summaryrefslogtreecommitdiff
path: root/lib/tdb_compat/tdb_compat.c
blob: 7fd3caf6b5acc53a930b3a1471efaea45b4cabb1 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
#include <tdb_compat.h>

/* Note: for the moment, we only need this file for TDB2, so we can
 * assume waf. */
#if BUILD_TDB2
TDB_DATA tdb_null = { NULL, 0 };

/* Proxy which sets waitflag to false so we never block. */
static int lock_nonblock(int fd, int rw, off_t off, off_t len, bool waitflag,
			 void *_orig)
{
	struct tdb_attribute_flock *orig = _orig;

	return orig->lock(fd, rw, off, len, false, orig->data);
}

enum TDB_ERROR tdb_transaction_start_nonblock(struct tdb_context *tdb)
{
	union tdb_attribute locking, orig;
	enum TDB_ERROR ecode;

	orig.base.attr = TDB_ATTRIBUTE_FLOCK;
	ecode = tdb_get_attribute(tdb, &orig);
	if (ecode != TDB_SUCCESS)
		return ecode;

	/* Replace locking function with our own. */
	locking = orig;
	locking.flock.data = &orig;
	locking.flock.lock = lock_nonblock;

	ecode = tdb_set_attribute(tdb, &locking);
	if (ecode != TDB_SUCCESS)
		return ecode;

	ecode = tdb_transaction_start(tdb);
	tdb_unset_attribute(tdb, TDB_ATTRIBUTE_FLOCK);
	return ecode;
}

enum TDB_ERROR tdb_chainlock_nonblock(struct tdb_context *tdb, TDB_DATA key)
{
	union tdb_attribute locking, orig;
	enum TDB_ERROR ecode;

	orig.base.attr = TDB_ATTRIBUTE_FLOCK;
	ecode = tdb_get_attribute(tdb, &orig);
	if (ecode != TDB_SUCCESS)
		return ecode;

	/* Replace locking function with our own. */
	locking = orig;
	locking.flock.data = &orig;
	locking.flock.lock = lock_nonblock;

	ecode = tdb_set_attribute(tdb, &locking);
	if (ecode != TDB_SUCCESS)
		return ecode;

	ecode = tdb_chainlock(tdb, key);
	tdb_unset_attribute(tdb, TDB_ATTRIBUTE_FLOCK);
	return ecode;
}

/* For TDB1 tdbs, read traverse vs normal matters: write traverse
   locks the entire thing! */
int64_t tdb_traverse_read_(struct tdb_context *tdb,
			   int (*fn)(struct tdb_context *,
						       TDB_DATA, TDB_DATA,
						       void *),
			   void *p)
{
	int64_t ret;

	if (tdb_get_flags(tdb) & TDB_RDONLY) {
		return tdb_traverse(tdb, fn, p);
	}

	tdb_add_flag(tdb, TDB_RDONLY);
	ret = tdb_traverse(tdb, fn, p);
	tdb_remove_flag(tdb, TDB_RDONLY);
	return ret;
}

/*
 * This handles TDB_CLEAR_IF_FIRST.
 */
static enum TDB_ERROR clear_if_first(int fd, void *unused)
{
	/* We hold a lock offset 4 always, so we can tell if anyone else is. */
	struct flock fl;

	fl.l_type = F_WRLCK;
	fl.l_whence = SEEK_SET;
	fl.l_start = 4; /* ACTIVE_LOCK */
	fl.l_len = 1;

	if (fcntl(fd, F_SETLK, &fl) == 0) {
		/* We must be first ones to open it w/ TDB_CLEAR_IF_FIRST! */
		if (ftruncate(fd, 0) != 0) {
			return TDB_ERR_IO;
		}
	}
	fl.l_type = F_RDLCK;
	if (fcntl(fd, F_SETLKW, &fl) != 0) {
		return TDB_ERR_IO;
	}
	return TDB_SUCCESS;
}

struct tdb_context *
tdb_open_compat_(const char *name, int hash_size,
		 int tdb_flags, int open_flags, mode_t mode,
		 void (*log_fn)(struct tdb_context *,
				enum tdb_log_level,
				enum TDB_ERROR,
				const char *message,
				void *data),
		 void *log_data)
{
	union tdb_attribute cif, log, hash, max_dead, hsize, *attr = NULL;

	if (!getenv("TDB_COMPAT_USE_TDB2")) {
		tdb_flags |= TDB_VERSION1;
	}

	if (log_fn) {
		log.log.base.attr = TDB_ATTRIBUTE_LOG;
		log.log.base.next = NULL;
		log.log.fn = log_fn;
		log.log.data = log_data;
		attr = &log;
	}

	if (tdb_flags & TDB_CLEAR_IF_FIRST) {
		cif.openhook.base.attr = TDB_ATTRIBUTE_OPENHOOK;
		cif.openhook.base.next = attr;
		cif.openhook.fn = clear_if_first;
		attr = &cif;
		tdb_flags &= ~TDB_CLEAR_IF_FIRST;
	}

	if (tdb_flags & TDB_INCOMPATIBLE_HASH) {
		if (tdb_flags & TDB_VERSION1) {
			hash.hash.base.attr = TDB_ATTRIBUTE_HASH;
			hash.hash.base.next = attr;
			hash.hash.fn = tdb1_incompatible_hash;
			attr = &hash;
		}
		tdb_flags &= ~TDB_INCOMPATIBLE_HASH;
	}

	if (tdb_flags & TDB_VOLATILE) {
		if (tdb_flags & TDB_VERSION1) {
			max_dead.base.attr = TDB_ATTRIBUTE_TDB1_MAX_DEAD;
			max_dead.base.next = attr;
			max_dead.tdb1_max_dead.max_dead = 5;
			attr = &max_dead;
		}
		tdb_flags &= ~TDB_VOLATILE;
	}

	if (hash_size && (tdb_flags & TDB_VERSION1) && (open_flags & O_CREAT)) {
		hsize.base.attr = TDB_ATTRIBUTE_TDB1_HASHSIZE;
		hsize.base.next = attr;
		hsize.tdb1_hashsize.hsize = hash_size;
		attr = &hsize;
	}

	/* Testsuite uses this to speed things up. */
	if (getenv("TDB_NO_FSYNC")) {
		tdb_flags |= TDB_NOSYNC;
	}

	return tdb_open(name, tdb_flags|TDB_ALLOW_NESTING, open_flags, mode,
			attr);
}

/* We only need these for the CLEAR_IF_FIRST lock. */
static int reacquire_cif_lock(struct tdb_context *tdb, bool *fail)
{
	struct flock fl;
	union tdb_attribute cif;

	cif.openhook.base.attr = TDB_ATTRIBUTE_OPENHOOK;
	cif.openhook.base.next = NULL;

	if (tdb_get_attribute(tdb, &cif) != TDB_SUCCESS
	    || cif.openhook.fn != clear_if_first) {
		return 0;
	}

	/* We hold a lock offset 4 always, so we can tell if anyone else is. */
	fl.l_type = F_RDLCK;
	fl.l_whence = SEEK_SET;
	fl.l_start = 4; /* ACTIVE_LOCK */
	fl.l_len = 1;
	if (fcntl(tdb_fd(tdb), F_SETLKW, &fl) != 0) {
		*fail = true;
		return -1;
	}
	return 0;
}

int tdb_reopen(struct tdb_context *tdb)
{
	bool unused;
	return reacquire_cif_lock(tdb, &unused);
}

int tdb_reopen_all(int parent_longlived)
{
	bool fail = false;

	if (parent_longlived) {
		return 0;
	}

	tdb_foreach(reacquire_cif_lock, &fail);
	if (fail)
		return -1;
	return 0;
}
#endif