Annotation of src/sys/kern/subr_pool.c, Revision 1.122.2.2
1.122.2.2! ad 1: /* $NetBSD: subr_pool.c,v 1.122.2.1 2006/09/11 00:12:01 ad Exp $ */
1.1 pk 2:
3: /*-
1.122.2.1 ad 4: * Copyright (c) 1997, 1999, 2000, 2002 The NetBSD Foundation, Inc.
1.1 pk 5: * All rights reserved.
6: *
7: * This code is derived from software contributed to The NetBSD Foundation
1.20 thorpej 8: * by Paul Kranenburg; by Jason R. Thorpe of the Numerical Aerospace
9: * Simulation Facility, NASA Ames Research Center.
1.1 pk 10: *
11: * Redistribution and use in source and binary forms, with or without
12: * modification, are permitted provided that the following conditions
13: * are met:
14: * 1. Redistributions of source code must retain the above copyright
15: * notice, this list of conditions and the following disclaimer.
16: * 2. Redistributions in binary form must reproduce the above copyright
17: * notice, this list of conditions and the following disclaimer in the
18: * documentation and/or other materials provided with the distribution.
19: * 3. All advertising materials mentioning features or use of this software
20: * must display the following acknowledgement:
1.13 christos 21: * This product includes software developed by the NetBSD
22: * Foundation, Inc. and its contributors.
1.1 pk 23: * 4. Neither the name of The NetBSD Foundation nor the names of its
24: * contributors may be used to endorse or promote products derived
25: * from this software without specific prior written permission.
26: *
27: * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
28: * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
29: * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
30: * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
31: * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
32: * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
33: * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
34: * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
35: * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
36: * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
37: * POSSIBILITY OF SUCH DAMAGE.
38: */
1.64 lukem 39:
40: #include <sys/cdefs.h>
1.122.2.2! ad 41: __KERNEL_RCSID(0, "$NetBSD: subr_pool.c,v 1.122.2.1 2006/09/11 00:12:01 ad Exp $");
1.24 scottr 42:
1.25 thorpej 43: #include "opt_pool.h"
1.24 scottr 44: #include "opt_poollog.h"
1.28 thorpej 45: #include "opt_lockdebug.h"
1.1 pk 46:
47: #include <sys/param.h>
48: #include <sys/systm.h>
49: #include <sys/proc.h>
50: #include <sys/errno.h>
51: #include <sys/kernel.h>
52: #include <sys/malloc.h>
53: #include <sys/lock.h>
54: #include <sys/pool.h>
1.20 thorpej 55: #include <sys/syslog.h>
1.3 pk 56:
57: #include <uvm/uvm.h>
58:
1.1 pk 59: /*
60: * Pool resource management utility.
1.3 pk 61: *
1.88 chs 62: * Memory is allocated in pages which are split into pieces according to
63: * the pool item size. Each page is kept on one of three lists in the
64: * pool structure: `pr_emptypages', `pr_fullpages' and `pr_partpages',
65: * for empty, full and partially-full pages respectively. The individual
66: * pool items are on a linked list headed by `ph_itemlist' in each page
67: * header. The memory for building the page list is either taken from
68: * the allocated pages themselves (for small pool items) or taken from
69: * an internal pool of page headers (`phpool').
1.1 pk 70: */
71:
1.3 pk 72: /* List of all pools */
1.102 chs 73: LIST_HEAD(,pool) pool_head = LIST_HEAD_INITIALIZER(pool_head);
1.3 pk 74:
75: /* Private pool for page header structures */
1.97 yamt 76: #define PHPOOL_MAX 8
77: static struct pool phpool[PHPOOL_MAX];
78: #define PHPOOL_FREELIST_NELEM(idx) (((idx) == 0) ? 0 : (1 << (idx)))
1.3 pk 79:
1.62 bjh21 80: #ifdef POOL_SUBPAGE
81: /* Pool of subpages for use by normal pools. */
82: static struct pool psppool;
83: #endif
84:
1.117 yamt 85: static SLIST_HEAD(, pool_allocator) pa_deferinitq =
86: SLIST_HEAD_INITIALIZER(pa_deferinitq);
87:
1.98 yamt 88: static void *pool_page_alloc_meta(struct pool *, int);
89: static void pool_page_free_meta(struct pool *, void *);
90:
91: /* allocator for pool metadata */
92: static struct pool_allocator pool_allocator_meta = {
1.117 yamt 93: pool_page_alloc_meta, pool_page_free_meta,
94: .pa_backingmapptr = &kmem_map,
1.98 yamt 95: };
96:
1.3 pk 97: /* # of seconds to retain page after last use */
98: int pool_inactive_time = 10;
99:
100: /* Next candidate for drainage (see pool_drain()) */
1.23 thorpej 101: static struct pool *drainpp;
102:
103: /* This spin lock protects both pool_head and drainpp. */
104: struct simplelock pool_head_slock = SIMPLELOCK_INITIALIZER;
1.3 pk 105:
1.99 yamt 106: typedef uint8_t pool_item_freelist_t;
107:
1.3 pk 108: struct pool_item_header {
109: /* Page headers */
1.88 chs 110: LIST_ENTRY(pool_item_header)
1.3 pk 111: ph_pagelist; /* pool page list */
1.88 chs 112: SPLAY_ENTRY(pool_item_header)
113: ph_node; /* Off-page page headers */
1.3 pk 114: caddr_t ph_page; /* this page's address */
115: struct timeval ph_time; /* last referenced */
1.97 yamt 116: union {
117: /* !PR_NOTOUCH */
118: struct {
1.102 chs 119: LIST_HEAD(, pool_item)
1.97 yamt 120: phu_itemlist; /* chunk list for this page */
121: } phu_normal;
122: /* PR_NOTOUCH */
123: struct {
124: uint16_t
125: phu_off; /* start offset in page */
1.99 yamt 126: pool_item_freelist_t
1.97 yamt 127: phu_firstfree; /* first free item */
1.99 yamt 128: /*
129: * XXX it might be better to use
130: * a simple bitmap and ffs(3)
131: */
1.97 yamt 132: } phu_notouch;
133: } ph_u;
134: uint16_t ph_nmissing; /* # of chunks in use */
1.3 pk 135: };
1.97 yamt 136: #define ph_itemlist ph_u.phu_normal.phu_itemlist
137: #define ph_off ph_u.phu_notouch.phu_off
138: #define ph_firstfree ph_u.phu_notouch.phu_firstfree
1.3 pk 139:
1.1 pk 140: struct pool_item {
1.3 pk 141: #ifdef DIAGNOSTIC
1.82 thorpej 142: u_int pi_magic;
1.33 chs 143: #endif
1.82 thorpej 144: #define PI_MAGIC 0xdeadbeefU
1.3 pk 145: /* Other entries use only this list entry */
1.102 chs 146: LIST_ENTRY(pool_item) pi_list;
1.3 pk 147: };
148:
1.53 thorpej 149: #define POOL_NEEDS_CATCHUP(pp) \
150: ((pp)->pr_nitems < (pp)->pr_minitems)
151:
1.43 thorpej 152: /*
153: * Pool cache management.
154: *
155: * Pool caches provide a way for constructed objects to be cached by the
156: * pool subsystem. This can lead to performance improvements by avoiding
157: * needless object construction/destruction; it is deferred until absolutely
158: * necessary.
159: *
160: * Caches are grouped into cache groups. Each cache group references
161: * up to 16 constructed objects. When a cache allocates an object
162: * from the pool, it calls the object's constructor and places it into
163: * a cache group. When a cache group frees an object back to the pool,
164: * it first calls the object's destructor. This allows the object to
165: * persist in constructed form while freed to the cache.
166: *
167: * Multiple caches may exist for each pool. This allows a single
168: * object type to have multiple constructed forms. The pool references
169: * each cache, so that when a pool is drained by the pagedaemon, it can
170: * drain each individual cache as well. Each time a cache is drained,
171: * the most idle cache group is freed to the pool in its entirety.
172: *
173: * Pool caches are layed on top of pools. By layering them, we can avoid
174: * the complexity of cache management for pools which would not benefit
175: * from it.
176: */
177:
178: /* The cache group pool. */
179: static struct pool pcgpool;
1.3 pk 180:
1.102 chs 181: static void pool_cache_reclaim(struct pool_cache *, struct pool_pagelist *,
182: struct pool_cache_grouplist *);
183: static void pcg_grouplist_free(struct pool_cache_grouplist *);
1.3 pk 184:
1.42 thorpej 185: static int pool_catchup(struct pool *);
1.55 thorpej 186: static void pool_prime_page(struct pool *, caddr_t,
187: struct pool_item_header *);
1.88 chs 188: static void pool_update_curpage(struct pool *);
1.66 thorpej 189:
1.113 yamt 190: static int pool_grow(struct pool *, int);
1.117 yamt 191: static void *pool_allocator_alloc(struct pool *, int);
192: static void pool_allocator_free(struct pool *, void *);
1.3 pk 193:
1.97 yamt 194: static void pool_print_pagelist(struct pool *, struct pool_pagelist *,
1.88 chs 195: void (*)(const char *, ...));
1.42 thorpej 196: static void pool_print1(struct pool *, const char *,
197: void (*)(const char *, ...));
1.3 pk 198:
1.88 chs 199: static int pool_chk_page(struct pool *, const char *,
200: struct pool_item_header *);
201:
1.3 pk 202: /*
1.52 thorpej 203: * Pool log entry. An array of these is allocated in pool_init().
1.3 pk 204: */
205: struct pool_log {
206: const char *pl_file;
207: long pl_line;
208: int pl_action;
1.25 thorpej 209: #define PRLOG_GET 1
210: #define PRLOG_PUT 2
1.3 pk 211: void *pl_addr;
1.1 pk 212: };
213:
1.86 matt 214: #ifdef POOL_DIAGNOSTIC
1.3 pk 215: /* Number of entries in pool log buffers */
1.17 thorpej 216: #ifndef POOL_LOGSIZE
217: #define POOL_LOGSIZE 10
218: #endif
219:
220: int pool_logsize = POOL_LOGSIZE;
1.1 pk 221:
1.110 perry 222: static inline void
1.42 thorpej 223: pr_log(struct pool *pp, void *v, int action, const char *file, long line)
1.3 pk 224: {
225: int n = pp->pr_curlogentry;
226: struct pool_log *pl;
227:
1.20 thorpej 228: if ((pp->pr_roflags & PR_LOGGING) == 0)
1.3 pk 229: return;
230:
231: /*
232: * Fill in the current entry. Wrap around and overwrite
233: * the oldest entry if necessary.
234: */
235: pl = &pp->pr_log[n];
236: pl->pl_file = file;
237: pl->pl_line = line;
238: pl->pl_action = action;
239: pl->pl_addr = v;
240: if (++n >= pp->pr_logsize)
241: n = 0;
242: pp->pr_curlogentry = n;
243: }
244:
245: static void
1.42 thorpej 246: pr_printlog(struct pool *pp, struct pool_item *pi,
247: void (*pr)(const char *, ...))
1.3 pk 248: {
249: int i = pp->pr_logsize;
250: int n = pp->pr_curlogentry;
251:
1.20 thorpej 252: if ((pp->pr_roflags & PR_LOGGING) == 0)
1.3 pk 253: return;
254:
255: /*
256: * Print all entries in this pool's log.
257: */
258: while (i-- > 0) {
259: struct pool_log *pl = &pp->pr_log[n];
260: if (pl->pl_action != 0) {
1.25 thorpej 261: if (pi == NULL || pi == pl->pl_addr) {
262: (*pr)("\tlog entry %d:\n", i);
263: (*pr)("\t\taction = %s, addr = %p\n",
264: pl->pl_action == PRLOG_GET ? "get" : "put",
265: pl->pl_addr);
266: (*pr)("\t\tfile: %s at line %lu\n",
267: pl->pl_file, pl->pl_line);
268: }
1.3 pk 269: }
270: if (++n >= pp->pr_logsize)
271: n = 0;
272: }
273: }
1.25 thorpej 274:
1.110 perry 275: static inline void
1.42 thorpej 276: pr_enter(struct pool *pp, const char *file, long line)
1.25 thorpej 277: {
278:
1.34 thorpej 279: if (__predict_false(pp->pr_entered_file != NULL)) {
1.25 thorpej 280: printf("pool %s: reentrancy at file %s line %ld\n",
281: pp->pr_wchan, file, line);
282: printf(" previous entry at file %s line %ld\n",
283: pp->pr_entered_file, pp->pr_entered_line);
284: panic("pr_enter");
285: }
286:
287: pp->pr_entered_file = file;
288: pp->pr_entered_line = line;
289: }
290:
1.110 perry 291: static inline void
1.42 thorpej 292: pr_leave(struct pool *pp)
1.25 thorpej 293: {
294:
1.34 thorpej 295: if (__predict_false(pp->pr_entered_file == NULL)) {
1.25 thorpej 296: printf("pool %s not entered?\n", pp->pr_wchan);
297: panic("pr_leave");
298: }
299:
300: pp->pr_entered_file = NULL;
301: pp->pr_entered_line = 0;
302: }
303:
1.110 perry 304: static inline void
1.42 thorpej 305: pr_enter_check(struct pool *pp, void (*pr)(const char *, ...))
1.25 thorpej 306: {
307:
308: if (pp->pr_entered_file != NULL)
309: (*pr)("\n\tcurrently entered from file %s line %ld\n",
310: pp->pr_entered_file, pp->pr_entered_line);
311: }
1.3 pk 312: #else
1.25 thorpej 313: #define pr_log(pp, v, action, file, line)
314: #define pr_printlog(pp, pi, pr)
315: #define pr_enter(pp, file, line)
316: #define pr_leave(pp)
317: #define pr_enter_check(pp, pr)
1.59 thorpej 318: #endif /* POOL_DIAGNOSTIC */
1.3 pk 319:
1.110 perry 320: static inline int
1.97 yamt 321: pr_item_notouch_index(const struct pool *pp, const struct pool_item_header *ph,
322: const void *v)
323: {
324: const char *cp = v;
325: int idx;
326:
327: KASSERT(pp->pr_roflags & PR_NOTOUCH);
328: idx = (cp - ph->ph_page - ph->ph_off) / pp->pr_size;
329: KASSERT(idx < pp->pr_itemsperpage);
330: return idx;
331: }
332:
1.99 yamt 333: #define PR_FREELIST_ALIGN(p) \
334: roundup((uintptr_t)(p), sizeof(pool_item_freelist_t))
335: #define PR_FREELIST(ph) ((pool_item_freelist_t *)PR_FREELIST_ALIGN((ph) + 1))
336: #define PR_INDEX_USED ((pool_item_freelist_t)-1)
337: #define PR_INDEX_EOL ((pool_item_freelist_t)-2)
1.97 yamt 338:
1.110 perry 339: static inline void
1.97 yamt 340: pr_item_notouch_put(const struct pool *pp, struct pool_item_header *ph,
341: void *obj)
342: {
343: int idx = pr_item_notouch_index(pp, ph, obj);
1.99 yamt 344: pool_item_freelist_t *freelist = PR_FREELIST(ph);
1.97 yamt 345:
346: KASSERT(freelist[idx] == PR_INDEX_USED);
347: freelist[idx] = ph->ph_firstfree;
348: ph->ph_firstfree = idx;
349: }
350:
1.110 perry 351: static inline void *
1.97 yamt 352: pr_item_notouch_get(const struct pool *pp, struct pool_item_header *ph)
353: {
354: int idx = ph->ph_firstfree;
1.99 yamt 355: pool_item_freelist_t *freelist = PR_FREELIST(ph);
1.97 yamt 356:
357: KASSERT(freelist[idx] != PR_INDEX_USED);
358: ph->ph_firstfree = freelist[idx];
359: freelist[idx] = PR_INDEX_USED;
360:
361: return ph->ph_page + ph->ph_off + idx * pp->pr_size;
362: }
363:
1.110 perry 364: static inline int
1.88 chs 365: phtree_compare(struct pool_item_header *a, struct pool_item_header *b)
366: {
1.121 yamt 367:
368: /*
369: * we consider pool_item_header with smaller ph_page bigger.
370: * (this unnatural ordering is for the benefit of pr_find_pagehead.)
371: */
372:
1.88 chs 373: if (a->ph_page < b->ph_page)
1.121 yamt 374: return (1);
375: else if (a->ph_page > b->ph_page)
1.88 chs 376: return (-1);
377: else
378: return (0);
379: }
380:
381: SPLAY_PROTOTYPE(phtree, pool_item_header, ph_node, phtree_compare);
382: SPLAY_GENERATE(phtree, pool_item_header, ph_node, phtree_compare);
383:
1.3 pk 384: /*
1.121 yamt 385: * Return the pool page header based on item address.
1.3 pk 386: */
1.110 perry 387: static inline struct pool_item_header *
1.121 yamt 388: pr_find_pagehead(struct pool *pp, void *v)
1.3 pk 389: {
1.88 chs 390: struct pool_item_header *ph, tmp;
1.3 pk 391:
1.121 yamt 392: if ((pp->pr_roflags & PR_NOALIGN) != 0) {
393: tmp.ph_page = (caddr_t)(uintptr_t)v;
394: ph = SPLAY_FIND(phtree, &pp->pr_phtree, &tmp);
395: if (ph == NULL) {
396: ph = SPLAY_ROOT(&pp->pr_phtree);
397: if (ph != NULL && phtree_compare(&tmp, ph) >= 0) {
398: ph = SPLAY_NEXT(phtree, &pp->pr_phtree, ph);
399: }
400: KASSERT(ph == NULL || phtree_compare(&tmp, ph) < 0);
401: }
402: } else {
403: caddr_t page =
404: (caddr_t)((uintptr_t)v & pp->pr_alloc->pa_pagemask);
405:
406: if ((pp->pr_roflags & PR_PHINPAGE) != 0) {
407: ph = (void *)(page + pp->pr_phoffset);
408: } else {
409: tmp.ph_page = page;
410: ph = SPLAY_FIND(phtree, &pp->pr_phtree, &tmp);
411: }
412: }
1.3 pk 413:
1.121 yamt 414: KASSERT(ph == NULL || ((pp->pr_roflags & PR_PHINPAGE) != 0) ||
415: (ph->ph_page <= (char *)v &&
416: (char *)v < ph->ph_page + pp->pr_alloc->pa_pagesz));
1.88 chs 417: return ph;
1.3 pk 418: }
419:
1.101 thorpej 420: static void
421: pr_pagelist_free(struct pool *pp, struct pool_pagelist *pq)
422: {
423: struct pool_item_header *ph;
424: int s;
425:
426: while ((ph = LIST_FIRST(pq)) != NULL) {
427: LIST_REMOVE(ph, ph_pagelist);
428: pool_allocator_free(pp, ph->ph_page);
429: if ((pp->pr_roflags & PR_PHINPAGE) == 0) {
430: s = splvm();
431: pool_put(pp->pr_phpool, ph);
432: splx(s);
433: }
434: }
435: }
436:
1.3 pk 437: /*
438: * Remove a page from the pool.
439: */
1.110 perry 440: static inline void
1.61 chs 441: pr_rmpage(struct pool *pp, struct pool_item_header *ph,
442: struct pool_pagelist *pq)
1.3 pk 443: {
444:
1.101 thorpej 445: LOCK_ASSERT(simple_lock_held(&pp->pr_slock));
1.91 yamt 446:
1.3 pk 447: /*
1.7 thorpej 448: * If the page was idle, decrement the idle page count.
1.3 pk 449: */
1.6 thorpej 450: if (ph->ph_nmissing == 0) {
451: #ifdef DIAGNOSTIC
452: if (pp->pr_nidle == 0)
453: panic("pr_rmpage: nidle inconsistent");
1.20 thorpej 454: if (pp->pr_nitems < pp->pr_itemsperpage)
455: panic("pr_rmpage: nitems inconsistent");
1.6 thorpej 456: #endif
457: pp->pr_nidle--;
458: }
1.7 thorpej 459:
1.20 thorpej 460: pp->pr_nitems -= pp->pr_itemsperpage;
461:
1.7 thorpej 462: /*
1.101 thorpej 463: * Unlink the page from the pool and queue it for release.
1.7 thorpej 464: */
1.88 chs 465: LIST_REMOVE(ph, ph_pagelist);
1.91 yamt 466: if ((pp->pr_roflags & PR_PHINPAGE) == 0)
467: SPLAY_REMOVE(phtree, &pp->pr_phtree, ph);
1.101 thorpej 468: LIST_INSERT_HEAD(pq, ph, ph_pagelist);
469:
1.7 thorpej 470: pp->pr_npages--;
471: pp->pr_npagefree++;
1.6 thorpej 472:
1.88 chs 473: pool_update_curpage(pp);
1.3 pk 474: }
475:
1.117 yamt 476: static boolean_t
477: pa_starved_p(struct pool_allocator *pa)
478: {
479:
480: if (pa->pa_backingmap != NULL) {
481: return vm_map_starved_p(pa->pa_backingmap);
482: }
483: return FALSE;
484: }
485:
486: static int
487: pool_reclaim_callback(struct callback_entry *ce, void *obj, void *arg)
488: {
489: struct pool *pp = obj;
490: struct pool_allocator *pa = pp->pr_alloc;
491:
492: KASSERT(&pp->pr_reclaimerentry == ce);
493: pool_reclaim(pp);
494: if (!pa_starved_p(pa)) {
495: return CALLBACK_CHAIN_ABORT;
496: }
497: return CALLBACK_CHAIN_CONTINUE;
498: }
499:
500: static void
501: pool_reclaim_register(struct pool *pp)
502: {
503: struct vm_map *map = pp->pr_alloc->pa_backingmap;
504: int s;
505:
506: if (map == NULL) {
507: return;
508: }
509:
510: s = splvm(); /* not necessary for INTRSAFE maps, but don't care. */
511: callback_register(&vm_map_to_kernel(map)->vmk_reclaim_callback,
512: &pp->pr_reclaimerentry, pp, pool_reclaim_callback);
513: splx(s);
514: }
515:
516: static void
517: pool_reclaim_unregister(struct pool *pp)
518: {
519: struct vm_map *map = pp->pr_alloc->pa_backingmap;
520: int s;
521:
522: if (map == NULL) {
523: return;
524: }
525:
526: s = splvm(); /* not necessary for INTRSAFE maps, but don't care. */
527: callback_unregister(&vm_map_to_kernel(map)->vmk_reclaim_callback,
528: &pp->pr_reclaimerentry);
529: splx(s);
530: }
531:
532: static void
533: pa_reclaim_register(struct pool_allocator *pa)
534: {
535: struct vm_map *map = *pa->pa_backingmapptr;
536: struct pool *pp;
537:
538: KASSERT(pa->pa_backingmap == NULL);
539: if (map == NULL) {
540: SLIST_INSERT_HEAD(&pa_deferinitq, pa, pa_q);
541: return;
542: }
543: pa->pa_backingmap = map;
544: TAILQ_FOREACH(pp, &pa->pa_list, pr_alloc_list) {
545: pool_reclaim_register(pp);
546: }
547: }
548:
1.3 pk 549: /*
1.94 simonb 550: * Initialize all the pools listed in the "pools" link set.
551: */
552: void
1.117 yamt 553: pool_subsystem_init(void)
1.94 simonb 554: {
1.117 yamt 555: struct pool_allocator *pa;
1.94 simonb 556: __link_set_decl(pools, struct link_pool_init);
557: struct link_pool_init * const *pi;
558:
559: __link_set_foreach(pi, pools)
560: pool_init((*pi)->pp, (*pi)->size, (*pi)->align,
561: (*pi)->align_offset, (*pi)->flags, (*pi)->wchan,
562: (*pi)->palloc);
1.117 yamt 563:
564: while ((pa = SLIST_FIRST(&pa_deferinitq)) != NULL) {
565: KASSERT(pa->pa_backingmapptr != NULL);
566: KASSERT(*pa->pa_backingmapptr != NULL);
567: SLIST_REMOVE_HEAD(&pa_deferinitq, pa_q);
568: pa_reclaim_register(pa);
569: }
1.94 simonb 570: }
571:
572: /*
1.3 pk 573: * Initialize the given pool resource structure.
574: *
575: * We export this routine to allow other kernel parts to declare
576: * static pools that must be initialized before malloc() is available.
577: */
578: void
1.42 thorpej 579: pool_init(struct pool *pp, size_t size, u_int align, u_int ioff, int flags,
1.66 thorpej 580: const char *wchan, struct pool_allocator *palloc)
1.3 pk 581: {
1.116 simonb 582: #ifdef DEBUG
583: struct pool *pp1;
584: #endif
1.92 enami 585: size_t trysize, phsize;
1.116 simonb 586: int off, slack, s;
1.3 pk 587:
1.99 yamt 588: KASSERT((1UL << (CHAR_BIT * sizeof(pool_item_freelist_t))) - 2 >=
589: PHPOOL_FREELIST_NELEM(PHPOOL_MAX - 1));
590:
1.116 simonb 591: #ifdef DEBUG
592: /*
593: * Check that the pool hasn't already been initialised and
594: * added to the list of all pools.
595: */
596: LIST_FOREACH(pp1, &pool_head, pr_poollist) {
597: if (pp == pp1)
598: panic("pool_init: pool %s already initialised",
599: wchan);
600: }
601: #endif
602:
1.25 thorpej 603: #ifdef POOL_DIAGNOSTIC
604: /*
605: * Always log if POOL_DIAGNOSTIC is defined.
606: */
607: if (pool_logsize != 0)
608: flags |= PR_LOGGING;
609: #endif
610:
1.66 thorpej 611: if (palloc == NULL)
612: palloc = &pool_allocator_kmem;
1.112 bjh21 613: #ifdef POOL_SUBPAGE
614: if (size > palloc->pa_pagesz) {
615: if (palloc == &pool_allocator_kmem)
616: palloc = &pool_allocator_kmem_fullpage;
617: else if (palloc == &pool_allocator_nointr)
618: palloc = &pool_allocator_nointr_fullpage;
619: }
1.66 thorpej 620: #endif /* POOL_SUBPAGE */
621: if ((palloc->pa_flags & PA_INITIALIZED) == 0) {
1.112 bjh21 622: if (palloc->pa_pagesz == 0)
1.66 thorpej 623: palloc->pa_pagesz = PAGE_SIZE;
624:
625: TAILQ_INIT(&palloc->pa_list);
626:
627: simple_lock_init(&palloc->pa_slock);
628: palloc->pa_pagemask = ~(palloc->pa_pagesz - 1);
629: palloc->pa_pageshift = ffs(palloc->pa_pagesz) - 1;
1.117 yamt 630:
631: if (palloc->pa_backingmapptr != NULL) {
632: pa_reclaim_register(palloc);
633: }
1.66 thorpej 634: palloc->pa_flags |= PA_INITIALIZED;
1.4 thorpej 635: }
1.3 pk 636:
637: if (align == 0)
638: align = ALIGN(1);
1.14 thorpej 639:
1.120 yamt 640: if ((flags & PR_NOTOUCH) == 0 && size < sizeof(struct pool_item))
1.14 thorpej 641: size = sizeof(struct pool_item);
1.3 pk 642:
1.78 thorpej 643: size = roundup(size, align);
1.66 thorpej 644: #ifdef DIAGNOSTIC
645: if (size > palloc->pa_pagesz)
1.121 yamt 646: panic("pool_init: pool item size (%zu) too large", size);
1.66 thorpej 647: #endif
1.35 pk 648:
1.3 pk 649: /*
650: * Initialize the pool structure.
651: */
1.88 chs 652: LIST_INIT(&pp->pr_emptypages);
653: LIST_INIT(&pp->pr_fullpages);
654: LIST_INIT(&pp->pr_partpages);
1.102 chs 655: LIST_INIT(&pp->pr_cachelist);
1.3 pk 656: pp->pr_curpage = NULL;
657: pp->pr_npages = 0;
658: pp->pr_minitems = 0;
659: pp->pr_minpages = 0;
660: pp->pr_maxpages = UINT_MAX;
1.20 thorpej 661: pp->pr_roflags = flags;
662: pp->pr_flags = 0;
1.35 pk 663: pp->pr_size = size;
1.3 pk 664: pp->pr_align = align;
665: pp->pr_wchan = wchan;
1.66 thorpej 666: pp->pr_alloc = palloc;
1.20 thorpej 667: pp->pr_nitems = 0;
668: pp->pr_nout = 0;
669: pp->pr_hardlimit = UINT_MAX;
670: pp->pr_hardlimit_warning = NULL;
1.31 thorpej 671: pp->pr_hardlimit_ratecap.tv_sec = 0;
672: pp->pr_hardlimit_ratecap.tv_usec = 0;
673: pp->pr_hardlimit_warning_last.tv_sec = 0;
674: pp->pr_hardlimit_warning_last.tv_usec = 0;
1.68 thorpej 675: pp->pr_drain_hook = NULL;
676: pp->pr_drain_hook_arg = NULL;
1.3 pk 677:
678: /*
679: * Decide whether to put the page header off page to avoid
1.92 enami 680: * wasting too large a part of the page or too big item.
681: * Off-page page headers go on a hash table, so we can match
682: * a returned item with its header based on the page address.
683: * We use 1/16 of the page size and about 8 times of the item
684: * size as the threshold (XXX: tune)
685: *
686: * However, we'll put the header into the page if we can put
687: * it without wasting any items.
688: *
689: * Silently enforce `0 <= ioff < align'.
1.3 pk 690: */
1.92 enami 691: pp->pr_itemoffset = ioff %= align;
692: /* See the comment below about reserved bytes. */
693: trysize = palloc->pa_pagesz - ((align - ioff) % align);
694: phsize = ALIGN(sizeof(struct pool_item_header));
1.121 yamt 695: if ((pp->pr_roflags & (PR_NOTOUCH | PR_NOALIGN)) == 0 &&
1.97 yamt 696: (pp->pr_size < MIN(palloc->pa_pagesz / 16, phsize << 3) ||
697: trysize / pp->pr_size == (trysize - phsize) / pp->pr_size)) {
1.3 pk 698: /* Use the end of the page for the page header */
1.20 thorpej 699: pp->pr_roflags |= PR_PHINPAGE;
1.92 enami 700: pp->pr_phoffset = off = palloc->pa_pagesz - phsize;
1.2 pk 701: } else {
1.3 pk 702: /* The page header will be taken from our page header pool */
703: pp->pr_phoffset = 0;
1.66 thorpej 704: off = palloc->pa_pagesz;
1.88 chs 705: SPLAY_INIT(&pp->pr_phtree);
1.2 pk 706: }
1.1 pk 707:
1.3 pk 708: /*
709: * Alignment is to take place at `ioff' within the item. This means
710: * we must reserve up to `align - 1' bytes on the page to allow
711: * appropriate positioning of each item.
712: */
713: pp->pr_itemsperpage = (off - ((align - ioff) % align)) / pp->pr_size;
1.43 thorpej 714: KASSERT(pp->pr_itemsperpage != 0);
1.97 yamt 715: if ((pp->pr_roflags & PR_NOTOUCH)) {
716: int idx;
717:
718: for (idx = 0; pp->pr_itemsperpage > PHPOOL_FREELIST_NELEM(idx);
719: idx++) {
720: /* nothing */
721: }
722: if (idx >= PHPOOL_MAX) {
723: /*
724: * if you see this panic, consider to tweak
725: * PHPOOL_MAX and PHPOOL_FREELIST_NELEM.
726: */
727: panic("%s: too large itemsperpage(%d) for PR_NOTOUCH",
728: pp->pr_wchan, pp->pr_itemsperpage);
729: }
730: pp->pr_phpool = &phpool[idx];
731: } else if ((pp->pr_roflags & PR_PHINPAGE) == 0) {
732: pp->pr_phpool = &phpool[0];
733: }
734: #if defined(DIAGNOSTIC)
735: else {
736: pp->pr_phpool = NULL;
737: }
738: #endif
1.3 pk 739:
740: /*
741: * Use the slack between the chunks and the page header
742: * for "cache coloring".
743: */
744: slack = off - pp->pr_itemsperpage * pp->pr_size;
745: pp->pr_maxcolor = (slack / align) * align;
746: pp->pr_curcolor = 0;
747:
748: pp->pr_nget = 0;
749: pp->pr_nfail = 0;
750: pp->pr_nput = 0;
751: pp->pr_npagealloc = 0;
752: pp->pr_npagefree = 0;
1.1 pk 753: pp->pr_hiwat = 0;
1.8 thorpej 754: pp->pr_nidle = 0;
1.3 pk 755:
1.59 thorpej 756: #ifdef POOL_DIAGNOSTIC
1.25 thorpej 757: if (flags & PR_LOGGING) {
758: if (kmem_map == NULL ||
759: (pp->pr_log = malloc(pool_logsize * sizeof(struct pool_log),
760: M_TEMP, M_NOWAIT)) == NULL)
1.20 thorpej 761: pp->pr_roflags &= ~PR_LOGGING;
1.3 pk 762: pp->pr_curlogentry = 0;
763: pp->pr_logsize = pool_logsize;
764: }
1.59 thorpej 765: #endif
1.25 thorpej 766:
767: pp->pr_entered_file = NULL;
768: pp->pr_entered_line = 0;
1.3 pk 769:
1.21 thorpej 770: simple_lock_init(&pp->pr_slock);
1.1 pk 771:
1.3 pk 772: /*
1.43 thorpej 773: * Initialize private page header pool and cache magazine pool if we
774: * haven't done so yet.
1.23 thorpej 775: * XXX LOCKING.
1.3 pk 776: */
1.97 yamt 777: if (phpool[0].pr_size == 0) {
778: int idx;
779: for (idx = 0; idx < PHPOOL_MAX; idx++) {
780: static char phpool_names[PHPOOL_MAX][6+1+6+1];
781: int nelem;
782: size_t sz;
783:
784: nelem = PHPOOL_FREELIST_NELEM(idx);
785: snprintf(phpool_names[idx], sizeof(phpool_names[idx]),
786: "phpool-%d", nelem);
787: sz = sizeof(struct pool_item_header);
788: if (nelem) {
789: sz = PR_FREELIST_ALIGN(sz)
1.99 yamt 790: + nelem * sizeof(pool_item_freelist_t);
1.97 yamt 791: }
792: pool_init(&phpool[idx], sz, 0, 0, 0,
1.98 yamt 793: phpool_names[idx], &pool_allocator_meta);
1.97 yamt 794: }
1.62 bjh21 795: #ifdef POOL_SUBPAGE
796: pool_init(&psppool, POOL_SUBPAGE, POOL_SUBPAGE, 0,
1.98 yamt 797: PR_RECURSIVE, "psppool", &pool_allocator_meta);
1.62 bjh21 798: #endif
1.43 thorpej 799: pool_init(&pcgpool, sizeof(struct pool_cache_group), 0, 0,
1.98 yamt 800: 0, "pcgpool", &pool_allocator_meta);
1.1 pk 801: }
802:
1.23 thorpej 803: /* Insert into the list of all pools. */
804: simple_lock(&pool_head_slock);
1.102 chs 805: LIST_INSERT_HEAD(&pool_head, pp, pr_poollist);
1.23 thorpej 806: simple_unlock(&pool_head_slock);
1.66 thorpej 807:
808: /* Insert this into the list of pools using this allocator. */
1.93 dbj 809: s = splvm();
1.66 thorpej 810: simple_lock(&palloc->pa_slock);
811: TAILQ_INSERT_TAIL(&palloc->pa_list, pp, pr_alloc_list);
812: simple_unlock(&palloc->pa_slock);
1.93 dbj 813: splx(s);
1.117 yamt 814: pool_reclaim_register(pp);
1.1 pk 815: }
816:
817: /*
818: * De-commision a pool resource.
819: */
820: void
1.42 thorpej 821: pool_destroy(struct pool *pp)
1.1 pk 822: {
1.101 thorpej 823: struct pool_pagelist pq;
1.3 pk 824: struct pool_item_header *ph;
1.93 dbj 825: int s;
1.43 thorpej 826:
1.101 thorpej 827: /* Remove from global pool list */
828: simple_lock(&pool_head_slock);
1.102 chs 829: LIST_REMOVE(pp, pr_poollist);
1.101 thorpej 830: if (drainpp == pp)
831: drainpp = NULL;
832: simple_unlock(&pool_head_slock);
833:
834: /* Remove this pool from its allocator's list of pools. */
1.117 yamt 835: pool_reclaim_unregister(pp);
1.93 dbj 836: s = splvm();
1.66 thorpej 837: simple_lock(&pp->pr_alloc->pa_slock);
838: TAILQ_REMOVE(&pp->pr_alloc->pa_list, pp, pr_alloc_list);
839: simple_unlock(&pp->pr_alloc->pa_slock);
1.93 dbj 840: splx(s);
1.66 thorpej 841:
1.101 thorpej 842: s = splvm();
843: simple_lock(&pp->pr_slock);
844:
1.102 chs 845: KASSERT(LIST_EMPTY(&pp->pr_cachelist));
1.3 pk 846:
847: #ifdef DIAGNOSTIC
1.20 thorpej 848: if (pp->pr_nout != 0) {
1.25 thorpej 849: pr_printlog(pp, NULL, printf);
1.80 provos 850: panic("pool_destroy: pool busy: still out: %u",
1.20 thorpej 851: pp->pr_nout);
1.3 pk 852: }
853: #endif
1.1 pk 854:
1.101 thorpej 855: KASSERT(LIST_EMPTY(&pp->pr_fullpages));
856: KASSERT(LIST_EMPTY(&pp->pr_partpages));
857:
1.3 pk 858: /* Remove all pages */
1.101 thorpej 859: LIST_INIT(&pq);
1.88 chs 860: while ((ph = LIST_FIRST(&pp->pr_emptypages)) != NULL)
1.101 thorpej 861: pr_rmpage(pp, ph, &pq);
862:
863: simple_unlock(&pp->pr_slock);
864: splx(s);
1.3 pk 865:
1.101 thorpej 866: pr_pagelist_free(pp, &pq);
1.3 pk 867:
1.59 thorpej 868: #ifdef POOL_DIAGNOSTIC
1.20 thorpej 869: if ((pp->pr_roflags & PR_LOGGING) != 0)
1.3 pk 870: free(pp->pr_log, M_TEMP);
1.59 thorpej 871: #endif
1.1 pk 872: }
873:
1.68 thorpej 874: void
875: pool_set_drain_hook(struct pool *pp, void (*fn)(void *, int), void *arg)
876: {
877:
878: /* XXX no locking -- must be used just after pool_init() */
879: #ifdef DIAGNOSTIC
880: if (pp->pr_drain_hook != NULL)
881: panic("pool_set_drain_hook(%s): already set", pp->pr_wchan);
882: #endif
883: pp->pr_drain_hook = fn;
884: pp->pr_drain_hook_arg = arg;
885: }
886:
1.88 chs 887: static struct pool_item_header *
1.55 thorpej 888: pool_alloc_item_header(struct pool *pp, caddr_t storage, int flags)
889: {
890: struct pool_item_header *ph;
891: int s;
892:
893: LOCK_ASSERT(simple_lock_held(&pp->pr_slock) == 0);
894:
895: if ((pp->pr_roflags & PR_PHINPAGE) != 0)
896: ph = (struct pool_item_header *) (storage + pp->pr_phoffset);
897: else {
1.85 pk 898: s = splvm();
1.97 yamt 899: ph = pool_get(pp->pr_phpool, flags);
1.55 thorpej 900: splx(s);
901: }
902:
903: return (ph);
904: }
1.1 pk 905:
906: /*
1.3 pk 907: * Grab an item from the pool; must be called at appropriate spl level
1.1 pk 908: */
1.3 pk 909: void *
1.59 thorpej 910: #ifdef POOL_DIAGNOSTIC
1.42 thorpej 911: _pool_get(struct pool *pp, int flags, const char *file, long line)
1.56 sommerfe 912: #else
913: pool_get(struct pool *pp, int flags)
914: #endif
1.1 pk 915: {
916: struct pool_item *pi;
1.3 pk 917: struct pool_item_header *ph;
1.55 thorpej 918: void *v;
1.1 pk 919:
1.2 pk 920: #ifdef DIAGNOSTIC
1.95 atatat 921: if (__predict_false(pp->pr_itemsperpage == 0))
922: panic("pool_get: pool %p: pr_itemsperpage is zero, "
923: "pool not initialized?", pp);
1.84 thorpej 924: if (__predict_false(curlwp == NULL && doing_shutdown == 0 &&
1.37 sommerfe 925: (flags & PR_WAITOK) != 0))
1.77 matt 926: panic("pool_get: %s: must have NOWAIT", pp->pr_wchan);
1.58 thorpej 927:
1.102 chs 928: #endif /* DIAGNOSTIC */
1.58 thorpej 929: #ifdef LOCKDEBUG
930: if (flags & PR_WAITOK)
1.119 yamt 931: ASSERT_SLEEPABLE(NULL, "pool_get(PR_WAITOK)");
1.56 sommerfe 932: #endif
1.1 pk 933:
1.21 thorpej 934: simple_lock(&pp->pr_slock);
1.25 thorpej 935: pr_enter(pp, file, line);
1.20 thorpej 936:
937: startover:
938: /*
939: * Check to see if we've reached the hard limit. If we have,
940: * and we can wait, then wait until an item has been returned to
941: * the pool.
942: */
943: #ifdef DIAGNOSTIC
1.34 thorpej 944: if (__predict_false(pp->pr_nout > pp->pr_hardlimit)) {
1.25 thorpej 945: pr_leave(pp);
1.21 thorpej 946: simple_unlock(&pp->pr_slock);
1.20 thorpej 947: panic("pool_get: %s: crossed hard limit", pp->pr_wchan);
948: }
949: #endif
1.34 thorpej 950: if (__predict_false(pp->pr_nout == pp->pr_hardlimit)) {
1.68 thorpej 951: if (pp->pr_drain_hook != NULL) {
952: /*
953: * Since the drain hook is going to free things
954: * back to the pool, unlock, call the hook, re-lock,
955: * and check the hardlimit condition again.
956: */
957: pr_leave(pp);
958: simple_unlock(&pp->pr_slock);
959: (*pp->pr_drain_hook)(pp->pr_drain_hook_arg, flags);
960: simple_lock(&pp->pr_slock);
961: pr_enter(pp, file, line);
962: if (pp->pr_nout < pp->pr_hardlimit)
963: goto startover;
964: }
965:
1.29 sommerfe 966: if ((flags & PR_WAITOK) && !(flags & PR_LIMITFAIL)) {
1.20 thorpej 967: /*
968: * XXX: A warning isn't logged in this case. Should
969: * it be?
970: */
971: pp->pr_flags |= PR_WANTED;
1.25 thorpej 972: pr_leave(pp);
1.40 sommerfe 973: ltsleep(pp, PSWP, pp->pr_wchan, 0, &pp->pr_slock);
1.25 thorpej 974: pr_enter(pp, file, line);
1.20 thorpej 975: goto startover;
976: }
1.31 thorpej 977:
978: /*
979: * Log a message that the hard limit has been hit.
980: */
981: if (pp->pr_hardlimit_warning != NULL &&
982: ratecheck(&pp->pr_hardlimit_warning_last,
983: &pp->pr_hardlimit_ratecap))
984: log(LOG_ERR, "%s\n", pp->pr_hardlimit_warning);
1.21 thorpej 985:
986: pp->pr_nfail++;
987:
1.25 thorpej 988: pr_leave(pp);
1.21 thorpej 989: simple_unlock(&pp->pr_slock);
1.20 thorpej 990: return (NULL);
991: }
992:
1.3 pk 993: /*
994: * The convention we use is that if `curpage' is not NULL, then
995: * it points at a non-empty bucket. In particular, `curpage'
996: * never points at a page header which has PR_PHINPAGE set and
997: * has no items in its bucket.
998: */
1.20 thorpej 999: if ((ph = pp->pr_curpage) == NULL) {
1.113 yamt 1000: int error;
1001:
1.20 thorpej 1002: #ifdef DIAGNOSTIC
1003: if (pp->pr_nitems != 0) {
1.21 thorpej 1004: simple_unlock(&pp->pr_slock);
1.20 thorpej 1005: printf("pool_get: %s: curpage NULL, nitems %u\n",
1006: pp->pr_wchan, pp->pr_nitems);
1.80 provos 1007: panic("pool_get: nitems inconsistent");
1.20 thorpej 1008: }
1009: #endif
1010:
1.21 thorpej 1011: /*
1012: * Call the back-end page allocator for more memory.
1013: * Release the pool lock, as the back-end page allocator
1014: * may block.
1015: */
1.25 thorpej 1016: pr_leave(pp);
1.113 yamt 1017: error = pool_grow(pp, flags);
1018: pr_enter(pp, file, line);
1019: if (error != 0) {
1.21 thorpej 1020: /*
1.55 thorpej 1021: * We were unable to allocate a page or item
1022: * header, but we released the lock during
1023: * allocation, so perhaps items were freed
1024: * back to the pool. Check for this case.
1.21 thorpej 1025: */
1026: if (pp->pr_curpage != NULL)
1027: goto startover;
1.15 pk 1028:
1.117 yamt 1029: pp->pr_nfail++;
1.25 thorpej 1030: pr_leave(pp);
1.117 yamt 1031: simple_unlock(&pp->pr_slock);
1032: return (NULL);
1.1 pk 1033: }
1.3 pk 1034:
1.20 thorpej 1035: /* Start the allocation process over. */
1036: goto startover;
1.3 pk 1037: }
1.97 yamt 1038: if (pp->pr_roflags & PR_NOTOUCH) {
1039: #ifdef DIAGNOSTIC
1040: if (__predict_false(ph->ph_nmissing == pp->pr_itemsperpage)) {
1041: pr_leave(pp);
1042: simple_unlock(&pp->pr_slock);
1043: panic("pool_get: %s: page empty", pp->pr_wchan);
1044: }
1045: #endif
1046: v = pr_item_notouch_get(pp, ph);
1047: #ifdef POOL_DIAGNOSTIC
1048: pr_log(pp, v, PRLOG_GET, file, line);
1049: #endif
1050: } else {
1.102 chs 1051: v = pi = LIST_FIRST(&ph->ph_itemlist);
1.97 yamt 1052: if (__predict_false(v == NULL)) {
1053: pr_leave(pp);
1054: simple_unlock(&pp->pr_slock);
1055: panic("pool_get: %s: page empty", pp->pr_wchan);
1056: }
1.20 thorpej 1057: #ifdef DIAGNOSTIC
1.97 yamt 1058: if (__predict_false(pp->pr_nitems == 0)) {
1059: pr_leave(pp);
1060: simple_unlock(&pp->pr_slock);
1061: printf("pool_get: %s: items on itemlist, nitems %u\n",
1062: pp->pr_wchan, pp->pr_nitems);
1063: panic("pool_get: nitems inconsistent");
1064: }
1.65 enami 1065: #endif
1.56 sommerfe 1066:
1.65 enami 1067: #ifdef POOL_DIAGNOSTIC
1.97 yamt 1068: pr_log(pp, v, PRLOG_GET, file, line);
1.65 enami 1069: #endif
1.3 pk 1070:
1.65 enami 1071: #ifdef DIAGNOSTIC
1.97 yamt 1072: if (__predict_false(pi->pi_magic != PI_MAGIC)) {
1073: pr_printlog(pp, pi, printf);
1074: panic("pool_get(%s): free list modified: "
1075: "magic=%x; page %p; item addr %p\n",
1076: pp->pr_wchan, pi->pi_magic, ph->ph_page, pi);
1077: }
1.3 pk 1078: #endif
1079:
1.97 yamt 1080: /*
1081: * Remove from item list.
1082: */
1.102 chs 1083: LIST_REMOVE(pi, pi_list);
1.97 yamt 1084: }
1.20 thorpej 1085: pp->pr_nitems--;
1086: pp->pr_nout++;
1.6 thorpej 1087: if (ph->ph_nmissing == 0) {
1088: #ifdef DIAGNOSTIC
1.34 thorpej 1089: if (__predict_false(pp->pr_nidle == 0))
1.6 thorpej 1090: panic("pool_get: nidle inconsistent");
1091: #endif
1092: pp->pr_nidle--;
1.88 chs 1093:
1094: /*
1095: * This page was previously empty. Move it to the list of
1096: * partially-full pages. This page is already curpage.
1097: */
1098: LIST_REMOVE(ph, ph_pagelist);
1099: LIST_INSERT_HEAD(&pp->pr_partpages, ph, ph_pagelist);
1.6 thorpej 1100: }
1.3 pk 1101: ph->ph_nmissing++;
1.97 yamt 1102: if (ph->ph_nmissing == pp->pr_itemsperpage) {
1.21 thorpej 1103: #ifdef DIAGNOSTIC
1.97 yamt 1104: if (__predict_false((pp->pr_roflags & PR_NOTOUCH) == 0 &&
1.102 chs 1105: !LIST_EMPTY(&ph->ph_itemlist))) {
1.25 thorpej 1106: pr_leave(pp);
1.21 thorpej 1107: simple_unlock(&pp->pr_slock);
1108: panic("pool_get: %s: nmissing inconsistent",
1109: pp->pr_wchan);
1110: }
1111: #endif
1.3 pk 1112: /*
1.88 chs 1113: * This page is now full. Move it to the full list
1114: * and select a new current page.
1.3 pk 1115: */
1.88 chs 1116: LIST_REMOVE(ph, ph_pagelist);
1117: LIST_INSERT_HEAD(&pp->pr_fullpages, ph, ph_pagelist);
1118: pool_update_curpage(pp);
1.1 pk 1119: }
1.3 pk 1120:
1121: pp->pr_nget++;
1.111 christos 1122: pr_leave(pp);
1.20 thorpej 1123:
1124: /*
1125: * If we have a low water mark and we are now below that low
1126: * water mark, add more items to the pool.
1127: */
1.53 thorpej 1128: if (POOL_NEEDS_CATCHUP(pp) && pool_catchup(pp) != 0) {
1.20 thorpej 1129: /*
1130: * XXX: Should we log a warning? Should we set up a timeout
1131: * to try again in a second or so? The latter could break
1132: * a caller's assumptions about interrupt protection, etc.
1133: */
1134: }
1135:
1.21 thorpej 1136: simple_unlock(&pp->pr_slock);
1.122.2.1 ad 1137: KASSERT((((vaddr_t)v + pp->pr_itemoffset) & (pp->pr_align - 1)) == 0);
1.1 pk 1138: return (v);
1139: }
1140:
1141: /*
1.43 thorpej 1142: * Internal version of pool_put(). Pool is already locked/entered.
1.1 pk 1143: */
1.43 thorpej 1144: static void
1.101 thorpej 1145: pool_do_put(struct pool *pp, void *v, struct pool_pagelist *pq)
1.1 pk 1146: {
1147: struct pool_item *pi = v;
1.3 pk 1148: struct pool_item_header *ph;
1149:
1.61 chs 1150: LOCK_ASSERT(simple_lock_held(&pp->pr_slock));
1151:
1.30 thorpej 1152: #ifdef DIAGNOSTIC
1.34 thorpej 1153: if (__predict_false(pp->pr_nout == 0)) {
1.30 thorpej 1154: printf("pool %s: putting with none out\n",
1155: pp->pr_wchan);
1156: panic("pool_put");
1157: }
1158: #endif
1.3 pk 1159:
1.121 yamt 1160: if (__predict_false((ph = pr_find_pagehead(pp, v)) == NULL)) {
1.25 thorpej 1161: pr_printlog(pp, NULL, printf);
1.3 pk 1162: panic("pool_put: %s: page header missing", pp->pr_wchan);
1163: }
1.28 thorpej 1164:
1165: #ifdef LOCKDEBUG
1166: /*
1167: * Check if we're freeing a locked simple lock.
1168: */
1169: simple_lock_freecheck((caddr_t)pi, ((caddr_t)pi) + pp->pr_size);
1170: #endif
1.3 pk 1171:
1172: /*
1173: * Return to item list.
1174: */
1.97 yamt 1175: if (pp->pr_roflags & PR_NOTOUCH) {
1176: pr_item_notouch_put(pp, ph, v);
1177: } else {
1.2 pk 1178: #ifdef DIAGNOSTIC
1.97 yamt 1179: pi->pi_magic = PI_MAGIC;
1.3 pk 1180: #endif
1.32 chs 1181: #ifdef DEBUG
1.97 yamt 1182: {
1183: int i, *ip = v;
1.32 chs 1184:
1.97 yamt 1185: for (i = 0; i < pp->pr_size / sizeof(int); i++) {
1186: *ip++ = PI_MAGIC;
1187: }
1.32 chs 1188: }
1189: #endif
1190:
1.102 chs 1191: LIST_INSERT_HEAD(&ph->ph_itemlist, pi, pi_list);
1.97 yamt 1192: }
1.79 thorpej 1193: KDASSERT(ph->ph_nmissing != 0);
1.3 pk 1194: ph->ph_nmissing--;
1195: pp->pr_nput++;
1.20 thorpej 1196: pp->pr_nitems++;
1197: pp->pr_nout--;
1.3 pk 1198:
1199: /* Cancel "pool empty" condition if it exists */
1200: if (pp->pr_curpage == NULL)
1201: pp->pr_curpage = ph;
1202:
1203: if (pp->pr_flags & PR_WANTED) {
1204: pp->pr_flags &= ~PR_WANTED;
1.15 pk 1205: if (ph->ph_nmissing == 0)
1206: pp->pr_nidle++;
1.3 pk 1207: wakeup((caddr_t)pp);
1208: return;
1209: }
1210:
1211: /*
1.88 chs 1212: * If this page is now empty, do one of two things:
1.21 thorpej 1213: *
1.88 chs 1214: * (1) If we have more pages than the page high water mark,
1.96 thorpej 1215: * free the page back to the system. ONLY CONSIDER
1.90 thorpej 1216: * FREEING BACK A PAGE IF WE HAVE MORE THAN OUR MINIMUM PAGE
1217: * CLAIM.
1.21 thorpej 1218: *
1.88 chs 1219: * (2) Otherwise, move the page to the empty page list.
1220: *
1221: * Either way, select a new current page (so we use a partially-full
1222: * page if one is available).
1.3 pk 1223: */
1224: if (ph->ph_nmissing == 0) {
1.6 thorpej 1225: pp->pr_nidle++;
1.90 thorpej 1226: if (pp->pr_npages > pp->pr_minpages &&
1227: (pp->pr_npages > pp->pr_maxpages ||
1.117 yamt 1228: pa_starved_p(pp->pr_alloc))) {
1.101 thorpej 1229: pr_rmpage(pp, ph, pq);
1.3 pk 1230: } else {
1.88 chs 1231: LIST_REMOVE(ph, ph_pagelist);
1232: LIST_INSERT_HEAD(&pp->pr_emptypages, ph, ph_pagelist);
1.3 pk 1233:
1.21 thorpej 1234: /*
1235: * Update the timestamp on the page. A page must
1236: * be idle for some period of time before it can
1237: * be reclaimed by the pagedaemon. This minimizes
1238: * ping-pong'ing for memory.
1239: */
1.118 kardel 1240: getmicrotime(&ph->ph_time);
1.1 pk 1241: }
1.88 chs 1242: pool_update_curpage(pp);
1.1 pk 1243: }
1.88 chs 1244:
1.21 thorpej 1245: /*
1.88 chs 1246: * If the page was previously completely full, move it to the
1247: * partially-full list and make it the current page. The next
1248: * allocation will get the item from this page, instead of
1249: * further fragmenting the pool.
1.21 thorpej 1250: */
1251: else if (ph->ph_nmissing == (pp->pr_itemsperpage - 1)) {
1.88 chs 1252: LIST_REMOVE(ph, ph_pagelist);
1253: LIST_INSERT_HEAD(&pp->pr_partpages, ph, ph_pagelist);
1.21 thorpej 1254: pp->pr_curpage = ph;
1255: }
1.43 thorpej 1256: }
1257:
1258: /*
1259: * Return resource to the pool; must be called at appropriate spl level
1260: */
1.59 thorpej 1261: #ifdef POOL_DIAGNOSTIC
1.43 thorpej 1262: void
1263: _pool_put(struct pool *pp, void *v, const char *file, long line)
1264: {
1.101 thorpej 1265: struct pool_pagelist pq;
1266:
1267: LIST_INIT(&pq);
1.43 thorpej 1268:
1269: simple_lock(&pp->pr_slock);
1270: pr_enter(pp, file, line);
1271:
1.56 sommerfe 1272: pr_log(pp, v, PRLOG_PUT, file, line);
1273:
1.101 thorpej 1274: pool_do_put(pp, v, &pq);
1.21 thorpej 1275:
1.25 thorpej 1276: pr_leave(pp);
1.21 thorpej 1277: simple_unlock(&pp->pr_slock);
1.101 thorpej 1278:
1.102 chs 1279: pr_pagelist_free(pp, &pq);
1.1 pk 1280: }
1.57 sommerfe 1281: #undef pool_put
1.59 thorpej 1282: #endif /* POOL_DIAGNOSTIC */
1.1 pk 1283:
1.56 sommerfe 1284: void
1285: pool_put(struct pool *pp, void *v)
1286: {
1.101 thorpej 1287: struct pool_pagelist pq;
1288:
1289: LIST_INIT(&pq);
1.56 sommerfe 1290:
1291: simple_lock(&pp->pr_slock);
1.101 thorpej 1292: pool_do_put(pp, v, &pq);
1293: simple_unlock(&pp->pr_slock);
1.56 sommerfe 1294:
1.102 chs 1295: pr_pagelist_free(pp, &pq);
1.56 sommerfe 1296: }
1.57 sommerfe 1297:
1.59 thorpej 1298: #ifdef POOL_DIAGNOSTIC
1.57 sommerfe 1299: #define pool_put(h, v) _pool_put((h), (v), __FILE__, __LINE__)
1.56 sommerfe 1300: #endif
1.74 thorpej 1301:
1302: /*
1.113 yamt 1303: * pool_grow: grow a pool by a page.
1304: *
1305: * => called with pool locked.
1306: * => unlock and relock the pool.
1307: * => return with pool locked.
1308: */
1309:
1310: static int
1311: pool_grow(struct pool *pp, int flags)
1312: {
1313: struct pool_item_header *ph = NULL;
1314: char *cp;
1315:
1316: simple_unlock(&pp->pr_slock);
1317: cp = pool_allocator_alloc(pp, flags);
1318: if (__predict_true(cp != NULL)) {
1319: ph = pool_alloc_item_header(pp, cp, flags);
1320: }
1321: if (__predict_false(cp == NULL || ph == NULL)) {
1322: if (cp != NULL) {
1323: pool_allocator_free(pp, cp);
1324: }
1325: simple_lock(&pp->pr_slock);
1326: return ENOMEM;
1327: }
1328:
1329: simple_lock(&pp->pr_slock);
1330: pool_prime_page(pp, cp, ph);
1331: pp->pr_npagealloc++;
1332: return 0;
1333: }
1334:
1335: /*
1.74 thorpej 1336: * Add N items to the pool.
1337: */
1338: int
1339: pool_prime(struct pool *pp, int n)
1340: {
1.75 simonb 1341: int newpages;
1.113 yamt 1342: int error = 0;
1.74 thorpej 1343:
1344: simple_lock(&pp->pr_slock);
1345:
1346: newpages = roundup(n, pp->pr_itemsperpage) / pp->pr_itemsperpage;
1347:
1348: while (newpages-- > 0) {
1.113 yamt 1349: error = pool_grow(pp, PR_NOWAIT);
1350: if (error) {
1.74 thorpej 1351: break;
1352: }
1353: pp->pr_minpages++;
1354: }
1355:
1356: if (pp->pr_minpages >= pp->pr_maxpages)
1357: pp->pr_maxpages = pp->pr_minpages + 1; /* XXX */
1358:
1359: simple_unlock(&pp->pr_slock);
1.113 yamt 1360: return error;
1.74 thorpej 1361: }
1.55 thorpej 1362:
1363: /*
1.3 pk 1364: * Add a page worth of items to the pool.
1.21 thorpej 1365: *
1366: * Note, we must be called with the pool descriptor LOCKED.
1.3 pk 1367: */
1.55 thorpej 1368: static void
1369: pool_prime_page(struct pool *pp, caddr_t storage, struct pool_item_header *ph)
1.3 pk 1370: {
1371: struct pool_item *pi;
1372: caddr_t cp = storage;
1.122.2.1 ad 1373: const unsigned int align = pp->pr_align;
1374: const unsigned int ioff = pp->pr_itemoffset;
1.55 thorpej 1375: int n;
1.36 pk 1376:
1.91 yamt 1377: LOCK_ASSERT(simple_lock_held(&pp->pr_slock));
1378:
1.66 thorpej 1379: #ifdef DIAGNOSTIC
1.121 yamt 1380: if ((pp->pr_roflags & PR_NOALIGN) == 0 &&
1381: ((uintptr_t)cp & (pp->pr_alloc->pa_pagesz - 1)) != 0)
1.36 pk 1382: panic("pool_prime_page: %s: unaligned page", pp->pr_wchan);
1.66 thorpej 1383: #endif
1.3 pk 1384:
1385: /*
1386: * Insert page header.
1387: */
1.88 chs 1388: LIST_INSERT_HEAD(&pp->pr_emptypages, ph, ph_pagelist);
1.102 chs 1389: LIST_INIT(&ph->ph_itemlist);
1.3 pk 1390: ph->ph_page = storage;
1391: ph->ph_nmissing = 0;
1.118 kardel 1392: getmicrotime(&ph->ph_time);
1.88 chs 1393: if ((pp->pr_roflags & PR_PHINPAGE) == 0)
1394: SPLAY_INSERT(phtree, &pp->pr_phtree, ph);
1.3 pk 1395:
1.6 thorpej 1396: pp->pr_nidle++;
1397:
1.3 pk 1398: /*
1399: * Color this page.
1400: */
1401: cp = (caddr_t)(cp + pp->pr_curcolor);
1402: if ((pp->pr_curcolor += align) > pp->pr_maxcolor)
1403: pp->pr_curcolor = 0;
1404:
1405: /*
1406: * Adjust storage to apply aligment to `pr_itemoffset' in each item.
1407: */
1408: if (ioff != 0)
1409: cp = (caddr_t)(cp + (align - ioff));
1410:
1.122.2.1 ad 1411: KASSERT((((vaddr_t)cp + ioff) & (align - 1)) == 0);
1412:
1.3 pk 1413: /*
1414: * Insert remaining chunks on the bucket list.
1415: */
1416: n = pp->pr_itemsperpage;
1.20 thorpej 1417: pp->pr_nitems += n;
1.3 pk 1418:
1.97 yamt 1419: if (pp->pr_roflags & PR_NOTOUCH) {
1.99 yamt 1420: pool_item_freelist_t *freelist = PR_FREELIST(ph);
1.97 yamt 1421: int i;
1422:
1.99 yamt 1423: ph->ph_off = cp - storage;
1.97 yamt 1424: ph->ph_firstfree = 0;
1425: for (i = 0; i < n - 1; i++)
1426: freelist[i] = i + 1;
1427: freelist[n - 1] = PR_INDEX_EOL;
1428: } else {
1429: while (n--) {
1430: pi = (struct pool_item *)cp;
1.78 thorpej 1431:
1.97 yamt 1432: KASSERT(((((vaddr_t)pi) + ioff) & (align - 1)) == 0);
1.3 pk 1433:
1.97 yamt 1434: /* Insert on page list */
1.102 chs 1435: LIST_INSERT_HEAD(&ph->ph_itemlist, pi, pi_list);
1.3 pk 1436: #ifdef DIAGNOSTIC
1.97 yamt 1437: pi->pi_magic = PI_MAGIC;
1.3 pk 1438: #endif
1.97 yamt 1439: cp = (caddr_t)(cp + pp->pr_size);
1.122.2.1 ad 1440:
1441: KASSERT((((vaddr_t)cp + ioff) & (align - 1)) == 0);
1.97 yamt 1442: }
1.3 pk 1443: }
1444:
1445: /*
1446: * If the pool was depleted, point at the new page.
1447: */
1448: if (pp->pr_curpage == NULL)
1449: pp->pr_curpage = ph;
1450:
1451: if (++pp->pr_npages > pp->pr_hiwat)
1452: pp->pr_hiwat = pp->pr_npages;
1453: }
1454:
1.20 thorpej 1455: /*
1.52 thorpej 1456: * Used by pool_get() when nitems drops below the low water mark. This
1.88 chs 1457: * is used to catch up pr_nitems with the low water mark.
1.20 thorpej 1458: *
1.21 thorpej 1459: * Note 1, we never wait for memory here, we let the caller decide what to do.
1.20 thorpej 1460: *
1.73 thorpej 1461: * Note 2, we must be called with the pool already locked, and we return
1.20 thorpej 1462: * with it locked.
1463: */
1464: static int
1.42 thorpej 1465: pool_catchup(struct pool *pp)
1.20 thorpej 1466: {
1467: int error = 0;
1468:
1.54 thorpej 1469: while (POOL_NEEDS_CATCHUP(pp)) {
1.113 yamt 1470: error = pool_grow(pp, PR_NOWAIT);
1471: if (error) {
1.20 thorpej 1472: break;
1473: }
1474: }
1.113 yamt 1475: return error;
1.20 thorpej 1476: }
1477:
1.88 chs 1478: static void
1479: pool_update_curpage(struct pool *pp)
1480: {
1481:
1482: pp->pr_curpage = LIST_FIRST(&pp->pr_partpages);
1483: if (pp->pr_curpage == NULL) {
1484: pp->pr_curpage = LIST_FIRST(&pp->pr_emptypages);
1485: }
1486: }
1487:
1.3 pk 1488: void
1.42 thorpej 1489: pool_setlowat(struct pool *pp, int n)
1.3 pk 1490: {
1.15 pk 1491:
1.21 thorpej 1492: simple_lock(&pp->pr_slock);
1493:
1.3 pk 1494: pp->pr_minitems = n;
1.15 pk 1495: pp->pr_minpages = (n == 0)
1496: ? 0
1.18 thorpej 1497: : roundup(n, pp->pr_itemsperpage) / pp->pr_itemsperpage;
1.20 thorpej 1498:
1499: /* Make sure we're caught up with the newly-set low water mark. */
1.75 simonb 1500: if (POOL_NEEDS_CATCHUP(pp) && pool_catchup(pp) != 0) {
1.20 thorpej 1501: /*
1502: * XXX: Should we log a warning? Should we set up a timeout
1503: * to try again in a second or so? The latter could break
1504: * a caller's assumptions about interrupt protection, etc.
1505: */
1506: }
1.21 thorpej 1507:
1508: simple_unlock(&pp->pr_slock);
1.3 pk 1509: }
1510:
1511: void
1.42 thorpej 1512: pool_sethiwat(struct pool *pp, int n)
1.3 pk 1513: {
1.15 pk 1514:
1.21 thorpej 1515: simple_lock(&pp->pr_slock);
1516:
1.15 pk 1517: pp->pr_maxpages = (n == 0)
1518: ? 0
1.18 thorpej 1519: : roundup(n, pp->pr_itemsperpage) / pp->pr_itemsperpage;
1.21 thorpej 1520:
1521: simple_unlock(&pp->pr_slock);
1.3 pk 1522: }
1523:
1.20 thorpej 1524: void
1.42 thorpej 1525: pool_sethardlimit(struct pool *pp, int n, const char *warnmess, int ratecap)
1.20 thorpej 1526: {
1527:
1.21 thorpej 1528: simple_lock(&pp->pr_slock);
1.20 thorpej 1529:
1530: pp->pr_hardlimit = n;
1531: pp->pr_hardlimit_warning = warnmess;
1.31 thorpej 1532: pp->pr_hardlimit_ratecap.tv_sec = ratecap;
1533: pp->pr_hardlimit_warning_last.tv_sec = 0;
1534: pp->pr_hardlimit_warning_last.tv_usec = 0;
1.20 thorpej 1535:
1536: /*
1.21 thorpej 1537: * In-line version of pool_sethiwat(), because we don't want to
1538: * release the lock.
1.20 thorpej 1539: */
1540: pp->pr_maxpages = (n == 0)
1541: ? 0
1542: : roundup(n, pp->pr_itemsperpage) / pp->pr_itemsperpage;
1.21 thorpej 1543:
1544: simple_unlock(&pp->pr_slock);
1.20 thorpej 1545: }
1.3 pk 1546:
1547: /*
1548: * Release all complete pages that have not been used recently.
1549: */
1.66 thorpej 1550: int
1.59 thorpej 1551: #ifdef POOL_DIAGNOSTIC
1.42 thorpej 1552: _pool_reclaim(struct pool *pp, const char *file, long line)
1.56 sommerfe 1553: #else
1554: pool_reclaim(struct pool *pp)
1555: #endif
1.3 pk 1556: {
1557: struct pool_item_header *ph, *phnext;
1.43 thorpej 1558: struct pool_cache *pc;
1.61 chs 1559: struct pool_pagelist pq;
1.102 chs 1560: struct pool_cache_grouplist pcgl;
1561: struct timeval curtime, diff;
1.3 pk 1562:
1.68 thorpej 1563: if (pp->pr_drain_hook != NULL) {
1564: /*
1565: * The drain hook must be called with the pool unlocked.
1566: */
1567: (*pp->pr_drain_hook)(pp->pr_drain_hook_arg, PR_NOWAIT);
1568: }
1569:
1.21 thorpej 1570: if (simple_lock_try(&pp->pr_slock) == 0)
1.66 thorpej 1571: return (0);
1.25 thorpej 1572: pr_enter(pp, file, line);
1.68 thorpej 1573:
1.88 chs 1574: LIST_INIT(&pq);
1.102 chs 1575: LIST_INIT(&pcgl);
1.3 pk 1576:
1.43 thorpej 1577: /*
1578: * Reclaim items from the pool's caches.
1579: */
1.102 chs 1580: LIST_FOREACH(pc, &pp->pr_cachelist, pc_poollist)
1581: pool_cache_reclaim(pc, &pq, &pcgl);
1.43 thorpej 1582:
1.118 kardel 1583: getmicrotime(&curtime);
1.21 thorpej 1584:
1.88 chs 1585: for (ph = LIST_FIRST(&pp->pr_emptypages); ph != NULL; ph = phnext) {
1586: phnext = LIST_NEXT(ph, ph_pagelist);
1.3 pk 1587:
1588: /* Check our minimum page claim */
1589: if (pp->pr_npages <= pp->pr_minpages)
1590: break;
1591:
1.88 chs 1592: KASSERT(ph->ph_nmissing == 0);
1593: timersub(&curtime, &ph->ph_time, &diff);
1.117 yamt 1594: if (diff.tv_sec < pool_inactive_time
1595: && !pa_starved_p(pp->pr_alloc))
1.88 chs 1596: continue;
1.21 thorpej 1597:
1.88 chs 1598: /*
1599: * If freeing this page would put us below
1600: * the low water mark, stop now.
1601: */
1602: if ((pp->pr_nitems - pp->pr_itemsperpage) <
1603: pp->pr_minitems)
1604: break;
1.21 thorpej 1605:
1.88 chs 1606: pr_rmpage(pp, ph, &pq);
1.3 pk 1607: }
1608:
1.25 thorpej 1609: pr_leave(pp);
1.21 thorpej 1610: simple_unlock(&pp->pr_slock);
1.102 chs 1611: if (LIST_EMPTY(&pq) && LIST_EMPTY(&pcgl))
1612: return 0;
1.66 thorpej 1613:
1.101 thorpej 1614: pr_pagelist_free(pp, &pq);
1.102 chs 1615: pcg_grouplist_free(&pcgl);
1.66 thorpej 1616: return (1);
1.3 pk 1617: }
1618:
1619: /*
1620: * Drain pools, one at a time.
1.21 thorpej 1621: *
1622: * Note, we must never be called from an interrupt context.
1.3 pk 1623: */
1624: void
1.42 thorpej 1625: pool_drain(void *arg)
1.3 pk 1626: {
1627: struct pool *pp;
1.23 thorpej 1628: int s;
1.3 pk 1629:
1.61 chs 1630: pp = NULL;
1.49 thorpej 1631: s = splvm();
1.23 thorpej 1632: simple_lock(&pool_head_slock);
1.61 chs 1633: if (drainpp == NULL) {
1.102 chs 1634: drainpp = LIST_FIRST(&pool_head);
1.61 chs 1635: }
1636: if (drainpp) {
1637: pp = drainpp;
1.102 chs 1638: drainpp = LIST_NEXT(pp, pr_poollist);
1.61 chs 1639: }
1640: simple_unlock(&pool_head_slock);
1.115 christos 1641: if (pp)
1642: pool_reclaim(pp);
1.61 chs 1643: splx(s);
1.3 pk 1644: }
1645:
1646: /*
1647: * Diagnostic helpers.
1648: */
1649: void
1.42 thorpej 1650: pool_print(struct pool *pp, const char *modif)
1.21 thorpej 1651: {
1652: int s;
1653:
1.49 thorpej 1654: s = splvm();
1.25 thorpej 1655: if (simple_lock_try(&pp->pr_slock) == 0) {
1656: printf("pool %s is locked; try again later\n",
1657: pp->pr_wchan);
1658: splx(s);
1659: return;
1660: }
1661: pool_print1(pp, modif, printf);
1.21 thorpej 1662: simple_unlock(&pp->pr_slock);
1663: splx(s);
1664: }
1665:
1.25 thorpej 1666: void
1.108 yamt 1667: pool_printall(const char *modif, void (*pr)(const char *, ...))
1668: {
1669: struct pool *pp;
1670:
1671: if (simple_lock_try(&pool_head_slock) == 0) {
1672: (*pr)("WARNING: pool_head_slock is locked\n");
1673: } else {
1674: simple_unlock(&pool_head_slock);
1675: }
1676:
1677: LIST_FOREACH(pp, &pool_head, pr_poollist) {
1678: pool_printit(pp, modif, pr);
1679: }
1680: }
1681:
1682: void
1.42 thorpej 1683: pool_printit(struct pool *pp, const char *modif, void (*pr)(const char *, ...))
1.25 thorpej 1684: {
1685:
1686: if (pp == NULL) {
1687: (*pr)("Must specify a pool to print.\n");
1688: return;
1689: }
1690:
1691: /*
1692: * Called from DDB; interrupts should be blocked, and all
1693: * other processors should be paused. We can skip locking
1694: * the pool in this case.
1695: *
1696: * We do a simple_lock_try() just to print the lock
1697: * status, however.
1698: */
1699:
1700: if (simple_lock_try(&pp->pr_slock) == 0)
1701: (*pr)("WARNING: pool %s is locked\n", pp->pr_wchan);
1702: else
1.107 yamt 1703: simple_unlock(&pp->pr_slock);
1.25 thorpej 1704:
1705: pool_print1(pp, modif, pr);
1706: }
1707:
1.21 thorpej 1708: static void
1.97 yamt 1709: pool_print_pagelist(struct pool *pp, struct pool_pagelist *pl,
1710: void (*pr)(const char *, ...))
1.88 chs 1711: {
1712: struct pool_item_header *ph;
1713: #ifdef DIAGNOSTIC
1714: struct pool_item *pi;
1715: #endif
1716:
1717: LIST_FOREACH(ph, pl, ph_pagelist) {
1718: (*pr)("\t\tpage %p, nmissing %d, time %lu,%lu\n",
1719: ph->ph_page, ph->ph_nmissing,
1720: (u_long)ph->ph_time.tv_sec,
1721: (u_long)ph->ph_time.tv_usec);
1722: #ifdef DIAGNOSTIC
1.97 yamt 1723: if (!(pp->pr_roflags & PR_NOTOUCH)) {
1.102 chs 1724: LIST_FOREACH(pi, &ph->ph_itemlist, pi_list) {
1.97 yamt 1725: if (pi->pi_magic != PI_MAGIC) {
1726: (*pr)("\t\t\titem %p, magic 0x%x\n",
1727: pi, pi->pi_magic);
1728: }
1.88 chs 1729: }
1730: }
1731: #endif
1732: }
1733: }
1734:
1735: static void
1.42 thorpej 1736: pool_print1(struct pool *pp, const char *modif, void (*pr)(const char *, ...))
1.3 pk 1737: {
1.25 thorpej 1738: struct pool_item_header *ph;
1.44 thorpej 1739: struct pool_cache *pc;
1740: struct pool_cache_group *pcg;
1741: int i, print_log = 0, print_pagelist = 0, print_cache = 0;
1.25 thorpej 1742: char c;
1743:
1744: while ((c = *modif++) != '\0') {
1745: if (c == 'l')
1746: print_log = 1;
1747: if (c == 'p')
1748: print_pagelist = 1;
1.44 thorpej 1749: if (c == 'c')
1750: print_cache = 1;
1.25 thorpej 1751: }
1752:
1753: (*pr)("POOL %s: size %u, align %u, ioff %u, roflags 0x%08x\n",
1754: pp->pr_wchan, pp->pr_size, pp->pr_align, pp->pr_itemoffset,
1755: pp->pr_roflags);
1.66 thorpej 1756: (*pr)("\talloc %p\n", pp->pr_alloc);
1.25 thorpej 1757: (*pr)("\tminitems %u, minpages %u, maxpages %u, npages %u\n",
1758: pp->pr_minitems, pp->pr_minpages, pp->pr_maxpages, pp->pr_npages);
1759: (*pr)("\titemsperpage %u, nitems %u, nout %u, hardlimit %u\n",
1760: pp->pr_itemsperpage, pp->pr_nitems, pp->pr_nout, pp->pr_hardlimit);
1761:
1762: (*pr)("\n\tnget %lu, nfail %lu, nput %lu\n",
1763: pp->pr_nget, pp->pr_nfail, pp->pr_nput);
1764: (*pr)("\tnpagealloc %lu, npagefree %lu, hiwat %u, nidle %lu\n",
1765: pp->pr_npagealloc, pp->pr_npagefree, pp->pr_hiwat, pp->pr_nidle);
1766:
1767: if (print_pagelist == 0)
1768: goto skip_pagelist;
1769:
1.88 chs 1770: if ((ph = LIST_FIRST(&pp->pr_emptypages)) != NULL)
1771: (*pr)("\n\tempty page list:\n");
1.97 yamt 1772: pool_print_pagelist(pp, &pp->pr_emptypages, pr);
1.88 chs 1773: if ((ph = LIST_FIRST(&pp->pr_fullpages)) != NULL)
1774: (*pr)("\n\tfull page list:\n");
1.97 yamt 1775: pool_print_pagelist(pp, &pp->pr_fullpages, pr);
1.88 chs 1776: if ((ph = LIST_FIRST(&pp->pr_partpages)) != NULL)
1777: (*pr)("\n\tpartial-page list:\n");
1.97 yamt 1778: pool_print_pagelist(pp, &pp->pr_partpages, pr);
1.88 chs 1779:
1.25 thorpej 1780: if (pp->pr_curpage == NULL)
1781: (*pr)("\tno current page\n");
1782: else
1783: (*pr)("\tcurpage %p\n", pp->pr_curpage->ph_page);
1784:
1785: skip_pagelist:
1786: if (print_log == 0)
1787: goto skip_log;
1788:
1789: (*pr)("\n");
1790: if ((pp->pr_roflags & PR_LOGGING) == 0)
1791: (*pr)("\tno log\n");
1.122 christos 1792: else {
1.25 thorpej 1793: pr_printlog(pp, NULL, pr);
1.122 christos 1794: }
1.3 pk 1795:
1.25 thorpej 1796: skip_log:
1.44 thorpej 1797: if (print_cache == 0)
1798: goto skip_cache;
1799:
1.102 chs 1800: #define PR_GROUPLIST(pcg) \
1801: (*pr)("\t\tgroup %p: avail %d\n", pcg, pcg->pcg_avail); \
1802: for (i = 0; i < PCG_NOBJECTS; i++) { \
1803: if (pcg->pcg_objects[i].pcgo_pa != \
1804: POOL_PADDR_INVALID) { \
1805: (*pr)("\t\t\t%p, 0x%llx\n", \
1806: pcg->pcg_objects[i].pcgo_va, \
1807: (unsigned long long) \
1808: pcg->pcg_objects[i].pcgo_pa); \
1809: } else { \
1810: (*pr)("\t\t\t%p\n", \
1811: pcg->pcg_objects[i].pcgo_va); \
1812: } \
1813: }
1814:
1815: LIST_FOREACH(pc, &pp->pr_cachelist, pc_poollist) {
1.103 chs 1816: (*pr)("\tcache %p\n", pc);
1.48 thorpej 1817: (*pr)("\t hits %lu misses %lu ngroups %lu nitems %lu\n",
1818: pc->pc_hits, pc->pc_misses, pc->pc_ngroups, pc->pc_nitems);
1.102 chs 1819: (*pr)("\t full groups:\n");
1.103 chs 1820: LIST_FOREACH(pcg, &pc->pc_fullgroups, pcg_list) {
1.102 chs 1821: PR_GROUPLIST(pcg);
1.103 chs 1822: }
1.102 chs 1823: (*pr)("\t partial groups:\n");
1.103 chs 1824: LIST_FOREACH(pcg, &pc->pc_partgroups, pcg_list) {
1.102 chs 1825: PR_GROUPLIST(pcg);
1.103 chs 1826: }
1.102 chs 1827: (*pr)("\t empty groups:\n");
1.103 chs 1828: LIST_FOREACH(pcg, &pc->pc_emptygroups, pcg_list) {
1.102 chs 1829: PR_GROUPLIST(pcg);
1.103 chs 1830: }
1.44 thorpej 1831: }
1.102 chs 1832: #undef PR_GROUPLIST
1.44 thorpej 1833:
1834: skip_cache:
1.88 chs 1835: pr_enter_check(pp, pr);
1836: }
1837:
1838: static int
1839: pool_chk_page(struct pool *pp, const char *label, struct pool_item_header *ph)
1840: {
1841: struct pool_item *pi;
1842: caddr_t page;
1843: int n;
1844:
1.121 yamt 1845: if ((pp->pr_roflags & PR_NOALIGN) == 0) {
1846: page = (caddr_t)((uintptr_t)ph & pp->pr_alloc->pa_pagemask);
1847: if (page != ph->ph_page &&
1848: (pp->pr_roflags & PR_PHINPAGE) != 0) {
1849: if (label != NULL)
1850: printf("%s: ", label);
1851: printf("pool(%p:%s): page inconsistency: page %p;"
1852: " at page head addr %p (p %p)\n", pp,
1853: pp->pr_wchan, ph->ph_page,
1854: ph, page);
1855: return 1;
1856: }
1.88 chs 1857: }
1.3 pk 1858:
1.97 yamt 1859: if ((pp->pr_roflags & PR_NOTOUCH) != 0)
1860: return 0;
1861:
1.102 chs 1862: for (pi = LIST_FIRST(&ph->ph_itemlist), n = 0;
1.88 chs 1863: pi != NULL;
1.102 chs 1864: pi = LIST_NEXT(pi,pi_list), n++) {
1.88 chs 1865:
1866: #ifdef DIAGNOSTIC
1867: if (pi->pi_magic != PI_MAGIC) {
1868: if (label != NULL)
1869: printf("%s: ", label);
1870: printf("pool(%s): free list modified: magic=%x;"
1.121 yamt 1871: " page %p; item ordinal %d; addr %p\n",
1.88 chs 1872: pp->pr_wchan, pi->pi_magic, ph->ph_page,
1.121 yamt 1873: n, pi);
1.88 chs 1874: panic("pool");
1875: }
1876: #endif
1.121 yamt 1877: if ((pp->pr_roflags & PR_NOALIGN) != 0) {
1878: continue;
1879: }
1880: page = (caddr_t)((uintptr_t)pi & pp->pr_alloc->pa_pagemask);
1.88 chs 1881: if (page == ph->ph_page)
1882: continue;
1883:
1884: if (label != NULL)
1885: printf("%s: ", label);
1886: printf("pool(%p:%s): page inconsistency: page %p;"
1887: " item ordinal %d; addr %p (p %p)\n", pp,
1888: pp->pr_wchan, ph->ph_page,
1889: n, pi, page);
1890: return 1;
1891: }
1892: return 0;
1.3 pk 1893: }
1894:
1.88 chs 1895:
1.3 pk 1896: int
1.42 thorpej 1897: pool_chk(struct pool *pp, const char *label)
1.3 pk 1898: {
1899: struct pool_item_header *ph;
1900: int r = 0;
1901:
1.21 thorpej 1902: simple_lock(&pp->pr_slock);
1.88 chs 1903: LIST_FOREACH(ph, &pp->pr_emptypages, ph_pagelist) {
1904: r = pool_chk_page(pp, label, ph);
1905: if (r) {
1906: goto out;
1907: }
1908: }
1909: LIST_FOREACH(ph, &pp->pr_fullpages, ph_pagelist) {
1910: r = pool_chk_page(pp, label, ph);
1911: if (r) {
1.3 pk 1912: goto out;
1913: }
1.88 chs 1914: }
1915: LIST_FOREACH(ph, &pp->pr_partpages, ph_pagelist) {
1916: r = pool_chk_page(pp, label, ph);
1917: if (r) {
1.3 pk 1918: goto out;
1919: }
1920: }
1.88 chs 1921:
1.3 pk 1922: out:
1.21 thorpej 1923: simple_unlock(&pp->pr_slock);
1.3 pk 1924: return (r);
1.43 thorpej 1925: }
1926:
1927: /*
1928: * pool_cache_init:
1929: *
1930: * Initialize a pool cache.
1931: *
1932: * NOTE: If the pool must be protected from interrupts, we expect
1933: * to be called at the appropriate interrupt priority level.
1934: */
1935: void
1936: pool_cache_init(struct pool_cache *pc, struct pool *pp,
1937: int (*ctor)(void *, void *, int),
1938: void (*dtor)(void *, void *),
1939: void *arg)
1940: {
1941:
1.102 chs 1942: LIST_INIT(&pc->pc_emptygroups);
1943: LIST_INIT(&pc->pc_fullgroups);
1944: LIST_INIT(&pc->pc_partgroups);
1.43 thorpej 1945: simple_lock_init(&pc->pc_slock);
1946:
1947: pc->pc_pool = pp;
1948:
1949: pc->pc_ctor = ctor;
1950: pc->pc_dtor = dtor;
1951: pc->pc_arg = arg;
1952:
1.48 thorpej 1953: pc->pc_hits = 0;
1954: pc->pc_misses = 0;
1955:
1956: pc->pc_ngroups = 0;
1957:
1958: pc->pc_nitems = 0;
1959:
1.43 thorpej 1960: simple_lock(&pp->pr_slock);
1.102 chs 1961: LIST_INSERT_HEAD(&pp->pr_cachelist, pc, pc_poollist);
1.43 thorpej 1962: simple_unlock(&pp->pr_slock);
1963: }
1964:
1965: /*
1966: * pool_cache_destroy:
1967: *
1968: * Destroy a pool cache.
1969: */
1970: void
1971: pool_cache_destroy(struct pool_cache *pc)
1972: {
1973: struct pool *pp = pc->pc_pool;
1974:
1975: /* First, invalidate the entire cache. */
1976: pool_cache_invalidate(pc);
1977:
1978: /* ...and remove it from the pool's cache list. */
1979: simple_lock(&pp->pr_slock);
1.102 chs 1980: LIST_REMOVE(pc, pc_poollist);
1.43 thorpej 1981: simple_unlock(&pp->pr_slock);
1982: }
1983:
1.110 perry 1984: static inline void *
1.87 thorpej 1985: pcg_get(struct pool_cache_group *pcg, paddr_t *pap)
1.43 thorpej 1986: {
1987: void *object;
1988: u_int idx;
1989:
1990: KASSERT(pcg->pcg_avail <= PCG_NOBJECTS);
1.45 thorpej 1991: KASSERT(pcg->pcg_avail != 0);
1.43 thorpej 1992: idx = --pcg->pcg_avail;
1993:
1.87 thorpej 1994: KASSERT(pcg->pcg_objects[idx].pcgo_va != NULL);
1995: object = pcg->pcg_objects[idx].pcgo_va;
1996: if (pap != NULL)
1997: *pap = pcg->pcg_objects[idx].pcgo_pa;
1998: pcg->pcg_objects[idx].pcgo_va = NULL;
1.43 thorpej 1999:
2000: return (object);
2001: }
2002:
1.110 perry 2003: static inline void
1.87 thorpej 2004: pcg_put(struct pool_cache_group *pcg, void *object, paddr_t pa)
1.43 thorpej 2005: {
2006: u_int idx;
2007:
2008: KASSERT(pcg->pcg_avail < PCG_NOBJECTS);
2009: idx = pcg->pcg_avail++;
2010:
1.87 thorpej 2011: KASSERT(pcg->pcg_objects[idx].pcgo_va == NULL);
2012: pcg->pcg_objects[idx].pcgo_va = object;
2013: pcg->pcg_objects[idx].pcgo_pa = pa;
1.43 thorpej 2014: }
2015:
1.102 chs 2016: static void
2017: pcg_grouplist_free(struct pool_cache_grouplist *pcgl)
2018: {
2019: struct pool_cache_group *pcg;
2020: int s;
2021:
2022: s = splvm();
2023: while ((pcg = LIST_FIRST(pcgl)) != NULL) {
2024: LIST_REMOVE(pcg, pcg_list);
2025: pool_put(&pcgpool, pcg);
2026: }
2027: splx(s);
2028: }
2029:
1.43 thorpej 2030: /*
1.87 thorpej 2031: * pool_cache_get{,_paddr}:
1.43 thorpej 2032: *
1.87 thorpej 2033: * Get an object from a pool cache (optionally returning
2034: * the physical address of the object).
1.43 thorpej 2035: */
2036: void *
1.87 thorpej 2037: pool_cache_get_paddr(struct pool_cache *pc, int flags, paddr_t *pap)
1.43 thorpej 2038: {
2039: struct pool_cache_group *pcg;
2040: void *object;
1.58 thorpej 2041:
2042: #ifdef LOCKDEBUG
2043: if (flags & PR_WAITOK)
1.119 yamt 2044: ASSERT_SLEEPABLE(NULL, "pool_cache_get(PR_WAITOK)");
1.58 thorpej 2045: #endif
1.43 thorpej 2046:
2047: simple_lock(&pc->pc_slock);
2048:
1.102 chs 2049: pcg = LIST_FIRST(&pc->pc_partgroups);
2050: if (pcg == NULL) {
2051: pcg = LIST_FIRST(&pc->pc_fullgroups);
2052: if (pcg != NULL) {
2053: LIST_REMOVE(pcg, pcg_list);
2054: LIST_INSERT_HEAD(&pc->pc_partgroups, pcg, pcg_list);
1.43 thorpej 2055: }
1.102 chs 2056: }
2057: if (pcg == NULL) {
1.43 thorpej 2058:
2059: /*
2060: * No groups with any available objects. Allocate
2061: * a new object, construct it, and return it to
2062: * the caller. We will allocate a group, if necessary,
2063: * when the object is freed back to the cache.
2064: */
1.48 thorpej 2065: pc->pc_misses++;
1.43 thorpej 2066: simple_unlock(&pc->pc_slock);
2067: object = pool_get(pc->pc_pool, flags);
2068: if (object != NULL && pc->pc_ctor != NULL) {
2069: if ((*pc->pc_ctor)(pc->pc_arg, object, flags) != 0) {
2070: pool_put(pc->pc_pool, object);
2071: return (NULL);
2072: }
2073: }
1.122.2.1 ad 2074: KASSERT((((vaddr_t)object + pc->pc_pool->pr_itemoffset) &
2075: (pc->pc_pool->pr_align - 1)) == 0);
1.87 thorpej 2076: if (object != NULL && pap != NULL) {
2077: #ifdef POOL_VTOPHYS
2078: *pap = POOL_VTOPHYS(object);
2079: #else
2080: *pap = POOL_PADDR_INVALID;
2081: #endif
2082: }
1.43 thorpej 2083: return (object);
2084: }
2085:
1.48 thorpej 2086: pc->pc_hits++;
2087: pc->pc_nitems--;
1.87 thorpej 2088: object = pcg_get(pcg, pap);
1.43 thorpej 2089:
1.102 chs 2090: if (pcg->pcg_avail == 0) {
2091: LIST_REMOVE(pcg, pcg_list);
2092: LIST_INSERT_HEAD(&pc->pc_emptygroups, pcg, pcg_list);
2093: }
1.43 thorpej 2094: simple_unlock(&pc->pc_slock);
2095:
1.122.2.1 ad 2096: KASSERT((((vaddr_t)object + pc->pc_pool->pr_itemoffset) &
2097: (pc->pc_pool->pr_align - 1)) == 0);
1.43 thorpej 2098: return (object);
2099: }
2100:
2101: /*
1.87 thorpej 2102: * pool_cache_put{,_paddr}:
1.43 thorpej 2103: *
1.87 thorpej 2104: * Put an object back to the pool cache (optionally caching the
2105: * physical address of the object).
1.43 thorpej 2106: */
2107: void
1.87 thorpej 2108: pool_cache_put_paddr(struct pool_cache *pc, void *object, paddr_t pa)
1.43 thorpej 2109: {
2110: struct pool_cache_group *pcg;
1.60 thorpej 2111: int s;
1.43 thorpej 2112:
1.109 christos 2113: if (__predict_false((pc->pc_pool->pr_flags & PR_WANTED) != 0)) {
2114: goto destruct;
2115: }
2116:
1.43 thorpej 2117: simple_lock(&pc->pc_slock);
2118:
1.102 chs 2119: pcg = LIST_FIRST(&pc->pc_partgroups);
2120: if (pcg == NULL) {
2121: pcg = LIST_FIRST(&pc->pc_emptygroups);
2122: if (pcg != NULL) {
2123: LIST_REMOVE(pcg, pcg_list);
2124: LIST_INSERT_HEAD(&pc->pc_partgroups, pcg, pcg_list);
1.43 thorpej 2125: }
1.102 chs 2126: }
2127: if (pcg == NULL) {
1.43 thorpej 2128:
2129: /*
2130: * No empty groups to free the object to. Attempt to
1.47 thorpej 2131: * allocate one.
1.43 thorpej 2132: */
1.47 thorpej 2133: simple_unlock(&pc->pc_slock);
1.60 thorpej 2134: s = splvm();
1.43 thorpej 2135: pcg = pool_get(&pcgpool, PR_NOWAIT);
1.60 thorpej 2136: splx(s);
1.102 chs 2137: if (pcg == NULL) {
1.109 christos 2138: destruct:
1.102 chs 2139:
2140: /*
2141: * Unable to allocate a cache group; destruct the object
2142: * and free it back to the pool.
2143: */
2144: pool_cache_destruct_object(pc, object);
2145: return;
1.43 thorpej 2146: }
1.102 chs 2147: memset(pcg, 0, sizeof(*pcg));
2148: simple_lock(&pc->pc_slock);
2149: pc->pc_ngroups++;
2150: LIST_INSERT_HEAD(&pc->pc_partgroups, pcg, pcg_list);
1.43 thorpej 2151: }
2152:
1.48 thorpej 2153: pc->pc_nitems++;
1.87 thorpej 2154: pcg_put(pcg, object, pa);
1.43 thorpej 2155:
1.102 chs 2156: if (pcg->pcg_avail == PCG_NOBJECTS) {
2157: LIST_REMOVE(pcg, pcg_list);
2158: LIST_INSERT_HEAD(&pc->pc_fullgroups, pcg, pcg_list);
2159: }
1.43 thorpej 2160: simple_unlock(&pc->pc_slock);
1.51 thorpej 2161: }
2162:
2163: /*
2164: * pool_cache_destruct_object:
2165: *
2166: * Force destruction of an object and its release back into
2167: * the pool.
2168: */
2169: void
2170: pool_cache_destruct_object(struct pool_cache *pc, void *object)
2171: {
2172:
2173: if (pc->pc_dtor != NULL)
2174: (*pc->pc_dtor)(pc->pc_arg, object);
2175: pool_put(pc->pc_pool, object);
1.43 thorpej 2176: }
2177:
1.102 chs 2178: static void
1.106 christos 2179: pool_do_cache_invalidate_grouplist(struct pool_cache_grouplist *pcgsl,
1.105 christos 2180: struct pool_cache *pc, struct pool_pagelist *pq,
1.106 christos 2181: struct pool_cache_grouplist *pcgdl)
1.102 chs 2182: {
1.106 christos 2183: struct pool_cache_group *pcg, *npcg;
1.102 chs 2184: void *object;
2185:
1.106 christos 2186: for (pcg = LIST_FIRST(pcgsl); pcg != NULL; pcg = npcg) {
1.102 chs 2187: npcg = LIST_NEXT(pcg, pcg_list);
2188: while (pcg->pcg_avail != 0) {
2189: pc->pc_nitems--;
2190: object = pcg_get(pcg, NULL);
2191: if (pc->pc_dtor != NULL)
2192: (*pc->pc_dtor)(pc->pc_arg, object);
2193: pool_do_put(pc->pc_pool, object, pq);
2194: }
1.103 chs 2195: pc->pc_ngroups--;
1.102 chs 2196: LIST_REMOVE(pcg, pcg_list);
1.106 christos 2197: LIST_INSERT_HEAD(pcgdl, pcg, pcg_list);
1.102 chs 2198: }
1.105 christos 2199: }
2200:
2201: static void
2202: pool_do_cache_invalidate(struct pool_cache *pc, struct pool_pagelist *pq,
2203: struct pool_cache_grouplist *pcgl)
2204: {
2205:
2206: LOCK_ASSERT(simple_lock_held(&pc->pc_slock));
2207: LOCK_ASSERT(simple_lock_held(&pc->pc_pool->pr_slock));
2208:
1.106 christos 2209: pool_do_cache_invalidate_grouplist(&pc->pc_fullgroups, pc, pq, pcgl);
2210: pool_do_cache_invalidate_grouplist(&pc->pc_partgroups, pc, pq, pcgl);
1.103 chs 2211:
2212: KASSERT(LIST_EMPTY(&pc->pc_partgroups));
2213: KASSERT(LIST_EMPTY(&pc->pc_fullgroups));
2214: KASSERT(pc->pc_nitems == 0);
1.102 chs 2215: }
2216:
1.43 thorpej 2217: /*
1.101 thorpej 2218: * pool_cache_invalidate:
1.43 thorpej 2219: *
1.101 thorpej 2220: * Invalidate a pool cache (destruct and release all of the
2221: * cached objects).
1.43 thorpej 2222: */
1.101 thorpej 2223: void
2224: pool_cache_invalidate(struct pool_cache *pc)
1.43 thorpej 2225: {
1.101 thorpej 2226: struct pool_pagelist pq;
1.102 chs 2227: struct pool_cache_grouplist pcgl;
1.101 thorpej 2228:
2229: LIST_INIT(&pq);
1.102 chs 2230: LIST_INIT(&pcgl);
1.101 thorpej 2231:
2232: simple_lock(&pc->pc_slock);
2233: simple_lock(&pc->pc_pool->pr_slock);
1.43 thorpej 2234:
1.102 chs 2235: pool_do_cache_invalidate(pc, &pq, &pcgl);
1.43 thorpej 2236:
1.101 thorpej 2237: simple_unlock(&pc->pc_pool->pr_slock);
2238: simple_unlock(&pc->pc_slock);
1.43 thorpej 2239:
1.102 chs 2240: pr_pagelist_free(pc->pc_pool, &pq);
2241: pcg_grouplist_free(&pcgl);
1.43 thorpej 2242: }
2243:
2244: /*
2245: * pool_cache_reclaim:
2246: *
2247: * Reclaim a pool cache for pool_reclaim().
2248: */
2249: static void
1.102 chs 2250: pool_cache_reclaim(struct pool_cache *pc, struct pool_pagelist *pq,
2251: struct pool_cache_grouplist *pcgl)
1.43 thorpej 2252: {
1.101 thorpej 2253:
2254: /*
2255: * We're locking in the wrong order (normally pool_cache -> pool,
2256: * but the pool is already locked when we get here), so we have
2257: * to use trylock. If we can't lock the pool_cache, it's not really
2258: * a big deal here.
2259: */
2260: if (simple_lock_try(&pc->pc_slock) == 0)
2261: return;
2262:
1.102 chs 2263: pool_do_cache_invalidate(pc, pq, pcgl);
1.43 thorpej 2264:
2265: simple_unlock(&pc->pc_slock);
1.3 pk 2266: }
1.66 thorpej 2267:
2268: /*
2269: * Pool backend allocators.
2270: *
2271: * Each pool has a backend allocator that handles allocation, deallocation,
2272: * and any additional draining that might be needed.
2273: *
2274: * We provide two standard allocators:
2275: *
2276: * pool_allocator_kmem - the default when no allocator is specified
2277: *
2278: * pool_allocator_nointr - used for pools that will not be accessed
2279: * in interrupt context.
2280: */
2281: void *pool_page_alloc(struct pool *, int);
2282: void pool_page_free(struct pool *, void *);
2283:
1.112 bjh21 2284: #ifdef POOL_SUBPAGE
2285: struct pool_allocator pool_allocator_kmem_fullpage = {
2286: pool_page_alloc, pool_page_free, 0,
1.117 yamt 2287: .pa_backingmapptr = &kmem_map,
1.112 bjh21 2288: };
2289: #else
1.66 thorpej 2290: struct pool_allocator pool_allocator_kmem = {
2291: pool_page_alloc, pool_page_free, 0,
1.117 yamt 2292: .pa_backingmapptr = &kmem_map,
1.66 thorpej 2293: };
1.112 bjh21 2294: #endif
1.66 thorpej 2295:
2296: void *pool_page_alloc_nointr(struct pool *, int);
2297: void pool_page_free_nointr(struct pool *, void *);
2298:
1.112 bjh21 2299: #ifdef POOL_SUBPAGE
2300: struct pool_allocator pool_allocator_nointr_fullpage = {
2301: pool_page_alloc_nointr, pool_page_free_nointr, 0,
1.117 yamt 2302: .pa_backingmapptr = &kernel_map,
1.112 bjh21 2303: };
2304: #else
1.66 thorpej 2305: struct pool_allocator pool_allocator_nointr = {
2306: pool_page_alloc_nointr, pool_page_free_nointr, 0,
1.117 yamt 2307: .pa_backingmapptr = &kernel_map,
1.66 thorpej 2308: };
1.112 bjh21 2309: #endif
1.66 thorpej 2310:
2311: #ifdef POOL_SUBPAGE
2312: void *pool_subpage_alloc(struct pool *, int);
2313: void pool_subpage_free(struct pool *, void *);
2314:
1.112 bjh21 2315: struct pool_allocator pool_allocator_kmem = {
2316: pool_subpage_alloc, pool_subpage_free, POOL_SUBPAGE,
1.117 yamt 2317: .pa_backingmapptr = &kmem_map,
1.112 bjh21 2318: };
2319:
2320: void *pool_subpage_alloc_nointr(struct pool *, int);
2321: void pool_subpage_free_nointr(struct pool *, void *);
2322:
2323: struct pool_allocator pool_allocator_nointr = {
2324: pool_subpage_alloc, pool_subpage_free, POOL_SUBPAGE,
1.117 yamt 2325: .pa_backingmapptr = &kmem_map,
1.66 thorpej 2326: };
2327: #endif /* POOL_SUBPAGE */
2328:
1.117 yamt 2329: static void *
2330: pool_allocator_alloc(struct pool *pp, int flags)
1.66 thorpej 2331: {
1.117 yamt 2332: struct pool_allocator *pa = pp->pr_alloc;
1.66 thorpej 2333: void *res;
2334:
1.117 yamt 2335: LOCK_ASSERT(!simple_lock_held(&pp->pr_slock));
1.66 thorpej 2336:
1.117 yamt 2337: res = (*pa->pa_alloc)(pp, flags);
2338: if (res == NULL && (flags & PR_WAITOK) == 0) {
1.66 thorpej 2339: /*
1.117 yamt 2340: * We only run the drain hook here if PR_NOWAIT.
2341: * In other cases, the hook will be run in
2342: * pool_reclaim().
1.66 thorpej 2343: */
1.117 yamt 2344: if (pp->pr_drain_hook != NULL) {
2345: (*pp->pr_drain_hook)(pp->pr_drain_hook_arg, flags);
2346: res = (*pa->pa_alloc)(pp, flags);
1.66 thorpej 2347: }
1.117 yamt 2348: }
2349: return res;
1.66 thorpej 2350: }
2351:
1.117 yamt 2352: static void
1.66 thorpej 2353: pool_allocator_free(struct pool *pp, void *v)
2354: {
2355: struct pool_allocator *pa = pp->pr_alloc;
2356:
1.91 yamt 2357: LOCK_ASSERT(!simple_lock_held(&pp->pr_slock));
2358:
1.66 thorpej 2359: (*pa->pa_free)(pp, v);
2360: }
2361:
2362: void *
2363: pool_page_alloc(struct pool *pp, int flags)
2364: {
2365: boolean_t waitok = (flags & PR_WAITOK) ? TRUE : FALSE;
2366:
1.100 yamt 2367: return ((void *) uvm_km_alloc_poolpage_cache(kmem_map, waitok));
1.66 thorpej 2368: }
2369:
2370: void
2371: pool_page_free(struct pool *pp, void *v)
2372: {
2373:
1.98 yamt 2374: uvm_km_free_poolpage_cache(kmem_map, (vaddr_t) v);
2375: }
2376:
2377: static void *
2378: pool_page_alloc_meta(struct pool *pp, int flags)
2379: {
2380: boolean_t waitok = (flags & PR_WAITOK) ? TRUE : FALSE;
2381:
1.100 yamt 2382: return ((void *) uvm_km_alloc_poolpage(kmem_map, waitok));
1.98 yamt 2383: }
2384:
2385: static void
2386: pool_page_free_meta(struct pool *pp, void *v)
2387: {
2388:
1.100 yamt 2389: uvm_km_free_poolpage(kmem_map, (vaddr_t) v);
1.66 thorpej 2390: }
2391:
2392: #ifdef POOL_SUBPAGE
2393: /* Sub-page allocator, for machines with large hardware pages. */
2394: void *
2395: pool_subpage_alloc(struct pool *pp, int flags)
2396: {
1.93 dbj 2397: void *v;
2398: int s;
2399: s = splvm();
2400: v = pool_get(&psppool, flags);
2401: splx(s);
2402: return v;
1.66 thorpej 2403: }
2404:
2405: void
2406: pool_subpage_free(struct pool *pp, void *v)
2407: {
1.93 dbj 2408: int s;
2409: s = splvm();
1.66 thorpej 2410: pool_put(&psppool, v);
1.93 dbj 2411: splx(s);
1.66 thorpej 2412: }
2413:
2414: /* We don't provide a real nointr allocator. Maybe later. */
2415: void *
1.112 bjh21 2416: pool_subpage_alloc_nointr(struct pool *pp, int flags)
1.66 thorpej 2417: {
2418:
2419: return (pool_subpage_alloc(pp, flags));
2420: }
2421:
2422: void
1.112 bjh21 2423: pool_subpage_free_nointr(struct pool *pp, void *v)
1.66 thorpej 2424: {
2425:
2426: pool_subpage_free(pp, v);
2427: }
1.112 bjh21 2428: #endif /* POOL_SUBPAGE */
1.66 thorpej 2429: void *
2430: pool_page_alloc_nointr(struct pool *pp, int flags)
2431: {
2432: boolean_t waitok = (flags & PR_WAITOK) ? TRUE : FALSE;
2433:
1.100 yamt 2434: return ((void *) uvm_km_alloc_poolpage_cache(kernel_map, waitok));
1.66 thorpej 2435: }
2436:
2437: void
2438: pool_page_free_nointr(struct pool *pp, void *v)
2439: {
2440:
1.98 yamt 2441: uvm_km_free_poolpage_cache(kernel_map, (vaddr_t) v);
1.66 thorpej 2442: }
CVSweb <webmaster@jp.NetBSD.org>