2 Unix SMB/CIFS implementation.
3 Samba database functions
4 Copyright (C) Andrew Tridgell 1999-2000
5 Copyright (C) Luke Kenneth Casson Leighton 2000
6 Copyright (C) Paul `Rusty' Russell 2000
7 Copyright (C) Jeremy Allison 2000
9 This program is free software; you can redistribute it and/or modify
10 it under the terms of the GNU General Public License as published by
11 the Free Software Foundation; either version 2 of the License, or
12 (at your option) any later version.
14 This program is distributed in the hope that it will be useful,
15 but WITHOUT ANY WARRANTY; without even the implied warranty of
16 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
17 GNU General Public License for more details.
19 You should have received a copy of the GNU General Public License
20 along with this program; if not, write to the Free Software
21 Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
43 #define TDB_MAGIC_FOOD "TDB file\n"
44 #define TDB_VERSION (0x26011967 + 6)
45 #define TDB_MAGIC (0x26011999U)
46 #define TDB_FREE_MAGIC (~TDB_MAGIC)
47 #define TDB_DEAD_MAGIC (0xFEE1DEAD)
48 #define TDB_ALIGNMENT 4
49 #define MIN_REC_SIZE (2*sizeof(struct list_struct) + TDB_ALIGNMENT)
50 #define DEFAULT_HASH_SIZE 131
51 #define TDB_PAGE_SIZE 0x2000
52 #define FREELIST_TOP (sizeof(struct tdb_header))
53 #define TDB_ALIGN(x,a) (((x) + (a)-1) & ~((a)-1))
54 #define TDB_BYTEREV(x) (((((x)&0xff)<<24)|((x)&0xFF00)<<8)|(((x)>>8)&0xFF00)|((x)>>24))
55 #define TDB_DEAD(r) ((r)->magic == TDB_DEAD_MAGIC)
56 #define TDB_BAD_MAGIC(r) ((r)->magic != TDB_MAGIC && !TDB_DEAD(r))
57 #define TDB_HASH_TOP(hash) (FREELIST_TOP + (BUCKET(hash)+1)*sizeof(tdb_off))
59 /* NB assumes there is a local variable called "tdb" that is the
60 * current context, also takes doubly-parenthesized print-style
62 #define TDB_LOG(x) (tdb->log_fn?((tdb->log_fn x),0) : 0)
73 #define MAP_FAILED ((void *)-1)
76 /* free memory if the pointer is valid and zero the pointer */
78 #define SAFE_FREE(x) do { if ((x) != NULL) {free((x)); (x)=NULL;} } while(0)
81 #define BUCKET(hash) ((hash) % tdb->header.hash_size)
84 /* all contexts, to ensure no double-opens (fcntl locks don't nest!) */
85 static TDB_CONTEXT *tdbs = NULL;
87 static int tdb_munmap(TDB_CONTEXT *tdb)
89 if (tdb->flags & TDB_INTERNAL)
94 int ret = munmap(tdb->map_ptr, tdb->map_size);
103 static void tdb_mmap(TDB_CONTEXT *tdb)
105 if (tdb->flags & TDB_INTERNAL)
109 if (!(tdb->flags & TDB_NOMMAP)) {
110 tdb->map_ptr = mmap(NULL, tdb->map_size,
111 PROT_READ|(tdb->read_only? 0:PROT_WRITE),
112 MAP_SHARED|MAP_FILE, tdb->fd, 0);
115 * NB. When mmap fails it returns MAP_FAILED *NOT* NULL !!!!
118 if (tdb->map_ptr == MAP_FAILED) {
120 TDB_LOG((tdb, 2, "tdb_mmap failed for size %d (%s)\n",
121 tdb->map_size, strerror(errno)));
131 /* Endian conversion: we only ever deal with 4 byte quantities */
132 static void *convert(void *buf, u32 size)
135 for (i = 0; i < size / 4; i++)
136 p[i] = TDB_BYTEREV(p[i]);
139 #define DOCONV() (tdb->flags & TDB_CONVERT)
140 #define CONVERT(x) (DOCONV() ? convert(&x, sizeof(x)) : &x)
142 /* the body of the database is made of one list_struct for the free space
143 plus a separate data list for each hash value */
145 tdb_off next; /* offset of the next record in the list */
146 tdb_len rec_len; /* total byte length of record */
147 tdb_len key_len; /* byte length of key */
148 tdb_len data_len; /* byte length of data */
149 u32 full_hash; /* the full 32 bit hash of the key */
150 u32 magic; /* try to catch errors */
151 /* the following union is implied:
153 char record[rec_len];
158 u32 totalsize; (tailer)
163 /* a byte range locking function - return 0 on success
164 this functions locks/unlocks 1 byte at the specified offset.
166 On error, errno is also set so that errors are passed back properly
167 through tdb_open(). */
168 static int tdb_brlock(TDB_CONTEXT *tdb, tdb_off offset,
169 int rw_type, int lck_type, int probe)
174 if (tdb->flags & TDB_NOLOCK)
176 if (tdb->read_only) {
182 fl.l_whence = SEEK_SET;
188 ret = fcntl(tdb->fd,lck_type,&fl);
189 } while (ret == -1 && errno == EINTR);
192 if (!probe && lck_type != F_SETLK) {
193 TDB_LOG((tdb, 5,"tdb_brlock failed (fd=%d) at offset %d rw_type=%d lck_type=%d\n",
194 tdb->fd, offset, rw_type, lck_type));
196 /* errno set by fcntl */
197 return TDB_ERRCODE(TDB_ERR_LOCK, -1);
202 /* lock a list in the database. list -1 is the alloc list */
203 static int tdb_lock(TDB_CONTEXT *tdb, int list, int ltype)
205 if (list < -1 || list >= (int)tdb->header.hash_size) {
206 TDB_LOG((tdb, 0,"tdb_lock: invalid list %d for ltype=%d\n",
210 if (tdb->flags & TDB_NOLOCK)
213 /* Since fcntl locks don't nest, we do a lock for the first one,
214 and simply bump the count for future ones */
215 if (tdb->locked[list+1].count == 0) {
216 if (!tdb->read_only && tdb->header.rwlocks) {
217 if (tdb_spinlock(tdb, list, ltype)) {
218 TDB_LOG((tdb, 0, "tdb_lock spinlock failed on list ltype=%d\n",
222 } else if (tdb_brlock(tdb,FREELIST_TOP+4*list,ltype,F_SETLKW, 0)) {
223 TDB_LOG((tdb, 0,"tdb_lock failed on list %d ltype=%d (%s)\n",
224 list, ltype, strerror(errno)));
227 tdb->locked[list+1].ltype = ltype;
229 tdb->locked[list+1].count++;
233 /* unlock the database: returns void because it's too late for errors. */
234 /* changed to return int it may be interesting to know there
235 has been an error --simo */
236 static int tdb_unlock(TDB_CONTEXT *tdb, int list, int ltype)
240 if (tdb->flags & TDB_NOLOCK)
244 if (list < -1 || list >= (int)tdb->header.hash_size) {
245 TDB_LOG((tdb, 0, "tdb_unlock: list %d invalid (%d)\n", list, tdb->header.hash_size));
249 if (tdb->locked[list+1].count==0) {
250 TDB_LOG((tdb, 0, "tdb_unlock: count is 0\n"));
254 if (tdb->locked[list+1].count == 1) {
255 /* Down to last nested lock: unlock underneath */
256 if (!tdb->read_only && tdb->header.rwlocks) {
257 ret = tdb_spinunlock(tdb, list, ltype);
259 ret = tdb_brlock(tdb, FREELIST_TOP+4*list, F_UNLCK, F_SETLKW, 0);
264 tdb->locked[list+1].count--;
267 TDB_LOG((tdb, 0,"tdb_unlock: An error occurred unlocking!\n"));
271 /* This is based on the hash algorithm from gdbm */
272 static u32 tdb_hash(TDB_DATA *key)
274 u32 value; /* Used to compute the hash value. */
275 u32 i; /* Used to cycle through random values. */
277 /* Set the initial value from the key size. */
278 for (value = 0x238F13AF * key->dsize, i=0; i < key->dsize; i++)
279 value = (value + (key->dptr[i] << (i*5 % 24)));
281 return (1103515243 * value + 12345);
284 /* check for an out of bounds access - if it is out of bounds then
285 see if the database has been expanded by someone else and expand
287 note that "len" is the minimum length needed for the db
289 static int tdb_oob(TDB_CONTEXT *tdb, tdb_off len, int probe)
292 if (len <= tdb->map_size)
294 if (tdb->flags & TDB_INTERNAL) {
296 TDB_LOG((tdb, 0,"tdb_oob len %d beyond internal malloc size %d\n",
297 (int)len, (int)tdb->map_size));
299 return TDB_ERRCODE(TDB_ERR_IO, -1);
302 if (fstat(tdb->fd, &st) == -1)
303 return TDB_ERRCODE(TDB_ERR_IO, -1);
305 if (st.st_size < (size_t)len) {
307 TDB_LOG((tdb, 0,"tdb_oob len %d beyond eof at %d\n",
308 (int)len, (int)st.st_size));
310 return TDB_ERRCODE(TDB_ERR_IO, -1);
313 /* Unmap, update size, remap */
314 if (tdb_munmap(tdb) == -1)
315 return TDB_ERRCODE(TDB_ERR_IO, -1);
316 tdb->map_size = st.st_size;
321 /* write a lump of data at a specified offset */
322 static int tdb_write(TDB_CONTEXT *tdb, tdb_off off, void *buf, tdb_len len)
324 if (tdb_oob(tdb, off + len, 0) != 0)
328 memcpy(off + (char *)tdb->map_ptr, buf, len);
330 else if (pwrite(tdb->fd, buf, len, off) != (ssize_t)len) {
332 else if (lseek(tdb->fd, off, SEEK_SET) != off
333 || write(tdb->fd, buf, len) != (ssize_t)len) {
335 TDB_LOG((tdb, 0,"tdb_write failed at %d len=%d (%s)\n",
336 off, len, strerror(errno)));
337 return TDB_ERRCODE(TDB_ERR_IO, -1);
342 /* read a lump of data at a specified offset, maybe convert */
343 static int tdb_read(TDB_CONTEXT *tdb,tdb_off off,void *buf,tdb_len len,int cv)
345 if (tdb_oob(tdb, off + len, 0) != 0)
349 memcpy(buf, off + (char *)tdb->map_ptr, len);
351 else if (pread(tdb->fd, buf, len, off) != (ssize_t)len) {
353 else if (lseek(tdb->fd, off, SEEK_SET) != off
354 || read(tdb->fd, buf, len) != (ssize_t)len) {
356 TDB_LOG((tdb, 0,"tdb_read failed at %d len=%d (%s)\n",
357 off, len, strerror(errno)));
358 return TDB_ERRCODE(TDB_ERR_IO, -1);
365 /* read a lump of data, allocating the space for it */
366 static char *tdb_alloc_read(TDB_CONTEXT *tdb, tdb_off offset, tdb_len len)
370 if (!(buf = malloc(len))) {
371 TDB_LOG((tdb, 0,"tdb_alloc_read malloc failed len=%d (%s)\n",
372 len, strerror(errno)));
373 return TDB_ERRCODE(TDB_ERR_OOM, buf);
375 if (tdb_read(tdb, offset, buf, len, 0) == -1) {
382 /* read/write a tdb_off */
383 static int ofs_read(TDB_CONTEXT *tdb, tdb_off offset, tdb_off *d)
385 return tdb_read(tdb, offset, (char*)d, sizeof(*d), DOCONV());
387 static int ofs_write(TDB_CONTEXT *tdb, tdb_off offset, tdb_off *d)
390 return tdb_write(tdb, offset, CONVERT(off), sizeof(*d));
393 /* read/write a record */
394 static int rec_read(TDB_CONTEXT *tdb, tdb_off offset, struct list_struct *rec)
396 if (tdb_read(tdb, offset, rec, sizeof(*rec),DOCONV()) == -1)
398 if (TDB_BAD_MAGIC(rec)) {
399 TDB_LOG((tdb, 0,"rec_read bad magic 0x%x at offset=%d\n", rec->magic, offset));
400 return TDB_ERRCODE(TDB_ERR_CORRUPT, -1);
402 return tdb_oob(tdb, rec->next+sizeof(*rec), 0);
404 static int rec_write(TDB_CONTEXT *tdb, tdb_off offset, struct list_struct *rec)
406 struct list_struct r = *rec;
407 return tdb_write(tdb, offset, CONVERT(r), sizeof(r));
410 /* read a freelist record and check for simple errors */
411 static int rec_free_read(TDB_CONTEXT *tdb, tdb_off off, struct list_struct *rec)
413 if (tdb_read(tdb, off, rec, sizeof(*rec),DOCONV()) == -1)
416 if (rec->magic == TDB_MAGIC) {
417 /* this happens when a app is showdown while deleting a record - we should
418 not completely fail when this happens */
419 TDB_LOG((tdb, 0,"rec_free_read non-free magic at offset=%d - fixing\n",
421 rec->magic = TDB_FREE_MAGIC;
422 if (tdb_write(tdb, off, rec, sizeof(*rec)) == -1)
426 if (rec->magic != TDB_FREE_MAGIC) {
427 TDB_LOG((tdb, 0,"rec_free_read bad magic 0x%x at offset=%d\n",
429 return TDB_ERRCODE(TDB_ERR_CORRUPT, -1);
431 if (tdb_oob(tdb, rec->next+sizeof(*rec), 0) != 0)
436 /* update a record tailer (must hold allocation lock) */
437 static int update_tailer(TDB_CONTEXT *tdb, tdb_off offset,
438 const struct list_struct *rec)
442 /* Offset of tailer from record header */
443 totalsize = sizeof(*rec) + rec->rec_len;
444 return ofs_write(tdb, offset + totalsize - sizeof(tdb_off),
448 static tdb_off tdb_dump_record(TDB_CONTEXT *tdb, tdb_off offset)
450 struct list_struct rec;
451 tdb_off tailer_ofs, tailer;
453 if (tdb_read(tdb, offset, (char *)&rec, sizeof(rec), DOCONV()) == -1) {
454 printf("ERROR: failed to read record at %u\n", offset);
458 printf(" rec: offset=%u next=%d rec_len=%d key_len=%d data_len=%d full_hash=0x%x magic=0x%x\n",
459 offset, rec.next, rec.rec_len, rec.key_len, rec.data_len, rec.full_hash, rec.magic);
461 tailer_ofs = offset + sizeof(rec) + rec.rec_len - sizeof(tdb_off);
462 if (ofs_read(tdb, tailer_ofs, &tailer) == -1) {
463 printf("ERROR: failed to read tailer at %u\n", tailer_ofs);
467 if (tailer != rec.rec_len + sizeof(rec)) {
468 printf("ERROR: tailer does not match record! tailer=%u totalsize=%u\n",
469 (unsigned)tailer, (unsigned)(rec.rec_len + sizeof(rec)));
474 static int tdb_dump_chain(TDB_CONTEXT *tdb, int i)
476 tdb_off rec_ptr, top;
478 top = TDB_HASH_TOP(i);
480 if (tdb_lock(tdb, i, F_WRLCK) != 0)
483 if (ofs_read(tdb, top, &rec_ptr) == -1)
484 return tdb_unlock(tdb, i, F_WRLCK);
487 printf("hash=%d\n", i);
490 rec_ptr = tdb_dump_record(tdb, rec_ptr);
493 return tdb_unlock(tdb, i, F_WRLCK);
496 void tdb_dump_all(TDB_CONTEXT *tdb)
499 for (i=0;i<tdb->header.hash_size;i++) {
500 tdb_dump_chain(tdb, i);
502 printf("freelist:\n");
503 tdb_dump_chain(tdb, -1);
506 int tdb_printfreelist(TDB_CONTEXT *tdb)
510 tdb_off offset, rec_ptr;
511 struct list_struct rec;
513 if ((ret = tdb_lock(tdb, -1, F_WRLCK)) != 0)
516 offset = FREELIST_TOP;
518 /* read in the freelist top */
519 if (ofs_read(tdb, offset, &rec_ptr) == -1) {
520 tdb_unlock(tdb, -1, F_WRLCK);
524 printf("freelist top=[0x%08x]\n", rec_ptr );
526 if (tdb_read(tdb, rec_ptr, (char *)&rec, sizeof(rec), DOCONV()) == -1) {
527 tdb_unlock(tdb, -1, F_WRLCK);
531 if (rec.magic != TDB_FREE_MAGIC) {
532 printf("bad magic 0x%08x in free list\n", rec.magic);
533 tdb_unlock(tdb, -1, F_WRLCK);
537 printf("entry offset=[0x%08x], rec.rec_len = [0x%08x (%d)]\n", rec.next, rec.rec_len, rec.rec_len );
538 total_free += rec.rec_len;
540 /* move to the next record */
543 printf("total rec_len = [0x%08x (%d)]\n", (int)total_free,
546 return tdb_unlock(tdb, -1, F_WRLCK);
549 /* Remove an element from the freelist. Must have alloc lock. */
550 static int remove_from_freelist(TDB_CONTEXT *tdb, tdb_off off, tdb_off next)
554 /* read in the freelist top */
555 last_ptr = FREELIST_TOP;
556 while (ofs_read(tdb, last_ptr, &i) != -1 && i != 0) {
558 /* We've found it! */
559 return ofs_write(tdb, last_ptr, &next);
561 /* Follow chain (next offset is at start of record) */
564 TDB_LOG((tdb, 0,"remove_from_freelist: not on list at off=%d\n", off));
565 return TDB_ERRCODE(TDB_ERR_CORRUPT, -1);
568 /* Add an element into the freelist. Merge adjacent records if
570 static int tdb_free(TDB_CONTEXT *tdb, tdb_off offset, struct list_struct *rec)
574 /* Allocation and tailer lock */
575 if (tdb_lock(tdb, -1, F_WRLCK) != 0)
578 /* set an initial tailer, so if we fail we don't leave a bogus record */
579 if (update_tailer(tdb, offset, rec) != 0) {
580 TDB_LOG((tdb, 0, "tdb_free: upfate_tailer failed!\n"));
584 /* Look right first (I'm an Australian, dammit) */
585 right = offset + sizeof(*rec) + rec->rec_len;
586 if (right + sizeof(*rec) <= tdb->map_size) {
587 struct list_struct r;
589 if (tdb_read(tdb, right, &r, sizeof(r), DOCONV()) == -1) {
590 TDB_LOG((tdb, 0, "tdb_free: right read failed at %u\n", right));
594 /* If it's free, expand to include it. */
595 if (r.magic == TDB_FREE_MAGIC) {
596 if (remove_from_freelist(tdb, right, r.next) == -1) {
597 TDB_LOG((tdb, 0, "tdb_free: right free failed at %u\n", right));
600 rec->rec_len += sizeof(r) + r.rec_len;
606 left = offset - sizeof(tdb_off);
607 if (left > TDB_HASH_TOP(tdb->header.hash_size-1)) {
608 struct list_struct l;
611 /* Read in tailer and jump back to header */
612 if (ofs_read(tdb, left, &leftsize) == -1) {
613 TDB_LOG((tdb, 0, "tdb_free: left offset read failed at %u\n", left));
616 left = offset - leftsize;
618 /* Now read in record */
619 if (tdb_read(tdb, left, &l, sizeof(l), DOCONV()) == -1) {
620 TDB_LOG((tdb, 0, "tdb_free: left read failed at %u (%u)\n", left, leftsize));
624 /* If it's free, expand to include it. */
625 if (l.magic == TDB_FREE_MAGIC) {
626 if (remove_from_freelist(tdb, left, l.next) == -1) {
627 TDB_LOG((tdb, 0, "tdb_free: left free failed at %u\n", left));
631 rec->rec_len += leftsize;
637 if (update_tailer(tdb, offset, rec) == -1) {
638 TDB_LOG((tdb, 0, "tdb_free: update_tailer failed at %u\n", offset));
642 /* Now, prepend to free list */
643 rec->magic = TDB_FREE_MAGIC;
645 if (ofs_read(tdb, FREELIST_TOP, &rec->next) == -1 ||
646 rec_write(tdb, offset, rec) == -1 ||
647 ofs_write(tdb, FREELIST_TOP, &offset) == -1) {
648 TDB_LOG((tdb, 0, "tdb_free record write failed at offset=%d\n", offset));
652 /* And we're done. */
653 tdb_unlock(tdb, -1, F_WRLCK);
657 tdb_unlock(tdb, -1, F_WRLCK);
662 /* expand a file. we prefer to use ftruncate, as that is what posix
663 says to use for mmap expansion */
664 static int expand_file(TDB_CONTEXT *tdb, tdb_off size, tdb_off addition)
667 #if HAVE_FTRUNCATE_EXTEND
668 if (ftruncate(tdb->fd, size+addition) != 0) {
669 TDB_LOG((tdb, 0, "expand_file ftruncate to %d failed (%s)\n",
670 size+addition, strerror(errno)));
677 if (pwrite(tdb->fd, &b, 1, (size+addition) - 1) != 1) {
679 if (lseek(tdb->fd, (size+addition) - 1, SEEK_SET) != (size+addition) - 1 ||
680 write(tdb->fd, &b, 1) != 1) {
682 TDB_LOG((tdb, 0, "expand_file to %d failed (%s)\n",
683 size+addition, strerror(errno)));
688 /* now fill the file with something. This ensures that the file isn't sparse, which would be
689 very bad if we ran out of disk. This must be done with write, not via mmap */
690 memset(buf, 0x42, sizeof(buf));
692 int n = addition>sizeof(buf)?sizeof(buf):addition;
694 int ret = pwrite(tdb->fd, buf, n, size);
697 if (lseek(tdb->fd, size, SEEK_SET) != size)
699 ret = write(tdb->fd, buf, n);
702 TDB_LOG((tdb, 0, "expand_file write of %d failed (%s)\n",
703 n, strerror(errno)));
713 /* expand the database at least size bytes by expanding the underlying
714 file and doing the mmap again if necessary */
715 static int tdb_expand(TDB_CONTEXT *tdb, tdb_off size)
717 struct list_struct rec;
720 if (tdb_lock(tdb, -1, F_WRLCK) == -1) {
721 TDB_LOG((tdb, 0, "lock failed in tdb_expand\n"));
725 /* must know about any previous expansions by another process */
726 tdb_oob(tdb, tdb->map_size + 1, 1);
728 /* always make room for at least 10 more records, and round
729 the database up to a multiple of TDB_PAGE_SIZE */
730 size = TDB_ALIGN(tdb->map_size + size*10, TDB_PAGE_SIZE) - tdb->map_size;
732 if (!(tdb->flags & TDB_INTERNAL))
736 * We must ensure the file is unmapped before doing this
737 * to ensure consistency with systems like OpenBSD where
738 * writes and mmaps are not consistent.
741 /* expand the file itself */
742 if (!(tdb->flags & TDB_INTERNAL)) {
743 if (expand_file(tdb, tdb->map_size, size) != 0)
747 tdb->map_size += size;
749 if (tdb->flags & TDB_INTERNAL)
750 tdb->map_ptr = realloc(tdb->map_ptr, tdb->map_size);
753 * We must ensure the file is remapped before adding the space
754 * to ensure consistency with systems like OpenBSD where
755 * writes and mmaps are not consistent.
758 /* We're ok if the mmap fails as we'll fallback to read/write */
762 /* form a new freelist record */
763 memset(&rec,'\0',sizeof(rec));
764 rec.rec_len = size - sizeof(rec);
766 /* link it into the free list */
767 offset = tdb->map_size - size;
768 if (tdb_free(tdb, offset, &rec) == -1)
771 tdb_unlock(tdb, -1, F_WRLCK);
774 tdb_unlock(tdb, -1, F_WRLCK);
778 /* allocate some space from the free list. The offset returned points
779 to a unconnected list_struct within the database with room for at
780 least length bytes of total data
782 0 is returned if the space could not be allocated
784 static tdb_off tdb_allocate(TDB_CONTEXT *tdb, tdb_len length,
785 struct list_struct *rec)
787 tdb_off rec_ptr, last_ptr, newrec_ptr;
788 struct list_struct newrec;
790 if (tdb_lock(tdb, -1, F_WRLCK) == -1)
793 /* Extra bytes required for tailer */
794 length += sizeof(tdb_off);
797 last_ptr = FREELIST_TOP;
799 /* read in the freelist top */
800 if (ofs_read(tdb, FREELIST_TOP, &rec_ptr) == -1)
803 /* keep looking until we find a freelist record big enough */
805 if (rec_free_read(tdb, rec_ptr, rec) == -1)
808 if (rec->rec_len >= length) {
809 /* found it - now possibly split it up */
810 if (rec->rec_len > length + MIN_REC_SIZE) {
811 /* Length of left piece */
812 length = TDB_ALIGN(length, TDB_ALIGNMENT);
814 /* Right piece to go on free list */
815 newrec.rec_len = rec->rec_len
816 - (sizeof(*rec) + length);
817 newrec_ptr = rec_ptr + sizeof(*rec) + length;
819 /* And left record is shortened */
820 rec->rec_len = length;
824 /* Remove allocated record from the free list */
825 if (ofs_write(tdb, last_ptr, &rec->next) == -1)
828 /* Update header: do this before we drop alloc
829 lock, otherwise tdb_free() might try to
830 merge with us, thinking we're free.
831 (Thanks Jeremy Allison). */
832 rec->magic = TDB_MAGIC;
833 if (rec_write(tdb, rec_ptr, rec) == -1)
836 /* Did we create new block? */
838 /* Update allocated record tailer (we
840 if (update_tailer(tdb, rec_ptr, rec) == -1)
843 /* Free new record */
844 if (tdb_free(tdb, newrec_ptr, &newrec) == -1)
848 /* all done - return the new record offset */
849 tdb_unlock(tdb, -1, F_WRLCK);
852 /* move to the next record */
856 /* we didn't find enough space. See if we can expand the
857 database and if we can then try again */
858 if (tdb_expand(tdb, length + sizeof(*rec)) == 0)
861 tdb_unlock(tdb, -1, F_WRLCK);
865 /* initialise a new database with a specified hash size */
866 static int tdb_new_database(TDB_CONTEXT *tdb, int hash_size)
868 struct tdb_header *newdb;
871 /* We make it up in memory, then write it out if not internal */
872 size = sizeof(struct tdb_header) + (hash_size+1)*sizeof(tdb_off);
873 if (!(newdb = calloc(size, 1)))
874 return TDB_ERRCODE(TDB_ERR_OOM, -1);
876 /* Fill in the header */
877 newdb->version = TDB_VERSION;
878 newdb->hash_size = hash_size;
880 newdb->rwlocks = size;
882 if (tdb->flags & TDB_INTERNAL) {
883 tdb->map_size = size;
884 tdb->map_ptr = (char *)newdb;
885 memcpy(&tdb->header, newdb, sizeof(tdb->header));
886 /* Convert the `ondisk' version if asked. */
890 if (lseek(tdb->fd, 0, SEEK_SET) == -1)
893 if (ftruncate(tdb->fd, 0) == -1)
896 /* This creates an endian-converted header, as if read from disk */
898 memcpy(&tdb->header, newdb, sizeof(tdb->header));
899 /* Don't endian-convert the magic food! */
900 memcpy(newdb->magic_food, TDB_MAGIC_FOOD, strlen(TDB_MAGIC_FOOD)+1);
901 if (write(tdb->fd, newdb, size) != size)
904 ret = tdb_create_rwlocks(tdb->fd, hash_size);
911 /* Returns 0 on fail. On success, return offset of record, and fills
913 static tdb_off tdb_find(TDB_CONTEXT *tdb, TDB_DATA key, u32 hash,
914 struct list_struct *r)
918 /* read in the hash top */
919 if (ofs_read(tdb, TDB_HASH_TOP(hash), &rec_ptr) == -1)
922 /* keep looking until we find the right record */
924 if (rec_read(tdb, rec_ptr, r) == -1)
927 if (!TDB_DEAD(r) && hash==r->full_hash && key.dsize==r->key_len) {
929 /* a very likely hit - read the key */
930 k = tdb_alloc_read(tdb, rec_ptr + sizeof(*r),
935 if (memcmp(key.dptr, k, key.dsize) == 0) {
943 return TDB_ERRCODE(TDB_ERR_NOEXIST, 0);
946 /* If they do lockkeys, check that this hash is one they locked */
947 static int tdb_keylocked(TDB_CONTEXT *tdb, u32 hash)
950 if (!tdb->lockedkeys)
952 for (i = 0; i < tdb->lockedkeys[0]; i++)
953 if (tdb->lockedkeys[i+1] == hash)
955 return TDB_ERRCODE(TDB_ERR_NOLOCK, 0);
958 /* As tdb_find, but if you succeed, keep the lock */
959 static tdb_off tdb_find_lock(TDB_CONTEXT *tdb, TDB_DATA key, int locktype,
960 struct list_struct *rec)
964 hash = tdb_hash(&key);
965 if (!tdb_keylocked(tdb, hash))
967 if (tdb_lock(tdb, BUCKET(hash), locktype) == -1)
969 if (!(rec_ptr = tdb_find(tdb, key, hash, rec)))
970 tdb_unlock(tdb, BUCKET(hash), locktype);
974 enum TDB_ERROR tdb_error(TDB_CONTEXT *tdb)
979 static struct tdb_errname {
980 enum TDB_ERROR ecode; const char *estring;
981 } emap[] = { {TDB_SUCCESS, "Success"},
982 {TDB_ERR_CORRUPT, "Corrupt database"},
983 {TDB_ERR_IO, "IO Error"},
984 {TDB_ERR_LOCK, "Locking error"},
985 {TDB_ERR_OOM, "Out of memory"},
986 {TDB_ERR_EXISTS, "Record exists"},
987 {TDB_ERR_NOLOCK, "Lock exists on other keys"},
988 {TDB_ERR_NOEXIST, "Record does not exist"} };
990 /* Error string for the last tdb error */
991 const char *tdb_errorstr(TDB_CONTEXT *tdb)
994 for (i = 0; i < sizeof(emap) / sizeof(struct tdb_errname); i++)
995 if (tdb->ecode == emap[i].ecode)
996 return emap[i].estring;
997 return "Invalid error code";
1000 /* update an entry in place - this only works if the new data size
1001 is <= the old data size and the key exists.
1002 on failure return -1
1004 static int tdb_update(TDB_CONTEXT *tdb, TDB_DATA key, TDB_DATA dbuf)
1006 struct list_struct rec;
1011 if (!(rec_ptr = tdb_find_lock(tdb, key, F_WRLCK, &rec)))
1014 /* must be long enough key, data and tailer */
1015 if (rec.rec_len < key.dsize + dbuf.dsize + sizeof(tdb_off)) {
1016 tdb->ecode = TDB_SUCCESS; /* Not really an error */
1020 if (tdb_write(tdb, rec_ptr + sizeof(rec) + rec.key_len,
1021 dbuf.dptr, dbuf.dsize) == -1)
1024 if (dbuf.dsize != rec.data_len) {
1026 rec.data_len = dbuf.dsize;
1027 ret = rec_write(tdb, rec_ptr, &rec);
1031 tdb_unlock(tdb, BUCKET(rec.full_hash), F_WRLCK);
1035 /* find an entry in the database given a key */
1036 TDB_DATA tdb_fetch(TDB_CONTEXT *tdb, TDB_DATA key)
1039 struct list_struct rec;
1042 /* find which hash bucket it is in */
1043 if (!(rec_ptr = tdb_find_lock(tdb,key,F_RDLCK,&rec)))
1046 ret.dptr = tdb_alloc_read(tdb, rec_ptr + sizeof(rec) + rec.key_len,
1048 ret.dsize = rec.data_len;
1049 tdb_unlock(tdb, BUCKET(rec.full_hash), F_RDLCK);
1053 /* check if an entry in the database exists
1055 note that 1 is returned if the key is found and 0 is returned if not found
1056 this doesn't match the conventions in the rest of this module, but is
1057 compatible with gdbm
1059 int tdb_exists(TDB_CONTEXT *tdb, TDB_DATA key)
1061 struct list_struct rec;
1063 if (tdb_find_lock(tdb, key, F_RDLCK, &rec) == 0)
1065 tdb_unlock(tdb, BUCKET(rec.full_hash), F_RDLCK);
1069 /* record lock stops delete underneath */
1070 static int lock_record(TDB_CONTEXT *tdb, tdb_off off)
1072 return off ? tdb_brlock(tdb, off, F_RDLCK, F_SETLKW, 0) : 0;
1075 Write locks override our own fcntl readlocks, so check it here.
1076 Note this is meant to be F_SETLK, *not* F_SETLKW, as it's not
1077 an error to fail to get the lock here.
1080 static int write_lock_record(TDB_CONTEXT *tdb, tdb_off off)
1082 struct tdb_traverse_lock *i;
1083 for (i = &tdb->travlocks; i; i = i->next)
1086 return tdb_brlock(tdb, off, F_WRLCK, F_SETLK, 1);
1090 Note this is meant to be F_SETLK, *not* F_SETLKW, as it's not
1091 an error to fail to get the lock here.
1094 static int write_unlock_record(TDB_CONTEXT *tdb, tdb_off off)
1096 return tdb_brlock(tdb, off, F_UNLCK, F_SETLK, 0);
1098 /* fcntl locks don't stack: avoid unlocking someone else's */
1099 static int unlock_record(TDB_CONTEXT *tdb, tdb_off off)
1101 struct tdb_traverse_lock *i;
1106 for (i = &tdb->travlocks; i; i = i->next)
1109 return (count == 1 ? tdb_brlock(tdb, off, F_UNLCK, F_SETLKW, 0) : 0);
1112 /* actually delete an entry in the database given the offset */
1113 static int do_delete(TDB_CONTEXT *tdb, tdb_off rec_ptr, struct list_struct*rec)
1115 tdb_off last_ptr, i;
1116 struct list_struct lastrec;
1118 if (tdb->read_only) return -1;
1120 if (write_lock_record(tdb, rec_ptr) == -1) {
1121 /* Someone traversing here: mark it as dead */
1122 rec->magic = TDB_DEAD_MAGIC;
1123 return rec_write(tdb, rec_ptr, rec);
1125 if (write_unlock_record(tdb, rec_ptr) != 0)
1128 /* find previous record in hash chain */
1129 if (ofs_read(tdb, TDB_HASH_TOP(rec->full_hash), &i) == -1)
1131 for (last_ptr = 0; i != rec_ptr; last_ptr = i, i = lastrec.next)
1132 if (rec_read(tdb, i, &lastrec) == -1)
1135 /* unlink it: next ptr is at start of record. */
1137 last_ptr = TDB_HASH_TOP(rec->full_hash);
1138 if (ofs_write(tdb, last_ptr, &rec->next) == -1)
1141 /* recover the space */
1142 if (tdb_free(tdb, rec_ptr, rec) == -1)
1147 /* Uses traverse lock: 0 = finish, -1 = error, other = record offset */
1148 static int tdb_next_lock(TDB_CONTEXT *tdb, struct tdb_traverse_lock *tlock,
1149 struct list_struct *rec)
1151 int want_next = (tlock->off != 0);
1153 /* No traversal allows if you've called tdb_lockkeys() */
1154 if (tdb->lockedkeys)
1155 return TDB_ERRCODE(TDB_ERR_NOLOCK, -1);
1157 /* Lock each chain from the start one. */
1158 for (; tlock->hash < tdb->header.hash_size; tlock->hash++) {
1159 if (tdb_lock(tdb, tlock->hash, F_WRLCK) == -1)
1162 /* No previous record? Start at top of chain. */
1164 if (ofs_read(tdb, TDB_HASH_TOP(tlock->hash),
1168 /* Otherwise unlock the previous record. */
1169 if (unlock_record(tdb, tlock->off) != 0)
1174 /* We have offset of old record: grab next */
1175 if (rec_read(tdb, tlock->off, rec) == -1)
1177 tlock->off = rec->next;
1180 /* Iterate through chain */
1181 while( tlock->off) {
1183 if (rec_read(tdb, tlock->off, rec) == -1)
1185 if (!TDB_DEAD(rec)) {
1186 /* Woohoo: we found one! */
1187 if (lock_record(tdb, tlock->off) != 0)
1191 /* Try to clean dead ones from old traverses */
1192 current = tlock->off;
1193 tlock->off = rec->next;
1194 if (do_delete(tdb, current, rec) != 0)
1197 tdb_unlock(tdb, tlock->hash, F_WRLCK);
1200 /* We finished iteration without finding anything */
1201 return TDB_ERRCODE(TDB_SUCCESS, 0);
1205 if (tdb_unlock(tdb, tlock->hash, F_WRLCK) != 0)
1206 TDB_LOG((tdb, 0, "tdb_next_lock: On error unlock failed!\n"));
1210 /* traverse the entire database - calling fn(tdb, key, data) on each element.
1211 return -1 on error or the record count traversed
1212 if fn is NULL then it is not called
1213 a non-zero return value from fn() indicates that the traversal should stop
1215 int tdb_traverse(TDB_CONTEXT *tdb, tdb_traverse_func fn, void *state)
1218 struct list_struct rec;
1219 struct tdb_traverse_lock tl = { NULL, 0, 0 };
1222 /* This was in the initializaton, above, but the IRIX compiler
1223 * did not like it. crh
1225 tl.next = tdb->travlocks.next;
1227 /* fcntl locks don't stack: beware traverse inside traverse */
1228 tdb->travlocks.next = &tl;
1230 /* tdb_next_lock places locks on the record returned, and its chain */
1231 while ((ret = tdb_next_lock(tdb, &tl, &rec)) > 0) {
1233 /* now read the full record */
1234 key.dptr = tdb_alloc_read(tdb, tl.off + sizeof(rec),
1235 rec.key_len + rec.data_len);
1238 if (tdb_unlock(tdb, tl.hash, F_WRLCK) != 0)
1240 if (unlock_record(tdb, tl.off) != 0)
1241 TDB_LOG((tdb, 0, "tdb_traverse: key.dptr == NULL and unlock_record failed!\n"));
1244 key.dsize = rec.key_len;
1245 dbuf.dptr = key.dptr + rec.key_len;
1246 dbuf.dsize = rec.data_len;
1248 /* Drop chain lock, call out */
1249 if (tdb_unlock(tdb, tl.hash, F_WRLCK) != 0) {
1253 if (fn && fn(tdb, key, dbuf, state)) {
1254 /* They want us to terminate traversal */
1256 if (unlock_record(tdb, tl.off) != 0) {
1257 TDB_LOG((tdb, 0, "tdb_traverse: unlock_record failed!\n"));;
1260 tdb->travlocks.next = tl.next;
1261 SAFE_FREE(key.dptr);
1264 SAFE_FREE(key.dptr);
1267 tdb->travlocks.next = tl.next;
1274 /* find the first entry in the database and return its key */
1275 TDB_DATA tdb_firstkey(TDB_CONTEXT *tdb)
1278 struct list_struct rec;
1280 /* release any old lock */
1281 if (unlock_record(tdb, tdb->travlocks.off) != 0)
1283 tdb->travlocks.off = tdb->travlocks.hash = 0;
1285 if (tdb_next_lock(tdb, &tdb->travlocks, &rec) <= 0)
1287 /* now read the key */
1288 key.dsize = rec.key_len;
1289 key.dptr =tdb_alloc_read(tdb,tdb->travlocks.off+sizeof(rec),key.dsize);
1290 if (tdb_unlock(tdb, BUCKET(tdb->travlocks.hash), F_WRLCK) != 0)
1291 TDB_LOG((tdb, 0, "tdb_firstkey: error occurred while tdb_unlocking!\n"));
1295 /* find the next entry in the database, returning its key */
1296 TDB_DATA tdb_nextkey(TDB_CONTEXT *tdb, TDB_DATA oldkey)
1299 TDB_DATA key = tdb_null;
1300 struct list_struct rec;
1303 /* Is locked key the old key? If so, traverse will be reliable. */
1304 if (tdb->travlocks.off) {
1305 if (tdb_lock(tdb,tdb->travlocks.hash,F_WRLCK))
1307 if (rec_read(tdb, tdb->travlocks.off, &rec) == -1
1308 || !(k = tdb_alloc_read(tdb,tdb->travlocks.off+sizeof(rec),
1310 || memcmp(k, oldkey.dptr, oldkey.dsize) != 0) {
1311 /* No, it wasn't: unlock it and start from scratch */
1312 if (unlock_record(tdb, tdb->travlocks.off) != 0)
1314 if (tdb_unlock(tdb, tdb->travlocks.hash, F_WRLCK) != 0)
1316 tdb->travlocks.off = 0;
1322 if (!tdb->travlocks.off) {
1323 /* No previous element: do normal find, and lock record */
1324 tdb->travlocks.off = tdb_find_lock(tdb, oldkey, F_WRLCK, &rec);
1325 if (!tdb->travlocks.off)
1327 tdb->travlocks.hash = BUCKET(rec.full_hash);
1328 if (lock_record(tdb, tdb->travlocks.off) != 0) {
1329 TDB_LOG((tdb, 0, "tdb_nextkey: lock_record failed (%s)!\n", strerror(errno)));
1333 oldhash = tdb->travlocks.hash;
1335 /* Grab next record: locks chain and returned record,
1336 unlocks old record */
1337 if (tdb_next_lock(tdb, &tdb->travlocks, &rec) > 0) {
1338 key.dsize = rec.key_len;
1339 key.dptr = tdb_alloc_read(tdb, tdb->travlocks.off+sizeof(rec),
1341 /* Unlock the chain of this new record */
1342 if (tdb_unlock(tdb, tdb->travlocks.hash, F_WRLCK) != 0)
1343 TDB_LOG((tdb, 0, "tdb_nextkey: WARNING tdb_unlock failed!\n"));
1345 /* Unlock the chain of old record */
1346 if (tdb_unlock(tdb, BUCKET(oldhash), F_WRLCK) != 0)
1347 TDB_LOG((tdb, 0, "tdb_nextkey: WARNING tdb_unlock failed!\n"));
1351 /* delete an entry in the database given a key */
1352 int tdb_delete(TDB_CONTEXT *tdb, TDB_DATA key)
1355 struct list_struct rec;
1358 if (!(rec_ptr = tdb_find_lock(tdb, key, F_WRLCK, &rec)))
1360 ret = do_delete(tdb, rec_ptr, &rec);
1361 if (tdb_unlock(tdb, BUCKET(rec.full_hash), F_WRLCK) != 0)
1362 TDB_LOG((tdb, 0, "tdb_delete: WARNING tdb_unlock failed!\n"));
1366 /* store an element in the database, replacing any existing element
1369 return 0 on success, -1 on failure
1371 int tdb_store(TDB_CONTEXT *tdb, TDB_DATA key, TDB_DATA dbuf, int flag)
1373 struct list_struct rec;
1379 /* find which hash bucket it is in */
1380 hash = tdb_hash(&key);
1381 if (!tdb_keylocked(tdb, hash))
1383 if (tdb_lock(tdb, BUCKET(hash), F_WRLCK) == -1)
1386 /* check for it existing, on insert. */
1387 if (flag == TDB_INSERT) {
1388 if (tdb_exists(tdb, key)) {
1389 tdb->ecode = TDB_ERR_EXISTS;
1393 /* first try in-place update, on modify or replace. */
1394 if (tdb_update(tdb, key, dbuf) == 0)
1396 if (flag == TDB_MODIFY && tdb->ecode == TDB_ERR_NOEXIST)
1399 /* reset the error code potentially set by the tdb_update() */
1400 tdb->ecode = TDB_SUCCESS;
1402 /* delete any existing record - if it doesn't exist we don't
1403 care. Doing this first reduces fragmentation, and avoids
1404 coalescing with `allocated' block before it's updated. */
1405 if (flag != TDB_INSERT)
1406 tdb_delete(tdb, key);
1408 /* Copy key+value *before* allocating free space in case malloc
1409 fails and we are left with a dead spot in the tdb. */
1411 if (!(p = (char *)malloc(key.dsize + dbuf.dsize))) {
1412 tdb->ecode = TDB_ERR_OOM;
1416 memcpy(p, key.dptr, key.dsize);
1417 memcpy(p+key.dsize, dbuf.dptr, dbuf.dsize);
1419 /* now we're into insert / modify / replace of a record which
1420 * we know could not be optimised by an in-place store (for
1421 * various reasons). */
1422 if (!(rec_ptr = tdb_allocate(tdb, key.dsize + dbuf.dsize, &rec)))
1425 /* Read hash top into next ptr */
1426 if (ofs_read(tdb, TDB_HASH_TOP(hash), &rec.next) == -1)
1429 rec.key_len = key.dsize;
1430 rec.data_len = dbuf.dsize;
1431 rec.full_hash = hash;
1432 rec.magic = TDB_MAGIC;
1434 /* write out and point the top of the hash chain at it */
1435 if (rec_write(tdb, rec_ptr, &rec) == -1
1436 || tdb_write(tdb, rec_ptr+sizeof(rec), p, key.dsize+dbuf.dsize)==-1
1437 || ofs_write(tdb, TDB_HASH_TOP(hash), &rec_ptr) == -1) {
1438 /* Need to tdb_unallocate() here */
1443 tdb_unlock(tdb, BUCKET(hash), F_WRLCK);
1450 static int tdb_already_open(dev_t device,
1455 for (i = tdbs; i; i = i->next) {
1456 if (i->device == device && i->inode == ino) {
1464 /* open the database, creating it if necessary
1466 The open_flags and mode are passed straight to the open call on the
1467 database file. A flags value of O_WRONLY is invalid. The hash size
1468 is advisory, use zero for a default value.
1470 Return is NULL on error, in which case errno is also set. Don't
1471 try to call tdb_error or tdb_errname, just do strerror(errno).
1473 @param name may be NULL for internal databases. */
1474 TDB_CONTEXT *tdb_open(const char *name, int hash_size, int tdb_flags,
1475 int open_flags, mode_t mode)
1477 return tdb_open_ex(name, hash_size, tdb_flags, open_flags, mode, NULL);
1481 TDB_CONTEXT *tdb_open_ex(const char *name, int hash_size, int tdb_flags,
1482 int open_flags, mode_t mode,
1483 tdb_log_func log_fn)
1487 int rev = 0, locked;
1491 if (!(tdb = calloc(1, sizeof *tdb))) {
1492 /* Can't log this */
1498 tdb->map_ptr = NULL;
1499 tdb->lockedkeys = NULL;
1500 tdb->flags = tdb_flags;
1501 tdb->open_flags = open_flags;
1502 tdb->log_fn = log_fn;
1504 if ((open_flags & O_ACCMODE) == O_WRONLY) {
1505 TDB_LOG((tdb, 0, "tdb_open_ex: can't open tdb %s write-only\n",
1512 hash_size = DEFAULT_HASH_SIZE;
1513 if ((open_flags & O_ACCMODE) == O_RDONLY) {
1515 /* read only databases don't do locking or clear if first */
1516 tdb->flags |= TDB_NOLOCK;
1517 tdb->flags &= ~TDB_CLEAR_IF_FIRST;
1520 /* internal databases don't mmap or lock, and start off cleared */
1521 if (tdb->flags & TDB_INTERNAL) {
1522 tdb->flags |= (TDB_NOLOCK | TDB_NOMMAP);
1523 tdb->flags &= ~TDB_CLEAR_IF_FIRST;
1524 if (tdb_new_database(tdb, hash_size) != 0) {
1525 TDB_LOG((tdb, 0, "tdb_open_ex: tdb_new_database failed!"));
1531 if ((tdb->fd = open(name, open_flags, mode)) == -1) {
1532 TDB_LOG((tdb, 5, "tdb_open_ex: could not open file %s: %s\n",
1533 name, strerror(errno)));
1534 goto fail; /* errno set by open(2) */
1537 /* ensure there is only one process initialising at once */
1538 if (tdb_brlock(tdb, GLOBAL_LOCK, F_WRLCK, F_SETLKW, 0) == -1) {
1539 TDB_LOG((tdb, 0, "tdb_open_ex: failed to get global lock on %s: %s\n",
1540 name, strerror(errno)));
1541 goto fail; /* errno set by tdb_brlock */
1544 /* we need to zero database if we are the only one with it open */
1545 if ((locked = (tdb_brlock(tdb, ACTIVE_LOCK, F_WRLCK, F_SETLK, 0) == 0))
1546 && (tdb_flags & TDB_CLEAR_IF_FIRST)) {
1547 open_flags |= O_CREAT;
1548 if (ftruncate(tdb->fd, 0) == -1) {
1549 TDB_LOG((tdb, 0, "tdb_open_ex: "
1550 "failed to truncate %s: %s\n",
1551 name, strerror(errno)));
1552 goto fail; /* errno set by ftruncate */
1556 if (read(tdb->fd, &tdb->header, sizeof(tdb->header)) != sizeof(tdb->header)
1557 || strcmp(tdb->header.magic_food, TDB_MAGIC_FOOD) != 0
1558 || (tdb->header.version != TDB_VERSION
1559 && !(rev = (tdb->header.version==TDB_BYTEREV(TDB_VERSION))))) {
1560 /* its not a valid database - possibly initialise it */
1561 if (!(open_flags & O_CREAT) || tdb_new_database(tdb, hash_size) == -1) {
1562 errno = EIO; /* ie bad format or something */
1565 rev = (tdb->flags & TDB_CONVERT);
1567 vp = (unsigned char *)&tdb->header.version;
1568 vertest = (((u32)vp[0]) << 24) | (((u32)vp[1]) << 16) |
1569 (((u32)vp[2]) << 8) | (u32)vp[3];
1570 tdb->flags |= (vertest==TDB_VERSION) ? TDB_BIGENDIAN : 0;
1572 tdb->flags &= ~TDB_CONVERT;
1574 tdb->flags |= TDB_CONVERT;
1575 convert(&tdb->header, sizeof(tdb->header));
1577 if (fstat(tdb->fd, &st) == -1)
1580 /* Is it already in the open list? If so, fail. */
1581 if (tdb_already_open(st.st_dev, st.st_ino)) {
1582 TDB_LOG((tdb, 2, "tdb_open_ex: "
1583 "%s (%d,%d) is already open in this process\n",
1584 name, st.st_dev, st.st_ino));
1589 if (!(tdb->name = (char *)strdup(name))) {
1594 tdb->map_size = st.st_size;
1595 tdb->device = st.st_dev;
1596 tdb->inode = st.st_ino;
1597 tdb->locked = calloc(tdb->header.hash_size+1, sizeof(tdb->locked[0]));
1599 TDB_LOG((tdb, 2, "tdb_open_ex: "
1600 "failed to allocate lock structure for %s\n",
1607 if (!tdb->read_only)
1608 if (tdb_clear_spinlocks(tdb) != 0) {
1609 TDB_LOG((tdb, 0, "tdb_open_ex: "
1610 "failed to clear spinlock\n"));
1613 if (tdb_brlock(tdb, ACTIVE_LOCK, F_UNLCK, F_SETLK, 0) == -1) {
1614 TDB_LOG((tdb, 0, "tdb_open_ex: "
1615 "failed to take ACTIVE_LOCK on %s: %s\n",
1616 name, strerror(errno)));
1620 /* leave this lock in place to indicate it's in use */
1621 if (tdb_brlock(tdb, ACTIVE_LOCK, F_RDLCK, F_SETLKW, 0) == -1)
1625 /* Internal (memory-only) databases skip all the code above to
1626 * do with disk files, and resume here by releasing their
1627 * global lock and hooking into the active list. */
1628 if (tdb_brlock(tdb, GLOBAL_LOCK, F_UNLCK, F_SETLKW, 0) == -1)
1635 { int save_errno = errno;
1641 if (tdb->flags & TDB_INTERNAL)
1642 SAFE_FREE(tdb->map_ptr);
1646 SAFE_FREE(tdb->name);
1648 if (close(tdb->fd) != 0)
1649 TDB_LOG((tdb, 5, "tdb_open_ex: failed to close tdb->fd on error!\n"));
1650 SAFE_FREE(tdb->locked);
1657 /* close a database */
1658 int tdb_close(TDB_CONTEXT *tdb)
1664 if (tdb->flags & TDB_INTERNAL)
1665 SAFE_FREE(tdb->map_ptr);
1669 SAFE_FREE(tdb->name);
1671 ret = close(tdb->fd);
1672 SAFE_FREE(tdb->locked);
1673 SAFE_FREE(tdb->lockedkeys);
1675 /* Remove from contexts list */
1676 for (i = &tdbs; *i; i = &(*i)->next) {
1683 memset(tdb, 0, sizeof(*tdb));
1689 /* lock/unlock entire database */
1690 int tdb_lockall(TDB_CONTEXT *tdb)
1694 /* There are no locks on read-only dbs */
1696 return TDB_ERRCODE(TDB_ERR_LOCK, -1);
1697 if (tdb->lockedkeys)
1698 return TDB_ERRCODE(TDB_ERR_NOLOCK, -1);
1699 for (i = 0; i < tdb->header.hash_size; i++)
1700 if (tdb_lock(tdb, i, F_WRLCK))
1703 /* If error, release locks we have... */
1704 if (i < tdb->header.hash_size) {
1707 for ( j = 0; j < i; j++)
1708 tdb_unlock(tdb, j, F_WRLCK);
1709 return TDB_ERRCODE(TDB_ERR_NOLOCK, -1);
1714 void tdb_unlockall(TDB_CONTEXT *tdb)
1717 for (i=0; i < tdb->header.hash_size; i++)
1718 tdb_unlock(tdb, i, F_WRLCK);
1721 int tdb_lockkeys(TDB_CONTEXT *tdb, u32 number, TDB_DATA keys[])
1725 /* Can't lock more keys if already locked */
1726 if (tdb->lockedkeys)
1727 return TDB_ERRCODE(TDB_ERR_NOLOCK, -1);
1728 if (!(tdb->lockedkeys = malloc(sizeof(u32) * (number+1))))
1729 return TDB_ERRCODE(TDB_ERR_OOM, -1);
1730 /* First number in array is # keys */
1731 tdb->lockedkeys[0] = number;
1733 /* Insertion sort by bucket */
1734 for (i = 0; i < number; i++) {
1735 hash = tdb_hash(&keys[i]);
1736 for (j = 0; j < i && BUCKET(tdb->lockedkeys[j+1]) < BUCKET(hash); j++);
1737 memmove(&tdb->lockedkeys[j+2], &tdb->lockedkeys[j+1], sizeof(u32) * (i-j));
1738 tdb->lockedkeys[j+1] = hash;
1740 /* Finally, lock in order */
1741 for (i = 0; i < number; i++)
1742 if (tdb_lock(tdb, i, F_WRLCK))
1745 /* If error, release locks we have... */
1747 for ( j = 0; j < i; j++)
1748 tdb_unlock(tdb, j, F_WRLCK);
1749 SAFE_FREE(tdb->lockedkeys);
1750 return TDB_ERRCODE(TDB_ERR_NOLOCK, -1);
1755 /* Unlock the keys previously locked by tdb_lockkeys() */
1756 void tdb_unlockkeys(TDB_CONTEXT *tdb)
1759 for (i = 0; i < tdb->lockedkeys[0]; i++)
1760 tdb_unlock(tdb, tdb->lockedkeys[i+1], F_WRLCK);
1761 SAFE_FREE(tdb->lockedkeys);
1764 /* lock/unlock one hash chain. This is meant to be used to reduce
1765 contention - it cannot guarantee how many records will be locked */
1766 int tdb_chainlock(TDB_CONTEXT *tdb, TDB_DATA key)
1768 return tdb_lock(tdb, BUCKET(tdb_hash(&key)), F_WRLCK);
1771 int tdb_chainunlock(TDB_CONTEXT *tdb, TDB_DATA key)
1773 return tdb_unlock(tdb, BUCKET(tdb_hash(&key)), F_WRLCK);
1777 /* register a loging function */
1778 void tdb_logging_function(TDB_CONTEXT *tdb, void (*fn)(TDB_CONTEXT *, int , const char *, ...))
1784 /* reopen a tdb - this is used after a fork to ensure that we have an independent
1785 seek pointer from our parent and to re-establish locks */
1786 int tdb_reopen(TDB_CONTEXT *tdb)
1790 if (tdb_munmap(tdb) != 0) {
1791 TDB_LOG((tdb, 0, "tdb_reopen: munmap failed (%s)\n", strerror(errno)));
1794 if (close(tdb->fd) != 0)
1795 TDB_LOG((tdb, 0, "tdb_reopen: WARNING closing tdb->fd failed!\n"));
1796 tdb->fd = open(tdb->name, tdb->open_flags & ~(O_CREAT|O_TRUNC), 0);
1797 if (tdb->fd == -1) {
1798 TDB_LOG((tdb, 0, "tdb_reopen: open failed (%s)\n", strerror(errno)));
1801 if (fstat(tdb->fd, &st) != 0) {
1802 TDB_LOG((tdb, 0, "tdb_reopen: fstat failed (%s)\n", strerror(errno)));
1805 if (st.st_ino != tdb->inode || st.st_dev != tdb->device) {
1806 TDB_LOG((tdb, 0, "tdb_reopen: file dev/inode has changed!\n"));
1810 if (tdb_brlock(tdb, ACTIVE_LOCK, F_RDLCK, F_SETLKW, 0) == -1) {
1811 TDB_LOG((tdb, 0, "tdb_reopen: failed to obtain active lock\n"));
1822 /* reopen all tdb's */
1823 int tdb_reopen_all(void)
1827 for (tdb=tdbs; tdb; tdb = tdb->next) {
1828 if (tdb_reopen(tdb) != 0) return -1;