1/*
2 * Copyright (C) 2008 The Android Open Source Project
3 * All rights reserved.
4 *
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
7 * are met:
8 *  * Redistributions of source code must retain the above copyright
9 *    notice, this list of conditions and the following disclaimer.
10 *  * Redistributions in binary form must reproduce the above copyright
11 *    notice, this list of conditions and the following disclaimer in
12 *    the documentation and/or other materials provided with the
13 *    distribution.
14 *
15 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
16 * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
17 * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
18 * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
19 * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
20 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
21 * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
22 * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
23 * AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
24 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
25 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
26 * SUCH DAMAGE.
27 */
28#include <string.h>
29#include <stdint.h>
30#include <machine/cpu-features.h>
31
32size_t strlen(const char *s)
33{
34    __builtin_prefetch(s);
35    __builtin_prefetch(s+32);
36
37    union {
38        const char      *b;
39        const uint32_t  *w;
40        uintptr_t       i;
41    } u;
42
43    // these are some scratch variables for the asm code below
44    uint32_t v, t;
45
46    // initialize the string length to zero
47    size_t l = 0;
48
49    // align the pointer to a 32-bit word boundary
50    u.b = s;
51    while (u.i & 0x3)  {
52        if (__builtin_expect(*u.b++ == 0, 0)) {
53            goto done;
54        }
55        l++;
56    }
57
58    // loop for each word, testing if it contains a zero byte
59    // if so, exit the loop and update the length.
60    // We need to process 32 bytes per loop to schedule PLD properly
61    // and achieve the maximum bus speed.
62    asm(
63        "ldr     %[v], [%[s]], #4           \n"
64        "sub     %[l], %[l], %[s]           \n"
65        "0:                                 \n"
66        "pld     [%[s], #64]                \n"
67        "sub     %[t], %[v], %[mask], lsr #7\n"
68        "and     %[t], %[t], %[mask]        \n"
69        "bics    %[t], %[t], %[v]           \n"
70        "it      eq                         \n"
71        "ldreq   %[v], [%[s]], #4           \n"
72#if !defined(__OPTIMIZE_SIZE__)
73        "bne     1f                         \n"
74        "sub     %[t], %[v], %[mask], lsr #7\n"
75        "and     %[t], %[t], %[mask]        \n"
76        "bics    %[t], %[t], %[v]           \n"
77        "it      eq                         \n"
78        "ldreq   %[v], [%[s]], #4           \n"
79        "bne     1f                         \n"
80        "sub     %[t], %[v], %[mask], lsr #7\n"
81        "and     %[t], %[t], %[mask]        \n"
82        "bics    %[t], %[t], %[v]           \n"
83        "it      eq                         \n"
84        "ldreq   %[v], [%[s]], #4           \n"
85        "bne     1f                         \n"
86        "sub     %[t], %[v], %[mask], lsr #7\n"
87        "and     %[t], %[t], %[mask]        \n"
88        "bics    %[t], %[t], %[v]           \n"
89        "it      eq                         \n"
90        "ldreq   %[v], [%[s]], #4           \n"
91        "bne     1f                         \n"
92        "sub     %[t], %[v], %[mask], lsr #7\n"
93        "and     %[t], %[t], %[mask]        \n"
94        "bics    %[t], %[t], %[v]           \n"
95        "it      eq                         \n"
96        "ldreq   %[v], [%[s]], #4           \n"
97        "bne     1f                         \n"
98        "sub     %[t], %[v], %[mask], lsr #7\n"
99        "and     %[t], %[t], %[mask]        \n"
100        "bics    %[t], %[t], %[v]           \n"
101        "it      eq                         \n"
102        "ldreq   %[v], [%[s]], #4           \n"
103        "bne     1f                         \n"
104        "sub     %[t], %[v], %[mask], lsr #7\n"
105        "and     %[t], %[t], %[mask]        \n"
106        "bics    %[t], %[t], %[v]           \n"
107        "it      eq                         \n"
108        "ldreq   %[v], [%[s]], #4           \n"
109        "bne     1f                         \n"
110        "sub     %[t], %[v], %[mask], lsr #7\n"
111        "and     %[t], %[t], %[mask]        \n"
112        "bics    %[t], %[t], %[v]           \n"
113        "it      eq                         \n"
114        "ldreq   %[v], [%[s]], #4           \n"
115#endif
116        "beq     0b                         \n"
117        "1:                                 \n"
118        "add     %[l], %[l], %[s]           \n"
119        "tst     %[v], #0xFF                \n"
120        "beq     2f                         \n"
121        "add     %[l], %[l], #1             \n"
122        "tst     %[v], #0xFF00              \n"
123        "beq     2f                         \n"
124        "add     %[l], %[l], #1             \n"
125        "tst     %[v], #0xFF0000            \n"
126        "it      ne                         \n"
127        "addne   %[l], %[l], #1             \n"
128        "2:                                 \n"
129        : [l]"=&r"(l), [v]"=&r"(v), [t]"=&r"(t), [s]"=&r"(u.b)
130        : "%[l]"(l), "%[s]"(u.b), [mask]"r"(0x80808080UL)
131        : "cc"
132    );
133
134done:
135    return l;
136}
137