Training courses

Kernel and Embedded Linux

Bootlin training courses

Embedded Linux, kernel,
Yocto Project, Buildroot, real-time,
graphics, boot time, debugging...

Bootlin logo

Elixir Cross Referencer

  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
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
/*	$NetBSD: mpool.c,v 1.23 2016/09/24 21:31:25 christos Exp $	*/

/*-
 * Copyright (c) 1990, 1993, 1994
 *	The Regents of the University of California.  All rights reserved.
 *
 * 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. 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.
 */

#if HAVE_NBTOOL_CONFIG_H
#include "nbtool_config.h"
#endif

#include <sys/cdefs.h>
__RCSID("$NetBSD: mpool.c,v 1.23 2016/09/24 21:31:25 christos Exp $");

#include "namespace.h"
#include <sys/queue.h>
#include <sys/stat.h>

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

#include <db.h>

#define	__MPOOLINTERFACE_PRIVATE
#include <mpool.h>

#ifdef __weak_alias
__weak_alias(mpool_close,_mpool_close)
__weak_alias(mpool_filter,_mpool_filter)
__weak_alias(mpool_get,_mpool_get)
__weak_alias(mpool_new,_mpool_new)
__weak_alias(mpool_newf,_mpool_newf)
__weak_alias(mpool_open,_mpool_open)
__weak_alias(mpool_put,_mpool_put)
__weak_alias(mpool_sync,_mpool_sync)
#endif

static BKT *mpool_bkt(MPOOL *);
static BKT *mpool_look(MPOOL *, pgno_t);
static int  mpool_write(MPOOL *, BKT *);

/*
 * mpool_open --
 *	Initialize a memory pool.
 */
/*ARGSUSED*/
MPOOL *
mpool_open(void *key, int fd, pgno_t pagesize, pgno_t maxcache)
{
	struct stat sb;
	MPOOL *mp;
	int entry;

	/*
	 * Get information about the file.
	 *
	 * XXX
	 * We don't currently handle pipes, although we should.
	 */
	if (fstat(fd, &sb))
		return NULL;
	if (!S_ISREG(sb.st_mode)) {
		errno = ESPIPE;
		return NULL;
	}

	/* Allocate and initialize the MPOOL cookie. */
	if ((mp = calloc(1, sizeof(*mp))) == NULL)
		return (NULL);
	TAILQ_INIT(&mp->lqh);
	for (entry = 0; entry < HASHSIZE; ++entry)
		TAILQ_INIT(&mp->hqh[entry]);
	mp->maxcache = maxcache;
	mp->npages = (pgno_t)(sb.st_size / pagesize);
	mp->pagesize = pagesize;
	mp->fd = fd;
	return mp;
}

/*
 * mpool_filter --
 *	Initialize input/output filters.
 */
void
mpool_filter(MPOOL *mp, void (*pgin)(void *, pgno_t, void *),
    void (*pgout)(void *, pgno_t, void *), void *pgcookie)
{
	mp->pgin = pgin;
	mp->pgout = pgout;
	mp->pgcookie = pgcookie;
}
	
/*
 * mpool_new --
 *	Get a new page of memory.
 */
void *
mpool_newf(MPOOL *mp, pgno_t *pgnoaddr, unsigned int flags)
{
	struct _hqh *head;
	BKT *bp;

	if (mp->npages == MAX_PAGE_NUMBER) {
		(void)fprintf(stderr, "mpool_new: page allocation overflow.\n");
		abort();
	}
#ifdef STATISTICS
	++mp->pagenew;
#endif
	/*
	 * Get a BKT from the cache.  Assign a new page number, attach
	 * it to the head of the hash chain, the tail of the lru chain,
	 * and return.
	 */
	if ((bp = mpool_bkt(mp)) == NULL)
		return NULL;

	if (flags == MPOOL_PAGE_REQUEST) {
		mp->npages++;
		bp->pgno = *pgnoaddr;
	} else
		bp->pgno = *pgnoaddr = mp->npages++;

	bp->flags = MPOOL_PINNED | MPOOL_INUSE;

	head = &mp->hqh[HASHKEY(bp->pgno)];
	TAILQ_INSERT_HEAD(head, bp, hq);
	TAILQ_INSERT_TAIL(&mp->lqh, bp, q);
	return bp->page;
}

void *
mpool_new(MPOOL *mp, pgno_t *pgnoaddr)
{
	return mpool_newf(mp, pgnoaddr, 0);
}

int
mpool_delete(MPOOL *mp, void *page)
{
	struct _hqh *head;
	BKT *bp;

	bp = (void *)((char *)page - sizeof(BKT));

#ifdef DEBUG
	if (!(bp->flags & MPOOL_PINNED)) {
	       (void)fprintf(stderr,
		   "%s: page %d not pinned\n", __func__, bp->pgno);
	       abort();
	}
#endif

	/* Remove from the hash and lru queues. */
	head = &mp->hqh[HASHKEY(bp->pgno)];
	TAILQ_REMOVE(head, bp, hq);
	TAILQ_REMOVE(&mp->lqh, bp, q);

	free(bp);
	return RET_SUCCESS;
}

/*
 * mpool_get
 *	Get a page.
 */
/*ARGSUSED*/
void *
mpool_get(MPOOL *mp, pgno_t pgno, unsigned int flags)
{
	struct _hqh *head;
	BKT *bp;
	off_t off;
	ssize_t nr;

	/* Check for attempt to retrieve a non-existent page. */
	if (pgno >= mp->npages) {
		errno = EINVAL;
		return NULL;
	}

#ifdef STATISTICS
	++mp->pageget;
#endif

	/* Check for a page that is cached. */
	if ((bp = mpool_look(mp, pgno)) != NULL) {
#ifdef DEBUG
		if (!(flags & MPOOL_IGNOREPIN) && bp->flags & MPOOL_PINNED) {
			(void)fprintf(stderr,
			    "mpool_get: page %d already pinned\n", bp->pgno);
			abort();
		}
#endif
		/*
		 * Move the page to the head of the hash chain and the tail
		 * of the lru chain.
		 */
		head = &mp->hqh[HASHKEY(bp->pgno)];
		TAILQ_REMOVE(head, bp, hq);
		TAILQ_INSERT_HEAD(head, bp, hq);
		TAILQ_REMOVE(&mp->lqh, bp, q);
		TAILQ_INSERT_TAIL(&mp->lqh, bp, q);

		/* Return a pinned page. */
		if (!(flags & MPOOL_IGNOREPIN))
			bp->flags |= MPOOL_PINNED;
		return bp->page;
	}

	/* Get a page from the cache. */
	if ((bp = mpool_bkt(mp)) == NULL)
		return NULL;

	/* Read in the contents. */
#ifdef STATISTICS
	++mp->pageread;
#endif
	off = mp->pagesize * pgno;
	if (off / mp->pagesize != pgno) {
		/* Run past the end of the file, or at least the part we
		   can address without large-file support?  */
		errno = E2BIG;
 		return NULL;
	}

	if ((nr = pread(mp->fd, bp->page, (size_t)mp->pagesize, off)) != (int)mp->pagesize) {
		if (nr > 0) {
			errno = EFTYPE;
			return NULL;
		} else if (nr == 0) {
			/*
			 * A zero-length reads, means you need to create a
			 * new page.
			 */
			memset(bp->page, 0, mp->pagesize);
		} else
			return NULL;
	}

	/* Set the page number, pin the page. */
	bp->pgno = pgno;
	if (!(flags & MPOOL_IGNOREPIN))
		bp->flags = MPOOL_PINNED;
bp->flags |= MPOOL_INUSE;

	/*
	 * Add the page to the head of the hash chain and the tail
	 * of the lru chain.
	 */
	head = &mp->hqh[HASHKEY(bp->pgno)];
	TAILQ_INSERT_HEAD(head, bp, hq);
	TAILQ_INSERT_TAIL(&mp->lqh, bp, q);

	/* Run through the user's filter. */
	if (mp->pgin != NULL)
		(mp->pgin)(mp->pgcookie, bp->pgno, bp->page);

	return bp->page;
}

/*
 * mpool_put
 *	Return a page.
 */
/*ARGSUSED*/
int
mpool_put(MPOOL *mp, void *page, u_int flags)
{
	BKT *bp;

#ifdef STATISTICS
	++mp->pageput;
#endif
	bp = (void *)((intptr_t)page - sizeof(BKT));
#ifdef DEBUG
	if (!(bp->flags & MPOOL_PINNED)) {
		(void)fprintf(stderr,
		    "mpool_put: page %d not pinned\n", bp->pgno);
		abort();
	}
#endif
	bp->flags &= ~MPOOL_PINNED;
	if (flags & MPOOL_DIRTY)
		bp->flags |= flags & MPOOL_DIRTY;
	return (RET_SUCCESS);
}

/*
 * mpool_close
 *	Close the buffer pool.
 */
int
mpool_close(MPOOL *mp)
{
	BKT *bp;

	/* Free up any space allocated to the lru pages. */
	while (!TAILQ_EMPTY(&mp->lqh)) {
		bp = TAILQ_FIRST(&mp->lqh);
		TAILQ_REMOVE(&mp->lqh, bp, q);
		free(bp);
	}

	/* Free the MPOOL cookie. */
	free(mp);
	return RET_SUCCESS;
}

/*
 * mpool_sync
 *	Sync the pool to disk.
 */
int
mpool_sync(MPOOL *mp)
{
	BKT *bp;

	/* Walk the lru chain, flushing any dirty pages to disk. */
	TAILQ_FOREACH(bp, &mp->lqh, q)
		if (bp->flags & MPOOL_DIRTY &&
		    mpool_write(mp, bp) == RET_ERROR)
			return RET_ERROR;

	/* Sync the file descriptor. */
	return fsync(mp->fd) ? RET_ERROR : RET_SUCCESS;
}

/*
 * mpool_bkt
 *	Get a page from the cache (or create one).
 */
static BKT *
mpool_bkt(MPOOL *mp)
{
	struct _hqh *head;
	BKT *bp;

	/* If under the max cached, always create a new page. */
	if (mp->curcache < mp->maxcache)
		goto new;

	/*
	 * If the cache is max'd out, walk the lru list for a buffer we
	 * can flush.  If we find one, write it (if necessary) and take it
	 * off any lists.  If we don't find anything we grow the cache anyway.
	 * The cache never shrinks.
	 */
	TAILQ_FOREACH(bp, &mp->lqh, q)
		if (!(bp->flags & MPOOL_PINNED)) {
			/* Flush if dirty. */
			if (bp->flags & MPOOL_DIRTY &&
			    mpool_write(mp, bp) == RET_ERROR)
				return NULL;
#ifdef STATISTICS
			++mp->pageflush;
#endif
			/* Remove from the hash and lru queues. */
			head = &mp->hqh[HASHKEY(bp->pgno)];
			TAILQ_REMOVE(head, bp, hq);
			TAILQ_REMOVE(&mp->lqh, bp, q);
#ifdef DEBUG
			{
				void *spage = bp->page;
				(void)memset(bp, 0xff,
				    (size_t)(sizeof(BKT) + mp->pagesize));
				bp->page = spage;
			}
#endif
			return bp;
		}

new:	if ((bp = calloc(1, (size_t)(sizeof(BKT) + mp->pagesize))) == NULL)
		return NULL;
#ifdef STATISTICS
	++mp->pagealloc;
#endif
#if defined(DEBUG) || defined(PURIFY)
	(void)memset(bp, 0xff, (size_t)(sizeof(BKT) + mp->pagesize));
#endif
	bp->page = (void *)((intptr_t)bp + sizeof(BKT));
	++mp->curcache;
	return bp;
}

/*
 * mpool_write
 *	Write a page to disk.
 */
static int
mpool_write(MPOOL *mp, BKT *bp)
{
	off_t off;

#ifdef STATISTICS
	++mp->pagewrite;
#endif

	/* Run through the user's filter. */
	if (mp->pgout)
		(mp->pgout)(mp->pgcookie, bp->pgno, bp->page);

	off = mp->pagesize * bp->pgno;
	if (off / mp->pagesize != bp->pgno) {
		/* Run past the end of the file, or at least the part we
		   can address without large-file support?  */
		errno = E2BIG;
		return RET_ERROR;
	}

	if (pwrite(mp->fd, bp->page, (size_t)mp->pagesize, off) !=
	    (ssize_t)mp->pagesize)
		return RET_ERROR;

	/*
	 * Re-run through the input filter since this page may soon be
	 * accessed via the cache, and whatever the user's output filter
	 * did may screw things up if we don't let the input filter
	 * restore the in-core copy.
	 */
	if (mp->pgin)
		(mp->pgin)(mp->pgcookie, bp->pgno, bp->page);

	bp->flags &= ~MPOOL_DIRTY;
	return RET_SUCCESS;
}

/*
 * mpool_look
 *	Lookup a page in the cache.
 */
static BKT *
mpool_look(MPOOL *mp, pgno_t pgno)
{
	struct _hqh *head;
	BKT *bp;

	head = &mp->hqh[HASHKEY(pgno)];
	TAILQ_FOREACH(bp, head, hq)
		if (bp->pgno == pgno) {
#ifdef STATISTICS
			++mp->cachehit;
#endif
			return bp;
		}
#ifdef STATISTICS
	++mp->cachemiss;
#endif
	return NULL;
}

#ifdef STATISTICS
/*
 * mpool_stat
 *	Print out cache statistics.
 */
void
mpool_stat(mp)
	MPOOL *mp;
{
	BKT *bp;
	int cnt;
	const char *sep;

	(void)fprintf(stderr, "%lu pages in the file\n", (u_long)mp->npages);
	(void)fprintf(stderr,
	    "page size %lu, cacheing %lu pages of %lu page max cache\n",
	    (u_long)mp->pagesize, (u_long)mp->curcache, (u_long)mp->maxcache);
	(void)fprintf(stderr, "%lu page puts, %lu page gets, %lu page new\n",
	    mp->pageput, mp->pageget, mp->pagenew);
	(void)fprintf(stderr, "%lu page allocs, %lu page flushes\n",
	    mp->pagealloc, mp->pageflush);
	if (mp->cachehit + mp->cachemiss)
		(void)fprintf(stderr,
		    "%.0f%% cache hit rate (%lu hits, %lu misses)\n", 
		    ((double)mp->cachehit / (mp->cachehit + mp->cachemiss))
		    * 100, mp->cachehit, mp->cachemiss);
	(void)fprintf(stderr, "%lu page reads, %lu page writes\n",
	    mp->pageread, mp->pagewrite);

	sep = "";
	cnt = 0;
	TAILQ_FOREACH(bp, &mp->lqh, q) {
		(void)fprintf(stderr, "%s%d", sep, bp->pgno);
		if (bp->flags & MPOOL_DIRTY)
			(void)fprintf(stderr, "d");
		if (bp->flags & MPOOL_PINNED)
			(void)fprintf(stderr, "P");
		if (++cnt == 10) {
			sep = "\n";
			cnt = 0;
		} else
			sep = ", ";
			
	}
	(void)fprintf(stderr, "\n");
}
#endif