[BACK]Return to ufs_dirhash.c CVS log [TXT][DIR] Up to [cvs.NetBSD.org] / src / sys / ufs / ufs

Annotation of src/sys/ufs/ufs/ufs_dirhash.c, Revision 1.10.2.1

1.10.2.1! rpaulo      1: /*     $NetBSD: ufs_dirhash.c,v 1.11 2006/03/19 17:50:42 matt Exp $    */
1.1       rumble      2:
                      3: /*
                      4:  * Copyright (c) 2001, 2002 Ian Dowse.  All rights reserved.
                      5:  *
                      6:  * Redistribution and use in source and binary forms, with or without
                      7:  * modification, are permitted provided that the following conditions
                      8:  * are met:
                      9:  * 1. Redistributions of source code must retain the above copyright
                     10:  *    notice, this list of conditions and the following disclaimer.
                     11:  * 2. Redistributions in binary form must reproduce the above copyright
                     12:  *    notice, this list of conditions and the following disclaimer in the
                     13:  *    documentation and/or other materials provided with the distribution.
                     14:  *
                     15:  * THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
                     16:  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
                     17:  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
                     18:  * ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
                     19:  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
                     20:  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
                     21:  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
                     22:  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
                     23:  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
                     24:  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
                     25:  * SUCH DAMAGE.
                     26:  *
                     27:  * $FreeBSD: src/sys/ufs/ufs/ufs_dirhash.c,v 1.3.2.8 2004/12/08 11:54:13 dwmalone Exp $
                     28:  */
                     29:
                     30: /*
                     31:  * This implements a hash-based lookup scheme for UFS directories.
                     32:  */
                     33:
                     34: #include <sys/param.h>
                     35: #include <sys/systm.h>
                     36: #include <sys/kernel.h>
                     37: #include <sys/malloc.h>
                     38: #include <sys/types.h>
                     39: #include <sys/hash.h>
                     40: #include <sys/proc.h>
                     41: #include <sys/buf.h>
                     42: #include <sys/vnode.h>
                     43: #include <sys/mount.h>
                     44: #include <sys/pool.h>
                     45: #include <sys/sysctl.h>
                     46:
                     47: #include <ufs/ufs/quota.h>
                     48: #include <ufs/ufs/inode.h>
                     49: #include <ufs/ufs/dir.h>
                     50: #include <ufs/ufs/dirhash.h>
                     51: #include <ufs/ufs/ufsmount.h>
                     52: #include <ufs/ufs/ufs_bswap.h>
                     53: #include <ufs/ufs/ufs_extern.h>
                     54:
                     55: #define WRAPINCR(val, limit)   (((val) + 1 == (limit)) ? 0 : ((val) + 1))
                     56: #define WRAPDECR(val, limit)   (((val) == 0) ? ((limit) - 1) : ((val) - 1))
                     57: #define OFSFMT(ip)             ((ip)->i_ump->um_maxsymlinklen <= 0)
                     58: #define BLKFREE2IDX(n)         ((n) > DH_NFSTATS ? DH_NFSTATS : (n))
                     59:
                     60: static MALLOC_DEFINE(M_DIRHASH, "UFS dirhash", "UFS directory hash tables");
                     61:
                     62: static int ufs_dirhashminblks = 5;
                     63: static int ufs_dirhashmaxmem = 2 * 1024 * 1024;
1.2       perry      64: static int ufs_dirhashmem;
1.1       rumble     65: static int ufs_dirhashcheck = 0;
                     66:
                     67: static int ufsdirhash_hash(struct dirhash *dh, const char *name, int namelen);
                     68: static void ufsdirhash_adjfree(struct dirhash *dh, doff_t offset, int diff,
                     69:           int dirblksiz);
                     70: static void ufsdirhash_delslot(struct dirhash *dh, int slot);
                     71: static int ufsdirhash_findslot(struct dirhash *dh, const char *name,
                     72:           int namelen, doff_t offset);
                     73: static doff_t ufsdirhash_getprev(struct direct *dp, doff_t offset,
                     74:           int dirblksiz);
                     75: static int ufsdirhash_recycle(int wanted);
                     76:
1.9       yamt       77: static POOL_INIT(ufsdirhash_pool, DH_NBLKOFF * sizeof(daddr_t), 0, 0, 0,
                     78:     "ufsdirhash", &pool_allocator_nointr);
1.1       rumble     79:
                     80: #define DIRHASHLIST_LOCK()             do { } while (0)
                     81: #define DIRHASHLIST_UNLOCK()           do { } while (0)
                     82: #define DIRHASH_LOCK(dh)               do { } while (0)
                     83: #define DIRHASH_UNLOCK(dh)             do { } while (0)
                     84: #define DIRHASH_BLKALLOC_WAITOK()      pool_get(&ufsdirhash_pool, PR_WAITOK)
                     85: #define DIRHASH_BLKFREE(ptr)           pool_put(&ufsdirhash_pool, ptr)
                     86:
                     87: /* Dirhash list; recently-used entries are near the tail. */
                     88: static TAILQ_HEAD(, dirhash) ufsdirhash_list;
                     89:
                     90: /*
                     91:  * Attempt to build up a hash table for the directory contents in
                     92:  * inode 'ip'. Returns 0 on success, or -1 of the operation failed.
                     93:  */
                     94: int
                     95: ufsdirhash_build(struct inode *ip)
                     96: {
                     97:        struct dirhash *dh;
                     98:        struct buf *bp = NULL;
                     99:        struct direct *ep;
                    100:        struct vnode *vp;
                    101:        doff_t bmask, pos;
                    102:        int dirblocks, i, j, memreqd, nblocks, narrays, nslots, slot;
1.2       perry     103:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
1.1       rumble    104:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    105:
                    106:        /* Check if we can/should use dirhash. */
                    107:        if (ip->i_dirhash == NULL) {
                    108:                if (ip->i_size < (ufs_dirhashminblks * dirblksiz) || OFSFMT(ip))
                    109:                        return (-1);
                    110:        } else {
                    111:                /* Hash exists, but sysctls could have changed. */
                    112:                if (ip->i_size < (ufs_dirhashminblks * dirblksiz) ||
                    113:                    ufs_dirhashmem > ufs_dirhashmaxmem) {
                    114:                        ufsdirhash_free(ip);
                    115:                        return (-1);
                    116:                }
                    117:                /* Check if hash exists and is intact (note: unlocked read). */
                    118:                if (ip->i_dirhash->dh_hash != NULL)
                    119:                        return (0);
                    120:                /* Free the old, recycled hash and build a new one. */
                    121:                ufsdirhash_free(ip);
                    122:        }
                    123:
                    124:        /* Don't hash removed directories. */
                    125:        if (ip->i_ffs_effnlink == 0)
                    126:                return (-1);
                    127:
                    128:        vp = ip->i_vnode;
                    129:        /* Allocate 50% more entries than this dir size could ever need. */
                    130:        KASSERT(ip->i_size >= dirblksiz);
                    131:        nslots = ip->i_size / DIRECTSIZ(1);
                    132:        nslots = (nslots * 3 + 1) / 2;
                    133:        narrays = howmany(nslots, DH_NBLKOFF);
                    134:        nslots = narrays * DH_NBLKOFF;
                    135:        dirblocks = howmany(ip->i_size, dirblksiz);
                    136:        nblocks = (dirblocks * 3 + 1) / 2;
                    137:
                    138:        memreqd = sizeof(*dh) + narrays * sizeof(*dh->dh_hash) +
                    139:            narrays * DH_NBLKOFF * sizeof(**dh->dh_hash) +
                    140:            nblocks * sizeof(*dh->dh_blkfree);
                    141:        DIRHASHLIST_LOCK();
                    142:        if (memreqd + ufs_dirhashmem > ufs_dirhashmaxmem) {
                    143:                DIRHASHLIST_UNLOCK();
                    144:                if (memreqd > ufs_dirhashmaxmem / 2)
                    145:                        return (-1);
                    146:
                    147:                /* Try to free some space. */
                    148:                if (ufsdirhash_recycle(memreqd) != 0)
                    149:                        return (-1);
                    150:                /* Enough was freed, and list has been locked. */
                    151:        }
                    152:        ufs_dirhashmem += memreqd;
                    153:        DIRHASHLIST_UNLOCK();
                    154:
                    155:        /*
                    156:         * Use non-blocking mallocs so that we will revert to a linear
                    157:         * lookup on failure rather than potentially blocking forever.
                    158:         */
                    159:        MALLOC(dh, struct dirhash *, sizeof *dh, M_DIRHASH, M_NOWAIT | M_ZERO);
                    160:        if (dh == NULL) {
                    161:                DIRHASHLIST_LOCK();
                    162:                ufs_dirhashmem -= memreqd;
                    163:                DIRHASHLIST_UNLOCK();
                    164:                return (-1);
                    165:        }
1.10.2.1! rpaulo    166:        dh->dh_hash = (doff_t **)malloc(narrays * sizeof(dh->dh_hash[0]),
1.1       rumble    167:            M_DIRHASH, M_NOWAIT | M_ZERO);
1.10.2.1! rpaulo    168:        dh->dh_blkfree = (u_int8_t *)malloc(nblocks * sizeof(dh->dh_blkfree[0]),
1.1       rumble    169:            M_DIRHASH, M_NOWAIT);
                    170:        if (dh->dh_hash == NULL || dh->dh_blkfree == NULL)
                    171:                goto fail;
                    172:        for (i = 0; i < narrays; i++) {
                    173:                if ((dh->dh_hash[i] = DIRHASH_BLKALLOC_WAITOK()) == NULL)
                    174:                        goto fail;
                    175:                for (j = 0; j < DH_NBLKOFF; j++)
                    176:                        dh->dh_hash[i][j] = DIRHASH_EMPTY;
                    177:        }
                    178:
                    179:        /* Initialise the hash table and block statistics. */
                    180:        dh->dh_narrays = narrays;
                    181:        dh->dh_hlen = nslots;
                    182:        dh->dh_nblk = nblocks;
                    183:        dh->dh_dirblks = dirblocks;
                    184:        for (i = 0; i < dirblocks; i++)
                    185:                dh->dh_blkfree[i] = dirblksiz / DIRALIGN;
                    186:        for (i = 0; i < DH_NFSTATS; i++)
                    187:                dh->dh_firstfree[i] = -1;
                    188:        dh->dh_firstfree[DH_NFSTATS] = 0;
                    189:        dh->dh_seqopt = 0;
                    190:        dh->dh_seqoff = 0;
                    191:        dh->dh_score = DH_SCOREINIT;
                    192:        ip->i_dirhash = dh;
                    193:
                    194:        bmask = VFSTOUFS(vp->v_mount)->um_mountp->mnt_stat.f_iosize - 1;
                    195:        pos = 0;
                    196:        while (pos < ip->i_size) {
1.8       yamt      197:                if ((curcpu()->ci_schedstate.spc_flags & SPCF_SHOULDYIELD)
                    198:                    != 0) {
                    199:                        preempt(1);
                    200:                }
1.1       rumble    201:                /* If necessary, get the next directory block. */
                    202:                if ((pos & bmask) == 0) {
                    203:                        if (bp != NULL)
                    204:                                brelse(bp);
1.10      yamt      205:                        if (ufs_blkatoff(vp, (off_t)pos, NULL, &bp) != 0)
1.1       rumble    206:                                goto fail;
                    207:                }
                    208:
                    209:                /* Add this entry to the hash. */
                    210:                ep = (struct direct *)((char *)bp->b_data + (pos & bmask));
                    211:                if (ep->d_reclen == 0 || ep->d_reclen >
                    212:                    dirblksiz - (pos & (dirblksiz - 1))) {
                    213:                        /* Corrupted directory. */
                    214:                        brelse(bp);
                    215:                        goto fail;
                    216:                }
                    217:                if (ep->d_ino != 0) {
                    218:                        /* Add the entry (simplified ufsdirhash_add). */
                    219:                        slot = ufsdirhash_hash(dh, ep->d_name, ep->d_namlen);
                    220:                        while (DH_ENTRY(dh, slot) != DIRHASH_EMPTY)
                    221:                                slot = WRAPINCR(slot, dh->dh_hlen);
                    222:                        dh->dh_hused++;
                    223:                        DH_ENTRY(dh, slot) = pos;
                    224:                        ufsdirhash_adjfree(dh, pos, -DIRSIZ(0, ep, needswap),
                    225:                            dirblksiz);
                    226:                }
                    227:                pos += ep->d_reclen;
                    228:        }
                    229:
                    230:        if (bp != NULL)
                    231:                brelse(bp);
                    232:        DIRHASHLIST_LOCK();
                    233:        TAILQ_INSERT_TAIL(&ufsdirhash_list, dh, dh_list);
                    234:        dh->dh_onlist = 1;
                    235:        DIRHASHLIST_UNLOCK();
                    236:        return (0);
                    237:
                    238: fail:
                    239:        if (dh->dh_hash != NULL) {
                    240:                for (i = 0; i < narrays; i++)
                    241:                        if (dh->dh_hash[i] != NULL)
                    242:                                DIRHASH_BLKFREE(dh->dh_hash[i]);
                    243:                FREE(dh->dh_hash, M_DIRHASH);
                    244:        }
                    245:        if (dh->dh_blkfree != NULL)
                    246:                FREE(dh->dh_blkfree, M_DIRHASH);
                    247:        FREE(dh, M_DIRHASH);
                    248:        ip->i_dirhash = NULL;
                    249:        DIRHASHLIST_LOCK();
                    250:        ufs_dirhashmem -= memreqd;
                    251:        DIRHASHLIST_UNLOCK();
                    252:        return (-1);
                    253: }
                    254:
                    255: /*
                    256:  * Free any hash table associated with inode 'ip'.
                    257:  */
                    258: void
                    259: ufsdirhash_free(struct inode *ip)
                    260: {
                    261:        struct dirhash *dh;
                    262:        int i, mem;
                    263:
                    264:        if ((dh = ip->i_dirhash) == NULL)
                    265:                return;
                    266:        DIRHASHLIST_LOCK();
                    267:        DIRHASH_LOCK(dh);
                    268:        if (dh->dh_onlist)
                    269:                TAILQ_REMOVE(&ufsdirhash_list, dh, dh_list);
                    270:        DIRHASH_UNLOCK(dh);
                    271:        DIRHASHLIST_UNLOCK();
                    272:
                    273:        /* The dirhash pointed to by 'dh' is exclusively ours now. */
                    274:
                    275:        mem = sizeof(*dh);
                    276:        if (dh->dh_hash != NULL) {
                    277:                for (i = 0; i < dh->dh_narrays; i++)
                    278:                        DIRHASH_BLKFREE(dh->dh_hash[i]);
                    279:                FREE(dh->dh_hash, M_DIRHASH);
                    280:                FREE(dh->dh_blkfree, M_DIRHASH);
                    281:                mem += dh->dh_narrays * sizeof(*dh->dh_hash) +
                    282:                    dh->dh_narrays * DH_NBLKOFF * sizeof(**dh->dh_hash) +
                    283:                    dh->dh_nblk * sizeof(*dh->dh_blkfree);
                    284:        }
                    285:        FREE(dh, M_DIRHASH);
                    286:        ip->i_dirhash = NULL;
                    287:
                    288:        DIRHASHLIST_LOCK();
                    289:        ufs_dirhashmem -= mem;
                    290:        DIRHASHLIST_UNLOCK();
                    291: }
                    292:
                    293: /*
                    294:  * Find the offset of the specified name within the given inode.
                    295:  * Returns 0 on success, ENOENT if the entry does not exist, or
                    296:  * EJUSTRETURN if the caller should revert to a linear search.
                    297:  *
                    298:  * If successful, the directory offset is stored in *offp, and a
                    299:  * pointer to a struct buf containing the entry is stored in *bpp. If
                    300:  * prevoffp is non-NULL, the offset of the previous entry within
                    301:  * the DIRBLKSIZ-sized block is stored in *prevoffp (if the entry
                    302:  * is the first in a block, the start of the block is used).
                    303:  */
                    304: int
                    305: ufsdirhash_lookup(struct inode *ip, const char *name, int namelen, doff_t *offp,
                    306:     struct buf **bpp, doff_t *prevoffp)
                    307: {
                    308:        struct dirhash *dh, *dh_next;
                    309:        struct direct *dp;
                    310:        struct vnode *vp;
                    311:        struct buf *bp;
                    312:        doff_t blkoff, bmask, offset, prevoff;
                    313:        int i, slot;
1.2       perry     314:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
1.1       rumble    315:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    316:
                    317:        if ((dh = ip->i_dirhash) == NULL)
                    318:                return (EJUSTRETURN);
                    319:        /*
                    320:         * Move this dirhash towards the end of the list if it has a
                    321:         * score higher than the next entry, and acquire the dh_mtx.
                    322:         * Optimise the case where it's already the last by performing
                    323:         * an unlocked read of the TAILQ_NEXT pointer.
                    324:         *
                    325:         * In both cases, end up holding just dh_mtx.
                    326:         */
                    327:        if (TAILQ_NEXT(dh, dh_list) != NULL) {
                    328:                DIRHASHLIST_LOCK();
                    329:                DIRHASH_LOCK(dh);
                    330:                /*
                    331:                 * If the new score will be greater than that of the next
                    332:                 * entry, then move this entry past it. With both mutexes
                    333:                 * held, dh_next won't go away, but its dh_score could
                    334:                 * change; that's not important since it is just a hint.
                    335:                 */
                    336:                if (dh->dh_hash != NULL &&
                    337:                    (dh_next = TAILQ_NEXT(dh, dh_list)) != NULL &&
                    338:                    dh->dh_score >= dh_next->dh_score) {
                    339:                        KASSERT(dh->dh_onlist);
                    340:                        TAILQ_REMOVE(&ufsdirhash_list, dh, dh_list);
                    341:                        TAILQ_INSERT_AFTER(&ufsdirhash_list, dh_next, dh,
                    342:                            dh_list);
                    343:                }
                    344:                DIRHASHLIST_UNLOCK();
                    345:        } else {
                    346:                /* Already the last, though that could change as we wait. */
                    347:                DIRHASH_LOCK(dh);
                    348:        }
                    349:        if (dh->dh_hash == NULL) {
                    350:                DIRHASH_UNLOCK(dh);
                    351:                ufsdirhash_free(ip);
                    352:                return (EJUSTRETURN);
                    353:        }
                    354:
                    355:        /* Update the score. */
                    356:        if (dh->dh_score < DH_SCOREMAX)
                    357:                dh->dh_score++;
                    358:
                    359:        vp = ip->i_vnode;
                    360:        bmask = VFSTOUFS(vp->v_mount)->um_mountp->mnt_stat.f_iosize - 1;
                    361:        blkoff = -1;
                    362:        bp = NULL;
                    363: restart:
                    364:        slot = ufsdirhash_hash(dh, name, namelen);
                    365:
                    366:        if (dh->dh_seqopt) {
                    367:                /*
                    368:                 * Sequential access optimisation. dh_seqoff contains the
                    369:                 * offset of the directory entry immediately following
                    370:                 * the last entry that was looked up. Check if this offset
                    371:                 * appears in the hash chain for the name we are looking for.
                    372:                 */
                    373:                for (i = slot; (offset = DH_ENTRY(dh, i)) != DIRHASH_EMPTY;
                    374:                    i = WRAPINCR(i, dh->dh_hlen))
                    375:                        if (offset == dh->dh_seqoff)
                    376:                                break;
                    377:                if (offset == dh->dh_seqoff) {
                    378:                        /*
                    379:                         * We found an entry with the expected offset. This
                    380:                         * is probably the entry we want, but if not, the
                    381:                         * code below will turn off seqoff and retry.
1.2       perry     382:                         */
1.1       rumble    383:                        slot = i;
1.2       perry     384:                } else
1.1       rumble    385:                        dh->dh_seqopt = 0;
                    386:        }
                    387:
                    388:        for (; (offset = DH_ENTRY(dh, slot)) != DIRHASH_EMPTY;
                    389:            slot = WRAPINCR(slot, dh->dh_hlen)) {
                    390:                if (offset == DIRHASH_DEL)
                    391:                        continue;
                    392:                DIRHASH_UNLOCK(dh);
                    393:
                    394:                if (offset < 0 || offset >= ip->i_size)
                    395:                        panic("ufsdirhash_lookup: bad offset in hash array");
                    396:                if ((offset & ~bmask) != blkoff) {
                    397:                        if (bp != NULL)
                    398:                                brelse(bp);
                    399:                        blkoff = offset & ~bmask;
1.10      yamt      400:                        if (ufs_blkatoff(vp, (off_t)blkoff, NULL, &bp) != 0)
1.1       rumble    401:                                return (EJUSTRETURN);
                    402:                }
                    403:                dp = (struct direct *)(bp->b_data + (offset & bmask));
                    404:                if (dp->d_reclen == 0 || dp->d_reclen >
                    405:                    dirblksiz - (offset & (dirblksiz - 1))) {
                    406:                        /* Corrupted directory. */
                    407:                        brelse(bp);
                    408:                        return (EJUSTRETURN);
                    409:                }
                    410:                if (dp->d_namlen == namelen &&
                    411:                    bcmp(dp->d_name, name, namelen) == 0) {
                    412:                        /* Found. Get the prev offset if needed. */
                    413:                        if (prevoffp != NULL) {
                    414:                                if (offset & (dirblksiz - 1)) {
                    415:                                        prevoff = ufsdirhash_getprev(dp,
                    416:                                            offset, dirblksiz);
                    417:                                        if (prevoff == -1) {
                    418:                                                brelse(bp);
                    419:                                                return (EJUSTRETURN);
                    420:                                        }
                    421:                                } else
                    422:                                        prevoff = offset;
                    423:                                *prevoffp = prevoff;
                    424:                        }
                    425:
                    426:                        /* Check for sequential access, and update offset. */
                    427:                        if (dh->dh_seqopt == 0 && dh->dh_seqoff == offset)
                    428:                                dh->dh_seqopt = 1;
                    429:                        dh->dh_seqoff = offset + DIRSIZ(0, dp, needswap);
                    430:
                    431:                        *bpp = bp;
                    432:                        *offp = offset;
                    433:                        return (0);
                    434:                }
                    435:
                    436:                DIRHASH_LOCK(dh);
                    437:                if (dh->dh_hash == NULL) {
                    438:                        DIRHASH_UNLOCK(dh);
                    439:                        if (bp != NULL)
                    440:                                brelse(bp);
                    441:                        ufsdirhash_free(ip);
                    442:                        return (EJUSTRETURN);
                    443:                }
                    444:                /*
                    445:                 * When the name doesn't match in the seqopt case, go back
                    446:                 * and search normally.
                    447:                 */
                    448:                if (dh->dh_seqopt) {
                    449:                        dh->dh_seqopt = 0;
                    450:                        goto restart;
                    451:                }
                    452:        }
                    453:        DIRHASH_UNLOCK(dh);
                    454:        if (bp != NULL)
                    455:                brelse(bp);
                    456:        return (ENOENT);
                    457: }
                    458:
                    459: /*
                    460:  * Find a directory block with room for 'slotneeded' bytes. Returns
                    461:  * the offset of the directory entry that begins the free space.
                    462:  * This will either be the offset of an existing entry that has free
                    463:  * space at the end, or the offset of an entry with d_ino == 0 at
                    464:  * the start of a DIRBLKSIZ block.
                    465:  *
                    466:  * To use the space, the caller may need to compact existing entries in
                    467:  * the directory. The total number of bytes in all of the entries involved
                    468:  * in the compaction is stored in *slotsize. In other words, all of
                    469:  * the entries that must be compacted are exactly contained in the
                    470:  * region beginning at the returned offset and spanning *slotsize bytes.
                    471:  *
                    472:  * Returns -1 if no space was found, indicating that the directory
                    473:  * must be extended.
                    474:  */
                    475: doff_t
                    476: ufsdirhash_findfree(struct inode *ip, int slotneeded, int *slotsize)
                    477: {
                    478:        struct direct *dp;
                    479:        struct dirhash *dh;
                    480:        struct buf *bp;
                    481:        doff_t pos, slotstart;
                    482:        int dirblock, error, freebytes, i;
1.2       perry     483:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
1.1       rumble    484:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    485:
                    486:        if ((dh = ip->i_dirhash) == NULL)
                    487:                return (-1);
                    488:        DIRHASH_LOCK(dh);
                    489:        if (dh->dh_hash == NULL) {
                    490:                DIRHASH_UNLOCK(dh);
                    491:                ufsdirhash_free(ip);
                    492:                return (-1);
                    493:        }
                    494:
                    495:        /* Find a directory block with the desired free space. */
                    496:        dirblock = -1;
                    497:        for (i = howmany(slotneeded, DIRALIGN); i <= DH_NFSTATS; i++)
                    498:                if ((dirblock = dh->dh_firstfree[i]) != -1)
                    499:                        break;
                    500:        if (dirblock == -1) {
                    501:                DIRHASH_UNLOCK(dh);
                    502:                return (-1);
                    503:        }
                    504:
                    505:        KASSERT(dirblock < dh->dh_nblk &&
                    506:            dh->dh_blkfree[dirblock] >= howmany(slotneeded, DIRALIGN));
                    507:        DIRHASH_UNLOCK(dh);
                    508:        pos = dirblock * dirblksiz;
1.10      yamt      509:        error = ufs_blkatoff(ip->i_vnode, (off_t)pos, (void *)&dp, &bp);
1.1       rumble    510:        if (error)
                    511:                return (-1);
                    512:        /* Find the first entry with free space. */
                    513:        for (i = 0; i < dirblksiz; ) {
                    514:                if (dp->d_reclen == 0) {
                    515:                        brelse(bp);
                    516:                        return (-1);
                    517:                }
                    518:                if (dp->d_ino == 0 || dp->d_reclen > DIRSIZ(0, dp, needswap))
                    519:                        break;
                    520:                i += dp->d_reclen;
                    521:                dp = (struct direct *)((char *)dp + dp->d_reclen);
                    522:        }
                    523:        if (i > dirblksiz) {
                    524:                brelse(bp);
                    525:                return (-1);
                    526:        }
                    527:        slotstart = pos + i;
                    528:
                    529:        /* Find the range of entries needed to get enough space */
                    530:        freebytes = 0;
                    531:        while (i < dirblksiz && freebytes < slotneeded) {
                    532:                freebytes += dp->d_reclen;
                    533:                if (dp->d_ino != 0)
                    534:                        freebytes -= DIRSIZ(0, dp, needswap);
                    535:                if (dp->d_reclen == 0) {
                    536:                        brelse(bp);
                    537:                        return (-1);
                    538:                }
                    539:                i += dp->d_reclen;
                    540:                dp = (struct direct *)((char *)dp + dp->d_reclen);
                    541:        }
                    542:        if (i > dirblksiz) {
                    543:                brelse(bp);
                    544:                return (-1);
                    545:        }
                    546:        if (freebytes < slotneeded)
                    547:                panic("ufsdirhash_findfree: free mismatch");
                    548:        brelse(bp);
                    549:        *slotsize = pos + i - slotstart;
                    550:        return (slotstart);
                    551: }
                    552:
                    553: /*
                    554:  * Return the start of the unused space at the end of a directory, or
                    555:  * -1 if there are no trailing unused blocks.
                    556:  */
                    557: doff_t
                    558: ufsdirhash_enduseful(struct inode *ip)
                    559: {
                    560:        struct dirhash *dh;
                    561:        int i;
                    562:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    563:
                    564:        if ((dh = ip->i_dirhash) == NULL)
                    565:                return (-1);
                    566:        DIRHASH_LOCK(dh);
                    567:        if (dh->dh_hash == NULL) {
                    568:                DIRHASH_UNLOCK(dh);
                    569:                ufsdirhash_free(ip);
                    570:                return (-1);
                    571:        }
                    572:
                    573:        if (dh->dh_blkfree[dh->dh_dirblks - 1] != dirblksiz / DIRALIGN) {
                    574:                DIRHASH_UNLOCK(dh);
                    575:                return (-1);
                    576:        }
                    577:
                    578:        for (i = dh->dh_dirblks - 1; i >= 0; i--)
                    579:                if (dh->dh_blkfree[i] != dirblksiz / DIRALIGN)
                    580:                        break;
                    581:        DIRHASH_UNLOCK(dh);
                    582:        return ((doff_t)(i + 1) * dirblksiz);
                    583: }
                    584:
                    585: /*
                    586:  * Insert information into the hash about a new directory entry. dirp
                    587:  * points to a struct direct containing the entry, and offset specifies
                    588:  * the offset of this entry.
                    589:  */
                    590: void
                    591: ufsdirhash_add(struct inode *ip, struct direct *dirp, doff_t offset)
                    592: {
                    593:        struct dirhash *dh;
                    594:        int slot;
                    595:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
                    596:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    597:
                    598:        if ((dh = ip->i_dirhash) == NULL)
                    599:                return;
                    600:        DIRHASH_LOCK(dh);
                    601:        if (dh->dh_hash == NULL) {
                    602:                DIRHASH_UNLOCK(dh);
                    603:                ufsdirhash_free(ip);
                    604:                return;
                    605:        }
                    606:
                    607:        KASSERT(offset < dh->dh_dirblks * dirblksiz);
                    608:        /*
                    609:         * Normal hash usage is < 66%. If the usage gets too high then
                    610:         * remove the hash entirely and let it be rebuilt later.
                    611:         */
                    612:        if (dh->dh_hused >= (dh->dh_hlen * 3) / 4) {
                    613:                DIRHASH_UNLOCK(dh);
                    614:                ufsdirhash_free(ip);
                    615:                return;
                    616:        }
                    617:
                    618:        /* Find a free hash slot (empty or deleted), and add the entry. */
                    619:        slot = ufsdirhash_hash(dh, dirp->d_name, dirp->d_namlen);
                    620:        while (DH_ENTRY(dh, slot) >= 0)
                    621:                slot = WRAPINCR(slot, dh->dh_hlen);
                    622:        if (DH_ENTRY(dh, slot) == DIRHASH_EMPTY)
                    623:                dh->dh_hused++;
                    624:        DH_ENTRY(dh, slot) = offset;
                    625:
                    626:        /* Update the per-block summary info. */
                    627:        ufsdirhash_adjfree(dh, offset, -DIRSIZ(0, dirp, needswap), dirblksiz);
                    628:        DIRHASH_UNLOCK(dh);
                    629: }
                    630:
                    631: /*
                    632:  * Remove the specified directory entry from the hash. The entry to remove
                    633:  * is defined by the name in `dirp', which must exist at the specified
                    634:  * `offset' within the directory.
                    635:  */
                    636: void
                    637: ufsdirhash_remove(struct inode *ip, struct direct *dirp, doff_t offset)
                    638: {
                    639:        struct dirhash *dh;
                    640:        int slot;
                    641:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
                    642:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    643:
                    644:        if ((dh = ip->i_dirhash) == NULL)
                    645:                return;
                    646:        DIRHASH_LOCK(dh);
                    647:        if (dh->dh_hash == NULL) {
                    648:                DIRHASH_UNLOCK(dh);
                    649:                ufsdirhash_free(ip);
                    650:                return;
                    651:        }
                    652:
                    653:        KASSERT(offset < dh->dh_dirblks * dirblksiz);
                    654:        /* Find the entry */
                    655:        slot = ufsdirhash_findslot(dh, dirp->d_name, dirp->d_namlen, offset);
                    656:
                    657:        /* Remove the hash entry. */
                    658:        ufsdirhash_delslot(dh, slot);
                    659:
                    660:        /* Update the per-block summary info. */
                    661:        ufsdirhash_adjfree(dh, offset, DIRSIZ(0, dirp, needswap), dirblksiz);
                    662:        DIRHASH_UNLOCK(dh);
                    663: }
                    664:
                    665: /*
                    666:  * Change the offset associated with a directory entry in the hash. Used
                    667:  * when compacting directory blocks.
                    668:  */
                    669: void
                    670: ufsdirhash_move(struct inode *ip, struct direct *dirp, doff_t oldoff,
                    671:     doff_t newoff)
                    672: {
                    673:        struct dirhash *dh;
                    674:        int slot;
                    675:
                    676:        if ((dh = ip->i_dirhash) == NULL)
                    677:                return;
                    678:        DIRHASH_LOCK(dh);
                    679:        if (dh->dh_hash == NULL) {
                    680:                DIRHASH_UNLOCK(dh);
                    681:                ufsdirhash_free(ip);
                    682:                return;
                    683:        }
                    684:
                    685:        KASSERT(oldoff < dh->dh_dirblks * ip->i_ump->um_dirblksiz &&
                    686:            newoff < dh->dh_dirblks * ip->i_ump->um_dirblksiz);
                    687:        /* Find the entry, and update the offset. */
                    688:        slot = ufsdirhash_findslot(dh, dirp->d_name, dirp->d_namlen, oldoff);
                    689:        DH_ENTRY(dh, slot) = newoff;
                    690:        DIRHASH_UNLOCK(dh);
                    691: }
                    692:
                    693: /*
                    694:  * Inform dirhash that the directory has grown by one block that
                    695:  * begins at offset (i.e. the new length is offset + DIRBLKSIZ).
                    696:  */
                    697: void
                    698: ufsdirhash_newblk(struct inode *ip, doff_t offset)
                    699: {
                    700:        struct dirhash *dh;
                    701:        int block;
                    702:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    703:
                    704:        if ((dh = ip->i_dirhash) == NULL)
                    705:                return;
                    706:        DIRHASH_LOCK(dh);
                    707:        if (dh->dh_hash == NULL) {
                    708:                DIRHASH_UNLOCK(dh);
                    709:                ufsdirhash_free(ip);
                    710:                return;
                    711:        }
                    712:
                    713:        KASSERT(offset == dh->dh_dirblks * dirblksiz);
                    714:        block = offset / dirblksiz;
                    715:        if (block >= dh->dh_nblk) {
                    716:                /* Out of space; must rebuild. */
                    717:                DIRHASH_UNLOCK(dh);
                    718:                ufsdirhash_free(ip);
                    719:                return;
                    720:        }
                    721:        dh->dh_dirblks = block + 1;
                    722:
                    723:        /* Account for the new free block. */
                    724:        dh->dh_blkfree[block] = dirblksiz / DIRALIGN;
                    725:        if (dh->dh_firstfree[DH_NFSTATS] == -1)
                    726:                dh->dh_firstfree[DH_NFSTATS] = block;
                    727:        DIRHASH_UNLOCK(dh);
                    728: }
                    729:
                    730: /*
                    731:  * Inform dirhash that the directory is being truncated.
                    732:  */
                    733: void
                    734: ufsdirhash_dirtrunc(struct inode *ip, doff_t offset)
                    735: {
                    736:        struct dirhash *dh;
                    737:        int block, i;
                    738:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    739:
                    740:        if ((dh = ip->i_dirhash) == NULL)
                    741:                return;
                    742:        DIRHASH_LOCK(dh);
                    743:        if (dh->dh_hash == NULL) {
                    744:                DIRHASH_UNLOCK(dh);
                    745:                ufsdirhash_free(ip);
                    746:                return;
                    747:        }
                    748:
                    749:        KASSERT(offset <= dh->dh_dirblks * dirblksiz);
                    750:        block = howmany(offset, dirblksiz);
                    751:        /*
                    752:         * If the directory shrinks to less than 1/8 of dh_nblk blocks
                    753:         * (about 20% of its original size due to the 50% extra added in
                    754:         * ufsdirhash_build) then free it, and let the caller rebuild
                    755:         * if necessary.
                    756:         */
                    757:        if (block < dh->dh_nblk / 8 && dh->dh_narrays > 1) {
                    758:                DIRHASH_UNLOCK(dh);
                    759:                ufsdirhash_free(ip);
                    760:                return;
                    761:        }
                    762:
                    763:        /*
                    764:         * Remove any `first free' information pertaining to the
                    765:         * truncated blocks. All blocks we're removing should be
                    766:         * completely unused.
                    767:         */
                    768:        if (dh->dh_firstfree[DH_NFSTATS] >= block)
                    769:                dh->dh_firstfree[DH_NFSTATS] = -1;
                    770:        for (i = block; i < dh->dh_dirblks; i++)
                    771:                if (dh->dh_blkfree[i] != dirblksiz / DIRALIGN)
                    772:                        panic("ufsdirhash_dirtrunc: blocks in use");
                    773:        for (i = 0; i < DH_NFSTATS; i++)
                    774:                if (dh->dh_firstfree[i] >= block)
                    775:                        panic("ufsdirhash_dirtrunc: first free corrupt");
                    776:        dh->dh_dirblks = block;
                    777:        DIRHASH_UNLOCK(dh);
                    778: }
                    779:
                    780: /*
                    781:  * Debugging function to check that the dirhash information about
                    782:  * a directory block matches its actual contents. Panics if a mismatch
                    783:  * is detected.
                    784:  *
1.3       christos  785:  * On entry, `sbuf' should point to the start of an in-core
1.1       rumble    786:  * DIRBLKSIZ-sized directory block, and `offset' should contain the
                    787:  * offset from the start of the directory of that block.
                    788:  */
                    789: void
1.3       christos  790: ufsdirhash_checkblock(struct inode *ip, char *sbuf, doff_t offset)
1.1       rumble    791: {
                    792:        struct dirhash *dh;
                    793:        struct direct *dp;
                    794:        int block, ffslot, i, nfree;
                    795:        const int needswap = UFS_MPNEEDSWAP(ip->i_ump);
                    796:        int dirblksiz = ip->i_ump->um_dirblksiz;
                    797:
                    798:        if (!ufs_dirhashcheck)
                    799:                return;
                    800:        if ((dh = ip->i_dirhash) == NULL)
                    801:                return;
                    802:        DIRHASH_LOCK(dh);
                    803:        if (dh->dh_hash == NULL) {
                    804:                DIRHASH_UNLOCK(dh);
                    805:                ufsdirhash_free(ip);
                    806:                return;
                    807:        }
                    808:
                    809:        block = offset / dirblksiz;
                    810:        if ((offset & (dirblksiz - 1)) != 0 || block >= dh->dh_dirblks)
                    811:                panic("ufsdirhash_checkblock: bad offset");
                    812:
                    813:        nfree = 0;
                    814:        for (i = 0; i < dirblksiz; i += dp->d_reclen) {
1.3       christos  815:                dp = (struct direct *)(sbuf + i);
1.1       rumble    816:                if (dp->d_reclen == 0 || i + dp->d_reclen > dirblksiz)
                    817:                        panic("ufsdirhash_checkblock: bad dir");
                    818:
                    819:                if (dp->d_ino == 0) {
                    820: #if 0
                    821:                        /*
                    822:                         * XXX entries with d_ino == 0 should only occur
                    823:                         * at the start of a DIRBLKSIZ block. However the
                    824:                         * ufs code is tolerant of such entries at other
                    825:                         * offsets, and fsck does not fix them.
                    826:                         */
                    827:                        if (i != 0)
                    828:                                panic("ufsdirhash_checkblock: bad dir inode");
                    829: #endif
                    830:                        nfree += dp->d_reclen;
                    831:                        continue;
                    832:                }
                    833:
                    834:                /* Check that the entry exists (will panic if it doesn't). */
                    835:                ufsdirhash_findslot(dh, dp->d_name, dp->d_namlen, offset + i);
                    836:
                    837:                nfree += dp->d_reclen - DIRSIZ(0, dp, needswap);
                    838:        }
                    839:        if (i != dirblksiz)
                    840:                panic("ufsdirhash_checkblock: bad dir end");
                    841:
                    842:        if (dh->dh_blkfree[block] * DIRALIGN != nfree)
                    843:                panic("ufsdirhash_checkblock: bad free count");
                    844:
                    845:        ffslot = BLKFREE2IDX(nfree / DIRALIGN);
                    846:        for (i = 0; i <= DH_NFSTATS; i++)
                    847:                if (dh->dh_firstfree[i] == block && i != ffslot)
                    848:                        panic("ufsdirhash_checkblock: bad first-free");
                    849:        if (dh->dh_firstfree[ffslot] == -1)
                    850:                panic("ufsdirhash_checkblock: missing first-free entry");
                    851:        DIRHASH_UNLOCK(dh);
                    852: }
                    853:
                    854: /*
                    855:  * Hash the specified filename into a dirhash slot.
                    856:  */
                    857: static int
                    858: ufsdirhash_hash(struct dirhash *dh, const char *name, int namelen)
                    859: {
                    860:        u_int32_t hash;
                    861:
                    862:        /*
                    863:         * We hash the name and then some other bit of data that is
                    864:         * invariant over the dirhash's lifetime. Otherwise names
                    865:         * differing only in the last byte are placed close to one
                    866:         * another in the table, which is bad for linear probing.
                    867:         */
                    868:        hash = hash32_buf(name, namelen, HASH32_BUF_INIT);
                    869:        hash = hash32_buf(&dh, sizeof(dh), hash);
                    870:        return (hash % dh->dh_hlen);
                    871: }
                    872:
                    873: /*
                    874:  * Adjust the number of free bytes in the block containing `offset'
                    875:  * by the value specified by `diff'.
                    876:  *
                    877:  * The caller must ensure we have exclusive access to `dh'; normally
                    878:  * that means that dh_mtx should be held, but this is also called
                    879:  * from ufsdirhash_build() where exclusive access can be assumed.
                    880:  */
                    881: static void
                    882: ufsdirhash_adjfree(struct dirhash *dh, doff_t offset, int diff, int dirblksiz)
                    883: {
                    884:        int block, i, nfidx, ofidx;
                    885:
                    886:        /* Update the per-block summary info. */
                    887:        block = offset / dirblksiz;
                    888:        KASSERT(block < dh->dh_nblk && block < dh->dh_dirblks);
                    889:        ofidx = BLKFREE2IDX(dh->dh_blkfree[block]);
                    890:        dh->dh_blkfree[block] = (int)dh->dh_blkfree[block] + (diff / DIRALIGN);
                    891:        nfidx = BLKFREE2IDX(dh->dh_blkfree[block]);
                    892:
                    893:        /* Update the `first free' list if necessary. */
                    894:        if (ofidx != nfidx) {
                    895:                /* If removing, scan forward for the next block. */
                    896:                if (dh->dh_firstfree[ofidx] == block) {
                    897:                        for (i = block + 1; i < dh->dh_dirblks; i++)
                    898:                                if (BLKFREE2IDX(dh->dh_blkfree[i]) == ofidx)
                    899:                                        break;
                    900:                        dh->dh_firstfree[ofidx] = (i < dh->dh_dirblks) ? i : -1;
                    901:                }
                    902:
                    903:                /* Make this the new `first free' if necessary */
                    904:                if (dh->dh_firstfree[nfidx] > block ||
                    905:                    dh->dh_firstfree[nfidx] == -1)
                    906:                        dh->dh_firstfree[nfidx] = block;
                    907:        }
                    908: }
                    909:
                    910: /*
                    911:  * Find the specified name which should have the specified offset.
                    912:  * Returns a slot number, and panics on failure.
                    913:  *
                    914:  * `dh' must be locked on entry and remains so on return.
                    915:  */
                    916: static int
                    917: ufsdirhash_findslot(struct dirhash *dh, const char *name, int namelen,
                    918:     doff_t offset)
                    919: {
                    920:        int slot;
                    921:
                    922:        /* Find the entry. */
                    923:        KASSERT(dh->dh_hused < dh->dh_hlen);
                    924:        slot = ufsdirhash_hash(dh, name, namelen);
                    925:        while (DH_ENTRY(dh, slot) != offset &&
                    926:            DH_ENTRY(dh, slot) != DIRHASH_EMPTY)
                    927:                slot = WRAPINCR(slot, dh->dh_hlen);
                    928:        if (DH_ENTRY(dh, slot) != offset)
                    929:                panic("ufsdirhash_findslot: '%.*s' not found", namelen, name);
                    930:
                    931:        return (slot);
                    932: }
                    933:
                    934: /*
                    935:  * Remove the entry corresponding to the specified slot from the hash array.
                    936:  *
                    937:  * `dh' must be locked on entry and remains so on return.
                    938:  */
                    939: static void
                    940: ufsdirhash_delslot(struct dirhash *dh, int slot)
                    941: {
                    942:        int i;
                    943:
                    944:        /* Mark the entry as deleted. */
                    945:        DH_ENTRY(dh, slot) = DIRHASH_DEL;
                    946:
                    947:        /* If this is the end of a chain of DIRHASH_DEL slots, remove them. */
                    948:        for (i = slot; DH_ENTRY(dh, i) == DIRHASH_DEL; )
                    949:                i = WRAPINCR(i, dh->dh_hlen);
                    950:        if (DH_ENTRY(dh, i) == DIRHASH_EMPTY) {
                    951:                i = WRAPDECR(i, dh->dh_hlen);
                    952:                while (DH_ENTRY(dh, i) == DIRHASH_DEL) {
                    953:                        DH_ENTRY(dh, i) = DIRHASH_EMPTY;
                    954:                        dh->dh_hused--;
                    955:                        i = WRAPDECR(i, dh->dh_hlen);
                    956:                }
                    957:                KASSERT(dh->dh_hused >= 0);
                    958:        }
                    959: }
                    960:
                    961: /*
                    962:  * Given a directory entry and its offset, find the offset of the
                    963:  * previous entry in the same DIRBLKSIZ-sized block. Returns an
                    964:  * offset, or -1 if there is no previous entry in the block or some
                    965:  * other problem occurred.
                    966:  */
                    967: static doff_t
                    968: ufsdirhash_getprev(struct direct *dirp, doff_t offset, int dirblksiz)
                    969: {
                    970:        struct direct *dp;
                    971:        char *blkbuf;
                    972:        doff_t blkoff, prevoff;
                    973:        int entrypos, i;
                    974:
                    975:        blkoff = offset & ~(dirblksiz - 1);     /* offset of start of block */
                    976:        entrypos = offset & (dirblksiz - 1);    /* entry relative to block */
                    977:        blkbuf = (char *)dirp - entrypos;
                    978:        prevoff = blkoff;
                    979:
                    980:        /* If `offset' is the start of a block, there is no previous entry. */
                    981:        if (entrypos == 0)
                    982:                return (-1);
                    983:
                    984:        /* Scan from the start of the block until we get to the entry. */
                    985:        for (i = 0; i < entrypos; i += dp->d_reclen) {
                    986:                dp = (struct direct *)(blkbuf + i);
                    987:                if (dp->d_reclen == 0 || i + dp->d_reclen > entrypos)
                    988:                        return (-1);    /* Corrupted directory. */
                    989:                prevoff = blkoff + i;
                    990:        }
                    991:        return (prevoff);
                    992: }
                    993:
                    994: /*
                    995:  * Try to free up `wanted' bytes by stealing memory from existing
                    996:  * dirhashes. Returns zero with list locked if successful.
                    997:  */
                    998: static int
                    999: ufsdirhash_recycle(int wanted)
                   1000: {
                   1001:        struct dirhash *dh;
                   1002:        doff_t **hash;
                   1003:        u_int8_t *blkfree;
                   1004:        int i, mem, narrays;
                   1005:
                   1006:        DIRHASHLIST_LOCK();
                   1007:        while (wanted + ufs_dirhashmem > ufs_dirhashmaxmem) {
                   1008:                /* Find a dirhash, and lock it. */
                   1009:                if ((dh = TAILQ_FIRST(&ufsdirhash_list)) == NULL) {
                   1010:                        DIRHASHLIST_UNLOCK();
                   1011:                        return (-1);
                   1012:                }
                   1013:                DIRHASH_LOCK(dh);
                   1014:                KASSERT(dh->dh_hash != NULL);
                   1015:
                   1016:                /* Decrement the score; only recycle if it becomes zero. */
                   1017:                if (--dh->dh_score > 0) {
                   1018:                        DIRHASH_UNLOCK(dh);
                   1019:                        DIRHASHLIST_UNLOCK();
                   1020:                        return (-1);
                   1021:                }
                   1022:
                   1023:                /* Remove it from the list and detach its memory. */
                   1024:                TAILQ_REMOVE(&ufsdirhash_list, dh, dh_list);
                   1025:                dh->dh_onlist = 0;
                   1026:                hash = dh->dh_hash;
                   1027:                dh->dh_hash = NULL;
                   1028:                blkfree = dh->dh_blkfree;
                   1029:                dh->dh_blkfree = NULL;
                   1030:                narrays = dh->dh_narrays;
                   1031:                mem = narrays * sizeof(*dh->dh_hash) +
                   1032:                    narrays * DH_NBLKOFF * sizeof(**dh->dh_hash) +
                   1033:                    dh->dh_nblk * sizeof(*dh->dh_blkfree);
                   1034:
                   1035:                /* Unlock everything, free the detached memory. */
                   1036:                DIRHASH_UNLOCK(dh);
                   1037:                DIRHASHLIST_UNLOCK();
                   1038:                for (i = 0; i < narrays; i++)
                   1039:                        DIRHASH_BLKFREE(hash[i]);
                   1040:                FREE(hash, M_DIRHASH);
                   1041:                FREE(blkfree, M_DIRHASH);
                   1042:
                   1043:                /* Account for the returned memory, and repeat if necessary. */
                   1044:                DIRHASHLIST_LOCK();
                   1045:                ufs_dirhashmem -= mem;
                   1046:        }
                   1047:        /* Success; return with list locked. */
                   1048:        return (0);
                   1049: }
                   1050:
                   1051: void
                   1052: ufsdirhash_init()
                   1053: {
                   1054: #ifdef _LKM
                   1055:        pool_init(&ufsdirhash_pool, DH_NBLKOFF * sizeof(daddr_t), 0, 0, 0,
                   1056:            "ufsdirhash", &pool_allocator_nointr);
                   1057: #endif
                   1058:        TAILQ_INIT(&ufsdirhash_list);
                   1059: }
                   1060:
                   1061: void
1.5       thorpej  1062: ufsdirhash_done(void)
1.1       rumble   1063: {
                   1064:        KASSERT(TAILQ_EMPTY(&ufsdirhash_list));
                   1065: #ifdef _LKM
                   1066:        pool_destroy(&ufsdirhash_pool);
                   1067: #endif
                   1068: }
                   1069:
                   1070: SYSCTL_SETUP(sysctl_vfs_ufs_setup, "sysctl vfs.ufs.dirhash subtree setup")
                   1071: {
1.4       atatat   1072:        const struct sysctlnode *rnode, *cnode;
1.1       rumble   1073:
                   1074:        sysctl_createv(clog, 0, NULL, &rnode,
                   1075:                       CTLFLAG_PERMANENT,
                   1076:                       CTLTYPE_NODE, "vfs", NULL,
                   1077:                       NULL, 0, NULL, 0,
                   1078:                       CTL_VFS, CTL_EOL);
                   1079:
                   1080:        sysctl_createv(clog, 0, &rnode, &rnode,
                   1081:                       CTLFLAG_PERMANENT,
                   1082:                       CTLTYPE_NODE, "ufs",
                   1083:                       SYSCTL_DESCR("ufs"),
                   1084:                       NULL, 0, NULL, 0,
                   1085:                       CTL_CREATE, CTL_EOL);
                   1086:
                   1087:        sysctl_createv(clog, 0, &rnode, &rnode,
                   1088:                       CTLFLAG_PERMANENT,
                   1089:                       CTLTYPE_NODE, "dirhash",
                   1090:                       SYSCTL_DESCR("dirhash"),
                   1091:                       NULL, 0, NULL, 0,
                   1092:                       CTL_CREATE, CTL_EOL);
                   1093:
                   1094:        sysctl_createv(clog, 0, &rnode, &cnode,
                   1095:                       CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
                   1096:                       CTLTYPE_INT, "minblocks",
                   1097:                       SYSCTL_DESCR("minimum hashed directory size in blocks"),
                   1098:                       NULL, 0, &ufs_dirhashminblks, 0,
                   1099:                       CTL_CREATE, CTL_EOL);
                   1100:
                   1101:        sysctl_createv(clog, 0, &rnode, &cnode,
                   1102:                       CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
                   1103:                       CTLTYPE_INT, "maxmem",
                   1104:                       SYSCTL_DESCR("maximum dirhash memory usage"),
                   1105:                       NULL, 0, &ufs_dirhashmaxmem, 0,
                   1106:                       CTL_CREATE, CTL_EOL);
                   1107:
                   1108:        sysctl_createv(clog, 0, &rnode, &cnode,
                   1109:                       CTLFLAG_PERMANENT|CTLFLAG_READONLY,
                   1110:                       CTLTYPE_INT, "memused",
                   1111:                       SYSCTL_DESCR("current dirhash memory usage"),
                   1112:                       NULL, 0, &ufs_dirhashmem, 0,
                   1113:                       CTL_CREATE, CTL_EOL);
                   1114:
                   1115:        sysctl_createv(clog, 0, &rnode, &cnode,
                   1116:                       CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
                   1117:                       CTLTYPE_INT, "docheck",
                   1118:                       SYSCTL_DESCR("enable extra sanity checks"),
                   1119:                       NULL, 0, &ufs_dirhashcheck, 0,
                   1120:                       CTL_CREATE, CTL_EOL);
                   1121: }

CVSweb <webmaster@jp.NetBSD.org>