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
|
From dd99689b821162293506e0344f163b82349a9298 Mon Sep 17 00:00:00 2001
From: Xue Liu <liuxue@loongson.cn>
Date: Sun, 29 Jan 2023 10:22:01 +0800
Subject: [PATCH 2/6] LoongArch: Optimize string functions strchr, strchrnull.
Change-Id: I8b274972642b6a1926d8fc176404bfd83344bc51
---
sysdeps/loongarch/lp64/strchr.S | 107 +++++++++++++++++++++++++++
sysdeps/loongarch/lp64/strchrnul.S | 115 +++++++++++++++++++++++++++++
2 files changed, 222 insertions(+)
create mode 100644 sysdeps/loongarch/lp64/strchr.S
create mode 100644 sysdeps/loongarch/lp64/strchrnul.S
diff --git a/sysdeps/loongarch/lp64/strchr.S b/sysdeps/loongarch/lp64/strchr.S
new file mode 100644
index 00000000..3d64c684
--- /dev/null
+++ b/sysdeps/loongarch/lp64/strchr.S
@@ -0,0 +1,107 @@
+/* Optimized strchr implementation for LoongArch.
+ Copyright (C) 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/>. */
+
+#include <sysdep.h>
+
+/* char * strchr (const char *s1, int c); */
+LEAF(strchr)
+ .align 6
+
+ li.w t4, 0x7
+ lu12i.w a2, 0x01010
+ bstrins.d a1, a1, 15, 8
+ andi t0, a0, 0x7
+
+ ori a2, a2, 0x101
+ andn t4, a0, t4
+ slli.w t1, t0, 3
+
+ ld.d t4, t4, 0
+
+ nor t8, zero, zero
+ bstrins.d a1, a1, 31, 16
+ srl.d t4, t4, t1
+
+ bstrins.d a1, a1, 63, 32
+ bstrins.d a2, a2, 63, 32
+ srl.d a7, t8, t1
+
+ li.w t1, 8
+ nor t8, a7, zero
+ slli.d a3, a2, 7
+ or t5, t8, t4
+ and t3, a7, a1
+
+ sub.w t1, t1, t0
+ nor a3, a3, zero
+ xor t2, t5, t3
+ sub.d a7, t5, a2
+ nor a6, t5, a3
+
+ sub.d a5, t2, a2
+ nor a4, t2, a3
+
+ and a6, a7, a6
+ and a5, a5, a4
+ or a7, a6, a5
+ bnez a7, L(_mc8_a)
+
+ add.d a0, a0, t1
+L(_aloop):
+ ld.d t4, a0, 0
+
+ xor t2, t4, a1
+ sub.d a7, t4, a2
+ nor a6, t4, a3
+ sub.d a5, t2, a2
+
+ nor a4, t2, a3
+ and a6, a7, a6
+ and a5, a5, a4
+ or a7, a6, a5
+ bnez a7, L(_mc8_a)
+
+ ld.d t4, a0, 8
+ addi.d a0, a0, 16
+ xor t2, t4, a1
+ sub.d a7, t4, a2
+ nor a6, t4, a3
+ sub.d a5, t2, a2
+
+ nor a4, t2, a3
+ and a6, a7, a6
+ and a5, a5, a4
+ or a7, a6, a5
+ beqz a7, L(_aloop)
+
+ addi.d a0, a0, -8
+L(_mc8_a):
+
+ ctz.d t0, a5
+ ctz.d t2, a6
+
+ srli.w t0, t0, 3
+ srli.w t2, t2, 3
+ sltu t1, t2, t0
+ add.d v0, a0, t0
+ masknez v0, v0, t1
+ jr ra
+END(strchr)
+
+libc_hidden_builtin_def (strchr)
+weak_alias (strchr, index)
diff --git a/sysdeps/loongarch/lp64/strchrnul.S b/sysdeps/loongarch/lp64/strchrnul.S
new file mode 100644
index 00000000..58b8b372
--- /dev/null
+++ b/sysdeps/loongarch/lp64/strchrnul.S
@@ -0,0 +1,115 @@
+/* Optimized strchrnul implementation for LoongArch.
+ Copyright (C) 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/>. */
+
+#include <sysdep.h>
+
+#define MOVZ(rd,rs,rt) \
+ masknez t6, rs, rt;\
+ maskeqz rd, rd, rt;\
+ or rd, rd, t6
+
+/* char *strchrnul(const char *s, int c); */
+LEAF(__strchrnul)
+ .align 6
+
+ li.w t4, 0x7
+ lu12i.w a2, 0x01010
+ bstrins.d a1, a1, 15, 8
+ andi t0, a0, 0x7
+
+ ori a2, a2, 0x101
+ andn t4, a0, t4
+ slli.w t1, t0, 3
+ ld.d t4, t4, 0
+
+ nor t8, zero, zero
+ bstrins.d a1, a1, 31, 16
+ srl.d t4, t4, t1
+
+ preld 0, a0, 32
+ bstrins.d a1, a1, 63, 32
+ bstrins.d a2, a2, 63, 32
+ srl.d a7, t8, t1
+
+ nor t8, a7, zero
+ slli.d a3, a2, 7
+ or t5, t8, t4
+ and t3, a7, a1
+
+ nor a3, a3, zero
+ xor t2, t5, t3
+ sub.d a7, t5, a2
+ nor a6, t5, a3
+
+ li.w t1, 8
+ sub.d a5, t2, a2
+ nor a4, t2, a3
+
+ and a6, a7, a6
+ and a5, a5, a4
+ or a7, a6, a5
+ bnez a7, L(_mc8_a)
+
+ sub.w t1, t1, t0
+ add.d a0, a0, t1
+L(_aloop):
+ ld.d t4, a0, 0
+
+ xor t2, t4, a1
+ sub.d a7, t4, a2
+ nor a6, t4, a3
+ sub.d a5, t2, a2
+
+ nor a4, t2, a3
+ and a6, a7, a6
+ and a5, a5, a4
+
+ or a7, a6, a5
+ bnez a7, L(_mc8_a)
+
+ ld.d t4, a0, 8
+ addi.d a0, a0, 16
+
+ xor t2, t4, a1
+ sub.d a7, t4, a2
+ nor a6, t4, a3
+ sub.d a5, t2, a2
+
+ nor a4, t2, a3
+ and a6, a7, a6
+ and a5, a5, a4
+
+ or a7, a6, a5
+ beqz a7, L(_aloop)
+
+ addi.d a0, a0, -8
+L(_mc8_a):
+ ctz.d t0, a5
+ ctz.d t2, a6
+
+ srli.w t0, t0, 3
+ srli.w t2, t2, 3
+ slt t1, t0, t2
+
+ MOVZ(t0,t2,t1)
+
+ add.d v0, a0, t0
+ jr ra
+END(__strchrnul)
+
+weak_alias(__strchrnul, strchrnul)
--
2.33.0
|
