xref: /titanic_50/usr/src/lib/libc/sparcv9/gen/strlen.s (revision 7257d1b4d25bfac0c802847390e98a464fd787ac)
1/*
2 * CDDL HEADER START
3 *
4 * The contents of this file are subject to the terms of the
5 * Common Development and Distribution License (the "License").
6 * You may not use this file except in compliance with the License.
7 *
8 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
9 * or http://www.opensolaris.org/os/licensing.
10 * See the License for the specific language governing permissions
11 * and limitations under the License.
12 *
13 * When distributing Covered Code, include this CDDL HEADER in each
14 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
15 * If applicable, add the following below this CDDL HEADER, with the
16 * fields enclosed by brackets "[]" replaced with your own identifying
17 * information: Portions Copyright [yyyy] [name of copyright owner]
18 *
19 * CDDL HEADER END
20 */
21
22/*
23 * Copyright 2008 Sun Microsystems, Inc.  All rights reserved.
24 * Use is subject to license terms.
25 */
26
27#pragma ident	"%Z%%M%	%I%	%E% SMI"
28
29	.file	"%M%"
30
31/*
32 * strlen(s)
33 *
34 * Given string s, return length (not including the terminating null).
35 *
36 * Fast assembler language version of the following C-program strlen
37 * which represents the `standard' for the C-library.
38 *
39 *	size_t
40 *	strlen(s)
41 *	register const char *s;
42 *	{
43 *		register const char *s0 = s + 1;
44 *
45 *		while (*s++ != '\0')
46 *			;
47 *		return (s - s0);
48 *	}
49 */
50
51#include <sys/asm_linkage.h>
52
53	! The object of strlen is to, as quickly as possible, find the
54	! null byte.  To this end, we attempt to get our string aligned
55	! and then blast across it using Alan Mycroft's algorithm for
56	! finding null bytes. If we are not aligned, the string is
57	! checked a byte at a time until it is.  Once this occurs,
58	! we can proceed word-wise across it.  Once a word with a
59	! zero byte has been found, we then check the word a byte
60	! at a time until we've located the zero byte, and return
61	! the proper length.
62
63	.align 32
64	ENTRY(strlen)
65	andcc		%o0, 3, %o4	! is src word aligned
66	bz,pt		%icc, .nowalgnd
67	mov		%o0, %o2
68
69	cmp		%o4, 2		! is src half-word aligned
70	be,a,pn		%icc, .s2algn
71	lduh		[%o2], %o1
72
73	ldub		[%o2], %o1
74	tst		%o1		! byte zero?
75	bz,pn		%icc, .done
76	cmp		%o4, 3		! src is byte aligned
77
78	be,pn		%icc, .nowalgnd
79	inc		1, %o2
80
81	lduh		[%o2], %o1
82
83.s2algn:
84	srl		%o1, 8, %o4
85	tst		%o4
86	bz,pn		%icc, .done
87	andcc		%o1, 0xff, %g0
88
89	bz,pn		%icc, .done
90	inc		1, %o2
91
92	inc		1, %o2
93
94.nowalgnd:
95	ld		[%o2], %o1
96	sethi		%hi(0x01010101), %o4
97	sethi		%hi(0x80808080), %o5
98	or		%o4, %lo(0x01010101), %o4
99	or		%o5, %lo(0x80808080), %o5
100
101	andn		%o5, %o1, %o3
102	sub		%o1, %o4, %g1
103	andcc		%o3, %g1, %g0
104	bnz,a,pn	%icc, .nullfound
105	sethi		%hi(0xff000000), %o4
106
107	ld		[%o2+4], %o1
108	inc		4, %o2
109
110.loop:						! this should be aligned to 32
111	inc		4, %o2
112	andn		%o5, %o1, %o3		! %o5 = ~word & 0x80808080
113	sub		%o1, %o4, %g1		! %g1 = word - 0x01010101
114	andcc		%o3, %g1, %g0
115	bz,a,pt		%icc, .loop
116	ld		[%o2], %o1
117
118	dec		4, %o2
119	sethi		%hi(0xff000000), %o4
120.nullfound:
121	andcc		%o1, %o4, %g0
122	bz,pn		%icc, .done		! first byte zero
123	srl		%o4, 8, %o4
124
125	andcc		%o1, %o4, %g0
126	bz,pn		%icc, .done		! second byte zero
127	inc		1, %o2
128
129	srl		%o4, 8, %o4
130	andcc		%o1, %o4, %g0
131	bz,pn		%icc, .done		! thrid byte zero
132	inc		1, %o2
133
134	inc		1, %o2			! fourth byte zero
135.done:
136	retl
137	sub		%o2, %o0, %o0
138	SET_SIZE(strlen)
139
140