm/unwind.h>
#include <asm/unistd.h>
#include <asm/tls.h>
+#include <asm/cache.h>
#include "entry-header.S"
@@ -165,7 +166,7 @@ ENDPROC(__und_invalid)
stmia r5, {r0 - r4}
.endm
- .align 5
+ .align L1_CACHE_SHIFT
__dabt_svc:
svc_entry
@@ -213,7 +214,7 @@ __dabt_svc:
UNWIND(.fnend )
ENDPROC(__dabt_svc)
- .align 5
+ .align L1_CACHE_SHIFT
__irq_svc:
svc_entry
@@ -257,7 +258,7 @@ svc_preempt:
b 1b
#endif
- .align 5
+ .align L1_CACHE_SHIFT
__und_svc:
#ifdef CONFIG_KPROBES
@ If a kprobe is about to simulate a "stmdb sp..." instruction,
@@ -303,7 +304,7 @@ __und_svc:
UNWIND(.fnend )
ENDPROC(__und_svc)
- .align 5
+ .align L1_CACHE_SHIFT
__pabt_svc:
svc_entry
@@ -339,7 +340,7 @@ __pabt_svc:
UNWIND(.fnend )
ENDPROC(__pabt_svc)
- .align 5
+ .align L1_CACHE_SHIFT
.LCcralign:
.word cr_alignment
#ifdef MULTI_DABORT
@@ -412,7 +413,7 @@ ENDPROC(__pabt_svc)
#endif
.endm
- .align 5
+ .align L1_CACHE_SHIFT
__dabt_usr:
usr_entry
kuser_cmpxchg_check
@@ -444,7 +445,7 @@ __dabt_usr:
UNWIND(.fnend )
ENDPROC(__dabt_usr)
- .align 5
+ .align L1_CACHE_SHIFT
__irq_usr:
usr_entry
kuser_cmpxchg_check
@@ -473,7 +474,7 @@ ENDPROC(__irq_usr)
.ltorg
- .align 5
+ .align L1_CACHE_SHIFT
__und_usr:
usr_entry
@@ -689,7 +690,7 @@ __und_usr_unknown:
b do_undefinstr
ENDPROC(__und_usr_unknown)
- .align 5
+ .align L1_CACHE_SHIFT
__pabt_usr:
usr_entry
@@ -803,7 +804,7 @@ ENDPROC(__switch_to)
#endif
.endm
- .align 5
+ .align L1_CACHE_SHIFT
.globl __kuser_helper_start
__kuser_helper_start:
@@ -843,7 +844,7 @@ __kuser_memory_barrier: @ 0xffff0fa0
smp_dmb
usr_ret lr
- .align 5
+ .align L1_CACHE_SHIFT
/*
* Reference prototype:
@@ -973,7 +974,7 @@ kuser_cmpxchg_fixup:
#endif
- .align 5
+ .align L1_CACHE_SHIFT
/*
* Reference prototype:
@@ -1051,7 +1052,7 @@ __kuser_helper_end:
* of which is copied into r0 for the mode specific abort handler.
*/
.macro vector_stub, name, mode, correction=0
- .align 5
+ .align L1_CACHE_SHIFT
vector_\name:
.if \correction
@@ -1182,7 +1183,7 @@ __stubs_start:
.long __und_invalid @ e
.long __und_invalid @ f
- .align 5
+ .align L1_CACHE_SHIFT
/*=============================================================================
* Undefined FIQs
@@ -1212,7 +1213,7 @@ vector_addrexcptn:
* We group all the following data together to optimise
* for CPUs with separate I & D caches.
*/
- .align 5
+ .align L1_CACHE_SHIFT
.LCvswi:
.word vector_swi
--- a/arch/arm/kernel/entry-common.S
+++ b/arch/arm/kernel/entry-common.S
@@ -10,13 +10,14 @@
#include <asm/unistd.h>
#include <asm/ftrace.h>
+#include <asm/cache.h>
#include <mach/entry-macro.S>
#include <asm/unwind.h>
#include "entry-header.S"
- .align 5
+ .align L1_CACHE_SHIFT
/*
* This is the fast syscall return path. We do as little as
* possible here, and this includes saving r0 back into the SVC
@@ -221,7 +222,7 @@ ftrace_stub:
#define A710(code...)
#endif
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(vector_swi)
sub sp, sp, #S_FRAME_SIZE
stmia sp, {r0 - r12} @ Calling r0 - r12
@@ -354,7 +355,7 @@ __sys_trace_return:
bl syscall_trace
b ret_slow_syscall
- .align 5
+ .align L1_CACHE_SHIFT
#ifdef CONFIG_ALIGNMENT_TRAP
.type __cr_alignment, #object
__cr_alignment:
--- a/arch/arm/kernel/head.S
+++ b/arch/arm/kernel/head.S
@@ -21,6 +21,7 @@
#include <asm/memory.h>
#include <asm/thread_info.h>
#include <asm/system.h>
+#include <asm/cache.h>
#if (PHYS_OFFSET & 0x001fffff)
#error "PHYS_OFFSET must be at an even 2MiB boundary!"
@@ -192,7 +193,7 @@ ENDPROC(__enable_mmu)
*
* other registers depend on the function called upon completion
*/
- .align 5
+ .align L1_CACHE_SHIFT
__turn_mmu_on:
mov r0, r0
mcr p15, 0, r0, c1, c0, 0 @ write control reg
--- a/arch/arm/kernel/vmlinux.lds.S
+++ b/arch/arm/kernel/vmlinux.lds.S
@@ -7,6 +7,7 @@
#include <asm/thread_info.h>
#include <asm/memory.h>
#include <asm/page.h>
+#include <asm/cache.h>
OUTPUT_ARCH(arm)
ENTRY(stext)
--- a/arch/arm/lib/copy_page.S
+++ b/arch/arm/lib/copy_page.S
@@ -17,7 +17,7 @@
#define COPY_COUNT (PAGE_SZ / (2 * L1_CACHE_BYTES) PLD( -1 ))
.text
- .align 5
+ .align L1_CACHE_SHIFT
/*
* StrongARM optimised copy_page routine
* now 1.78bytes/cycle, was 1.60 bytes/cycle (50MHz bus -> 89MB/s)
--- a/arch/arm/lib/memchr.S
+++ b/arch/arm/lib/memchr.S
@@ -11,9 +11,10 @@
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(memchr)
1: subs r2, r2, #1
bmi 2f
--- a/arch/arm/lib/memset.S
+++ b/arch/arm/lib/memset.S
@@ -11,9 +11,10 @@
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
.word 0
1: subs r2, r2, #4 @ 1 do we have enough
--- a/arch/arm/lib/memzero.S
+++ b/arch/arm/lib/memzero.S
@@ -9,9 +9,10 @@
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
.word 0
/*
* Align the pointer in r0. r3 contains the number of bytes that we are
--- a/arch/arm/lib/strchr.S
+++ b/arch/arm/lib/strchr.S
@@ -11,9 +11,10 @@
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(strchr)
and r1, r1, #0xff
1: ldrb r2, [r0], #1
--- a/arch/arm/lib/strncpy_from_user.S
+++ b/arch/arm/lib/strncpy_from_user.S
@@ -10,9 +10,10 @@
#include <linux/linkage.h>
#include <asm/assembler.h>
#include <asm/errno.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
/*
* Copy a string from user space to kernel space.
--- a/arch/arm/lib/strnlen_user.S
+++ b/arch/arm/lib/strnlen_user.S
@@ -10,9 +10,10 @@
#include <linux/linkage.h>
#include <asm/assembler.h>
#include <asm/errno.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
/* Prototype: unsigned long __strnlen_user(const char *str, long n)
* Purpose : get length of a string in user memory
--- a/arch/arm/lib/strrchr.S
+++ b/arch/arm/lib/strrchr.S
@@ -11,9 +11,10 @@
*/
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
.text
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(strrchr)
mov r3, #0
1: ldrb r2, [r0], #1
--- a/arch/arm/mm/abort-ev4.S
+++ b/arch/arm/mm/abort-ev4.S
@@ -1,5 +1,6 @@
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
/*
* Function: v4_early_abort
*
@@ -17,7 +18,7 @@
* abort here if the I-TLB and D-TLB aren't seeing the same
* picture. Unfortunately, this does happen. We live with it.
*/
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(v4_early_abort)
mrc p15, 0, r1, c5, c0, 0 @ get FSR
mrc p15, 0, r0, c6, c0, 0 @ get FAR
--- a/arch/arm/mm/abort-nommu.S
+++ b/arch/arm/mm/abort-nommu.S
@@ -1,5 +1,6 @@
#include <linux/linkage.h>
#include <asm/assembler.h>
+#include <asm/cache.h>
/*
* Function: nommu_early_abort
*
@@ -12,7 +13,7 @@
* Note: There is no FSR/FAR on !CPU_CP15_MMU cores.
* Just fill zero into the registers.
*/
- .align 5
+ .align L1_CACHE_SHIFT
ENTRY(nommu_early_abort)
mov r0, #0 @ clear r0, r1 (no FSR/FAR)
mov r1, #0
|
