about summary refs log tree commit diff
path: root/sysdeps/x86_64/multiarch/memchr-avx2.S
blob: 77a95231681df3e3bb3b5e1877af46fc7d6e0b4e (plain) (blame)
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
/* memchr/wmemchr optimized with AVX2.
   Copyright (C) 2017-2021 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/>.  */

#if IS_IN (libc)

# include <sysdep.h>

# ifndef MEMCHR
#  define MEMCHR	__memchr_avx2
# endif

# ifdef USE_AS_WMEMCHR
#  define VPCMPEQ	vpcmpeqd
# else
#  define VPCMPEQ	vpcmpeqb
# endif

# ifndef VZEROUPPER
#  define VZEROUPPER	vzeroupper
# endif

# define VEC_SIZE 32

	.section .text.avx,"ax",@progbits
ENTRY (MEMCHR)
# ifndef USE_AS_RAWMEMCHR
	/* Check for zero length.  */
	test	%RDX_LP, %RDX_LP
	jz	L(null)
# endif
	movl	%edi, %ecx
	/* Broadcast CHAR to YMM0.  */
	vmovd	%esi, %xmm0
# ifdef USE_AS_WMEMCHR
	shl	$2, %RDX_LP
	vpbroadcastd %xmm0, %ymm0
# else
#  ifdef __ILP32__
	/* Clear the upper 32 bits.  */
	movl	%edx, %edx
#  endif
	vpbroadcastb %xmm0, %ymm0
# endif
	/* Check if we may cross page boundary with one vector load.  */
	andl	$(2 * VEC_SIZE - 1), %ecx
	cmpl	$VEC_SIZE, %ecx
	ja	L(cros_page_boundary)

	/* Check the first VEC_SIZE bytes.  */
	VPCMPEQ (%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax

# ifndef USE_AS_RAWMEMCHR
	jnz	L(first_vec_x0_check)
	/* Adjust length and check the end of data.  */
	subq	$VEC_SIZE, %rdx
	jbe	L(zero)
# else
	jnz	L(first_vec_x0)
# endif

	/* Align data for aligned loads in the loop.  */
	addq	$VEC_SIZE, %rdi
	andl	$(VEC_SIZE - 1), %ecx
	andq	$-VEC_SIZE, %rdi

# ifndef USE_AS_RAWMEMCHR
	/* Adjust length.  */
	addq	%rcx, %rdx

	subq	$(VEC_SIZE * 4), %rdx
	jbe	L(last_4x_vec_or_less)
# endif
	jmp	L(more_4x_vec)

	.p2align 4
L(cros_page_boundary):
	andl	$(VEC_SIZE - 1), %ecx
	andq	$-VEC_SIZE, %rdi
	VPCMPEQ (%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	/* Remove the leading bytes.  */
	sarl	%cl, %eax
	testl	%eax, %eax
	jz	L(aligned_more)
	tzcntl	%eax, %eax
# ifndef USE_AS_RAWMEMCHR
	/* Check the end of data.  */
	cmpq	%rax, %rdx
	jbe	L(zero)
# endif
	addq	%rdi, %rax
	addq	%rcx, %rax
	VZEROUPPER
	ret

	.p2align 4
L(aligned_more):
# ifndef USE_AS_RAWMEMCHR
        /* Calculate "rdx + rcx - VEC_SIZE" with "rdx - (VEC_SIZE - rcx)"
	   instead of "(rdx + rcx) - VEC_SIZE" to void possible addition
	   overflow.  */
	negq	%rcx
	addq	$VEC_SIZE, %rcx

	/* Check the end of data.  */
	subq	%rcx, %rdx
	jbe	L(zero)
# endif

	addq	$VEC_SIZE, %rdi

# ifndef USE_AS_RAWMEMCHR
	subq	$(VEC_SIZE * 4), %rdx
	jbe	L(last_4x_vec_or_less)
# endif

L(more_4x_vec):
	/* Check the first 4 * VEC_SIZE.  Only one VEC_SIZE at a time
	   since data is only aligned to VEC_SIZE.  */
	VPCMPEQ (%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x0)

	VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x1)

	VPCMPEQ (VEC_SIZE * 2)(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x2)

	VPCMPEQ (VEC_SIZE * 3)(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x3)

	addq	$(VEC_SIZE * 4), %rdi

# ifndef USE_AS_RAWMEMCHR
	subq	$(VEC_SIZE * 4), %rdx
	jbe	L(last_4x_vec_or_less)
# endif

	/* Align data to 4 * VEC_SIZE.  */
	movq	%rdi, %rcx
	andl	$(4 * VEC_SIZE - 1), %ecx
	andq	$-(4 * VEC_SIZE), %rdi

# ifndef USE_AS_RAWMEMCHR
	/* Adjust length.  */
	addq	%rcx, %rdx
# endif

	.p2align 4
L(loop_4x_vec):
	/* Compare 4 * VEC at a time forward.  */
	VPCMPEQ (%rdi), %ymm0, %ymm1
	VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm2
	VPCMPEQ (VEC_SIZE * 2)(%rdi), %ymm0, %ymm3
	VPCMPEQ (VEC_SIZE * 3)(%rdi), %ymm0, %ymm4

	vpor	%ymm1, %ymm2, %ymm5
	vpor	%ymm3, %ymm4, %ymm6
	vpor	%ymm5, %ymm6, %ymm5

	vpmovmskb %ymm5, %eax
	testl	%eax, %eax
	jnz	L(4x_vec_end)

	addq	$(VEC_SIZE * 4), %rdi

# ifdef USE_AS_RAWMEMCHR
	jmp	L(loop_4x_vec)
# else
	subq	$(VEC_SIZE * 4), %rdx
	ja	L(loop_4x_vec)

L(last_4x_vec_or_less):
	/* Less than 4 * VEC and aligned to VEC_SIZE.  */
	addl	$(VEC_SIZE * 2), %edx
	jle	L(last_2x_vec)

	VPCMPEQ (%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x0)

	VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x1)

	VPCMPEQ (VEC_SIZE * 2)(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax

	jnz	L(first_vec_x2_check)
	subl	$VEC_SIZE, %edx
	jle	L(zero)

	VPCMPEQ (VEC_SIZE * 3)(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax

	jnz	L(first_vec_x3_check)
	xorl	%eax, %eax
	VZEROUPPER
	ret

	.p2align 4
L(last_2x_vec):
	addl	$(VEC_SIZE * 2), %edx
	VPCMPEQ (%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax

	jnz	L(first_vec_x0_check)
	subl	$VEC_SIZE, %edx
	jle	L(zero)

	VPCMPEQ VEC_SIZE(%rdi), %ymm0, %ymm1
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x1_check)
	xorl	%eax, %eax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x0_check):
	tzcntl	%eax, %eax
	/* Check the end of data.  */
	cmpq	%rax, %rdx
	jbe	L(zero)
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x1_check):
	tzcntl	%eax, %eax
	/* Check the end of data.  */
	cmpq	%rax, %rdx
	jbe	L(zero)
	addq	$VEC_SIZE, %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x2_check):
	tzcntl	%eax, %eax
	/* Check the end of data.  */
	cmpq	%rax, %rdx
	jbe	L(zero)
	addq	$(VEC_SIZE * 2), %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x3_check):
	tzcntl	%eax, %eax
	/* Check the end of data.  */
	cmpq	%rax, %rdx
	jbe	L(zero)
	addq	$(VEC_SIZE * 3), %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(zero):
	VZEROUPPER
L(null):
	xorl	%eax, %eax
	ret
# endif

	.p2align 4
L(first_vec_x0):
	tzcntl	%eax, %eax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x1):
	tzcntl	%eax, %eax
	addq	$VEC_SIZE, %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(first_vec_x2):
	tzcntl	%eax, %eax
	addq	$(VEC_SIZE * 2), %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

	.p2align 4
L(4x_vec_end):
	vpmovmskb %ymm1, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x0)
	vpmovmskb %ymm2, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x1)
	vpmovmskb %ymm3, %eax
	testl	%eax, %eax
	jnz	L(first_vec_x2)
	vpmovmskb %ymm4, %eax
	testl	%eax, %eax
L(first_vec_x3):
	tzcntl	%eax, %eax
	addq	$(VEC_SIZE * 3), %rax
	addq	%rdi, %rax
	VZEROUPPER
	ret

END (MEMCHR)
#endif