du.c revision 144840 
1/*
2 * Copyright (c) 1989, 1993, 1994
3 *	The Regents of the University of California.  All rights reserved.
4 *
5 * This code is derived from software contributed to Berkeley by
6 * Chris Newcomb.
7 *
8 * Redistribution and use in source and binary forms, with or without
9 * modification, are permitted provided that the following conditions
10 * are met:
11 * 1. Redistributions of source code must retain the above copyright
12 *    notice, this list of conditions and the following disclaimer.
13 * 2. Redistributions in binary form must reproduce the above copyright
14 *    notice, this list of conditions and the following disclaimer in the
15 *    documentation and/or other materials provided with the distribution.
16 * 3. All advertising materials mentioning features or use of this software
17 *    must display the following acknowledgement:
18 *	This product includes software developed by the University of
19 *	California, Berkeley and its contributors.
20 * 4. Neither the name of the University nor the names of its contributors
21 *    may be used to endorse or promote products derived from this software
22 *    without specific prior written permission.
23 *
24 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
25 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
26 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
27 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
28 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
29 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
30 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
31 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
32 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
33 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
34 * SUCH DAMAGE.
35 */
36
37#ifndef lint
38static const char copyright[] =
39"@(#) Copyright (c) 1989, 1993, 1994\n\
40	The Regents of the University of California.  All rights reserved.\n";
41#endif /* not lint */
42
43#ifndef lint
44#if 0
45static const char sccsid[] = "@(#)du.c	8.5 (Berkeley) 5/4/95";
46#endif
47#endif /* not lint */
48#include <sys/cdefs.h>
49__FBSDID("$FreeBSD: head/usr.bin/du/du.c 144840 2005-04-09 14:31:41Z stefanf $");
50
51#include <sys/param.h>
52#include <sys/queue.h>
53#include <sys/stat.h>
54
55#include <err.h>
56#include <errno.h>
57#include <fnmatch.h>
58#include <fts.h>
59#include <libutil.h>
60#include <locale.h>
61#include <stdint.h>
62#include <stdio.h>
63#include <stdlib.h>
64#include <string.h>
65#include <sysexits.h>
66#include <unistd.h>
67
68SLIST_HEAD(ignhead, ignentry) ignores;
69struct ignentry {
70	char			*mask;
71	SLIST_ENTRY(ignentry)	next;
72};
73
74static int	linkchk(FTSENT *);
75static void	usage(void);
76void		prthumanval(int64_t);
77void		ignoreadd(const char *);
78void		ignoreclean(void);
79int		ignorep(FTSENT *);
80
81int
82main(int argc, char *argv[])
83{
84	FTS		*fts;
85	FTSENT		*p;
86	off_t		savednumber = 0;
87	long		blocksize;
88	int		ftsoptions;
89	int		listall;
90	int		depth;
91	int		Hflag, Lflag, Pflag, aflag, sflag, dflag, cflag, hflag, ch, notused, rval;
92	char 		**save;
93	static char	dot[] = ".";
94
95	setlocale(LC_ALL, "");
96
97	Hflag = Lflag = Pflag = aflag = sflag = dflag = cflag = hflag = 0;
98
99	save = argv;
100	ftsoptions = 0;
101	depth = INT_MAX;
102	SLIST_INIT(&ignores);
103
104	while ((ch = getopt(argc, argv, "HI:LPasd:chkmrx")) != -1)
105		switch (ch) {
106			case 'H':
107				Hflag = 1;
108				break;
109			case 'I':
110				ignoreadd(optarg);
111				break;
112			case 'L':
113				if (Pflag)
114					usage();
115				Lflag = 1;
116				break;
117			case 'P':
118				if (Lflag)
119					usage();
120				Pflag = 1;
121				break;
122			case 'a':
123				aflag = 1;
124				break;
125			case 's':
126				sflag = 1;
127				break;
128			case 'd':
129				dflag = 1;
130				errno = 0;
131				depth = atoi(optarg);
132				if (errno == ERANGE || depth < 0) {
133					warnx("invalid argument to option d: %s", optarg);
134					usage();
135				}
136				break;
137			case 'c':
138				cflag = 1;
139				break;
140			case 'h':
141				putenv("BLOCKSIZE=512");
142				hflag = 1;
143				break;
144			case 'k':
145				hflag = 0;
146				putenv("BLOCKSIZE=1024");
147				break;
148			case 'm':
149				hflag = 0;
150				putenv("BLOCKSIZE=1048576");
151				break;
152			case 'r':		 /* Compatibility. */
153				break;
154			case 'x':
155				ftsoptions |= FTS_XDEV;
156				break;
157			case '?':
158			default:
159				usage();
160		}
161
162	argc -= optind;
163	argv += optind;
164
165	/*
166	 * XXX
167	 * Because of the way that fts(3) works, logical walks will not count
168	 * the blocks actually used by symbolic links.  We rationalize this by
169	 * noting that users computing logical sizes are likely to do logical
170	 * copies, so not counting the links is correct.  The real reason is
171	 * that we'd have to re-implement the kernel's symbolic link traversing
172	 * algorithm to get this right.  If, for example, you have relative
173	 * symbolic links referencing other relative symbolic links, it gets
174	 * very nasty, very fast.  The bottom line is that it's documented in
175	 * the man page, so it's a feature.
176	 */
177
178	if (Hflag + Lflag + Pflag > 1)
179		usage();
180
181	if (Hflag + Lflag + Pflag == 0)
182		Pflag = 1;			/* -P (physical) is default */
183
184	if (Hflag)
185		ftsoptions |= FTS_COMFOLLOW;
186
187	if (Lflag)
188		ftsoptions |= FTS_LOGICAL;
189
190	if (Pflag)
191		ftsoptions |= FTS_PHYSICAL;
192
193	listall = 0;
194
195	if (aflag) {
196		if (sflag || dflag)
197			usage();
198		listall = 1;
199	} else if (sflag) {
200		if (dflag)
201			usage();
202		depth = 0;
203	}
204
205	if (!*argv) {
206		argv = save;
207		argv[0] = dot;
208		argv[1] = NULL;
209	}
210
211	(void) getbsize(&notused, &blocksize);
212	blocksize /= 512;
213
214	rval = 0;
215
216	if ((fts = fts_open(argv, ftsoptions, NULL)) == NULL)
217		err(1, "fts_open");
218
219	while ((p = fts_read(fts)) != NULL) {
220		switch (p->fts_info) {
221			case FTS_D:			/* Ignore. */
222				if (ignorep(p))
223					fts_set(fts, p, FTS_SKIP);
224				break;
225			case FTS_DP:
226				if (ignorep(p))
227					break;
228
229				p->fts_parent->fts_bignum +=
230				    p->fts_bignum += p->fts_statp->st_blocks;
231
232				if (p->fts_level <= depth) {
233					if (hflag) {
234						(void) prthumanval(howmany(p->fts_bignum, blocksize));
235						(void) printf("\t%s\n", p->fts_path);
236					} else {
237					(void) printf("%jd\t%s\n",
238					    (intmax_t)howmany(p->fts_bignum, blocksize),
239					    p->fts_path);
240					}
241				}
242				break;
243			case FTS_DC:			/* Ignore. */
244				break;
245			case FTS_DNR:			/* Warn, continue. */
246			case FTS_ERR:
247			case FTS_NS:
248				warnx("%s: %s", p->fts_path, strerror(p->fts_errno));
249				rval = 1;
250				break;
251			default:
252				if (ignorep(p))
253					break;
254
255				if (p->fts_statp->st_nlink > 1 && linkchk(p))
256					break;
257
258				if (listall || p->fts_level == 0) {
259					if (hflag) {
260						(void) prthumanval(howmany(p->fts_statp->st_blocks,
261							blocksize));
262						(void) printf("\t%s\n", p->fts_path);
263					} else {
264						(void) printf("%jd\t%s\n",
265							(intmax_t)howmany(p->fts_statp->st_blocks, blocksize),
266							p->fts_path);
267					}
268				}
269
270				p->fts_parent->fts_bignum += p->fts_statp->st_blocks;
271		}
272		savednumber = p->fts_parent->fts_bignum;
273	}
274
275	if (errno)
276		err(1, "fts_read");
277
278	if (cflag) {
279		if (hflag) {
280			(void) prthumanval(howmany(savednumber, blocksize));
281			(void) printf("\ttotal\n");
282		} else {
283			(void) printf("%jd\ttotal\n", (intmax_t)howmany(savednumber, blocksize));
284		}
285	}
286
287	ignoreclean();
288	exit(rval);
289}
290
291static int
292linkchk(FTSENT *p)
293{
294	struct links_entry {
295		struct links_entry *next;
296		struct links_entry *previous;
297		int	 links;
298		dev_t	 dev;
299		ino_t	 ino;
300	};
301	static const size_t links_hash_initial_size = 8192;
302	static struct links_entry **buckets;
303	static struct links_entry *free_list;
304	static size_t number_buckets;
305	static unsigned long number_entries;
306	static char stop_allocating;
307	struct links_entry *le, **new_buckets;
308	struct stat *st;
309	size_t i, new_size;
310	int hash;
311
312	st = p->fts_statp;
313
314	/* If necessary, initialize the hash table. */
315	if (buckets == NULL) {
316		number_buckets = links_hash_initial_size;
317		buckets = malloc(number_buckets * sizeof(buckets[0]));
318		if (buckets == NULL)
319			errx(1, "No memory for hardlink detection");
320		for (i = 0; i < number_buckets; i++)
321			buckets[i] = NULL;
322	}
323
324	/* If the hash table is getting too full, enlarge it. */
325	if (number_entries > number_buckets * 10 && !stop_allocating) {
326		new_size = number_buckets * 2;
327		new_buckets = malloc(new_size * sizeof(struct links_entry *));
328
329		/* Try releasing the free list to see if that helps. */
330		if (new_buckets == NULL && free_list != NULL) {
331			while (free_list != NULL) {
332				le = free_list;
333				free_list = le->next;
334				free(le);
335			}
336			new_buckets = malloc(new_size * sizeof(new_buckets[0]));
337		}
338
339		if (new_buckets == NULL) {
340			stop_allocating = 1;
341			warnx("No more memory for tracking hard links");
342		} else {
343			memset(new_buckets, 0,
344			    new_size * sizeof(struct links_entry *));
345			for (i = 0; i < number_buckets; i++) {
346				while (buckets[i] != NULL) {
347					/* Remove entry from old bucket. */
348					le = buckets[i];
349					buckets[i] = le->next;
350
351					/* Add entry to new bucket. */
352					hash = (le->dev ^ le->ino) % new_size;
353
354					if (new_buckets[hash] != NULL)
355						new_buckets[hash]->previous =
356						    le;
357					le->next = new_buckets[hash];
358					le->previous = NULL;
359					new_buckets[hash] = le;
360				}
361			}
362			free(buckets);
363			buckets = new_buckets;
364			number_buckets = new_size;
365		}
366	}
367
368	/* Try to locate this entry in the hash table. */
369	hash = ( st->st_dev ^ st->st_ino ) % number_buckets;
370	for (le = buckets[hash]; le != NULL; le = le->next) {
371		if (le->dev == st->st_dev && le->ino == st->st_ino) {
372			/*
373			 * Save memory by releasing an entry when we've seen
374			 * all of it's links.
375			 */
376			if (--le->links <= 0) {
377				if (le->previous != NULL)
378					le->previous->next = le->next;
379				if (le->next != NULL)
380					le->next->previous = le->previous;
381				if (buckets[hash] == le)
382					buckets[hash] = le->next;
383				number_entries--;
384				/* Recycle this node through the free list */
385				if (stop_allocating) {
386					free(le);
387				} else {
388					le->next = free_list;
389					free_list = le;
390				}
391			}
392			return (1);
393		}
394	}
395
396	if (stop_allocating)
397		return (0);
398
399	/* Add this entry to the links cache. */
400	if (free_list != NULL) {
401		/* Pull a node from the free list if we can. */
402		le = free_list;
403		free_list = le->next;
404	} else
405		/* Malloc one if we have to. */
406		le = malloc(sizeof(struct links_entry));
407	if (le == NULL) {
408		stop_allocating = 1;
409		warnx("No more memory for tracking hard links");
410		return (0);
411	}
412	le->dev = st->st_dev;
413	le->ino = st->st_ino;
414	le->links = st->st_nlink - 1;
415	number_entries++;
416	le->next = buckets[hash];
417	le->previous = NULL;
418	if (buckets[hash] != NULL)
419		buckets[hash]->previous = le;
420	buckets[hash] = le;
421	return (0);
422}
423
424void
425prthumanval(int64_t bytes)
426{
427	char buf[5];
428
429	bytes *= DEV_BSIZE;
430
431	humanize_number(buf, sizeof(buf), bytes, "", HN_AUTOSCALE,
432	    HN_B | HN_NOSPACE | HN_DECIMAL);
433
434	(void)printf("%4s", buf);
435}
436
437static void
438usage(void)
439{
440	(void)fprintf(stderr,
441		"usage: du [-H | -L | -P] [-a | -s | -d depth] [-c] [-h | -k | -m] [-x] [-I mask] [file ...]\n");
442	exit(EX_USAGE);
443}
444
445void
446ignoreadd(const char *mask)
447{
448	struct ignentry *ign;
449
450	ign = calloc(1, sizeof(*ign));
451	if (ign == NULL)
452		errx(1, "cannot allocate memory");
453	ign->mask = strdup(mask);
454	if (ign->mask == NULL)
455		errx(1, "cannot allocate memory");
456	SLIST_INSERT_HEAD(&ignores, ign, next);
457}
458
459void
460ignoreclean(void)
461{
462	struct ignentry *ign;
463
464	while (!SLIST_EMPTY(&ignores)) {
465		ign = SLIST_FIRST(&ignores);
466		SLIST_REMOVE_HEAD(&ignores, next);
467		free(ign->mask);
468		free(ign);
469	}
470}
471
472int
473ignorep(FTSENT *ent)
474{
475	struct ignentry *ign;
476
477	SLIST_FOREACH(ign, &ignores, next)
478		if (fnmatch(ign->mask, ent->fts_name, 0) != FNM_NOMATCH)
479			return 1;
480	return 0;
481}
482