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
|
/* strcmp implementation for ARMv7-A, optimized for Cortex-A15.
Copyright (C) 2012-2025 Free Software Foundation, Inc.
This file is part of the GNU C Library.
The GNU C Library is free software; you can redistribute it and/or
modify it under the terms of the GNU Lesser General Public
License as published by the Free Software Foundation; either
version 2.1 of the License, or (at your option) any later version.
The GNU C Library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
Lesser General Public License for more details.
You should have received a copy of the GNU Lesser General Public
License along with the GNU C Library. If not, see
<https://www.gnu.org/licenses/>. */
#include <arm-features.h>
#include <sysdep.h>
/* Implementation of strcmp for ARMv7 when DSP instructions are
available. Use ldrd to support wider loads, provided the data
is sufficiently aligned. Use saturating arithmetic to optimize
the compares. */
/* Build Options:
STRCMP_PRECHECK: Run a quick pre-check of the first byte in the
string. If comparing completely random strings the pre-check will
save time, since there is a very high probability of a mismatch in
the first character: we save significant overhead if this is the
common case. However, if strings are likely to be identical (e.g.
because we're verifying a hit in a hash table), then this check
is largely redundant. */
#define STRCMP_PRECHECK 1
.syntax unified
#ifdef __ARM_BIG_ENDIAN
# define S2LO lsl
# define S2LOEQ lsleq
# define S2HI lsr
# define MSB 0x000000ff
# define LSB 0xff000000
# define BYTE0_OFFSET 24
# define BYTE1_OFFSET 16
# define BYTE2_OFFSET 8
# define BYTE3_OFFSET 0
#else /* not __ARM_BIG_ENDIAN */
# define S2LO lsr
# define S2LOEQ lsreq
# define S2HI lsl
# define BYTE0_OFFSET 0
# define BYTE1_OFFSET 8
# define BYTE2_OFFSET 16
# define BYTE3_OFFSET 24
# define MSB 0xff000000
# define LSB 0x000000ff
#endif /* not __ARM_BIG_ENDIAN */
/* Parameters and result. */
#define src1 r0
#define src2 r1
#define result r0 /* Overlaps src1. */
/* Internal variables. */
#define tmp1 r4
#define tmp2 r5
#define const_m1 r12
/* Additional internal variables for 64-bit aligned data. */
#define data1a r2
#define data1b r3
#define data2a r6
#define data2b r7
#define syndrome_a tmp1
#define syndrome_b tmp2
/* Additional internal variables for 32-bit aligned data. */
#define data1 r2
#define data2 r3
#define syndrome tmp2
.thumb
/* In Thumb code we can't use MVN with a register shift, but we do have ORN. */
.macro prepare_mask mask_reg, nbits_reg
S2HI \mask_reg, const_m1, \nbits_reg
.endm
.macro apply_mask data_reg, mask_reg
orn \data_reg, \data_reg, \mask_reg
.endm
/* Macro to compute and return the result value for word-aligned
cases. */
.macro strcmp_epilogue_aligned synd d1 d2 restore_r6
#ifdef __ARM_BIG_ENDIAN
/* If data1 contains a zero byte, then syndrome will contain a 1 in
bit 7 of that byte. Otherwise, the highest set bit in the
syndrome will highlight the first different bit. It is therefore
sufficient to extract the eight bits starting with the syndrome
bit. */
clz tmp1, \synd
lsl r1, \d2, tmp1
.if \restore_r6
ldrd r6, r7, [sp, #8]
.endif
lsl \d1, \d1, tmp1
lsr result, \d1, #24
ldrd r4, r5, [sp], #16
cfi_remember_state
cfi_def_cfa_offset (0)
cfi_restore (r4)
cfi_restore (r5)
cfi_restore (r6)
cfi_restore (r7)
sub result, result, r1, lsr #24
bx lr
#else
/* To use the big-endian trick we'd have to reverse all three words.
that's slower than this approach. */
rev \synd, \synd
clz tmp1, \synd
bic tmp1, tmp1, #7
lsr r1, \d2, tmp1
.if \restore_r6
ldrd r6, r7, [sp, #8]
.endif
lsr \d1, \d1, tmp1
and result, \d1, #255
and r1, r1, #255
ldrd r4, r5, [sp], #16
cfi_remember_state
cfi_def_cfa_offset (0)
cfi_restore (r4)
cfi_restore (r5)
cfi_restore (r6)
cfi_restore (r7)
sub result, result, r1
bx lr
#endif
.endm
.text
.p2align 5
.Lstrcmp_start_addr:
#if STRCMP_PRECHECK == 1
.Lfastpath_exit:
sub r0, r2, r3
bx lr
nop
#endif
ENTRY (strcmp)
#if STRCMP_PRECHECK == 1
ldrb r2, [src1]
ldrb r3, [src2]
cmp r2, #1
it cs
cmpcs r2, r3
bne .Lfastpath_exit
#endif
strd r4, r5, [sp, #-16]!
cfi_def_cfa_offset (16)
cfi_offset (r4, -16)
cfi_offset (r5, -12)
orr tmp1, src1, src2
strd r6, r7, [sp, #8]
cfi_offset (r6, -8)
cfi_offset (r7, -4)
mvn const_m1, #0
lsl r2, tmp1, #29
cbz r2, .Lloop_aligned8
.Lnot_aligned:
eor tmp1, src1, src2
tst tmp1, #7
bne .Lmisaligned8
/* Deal with mutual misalignment by aligning downwards and then
masking off the unwanted loaded data to prevent a difference. */
and tmp1, src1, #7
bic src1, src1, #7
and tmp2, tmp1, #3
bic src2, src2, #7
lsl tmp2
|