about summary refs log tree commit diff
path: root/db2/db/db_overflow.c
blob: 2340e9e3581fde62ef4d67a172e2303de8abb322 (plain) (blame)
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
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
/*-
 * See the file LICENSE for redistribution information.
 *
 * Copyright (c) 1996, 1997
 *	Sleepycat Software.  All rights reserved.
 */
/*
 * Copyright (c) 1990, 1993, 1994, 1995, 1996
 *	Keith Bostic.  All rights reserved.
 */
/*
 * Copyright (c) 1990, 1993, 1994, 1995
 *	The Regents of the University of California.  All rights reserved.
 *
 * This code is derived from software contributed to Berkeley by
 * Mike Olson.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */

#include "config.h"

#ifndef lint
static const char sccsid[] = "@(#)db_overflow.c	10.4 (Sleepycat) 7/2/97";
#endif /* not lint */

#ifndef NO_SYSTEM_INCLUDES
#include <sys/types.h>

#include <errno.h>
#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#endif

#include "db_int.h"
#include "db_page.h"
#include "db_am.h"
#include "common_ext.h"

/*
 * Big key/data code.
 *
 * Big key and data entries are stored on linked lists of pages.  The initial
 * reference is a structure with the total length of the item and the page
 * number where it begins.  Each entry in the linked list contains a pointer
 * to the next page of data, and so on.
 */

/*
 * __db_goff --
 *	Get an offpage item.
 *
 * PUBLIC: int __db_goff __P((DB *, DBT *,
 * PUBLIC:     u_int32_t, db_pgno_t, void **, u_int32_t *));
 */
int
__db_goff(dbp, dbt, tlen, pgno, bpp, bpsz)
	DB *dbp;
	DBT *dbt;
	u_int32_t tlen;
	db_pgno_t pgno;
	void **bpp;
	u_int32_t *bpsz;
{
	PAGE *h;
	db_indx_t bytes;
	int ret;
	u_int32_t curoff, needed, start;
	u_int8_t *p, *src;

	/*
	 * Check if the buffer is big enough; if it is not and we are
	 * allowed to malloc space, then we'll malloc it.  If we are
	 * not (DB_DBT_USERMEM), then we'll set the dbt and return
	 * appropriately.
	 */
	if (F_ISSET(dbt, DB_DBT_PARTIAL)) {
		start = dbt->doff;
		needed = dbt->dlen;
	} else {
		start = 0;
		needed = tlen;
	}

	/*
	 * Allocate any necessary memory.
	 *
	 * XXX: Never allocate 0 bytes;
	 */
	if (F_ISSET(dbt, DB_DBT_USERMEM)) {
		if (needed > dbt->ulen) {
			dbt->size = needed;
			return (ENOMEM);
		}
	} else if (F_ISSET(dbt, DB_DBT_MALLOC)) {
		dbt->data = dbp->db_malloc == NULL ?
		    (void *)malloc(needed + 1) :
		    (void *)dbp->db_malloc(needed + 1);
		if (dbt->data == NULL)
			return (ENOMEM);
	} else if (*bpsz == 0 || *bpsz < needed) {
		*bpp = (*bpp == NULL ?
		    (void *)malloc(needed + 1) :
		    (void *)realloc(*bpp, needed + 1));
		if (*bpp == NULL)
			return (ENOMEM);
		*bpsz = needed + 1;
		dbt->data = *bpp;
	} else
		dbt->data = *bpp;

	/*
	 * Step through the linked list of pages, copying the data on each
	 * one into the buffer.  Never copy more than the total data length.
	 */
	dbt->size = needed;
	for (curoff = 0, p = dbt->data; pgno != P_INVALID && needed > 0;) {
		if ((ret = memp_fget(dbp->mpf, &pgno, 0, &h)) != 0) {
			(void)__db_pgerr(dbp, pgno);
			return (ret);
		}
		/* Check if we need any bytes from this page. */
		if (curoff + OV_LEN(h) >= start) {
			src = (u_int8_t *)h + P_OVERHEAD;
			bytes = OV_LEN(h);
			if (start > curoff) {
				src += start - curoff;
				bytes -= start - curoff;
			}
			if (bytes > needed)
				bytes = needed;
			memcpy(p, src, bytes);
			p += bytes;
			needed -= bytes;
		}
		curoff += OV_LEN(h);
		pgno = h->next_pgno;
		memp_fput(dbp->mpf, h, 0);
	}
	return (0);
}

/*
 * __db_poff --
 *	Put an offpage item.
 *
 * PUBLIC: int __db_poff __P((DB *, const DBT *, db_pgno_t *,
 * PUBLIC:     int (*)(DB *, u_int32_t, PAGE **)));
 */
int
__db_poff(dbp, dbt, pgnop, newfunc)
	DB *dbp;
	const DBT *dbt;
	db_pgno_t *pgnop;
	int (*newfunc) __P((DB *, u_int32_t, PAGE **));
{
	PAGE *pagep, *lastp;
	DB_LSN new_lsn, null_lsn;
	DBT tmp_dbt;
	db_indx_t pagespace;
	u_int32_t sz;
	u_int8_t *p;
	int ret;

	/*
	 * Allocate pages and copy the key/data item into them.  Calculate the
	 * number of bytes we get for pages we fill completely with a single
	 * item.
	 */
	pagespace = P_MAXSPACE(dbp->pgsize);

	lastp = NULL;
	for (p = dbt->data,
	    sz = dbt->size; sz > 0; p += pagespace, sz -= pagespace) {
		/*
		 * Reduce pagespace so we terminate the loop correctly and
		 * don't copy too much data.
		 */
		if (sz < pagespace)
			pagespace = sz;

		/*
		 * Allocate and initialize a new page and copy all or part of
		 * the item onto the page.  If sz is less than pagespace, we
		 * have a partial record.
		 */
		if ((ret = newfunc(dbp, P_OVERFLOW, &pagep)) != 0)
			return (ret);
		if (DB_LOGGING(dbp)) {
			tmp_dbt.data = p;
			tmp_dbt.size = pagespace;
			ZERO_LSN(null_lsn);
			if ((ret = __db_big_log(dbp->dbenv->lg_info, dbp->txn,
			    &new_lsn, 0, DB_ADD_BIG, dbp->log_fileid,
			    PGNO(pagep), lastp ? PGNO(lastp) : PGNO_INVALID,
			    PGNO_INVALID, &tmp_dbt, &LSN(pagep),
			    lastp == NULL ? &null_lsn : &LSN(lastp),
			    &null_lsn)) != 0)
				return (ret);

			/* Move lsn onto page. */
			if (lastp)
				LSN(lastp) = new_lsn;
			LSN(pagep) = new_lsn;
		}

		P_INIT(pagep, dbp->pgsize,
		    PGNO(pagep), PGNO_INVALID, PGNO_INVALID, 0, P_OVERFLOW);
		OV_LEN(pagep) = pagespace;
		OV_REF(pagep) = 1;
		memcpy((u_int8_t *)pagep + P_OVERHEAD, p, pagespace);

		/*
		 * If this is the first entry, update the user's info.
		 * Otherwise, update the entry on the last page filled
		 * in and release that page.
		 */
		if (lastp == NULL)
			*pgnop = PGNO(pagep);
		else {
			lastp->next_pgno = PGNO(pagep);
			pagep->prev_pgno = PGNO(lastp);
			(void)memp_fput(dbp->mpf, lastp, DB_MPOOL_DIRTY);
		}
		lastp = pagep;
	}
	(void)memp_fput(dbp->mpf, lastp, DB_MPOOL_DIRTY);
	return (0);
}

/*
 * __db_ioff --
 *	Increment the reference count on an overflow page.
 *
 * PUBLIC: int __db_ioff __P((DB *, db_pgno_t));
 */
int
__db_ioff(dbp, pgno)
	DB *dbp;
	db_pgno_t pgno;
{
	PAGE *h;
	int ret;

	if ((ret = memp_fget(dbp->mpf, &pgno, 0, &h)) != 0) {
		(void)__db_pgerr(dbp, pgno);
		return (ret);
	}

	++OV_REF(h);
	if (DB_LOGGING(dbp) && (ret = __db_ovref_log(dbp->dbenv->lg_info,
	    dbp->txn, &LSN(h), 0, dbp->log_fileid, h->pgno, &LSN(h))) != 0)
		return (ret);

	(void)memp_fput(dbp->mpf, h, DB_MPOOL_DIRTY);
	return (0);
}

/*
 * __db_doff --
 *	Delete an offpage chain of overflow pages.
 *
 * PUBLIC: int __db_doff __P((DB *, db_pgno_t, int (*)(DB *, PAGE *)));
 */
int
__db_doff(dbp, pgno, freefunc)
	DB *dbp;
	db_pgno_t pgno;
	int (*freefunc) __P((DB *, PAGE *));
{
	PAGE *pagep;
	DB_LSN null_lsn;
	DBT tmp_dbt;
	int ret;

	do {
		if ((ret = memp_fget(dbp->mpf, &pgno, 0, &pagep)) != 0) {
			(void)__db_pgerr(dbp, pgno);
			return (ret);
		}

		/*
		 * If it's an overflow page and it's referenced by more than
		 * one key/data item, decrement the reference count and return.
		 */
		if (TYPE(pagep) == P_OVERFLOW && OV_REF(pagep) > 1) {
			--OV_REF(pagep);
			(void)memp_fput(dbp->mpf, pagep, DB_MPOOL_DIRTY);
			return (0);
		}

		if (DB_LOGGING(dbp)) {
			tmp_dbt.data = (u_int8_t *)pagep + P_OVERHEAD;
			tmp_dbt.size = OV_LEN(pagep);
			ZERO_LSN(null_lsn);
			if ((ret = __db_big_log(dbp->dbenv->lg_info, dbp->txn,
			    &LSN(pagep), 0, DB_REM_BIG, dbp->log_fileid,
			    PGNO(pagep), PREV_PGNO(pagep), NEXT_PGNO(pagep),
			    &tmp_dbt, &LSN(pagep), &null_lsn, &null_lsn)) != 0)
				return (ret);
		}
		pgno = pagep->next_pgno;
		if ((ret = freefunc(dbp, pagep)) != 0)
			return (ret);
	} while (pgno != PGNO_INVALID);

	return (0);
}

/*
 * __db_moff --
 *	Match on overflow pages.
 *
 * Given a starting page number and a key, return <0, 0, >0 to indicate if the
 * key on the page is less than, equal to or greater than the key specified.
 *
 * PUBLIC: int __db_moff __P((DB *, const DBT *, db_pgno_t));
 */
int
__db_moff(dbp, dbt, pgno)
	DB *dbp;
	const DBT *dbt;
	db_pgno_t pgno;
{
	PAGE *pagep;
	u_int32_t cmp_bytes, key_left;
	int ret;
	u_int8_t *p1, *p2;

	/* While there are both keys to compare. */
	for (ret = 0, p1 = dbt->data,
	    key_left = dbt->size; key_left > 0 && pgno != PGNO_INVALID;) {
		if (memp_fget(dbp->mpf, &pgno, 0, &pagep) != 0) {
			(void)__db_pgerr(dbp, pgno);
			return (0);	/* No system error return. */
		}

		cmp_bytes = OV_LEN(pagep) < key_left ? OV_LEN(pagep) : key_left;
		key_left -= cmp_bytes;
		for (p2 =
		    (u_int8_t *)pagep + P_OVERHEAD; cmp_bytes-- > 0; ++p1, ++p2)
			if (*p1 != *p2) {
				ret = (long)*p1 - (long)*p2;
				break;
			}
		pgno = NEXT_PGNO(pagep);
		(void)memp_fput(dbp->mpf, pagep, 0);
		if (ret != 0)
			return (ret);
	}
	if (key_left > 0)		/* DBT is longer than page key. */
		return (-1);
	if (pgno != PGNO_INVALID)	/* DBT is shorter than page key. */
		return (1);
	return (0);
}