xref: /openbsd-src/lib/libc/gen/opendir.c (revision b725ae7711052a2233e31a66fefb8a752c388d7a)
1 /*
2  * Copyright (c) 1983, 1993
3  *	The Regents of the University of California.  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  * 1. Redistributions of source code must retain the above copyright
9  *    notice, this list of conditions and the following disclaimer.
10  * 2. Redistributions in binary form must reproduce the above copyright
11  *    notice, this list of conditions and the following disclaimer in the
12  *    documentation and/or other materials provided with the distribution.
13  * 3. Neither the name of the University nor the names of its contributors
14  *    may be used to endorse or promote products derived from this software
15  *    without specific prior written permission.
16  *
17  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
18  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
21  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
27  * SUCH DAMAGE.
28  */
29 
30 #if defined(LIBC_SCCS) && !defined(lint)
31 static char rcsid[] = "$OpenBSD: opendir.c,v 1.10 2004/05/18 02:05:52 jfb Exp $";
32 #endif /* LIBC_SCCS and not lint */
33 
34 #include <sys/param.h>
35 #include <sys/mount.h>
36 #include <sys/stat.h>
37 
38 #include <dirent.h>
39 #include <errno.h>
40 #include <fcntl.h>
41 #include <limits.h>
42 #include <stdlib.h>
43 #include <string.h>
44 #include <unistd.h>
45 
46 static int direntcmp(const void *, const void *);
47 
48 /*
49  * Comparison function for sorting dirent structures that never returns 0;
50  * this causes qsort() to emulate a stable sort.
51  */
52 static int
53 direntcmp(const void *d1, const void *d2)
54 {
55 	int i;
56 
57 	i = strcmp((*(struct dirent **)d1)->d_name,
58 	    (*(struct dirent **)d2)->d_name);
59 	return (i != 0 ? i : (char *)d2 - (char *)d1);
60 }
61 
62 /*
63  * Open a directory.
64  */
65 DIR *
66 opendir(const char *name)
67 {
68 
69 	return (__opendir2(name, DTF_HIDEW|DTF_NODUP));
70 }
71 
72 DIR *
73 __opendir2(const char *name, int flags)
74 {
75 	DIR *dirp;
76 	int fd;
77 	struct stat sb;
78 	int pagesz;
79 	int incr;
80 	int unionstack;
81 
82 	if ((fd = open(name, O_RDONLY | O_NONBLOCK)) == -1)
83 		return (NULL);
84 	if (fstat(fd, &sb) || !S_ISDIR(sb.st_mode)) {
85 		errno = ENOTDIR;
86 		close(fd);
87 		return (NULL);
88 	}
89 	if (fcntl(fd, F_SETFD, FD_CLOEXEC) == -1 ||
90 	    (dirp = (DIR *)malloc(sizeof(DIR))) == NULL) {
91 		close(fd);
92 		return (NULL);
93 	}
94 
95 	/*
96 	 * If the machine's page size is an exact multiple of DIRBLKSIZ,
97 	 * use a buffer that is cluster boundary aligned.
98 	 * Hopefully this can be a big win someday by allowing page trades
99 	 * to user space to be done by getdirentries()
100 	 */
101 	if (((pagesz = getpagesize()) % DIRBLKSIZ) == 0)
102 		incr = pagesz;
103 	else
104 		incr = DIRBLKSIZ;
105 
106 	/*
107 	 * Determine whether this directory is the top of a union stack.
108 	 */
109 	if (flags & DTF_NODUP) {
110 		struct statfs sfb;
111 
112 		if (fstatfs(fd, &sfb) < 0) {
113 			free(dirp);
114 			close(fd);
115 			return (NULL);
116 		}
117 		unionstack = strncmp(sfb.f_fstypename, MOUNT_UNION, MFSNAMELEN) == 0 ||
118 			     (sfb.f_flags & MNT_UNION);
119 	} else {
120 		unionstack = 0;
121 	}
122 
123 	if (unionstack) {
124 		int len = 0;
125 		int space = 0;
126 		char *buf = 0;
127 		char *ddptr = 0;
128 		char *ddeptr;
129 		int n;
130 		struct dirent **dpv;
131 
132 		/*
133 		 * The strategy here is to read all the directory
134 		 * entries into a buffer, sort the buffer, and
135 		 * remove duplicate entries by setting the inode
136 		 * number to zero.
137 		 */
138 
139 		do {
140 			/*
141 			 * Always make at least DIRBLKSIZ bytes
142 			 * available to getdirentries
143 			 */
144 			if (space < DIRBLKSIZ) {
145 				char *nbuf;
146 
147 				space += incr;
148 				len += incr;
149 				nbuf = realloc(buf, len);
150 				if (nbuf == NULL) {
151 					if (buf)
152 						free(buf);
153 					free(dirp);
154 					close(fd);
155 					return (NULL);
156 				}
157 				buf = nbuf;
158 				ddptr = buf + (len - space);
159 			}
160 
161 			n = getdirentries(fd, ddptr, space, &dirp->dd_seek);
162 			if (n > 0) {
163 				ddptr += n;
164 				space -= n;
165 			}
166 		} while (n > 0);
167 
168 		ddeptr = ddptr;
169 		flags |= __DTF_READALL;
170 
171 		/*
172 		 * Re-open the directory.
173 		 * This has the effect of rewinding back to the
174 		 * top of the union stack and is needed by
175 		 * programs which plan to fchdir to a descriptor
176 		 * which has also been read -- see fts.c.
177 		 */
178 		if (flags & DTF_REWIND) {
179 			(void) close(fd);
180 			if ((fd = open(name, O_RDONLY)) == -1) {
181 				free(buf);
182 				free(dirp);
183 				return (NULL);
184 			}
185 		}
186 
187 		/*
188 		 * There is now a buffer full of (possibly) duplicate
189 		 * names.
190 		 */
191 		dirp->dd_buf = buf;
192 
193 		/*
194 		 * Go round this loop twice...
195 		 *
196 		 * Scan through the buffer, counting entries.
197 		 * On the second pass, save pointers to each one.
198 		 * Then sort the pointers and remove duplicate names.
199 		 */
200 		for (dpv = 0;;) {
201 			for (n = 0, ddptr = buf; ddptr < ddeptr;) {
202 				struct dirent *dp;
203 
204 				dp = (struct dirent *) ddptr;
205 				if ((long)dp & 03)
206 					break;
207 				if ((dp->d_reclen <= 0) ||
208 				    (dp->d_reclen > (ddeptr + 1 - ddptr)))
209 					break;
210 				ddptr += dp->d_reclen;
211 				if (dp->d_fileno) {
212 					if (dpv)
213 						dpv[n] = dp;
214 					n++;
215 				}
216 			}
217 
218 			if (dpv) {
219 				struct dirent *xp;
220 
221 				/*
222 				 * This sort must be stable.
223 				 */
224 				qsort(dpv, n, sizeof(*dpv), direntcmp);
225 
226 				dpv[n] = NULL;
227 				xp = NULL;
228 
229 				/*
230 				 * Scan through the buffer in sort order,
231 				 * zapping the inode number of any
232 				 * duplicate names.
233 				 */
234 				for (n = 0; dpv[n]; n++) {
235 					struct dirent *dp = dpv[n];
236 
237 					if ((xp == NULL) ||
238 					    strcmp(dp->d_name, xp->d_name))
239 						xp = dp;
240 					else
241 						dp->d_fileno = 0;
242 					if (dp->d_type == DT_WHT &&
243 					    (flags & DTF_HIDEW))
244 						dp->d_fileno = 0;
245 				}
246 
247 				free(dpv);
248 				break;
249 			} else {
250 				if (n+1 > SIZE_T_MAX / sizeof(struct dirent *))
251 					break;
252 				dpv = malloc((n+1) * sizeof(struct dirent *));
253 				if (dpv == NULL)
254 					break;
255 			}
256 		}
257 
258 		dirp->dd_len = len;
259 		dirp->dd_size = ddptr - dirp->dd_buf;
260 	} else {
261 		dirp->dd_len = incr;
262 		dirp->dd_buf = malloc(dirp->dd_len);
263 		if (dirp->dd_buf == NULL) {
264 			free(dirp);
265 			close (fd);
266 			return (NULL);
267 		}
268 		dirp->dd_seek = 0;
269 		flags &= ~DTF_REWIND;
270 	}
271 
272 	dirp->dd_loc = 0;
273 	dirp->dd_fd = fd;
274 	dirp->dd_flags = flags;
275 
276 	/*
277 	 * Set up seek point for rewinddir.
278 	 */
279 	dirp->dd_rewind = telldir(dirp);
280 
281 	return (dirp);
282 }
283