diff options
Diffstat (limited to 'fs/quota/dquot.c')
| -rw-r--r-- | fs/quota/dquot.c | 2095 |
1 files changed, 1214 insertions, 881 deletions
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index fbad622841f9..376739f6420e 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * Implementation of the diskquota system for the LINUX operating system. QUOTA * is implemented using the BSD system call interface as the means of @@ -8,7 +9,7 @@ * on the Melbourne quota system as used on BSD derived systems. The internal * implementation is based on one of the several variants of the LINUX * inode-subsystem with added complexity of the diskquota system. - * + * * Author: Marco van Wieringen <mvw@planets.elm.net> * * Fixes: Dmitry Gorodchanin <pgmdsg@ibi.com>, 11 Feb 96 @@ -50,7 +51,7 @@ * Added journalled quota support, fix lock inversion problems * Jan Kara, <jack@suse.cz>, 2003,2004 * - * (C) Copyright 1994 - 1997 Marco van Wieringen + * (C) Copyright 1994 - 1997 Marco van Wieringen */ #include <linux/errno.h> @@ -72,65 +73,65 @@ #include <linux/proc_fs.h> #include <linux/security.h> #include <linux/sched.h> +#include <linux/cred.h> #include <linux/kmod.h> #include <linux/namei.h> #include <linux/capability.h> #include <linux/quotaops.h> -#include "../internal.h" /* ugh */ +#include <linux/blkdev.h> +#include <linux/sched/mm.h> #include <linux/uaccess.h> /* - * There are three quota SMP locks. dq_list_lock protects all lists with quotas - * and quota formats. - * dq_data_lock protects data from dq_dqb and also mem_dqinfo structures and - * also guards consistency of dquot->dq_dqb with inode->i_blocks, i_bytes. - * i_blocks and i_bytes updates itself are guarded by i_lock acquired directly - * in inode_add_bytes() and inode_sub_bytes(). dq_state_lock protects - * modifications of quota state (on quotaon and quotaoff) and readers who care - * about latest values take it as well. + * There are five quota SMP locks: + * * dq_list_lock protects all lists with quotas and quota formats. + * * dquot->dq_dqb_lock protects data from dq_dqb + * * inode->i_lock protects inode->i_blocks, i_bytes and also guards + * consistency of dquot->dq_dqb with inode->i_blocks, i_bytes so that + * dquot_transfer() can stabilize amount it transfers + * * dq_data_lock protects mem_dqinfo structures and modifications of dquot + * pointers in the inode + * * dq_state_lock protects modifications of quota state (on quotaon and + * quotaoff) and readers who care about latest values take it as well. * - * The spinlock ordering is hence: dq_data_lock > dq_list_lock > i_lock, + * The spinlock ordering is hence: + * dq_data_lock > dq_list_lock > i_lock > dquot->dq_dqb_lock, * dq_list_lock > dq_state_lock * * Note that some things (eg. sb pointer, type, id) doesn't change during * the life of the dquot structure and so needn't to be protected by a lock * - * Any operation working on dquots via inode pointers must hold dqptr_sem. If - * operation is just reading pointers from inode (or not using them at all) the - * read lock is enough. If pointers are altered function must hold write lock. + * Operation accessing dquots via inode pointers are protected by dquot_srcu. + * Operation of reading pointer needs srcu_read_lock(&dquot_srcu), and + * synchronize_srcu(&dquot_srcu) is called after clearing pointers from + * inode and before dropping dquot references to avoid use of dquots after + * they are freed. dq_data_lock is used to serialize the pointer setting and + * clearing operations. * Special care needs to be taken about S_NOQUOTA inode flag (marking that * inode is a quota file). Functions adding pointers from inode to dquots have - * to check this flag under dqptr_sem and then (if S_NOQUOTA is not set) they - * have to do all pointer modifications before dropping dqptr_sem. This makes + * to check this flag under dq_data_lock and then (if S_NOQUOTA is not set) they + * have to do all pointer modifications before dropping dq_data_lock. This makes * sure they cannot race with quotaon which first sets S_NOQUOTA flag and * then drops all pointers to dquots from an inode. * - * Each dquot has its dq_lock mutex. Locked dquots might not be referenced - * from inodes (dquot_alloc_space() and such don't check the dq_lock). - * Currently dquot is locked only when it is being read to memory (or space for - * it is being allocated) on the first dqget() and when it is being released on - * the last dqput(). The allocation and release oparations are serialized by - * the dq_lock and by checking the use count in dquot_release(). Write - * operations on dquots don't hold dq_lock as they copy data under dq_data_lock - * spinlock to internal buffers before writing. + * Each dquot has its dq_lock mutex. Dquot is locked when it is being read to + * memory (or space for it is being allocated) on the first dqget(), when it is + * being written out, and when it is being released on the last dqput(). The + * allocation and release operations are serialized by the dq_lock and by + * checking the use count in dquot_release(). * * Lock ordering (including related VFS locks) is the following: - * dqonoff_mutex > i_mutex > journal_lock > dqptr_sem > dquot->dq_lock > - * dqio_mutex - * dqonoff_mutex > i_mutex comes from dquot_quota_sync, dquot_enable, etc. - * The lock ordering of dqptr_sem imposed by quota code is only dqonoff_sem > - * dqptr_sem. But filesystem has to count with the fact that functions such as - * dquot_alloc_space() acquire dqptr_sem and they usually have to be called - * from inside a transaction to keep filesystem consistency after a crash. Also - * filesystems usually want to do some IO on dquot from ->mark_dirty which is - * called with dqptr_sem held. + * s_umount > i_mutex > journal_lock > dquot->dq_lock > dqio_sem */ static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_list_lock); static __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_state_lock); __cacheline_aligned_in_smp DEFINE_SPINLOCK(dq_data_lock); EXPORT_SYMBOL(dq_data_lock); +DEFINE_STATIC_SRCU(dquot_srcu); + +static DECLARE_WAIT_QUEUE_HEAD(dquot_ref_wq); void __quota_error(struct super_block *sb, const char *func, const char *fmt, ...) @@ -161,13 +162,15 @@ static struct quota_module_name module_names[] = INIT_QUOTA_MODULE_NAMES; /* SLAB cache for dquot structures */ static struct kmem_cache *dquot_cachep; -int register_quota_format(struct quota_format_type *fmt) +/* workqueue for work quota_release_work*/ +static struct workqueue_struct *quota_unbound_wq; + +void register_quota_format(struct quota_format_type *fmt) { spin_lock(&dq_list_lock); fmt->qf_next = quota_formats; quota_formats = fmt; spin_unlock(&dq_list_lock); - return 0; } EXPORT_SYMBOL(register_quota_format); @@ -197,7 +200,7 @@ static struct quota_format_type *find_quota_format(int id) int qm; spin_unlock(&dq_list_lock); - + for (qm = 0; module_names[qm].qm_fmt_id && module_names[qm].qm_fmt_id != id; qm++) ; @@ -223,18 +226,31 @@ static void put_quota_format(struct quota_format_type *fmt) /* * Dquot List Management: - * The quota code uses three lists for dquot management: the inuse_list, - * free_dquots, and dquot_hash[] array. A single dquot structure may be - * on all three lists, depending on its current state. + * The quota code uses five lists for dquot management: the inuse_list, + * releasing_dquots, free_dquots, dqi_dirty_list, and dquot_hash[] array. + * A single dquot structure may be on some of those lists, depending on + * its current state. * * All dquots are placed to the end of inuse_list when first created, and this * list is used for invalidate operation, which must look at every dquot. * - * Unused dquots (dq_count == 0) are added to the free_dquots list when freed, - * and this list is searched whenever we need an available dquot. Dquots are - * removed from the list as soon as they are used again, and - * dqstats.free_dquots gives the number of dquots on the list. When - * dquot is invalidated it's completely released from memory. + * When the last reference of a dquot is dropped, the dquot is added to + * releasing_dquots. We'll then queue work item which will call + * synchronize_srcu() and after that perform the final cleanup of all the + * dquots on the list. Each cleaned up dquot is moved to free_dquots list. + * Both releasing_dquots and free_dquots use the dq_free list_head in the dquot + * struct. + * + * Unused and cleaned up dquots are in the free_dquots list and this list is + * searched whenever we need an available dquot. Dquots are removed from the + * list as soon as they are used again and dqstats.free_dquots gives the number + * of dquots on the list. When dquot is invalidated it's completely released + * from memory. + * + * Dirty dquots are added to the dqi_dirty_list of quota_info when mark + * dirtied, and this list is searched when writing dirty dquots back to + * quota file. Note that some filesystems do dirty dquot tracking on their + * own (e.g. in a journal) and thus don't use dqi_dirty_list. * * Dquots with a specific identity (device, type and id) are placed on * one of the dquot_hash[] hash chains. The provides an efficient search @@ -243,6 +259,7 @@ static void put_quota_format(struct quota_format_type *fmt) static LIST_HEAD(inuse_list); static LIST_HEAD(free_dquots); +static LIST_HEAD(releasing_dquots); static unsigned int dq_hash_bits, dq_hash_mask; static struct hlist_head *dquot_hash; @@ -250,7 +267,11 @@ struct dqstats dqstats; EXPORT_SYMBOL(dqstats); static qsize_t inode_get_rsv_space(struct inode *inode); -static void __dquot_initialize(struct inode *inode, int type); +static qsize_t __inode_get_rsv_space(struct inode *inode); +static int __dquot_initialize(struct inode *inode, int type); + +static void quota_release_workfn(struct work_struct *work); +static DECLARE_DELAYED_WORK(quota_release_work, quota_release_workfn); static inline unsigned int hashfn(const struct super_block *sb, struct kqid qid) @@ -281,14 +302,12 @@ static inline void remove_dquot_hash(struct dquot *dquot) static struct dquot *find_dquot(unsigned int hashent, struct super_block *sb, struct kqid qid) { - struct hlist_node *node; struct dquot *dquot; - hlist_for_each (node, dquot_hash+hashent) { - dquot = hlist_entry(node, struct dquot, dq_hash); + hlist_for_each_entry(dquot, dquot_hash+hashent, dq_hash) if (dquot->dq_sb == sb && qid_eq(dquot->dq_id, qid)) return dquot; - } + return NULL; } @@ -299,12 +318,21 @@ static inline void put_dquot_last(struct dquot *dquot) dqstats_inc(DQST_FREE_DQUOTS); } +static inline void put_releasing_dquots(struct dquot *dquot) +{ + list_add_tail(&dquot->dq_free, &releasing_dquots); + set_bit(DQ_RELEASING_B, &dquot->dq_flags); +} + static inline void remove_free_dquot(struct dquot *dquot) { if (list_empty(&dquot->dq_free)) return; list_del_init(&dquot->dq_free); - dqstats_dec(DQST_FREE_DQUOTS); + if (!test_bit(DQ_RELEASING_B, &dquot->dq_flags)) + dqstats_dec(DQST_FREE_DQUOTS); + else + clear_bit(DQ_RELEASING_B, &dquot->dq_flags); } static inline void put_inuse(struct dquot *dquot) @@ -330,6 +358,11 @@ static void wait_on_dquot(struct dquot *dquot) mutex_unlock(&dquot->dq_lock); } +static inline int dquot_active(struct dquot *dquot) +{ + return test_bit(DQ_ACTIVE_B, &dquot->dq_flags); +} + static inline int dquot_dirty(struct dquot *dquot) { return test_bit(DQ_MOD_B, &dquot->dq_flags); @@ -345,8 +378,14 @@ int dquot_mark_dquot_dirty(struct dquot *dquot) { int ret = 1; + if (!dquot_active(dquot)) + return 0; + + if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NOLIST_DIRTY) + return test_and_set_bit(DQ_MOD_B, &dquot->dq_flags); + /* If quota is dirty already, we don't have to acquire dq_list_lock */ - if (test_bit(DQ_MOD_B, &dquot->dq_flags)) + if (dquot_dirty(dquot)) return 1; spin_lock(&dq_list_lock); @@ -361,16 +400,18 @@ int dquot_mark_dquot_dirty(struct dquot *dquot) EXPORT_SYMBOL(dquot_mark_dquot_dirty); /* Dirtify all the dquots - this can block when journalling */ -static inline int mark_all_dquot_dirty(struct dquot * const *dquot) +static inline int mark_all_dquot_dirty(struct dquot __rcu * const *dquots) { int ret, err, cnt; + struct dquot *dquot; ret = err = 0; for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (dquot[cnt]) + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (dquot) /* Even in case of error we have to continue */ - ret = mark_dquot_dirty(dquot[cnt]); - if (!err) + ret = mark_dquot_dirty(dquot); + if (!err && ret < 0) err = ret; } return err; @@ -384,18 +425,26 @@ static inline void dqput_all(struct dquot **dquot) dqput(dquot[cnt]); } -/* This function needs dq_list_lock */ static inline int clear_dquot_dirty(struct dquot *dquot) { - if (!test_and_clear_bit(DQ_MOD_B, &dquot->dq_flags)) + if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NOLIST_DIRTY) + return test_and_clear_bit(DQ_MOD_B, &dquot->dq_flags); + + spin_lock(&dq_list_lock); + if (!test_and_clear_bit(DQ_MOD_B, &dquot->dq_flags)) { + spin_unlock(&dq_list_lock); return 0; + } list_del_init(&dquot->dq_dirty); + spin_unlock(&dq_list_lock); return 1; } void mark_info_dirty(struct super_block *sb, int type) { - set_bit(DQF_INFO_DIRTY_B, &sb_dqopt(sb)->info[type].dqi_flags); + spin_lock(&dq_data_lock); + sb_dqopt(sb)->info[type].dqi_flags |= DQF_INFO_DIRTY; + spin_unlock(&dq_data_lock); } EXPORT_SYMBOL(mark_info_dirty); @@ -406,17 +455,21 @@ EXPORT_SYMBOL(mark_info_dirty); int dquot_acquire(struct dquot *dquot) { int ret = 0, ret2 = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dquot->dq_lock); - mutex_lock(&dqopt->dqio_mutex); - if (!test_bit(DQ_READ_B, &dquot->dq_flags)) + memalloc = memalloc_nofs_save(); + if (!test_bit(DQ_READ_B, &dquot->dq_flags)) { ret = dqopt->ops[dquot->dq_id.type]->read_dqblk(dquot); - if (ret < 0) - goto out_iolock; + if (ret < 0) + goto out_iolock; + } + /* Make sure flags update is visible after dquot has been filled */ + smp_mb__before_atomic(); set_bit(DQ_READ_B, &dquot->dq_flags); /* Instantiate dquot if needed */ - if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && !dquot->dq_off) { + if (!dquot_active(dquot) && !dquot->dq_off) { ret = dqopt->ops[dquot->dq_id.type]->commit_dqblk(dquot); /* Write the info if needed */ if (info_dirty(&dqopt->info[dquot->dq_id.type])) { @@ -430,9 +483,14 @@ int dquot_acquire(struct dquot *dquot) goto out_iolock; } } + /* + * Make sure flags update is visible after on-disk struct has been + * allocated. Paired with smp_rmb() in dqget(). + */ + smp_mb__before_atomic(); set_bit(DQ_ACTIVE_B, &dquot->dq_flags); out_iolock: - mutex_unlock(&dqopt->dqio_mutex); + memalloc_nofs_restore(memalloc); mutex_unlock(&dquot->dq_lock); return ret; } @@ -444,23 +502,22 @@ EXPORT_SYMBOL(dquot_acquire); int dquot_commit(struct dquot *dquot) { int ret = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); - mutex_lock(&dqopt->dqio_mutex); - spin_lock(&dq_list_lock); - if (!clear_dquot_dirty(dquot)) { - spin_unlock(&dq_list_lock); - goto out_sem; - } - spin_unlock(&dq_list_lock); + mutex_lock(&dquot->dq_lock); + memalloc = memalloc_nofs_save(); + if (!clear_dquot_dirty(dquot)) + goto out_lock; /* Inactive dquot can be only if there was error during read/init * => we have better not writing it */ - if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) + if (dquot_active(dquot)) ret = dqopt->ops[dquot->dq_id.type]->commit_dqblk(dquot); else ret = -EIO; -out_sem: - mutex_unlock(&dqopt->dqio_mutex); +out_lock: + memalloc_nofs_restore(memalloc); + mutex_unlock(&dquot->dq_lock); return ret; } EXPORT_SYMBOL(dquot_commit); @@ -471,13 +528,14 @@ EXPORT_SYMBOL(dquot_commit); int dquot_release(struct dquot *dquot) { int ret = 0, ret2 = 0; + unsigned int memalloc; struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); mutex_lock(&dquot->dq_lock); + memalloc = memalloc_nofs_save(); /* Check whether we are not racing with some other dqget() */ - if (atomic_read(&dquot->dq_count) > 1) + if (dquot_is_busy(dquot)) goto out_dqlock; - mutex_lock(&dqopt->dqio_mutex); if (dqopt->ops[dquot->dq_id.type]->release_dqblk) { ret = dqopt->ops[dquot->dq_id.type]->release_dqblk(dquot); /* Write the info */ @@ -489,8 +547,8 @@ int dquot_release(struct dquot *dquot) ret = ret2; } clear_bit(DQ_ACTIVE_B, &dquot->dq_flags); - mutex_unlock(&dqopt->dqio_mutex); out_dqlock: + memalloc_nofs_restore(memalloc); mutex_unlock(&dquot->dq_lock); return ret; } @@ -518,6 +576,8 @@ static void invalidate_dquots(struct super_block *sb, int type) struct dquot *dquot, *tmp; restart: + flush_delayed_work("a_release_work); + spin_lock(&dq_list_lock); list_for_each_entry_safe(dquot, tmp, &inuse_list, dq_inuse) { if (dquot->dq_sb != sb) @@ -526,22 +586,18 @@ restart: continue; /* Wait for dquot users */ if (atomic_read(&dquot->dq_count)) { - DEFINE_WAIT(wait); - atomic_inc(&dquot->dq_count); - prepare_to_wait(&dquot->dq_wait_unused, &wait, - TASK_UNINTERRUPTIBLE); spin_unlock(&dq_list_lock); - /* Once dqput() wakes us up, we know it's time to free + /* + * Once dqput() wakes us up, we know it's time to free * the dquot. * IMPORTANT: we rely on the fact that there is always * at most one process waiting for dquot to free. * Otherwise dq_count would be > 1 and we would never * wake up. */ - if (atomic_read(&dquot->dq_count) > 1) - schedule(); - finish_wait(&dquot->dq_wait_unused, &wait); + wait_event(dquot_ref_wq, + atomic_read(&dquot->dq_count) == 1); dqput(dquot); /* At this moment dquot() need not exist (it could be * reclaimed by prune_dqcache(). Hence we must @@ -549,6 +605,15 @@ restart: goto restart; } /* + * The last user already dropped its reference but dquot didn't + * get fully cleaned up yet. Restart the scan which flushes the + * work cleaning up released dquots. + */ + if (test_bit(DQ_RELEASING_B, &dquot->dq_flags)) { + spin_unlock(&dq_list_lock); + goto restart; + } + /* * Quota now has no users and it has been written on last * dqput() */ @@ -568,22 +633,30 @@ int dquot_scan_active(struct super_block *sb, struct dquot *dquot, *old_dquot = NULL; int ret = 0; - mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); + WARN_ON_ONCE(!rwsem_is_locked(&sb->s_umount)); + spin_lock(&dq_list_lock); list_for_each_entry(dquot, &inuse_list, dq_inuse) { - if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) + if (!dquot_active(dquot)) continue; if (dquot->dq_sb != sb) continue; /* Now we have active dquot so we can just increase use count */ atomic_inc(&dquot->dq_count); spin_unlock(&dq_list_lock); - dqstats_inc(DQST_LOOKUPS); dqput(old_dquot); old_dquot = dquot; - ret = fn(dquot, priv); - if (ret < 0) - goto out; + /* + * ->release_dquot() can be racing with us. Our reference + * protects us from new calls to it so just wait for any + * outstanding call and recheck the DQ_ACTIVE_B after that. + */ + wait_on_dquot(dquot); + if (dquot_active(dquot)) { + ret = fn(dquot, priv); + if (ret < 0) + goto out; + } spin_lock(&dq_list_lock); /* We are safe to continue now because our dquot could not * be moved out of the inuse list while we hold the reference */ @@ -591,45 +664,64 @@ int dquot_scan_active(struct super_block *sb, spin_unlock(&dq_list_lock); out: dqput(old_dquot); - mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); return ret; } EXPORT_SYMBOL(dquot_scan_active); +static inline int dquot_write_dquot(struct dquot *dquot) +{ + int ret = dquot->dq_sb->dq_op->write_dquot(dquot); + if (ret < 0) { + quota_error(dquot->dq_sb, "Can't write quota structure " + "(error %d). Quota may get out of sync!", ret); + /* Clear dirty bit anyway to avoid infinite loop. */ + clear_dquot_dirty(dquot); + } + return ret; +} + /* Write all dquot structures to quota files */ int dquot_writeback_dquots(struct super_block *sb, int type) { - struct list_head *dirty; + struct list_head dirty; struct dquot *dquot; struct quota_info *dqopt = sb_dqopt(sb); int cnt; int err, ret = 0; - mutex_lock(&dqopt->dqonoff_mutex); + WARN_ON_ONCE(!rwsem_is_locked(&sb->s_umount)); + + flush_delayed_work("a_release_work); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (type != -1 && cnt != type) continue; if (!sb_has_quota_active(sb, cnt)) continue; spin_lock(&dq_list_lock); - dirty = &dqopt->info[cnt].dqi_dirty_list; - while (!list_empty(dirty)) { - dquot = list_first_entry(dirty, struct dquot, + /* Move list away to avoid livelock. */ + list_replace_init(&dqopt->info[cnt].dqi_dirty_list, &dirty); + while (!list_empty(&dirty)) { + dquot = list_first_entry(&dirty, struct dquot, dq_dirty); - /* Dirty and inactive can be only bad dquot... */ - if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) { - clear_dquot_dirty(dquot); + + WARN_ON(!dquot_active(dquot)); + /* If the dquot is releasing we should not touch it */ + if (test_bit(DQ_RELEASING_B, &dquot->dq_flags)) { + spin_unlock(&dq_list_lock); + flush_delayed_work("a_release_work); + spin_lock(&dq_list_lock); continue; } + /* Now we have active dquot from which someone is * holding reference so we can safely just increase * use count */ - atomic_inc(&dquot->dq_count); + dqgrab(dquot); spin_unlock(&dq_list_lock); - dqstats_inc(DQST_LOOKUPS); - err = sb->dq_op->write_dquot(dquot); - if (!ret && err) - err = ret; + err = dquot_write_dquot(dquot); + if (err && !ret) + ret = err; dqput(dquot); spin_lock(&dq_list_lock); } @@ -641,7 +733,6 @@ int dquot_writeback_dquots(struct super_block *sb, int type) && info_dirty(&dqopt->info[cnt])) sb->dq_op->write_info(sb, cnt); dqstats_inc(DQST_SYNCS); - mutex_unlock(&dqopt->dqonoff_mutex); return ret; } @@ -663,80 +754,107 @@ int dquot_quota_sync(struct super_block *sb, int type) /* This is not very clever (and fast) but currently I don't know about * any other simple way of getting quota data to disk and we must get * them there for userspace to be visible... */ - if (sb->s_op->sync_fs) - sb->s_op->sync_fs(sb, 1); - sync_blockdev(sb->s_bdev); + if (sb->s_op->sync_fs) { + ret = sb->s_op->sync_fs(sb, 1); + if (ret) + return ret; + } + ret = sync_blockdev(sb->s_bdev); + if (ret) + return ret; /* * Now when everything is written we can discard the pagecache so * that userspace sees the changes. */ - mutex_lock(&dqopt->dqonoff_mutex); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (type != -1 && cnt != type) continue; if (!sb_has_quota_active(sb, cnt)) continue; - mutex_lock(&dqopt->files[cnt]->i_mutex); + inode_lock(dqopt->files[cnt]); truncate_inode_pages(&dqopt->files[cnt]->i_data, 0); - mutex_unlock(&dqopt->files[cnt]->i_mutex); + inode_unlock(dqopt->files[cnt]); } - mutex_unlock(&dqopt->dqonoff_mutex); return 0; } EXPORT_SYMBOL(dquot_quota_sync); -/* Free unused dquots from cache */ -static void prune_dqcache(int count) +static unsigned long +dqcache_shrink_scan(struct shrinker *shrink, struct shrink_control *sc) { - struct list_head *head; struct dquot *dquot; + unsigned long freed = 0; - head = free_dquots.prev; - while (head != &free_dquots && count) { - dquot = list_entry(head, struct dquot, dq_free); + spin_lock(&dq_list_lock); + while (!list_empty(&free_dquots) && sc->nr_to_scan) { + dquot = list_first_entry(&free_dquots, struct dquot, dq_free); remove_dquot_hash(dquot); remove_free_dquot(dquot); remove_inuse(dquot); do_destroy_dquot(dquot); - count--; - head = free_dquots.prev; + sc->nr_to_scan--; + freed++; } + spin_unlock(&dq_list_lock); + return freed; +} + +static unsigned long +dqcache_shrink_count(struct shrinker *shrink, struct shrink_control *sc) +{ + return vfs_pressure_ratio( + percpu_counter_read_positive(&dqstats.counter[DQST_FREE_DQUOTS])); } /* - * This is called from kswapd when we think we need some - * more memory + * Safely release dquot and put reference to dquot. */ -static int shrink_dqcache_memory(struct shrinker *shrink, - struct shrink_control *sc) +static void quota_release_workfn(struct work_struct *work) { - int nr = sc->nr_to_scan; + struct dquot *dquot; + struct list_head rls_head; - if (nr) { - spin_lock(&dq_list_lock); - prune_dqcache(nr); - spin_unlock(&dq_list_lock); + spin_lock(&dq_list_lock); + /* Exchange the list head to avoid livelock. */ + list_replace_init(&releasing_dquots, &rls_head); + spin_unlock(&dq_list_lock); + synchronize_srcu(&dquot_srcu); + +restart: + spin_lock(&dq_list_lock); + while (!list_empty(&rls_head)) { + dquot = list_first_entry(&rls_head, struct dquot, dq_free); + WARN_ON_ONCE(atomic_read(&dquot->dq_count)); + /* + * Note that DQ_RELEASING_B protects us from racing with + * invalidate_dquots() calls so we are safe to work with the + * dquot even after we drop dq_list_lock. + */ + if (dquot_dirty(dquot)) { + spin_unlock(&dq_list_lock); + /* Commit dquot before releasing */ + dquot_write_dquot(dquot); + goto restart; + } + if (dquot_active(dquot)) { + spin_unlock(&dq_list_lock); + dquot->dq_sb->dq_op->release_dquot(dquot); + goto restart; + } + /* Dquot is inactive and clean, now move it to free list */ + remove_free_dquot(dquot); + put_dquot_last(dquot); } - return ((unsigned) - percpu_counter_read_positive(&dqstats.counter[DQST_FREE_DQUOTS]) - /100) * sysctl_vfs_cache_pressure; + spin_unlock(&dq_list_lock); } -static struct shrinker dqcache_shrinker = { - .shrink = shrink_dqcache_memory, - .seeks = DEFAULT_SEEKS, -}; - /* * Put reference to dquot - * NOTE: If you change this function please check whether dqput_blocks() works right... */ void dqput(struct dquot *dquot) { - int ret; - if (!dquot) return; #ifdef CONFIG_QUOTA_DEBUG @@ -748,7 +866,7 @@ void dqput(struct dquot *dquot) } #endif dqstats_inc(DQST_DROPS); -we_slept: + spin_lock(&dq_list_lock); if (atomic_read(&dquot->dq_count) > 1) { /* We have more than one user... nothing to do */ @@ -756,43 +874,17 @@ we_slept: /* Releasing dquot during quotaoff phase? */ if (!sb_has_quota_active(dquot->dq_sb, dquot->dq_id.type) && atomic_read(&dquot->dq_count) == 1) - wake_up(&dquot->dq_wait_unused); + wake_up(&dquot_ref_wq); spin_unlock(&dq_list_lock); return; } + /* Need to release dquot? */ - if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && dquot_dirty(dquot)) { - spin_unlock(&dq_list_lock); - /* Commit dquot before releasing */ - ret = dquot->dq_sb->dq_op->write_dquot(dquot); - if (ret < 0) { - quota_error(dquot->dq_sb, "Can't write quota structure" - " (error %d). Quota may get out of sync!", - ret); - /* - * We clear dirty bit anyway, so that we avoid - * infinite loop here - */ - spin_lock(&dq_list_lock); - clear_dquot_dirty(dquot); - spin_unlock(&dq_list_lock); - } - goto we_slept; - } - /* Clear flag in case dquot was inactive (something bad happened) */ - clear_dquot_dirty(dquot); - if (test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) { - spin_unlock(&dq_list_lock); - dquot->dq_sb->dq_op->release_dquot(dquot); - goto we_slept; - } + WARN_ON_ONCE(!list_empty(&dquot->dq_free)); + put_releasing_dquots(dquot); atomic_dec(&dquot->dq_count); -#ifdef CONFIG_QUOTA_DEBUG - /* sanity check */ - BUG_ON(!list_empty(&dquot->dq_free)); -#endif - put_dquot_last(dquot); spin_unlock(&dq_list_lock); + queue_delayed_work(quota_unbound_wq, "a_release_work, 1); } EXPORT_SYMBOL(dqput); @@ -815,10 +907,10 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type) INIT_LIST_HEAD(&dquot->dq_inuse); INIT_HLIST_NODE(&dquot->dq_hash); INIT_LIST_HEAD(&dquot->dq_dirty); - init_waitqueue_head(&dquot->dq_wait_unused); dquot->dq_sb = sb; dquot->dq_id = make_kqid_invalid(type); atomic_set(&dquot->dq_count, 1); + spin_lock_init(&dquot->dq_dqb_lock); return dquot; } @@ -834,16 +926,20 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type) struct dquot *dqget(struct super_block *sb, struct kqid qid) { unsigned int hashent = hashfn(sb, qid); - struct dquot *dquot = NULL, *empty = NULL; + struct dquot *dquot, *empty = NULL; + + if (!qid_has_mapping(sb->s_user_ns, qid)) + return ERR_PTR(-EINVAL); if (!sb_has_quota_active(sb, qid.type)) - return NULL; + return ERR_PTR(-ESRCH); we_slept: spin_lock(&dq_list_lock); spin_lock(&dq_state_lock); if (!sb_has_quota_active(sb, qid.type)) { spin_unlock(&dq_state_lock); spin_unlock(&dq_list_lock); + dquot = ERR_PTR(-ESRCH); goto out; } spin_unlock(&dq_state_lock); @@ -875,18 +971,26 @@ we_slept: dqstats_inc(DQST_LOOKUPS); } /* Wait for dq_lock - after this we know that either dquot_release() is - * already finished or it will be canceled due to dq_count > 1 test */ + * already finished or it will be canceled due to dq_count > 0 test */ wait_on_dquot(dquot); /* Read the dquot / allocate space in quota file */ - if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && - sb->dq_op->acquire_dquot(dquot) < 0) { - dqput(dquot); - dquot = NULL; - goto out; + if (!dquot_active(dquot)) { + int err; + + err = sb->dq_op->acquire_dquot(dquot); + if (err < 0) { + dqput(dquot); + dquot = ERR_PTR(err); + goto out; + } } -#ifdef CONFIG_QUOTA_DEBUG - BUG_ON(!dquot->dq_sb); /* Has somebody invalidated entry under us? */ -#endif + /* + * Make sure following reads see filled structure - paired with + * smp_mb__before_atomic() in dquot_acquire(). + */ + smp_rmb(); + /* Has somebody invalidated entry under us? */ + WARN_ON_ONCE(hlist_unhashed(&dquot->dq_hash)); out: if (empty) do_destroy_dquot(empty); @@ -895,32 +999,41 @@ out: } EXPORT_SYMBOL(dqget); +static inline struct dquot __rcu **i_dquot(struct inode *inode) +{ + return inode->i_sb->s_op->get_dquots(inode); +} + static int dqinit_needed(struct inode *inode, int type) { + struct dquot __rcu * const *dquots; int cnt; if (IS_NOQUOTA(inode)) return 0; + + dquots = i_dquot(inode); if (type != -1) - return !inode->i_dquot[type]; + return !dquots[type]; for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (!inode->i_dquot[cnt]) + if (!dquots[cnt]) return 1; return 0; } -/* This routine is guarded by dqonoff_mutex mutex */ -static void add_dquot_ref(struct super_block *sb, int type) +/* This routine is guarded by s_umount semaphore */ +static int add_dquot_ref(struct super_block *sb, int type) { struct inode *inode, *old_inode = NULL; #ifdef CONFIG_QUOTA_DEBUG int reserved = 0; #endif + int err = 0; - spin_lock(&inode_sb_list_lock); + spin_lock(&sb->s_inode_list_lock); list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { spin_lock(&inode->i_lock); - if ((inode->i_state & (I_FREEING|I_WILL_FREE|I_NEW)) || + if ((inode_state_read(inode) & (I_FREEING | I_WILL_FREE | I_NEW)) || !atomic_read(&inode->i_writecount) || !dqinit_needed(inode, type)) { spin_unlock(&inode->i_lock); @@ -928,29 +1041,34 @@ static void add_dquot_ref(struct super_block *sb, int type) } __iget(inode); spin_unlock(&inode->i_lock); - spin_unlock(&inode_sb_list_lock); + spin_unlock(&sb->s_inode_list_lock); #ifdef CONFIG_QUOTA_DEBUG if (unlikely(inode_get_rsv_space(inode) > 0)) reserved = 1; #endif iput(old_inode); - __dquot_initialize(inode, type); + err = __dquot_initialize(inode, type); + if (err) { + iput(inode); + goto out; + } /* * We hold a reference to 'inode' so it couldn't have been * removed from s_inodes list while we dropped the - * inode_sb_list_lock We cannot iput the inode now as we can be + * s_inode_list_lock. We cannot iput the inode now as we can be * holding the last reference and we cannot iput it under - * inode_sb_list_lock. So we keep the reference and iput it + * s_inode_list_lock. So we keep the reference and iput it * later. */ old_inode = inode; - spin_lock(&inode_sb_list_lock); + cond_resched(); + spin_lock(&sb->s_inode_list_lock); } - spin_unlock(&inode_sb_list_lock); + spin_unlock(&sb->s_inode_list_lock); iput(old_inode); - +out: #ifdef CONFIG_QUOTA_DEBUG if (reserved) { quota_error(sb, "Writes happened before quota was turned on " @@ -958,92 +1076,42 @@ static void add_dquot_ref(struct super_block *sb, int type) "Please run quotacheck(8)"); } #endif + return err; } -/* - * Return 0 if dqput() won't block. - * (note that 1 doesn't necessarily mean blocking) - */ -static inline int dqput_blocks(struct dquot *dquot) -{ - if (atomic_read(&dquot->dq_count) <= 1) - return 1; - return 0; -} - -/* - * Remove references to dquots from inode and add dquot to list for freeing - * if we have the last reference to dquot - * We can't race with anybody because we hold dqptr_sem for writing... - */ -static int remove_inode_dquot_ref(struct inode *inode, int type, - struct list_head *tofree_head) -{ - struct dquot *dquot = inode->i_dquot[type]; - - inode->i_dquot[type] = NULL; - if (dquot) { - if (dqput_blocks(dquot)) { -#ifdef CONFIG_QUOTA_DEBUG - if (atomic_read(&dquot->dq_count) != 1) - quota_error(inode->i_sb, "Adding dquot with " - "dq_count %d to dispose list", - atomic_read(&dquot->dq_count)); -#endif - spin_lock(&dq_list_lock); - /* As dquot must have currently users it can't be on - * the free list... */ - list_add(&dquot->dq_free, tofree_head); - spin_unlock(&dq_list_lock); - return 1; - } - else - dqput(dquot); /* We have guaranteed we won't block */ - } - return 0; -} - -/* - * Free list of dquots - * Dquots are removed from inodes and no new references can be got so we are - * the only ones holding reference - */ -static void put_dquot_list(struct list_head *tofree_head) -{ - struct list_head *act_head; - struct dquot *dquot; - - act_head = tofree_head->next; - while (act_head != tofree_head) { - dquot = list_entry(act_head, struct dquot, dq_free); - act_head = act_head->next; - /* Remove dquot from the list so we won't have problems... */ - list_del_init(&dquot->dq_free); - dqput(dquot); - } -} - -static void remove_dquot_ref(struct super_block *sb, int type, - struct list_head *tofree_head) +static void remove_dquot_ref(struct super_block *sb, int type) { struct inode *inode; +#ifdef CONFIG_QUOTA_DEBUG int reserved = 0; +#endif - spin_lock(&inode_sb_list_lock); + spin_lock(&sb->s_inode_list_lock); list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { /* * We have to scan also I_NEW inodes because they can already * have quota pointer initialized. Luckily, we need to touch * only quota pointers and these have separate locking - * (dqptr_sem). + * (dq_data_lock). */ + spin_lock(&dq_data_lock); if (!IS_NOQUOTA(inode)) { + struct dquot __rcu **dquots = i_dquot(inode); + struct dquot *dquot = srcu_dereference_check( + dquots[type], &dquot_srcu, + lockdep_is_held(&dq_data_lock)); + +#ifdef CONFIG_QUOTA_DEBUG if (unlikely(inode_get_rsv_space(inode) > 0)) reserved = 1; - remove_inode_dquot_ref(inode, type, tofree_head); +#endif + rcu_assign_pointer(dquots[type], NULL); + if (dquot) + dqput(dquot); } + spin_unlock(&dq_data_lock); } - spin_unlock(&inode_sb_list_lock); + spin_unlock(&sb->s_inode_list_lock); #ifdef CONFIG_QUOTA_DEBUG if (reserved) { printk(KERN_WARNING "VFS (%s): Writes happened after quota" @@ -1056,42 +1124,8 @@ static void remove_dquot_ref(struct super_block *sb, int type, /* Gather all references from inodes and drop them */ static void drop_dquot_ref(struct super_block *sb, int type) { - LIST_HEAD(tofree_head); - - if (sb->dq_op) { - down_write(&sb_dqopt(sb)->dqptr_sem); - remove_dquot_ref(sb, type, &tofree_head); - up_write(&sb_dqopt(sb)->dqptr_sem); - put_dquot_list(&tofree_head); - } -} - -static inline void dquot_incr_inodes(struct dquot *dquot, qsize_t number) -{ - dquot->dq_dqb.dqb_curinodes += number; -} - -static inline void dquot_incr_space(struct dquot *dquot, qsize_t number) -{ - dquot->dq_dqb.dqb_curspace += number; -} - -static inline void dquot_resv_space(struct dquot *dquot, qsize_t number) -{ - dquot->dq_dqb.dqb_rsvspace += number; -} - -/* - * Claim reserved quota space - */ -static void dquot_claim_reserved_space(struct dquot *dquot, qsize_t number) -{ - if (dquot->dq_dqb.dqb_rsvspace < number) { - WARN_ON_ONCE(1); - number = dquot->dq_dqb.dqb_rsvspace; - } - dquot->dq_dqb.dqb_curspace += number; - dquot->dq_dqb.dqb_rsvspace -= number; + if (sb->dq_op) + remove_dquot_ref(sb, type); } static inline @@ -1103,6 +1137,10 @@ void dquot_free_reserved_space(struct dquot *dquot, qsize_t number) WARN_ON_ONCE(1); dquot->dq_dqb.dqb_rsvspace = 0; } + if (dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace <= + dquot->dq_dqb.dqb_bsoftlimit) + dquot->dq_dqb.dqb_btime = (time64_t) 0; + clear_bit(DQ_BLKS_B, &dquot->dq_flags); } static void dquot_decr_inodes(struct dquot *dquot, qsize_t number) @@ -1113,7 +1151,7 @@ static void dquot_decr_inodes(struct dquot *dquot, qsize_t number) else dquot->dq_dqb.dqb_curinodes = 0; if (dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit) - dquot->dq_dqb.dqb_itime = (time_t) 0; + dquot->dq_dqb.dqb_itime = (time64_t) 0; clear_bit(DQ_INODES_B, &dquot->dq_flags); } @@ -1124,8 +1162,9 @@ static void dquot_decr_space(struct dquot *dquot, qsize_t number) dquot->dq_dqb.dqb_curspace -= number; else dquot->dq_dqb.dqb_curspace = 0; - if (dquot->dq_dqb.dqb_curspace <= dquot->dq_dqb.dqb_bsoftlimit) - dquot->dq_dqb.dqb_btime = (time_t) 0; + if (dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace <= + dquot->dq_dqb.dqb_bsoftlimit) + dquot->dq_dqb.dqb_btime = (time64_t) 0; clear_bit(DQ_BLKS_B, &dquot->dq_flags); } @@ -1160,8 +1199,8 @@ static int need_print_warning(struct dquot_warn *warn) return uid_eq(current_fsuid(), warn->w_dq_id.uid); case GRPQUOTA: return in_group_p(warn->w_dq_id.gid); - case PRJQUOTA: /* Never taken... Just make gcc happy */ - return 0; + case PRJQUOTA: + return 1; } return 0; } @@ -1249,95 +1288,120 @@ static int ignore_hardlimit(struct dquot *dquot) return capable(CAP_SYS_RESOURCE) && (info->dqi_format->qf_fmt_id != QFMT_VFS_OLD || - !(info->dqi_flags & V1_DQF_RSQUASH)); + !(info->dqi_flags & DQF_ROOT_SQUASH)); } -/* needs dq_data_lock */ -static int check_idq(struct dquot *dquot, qsize_t inodes, - struct dquot_warn *warn) +static int dquot_add_inodes(struct dquot *dquot, qsize_t inodes, + struct dquot_warn *warn) { - qsize_t newinodes = dquot->dq_dqb.dqb_curinodes + inodes; + qsize_t newinodes; + int ret = 0; + spin_lock(&dquot->dq_dqb_lock); + newinodes = dquot->dq_dqb.dqb_curinodes + inodes; if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_id.type) || test_bit(DQ_FAKE_B, &dquot->dq_flags)) - return 0; + goto add; if (dquot->dq_dqb.dqb_ihardlimit && newinodes > dquot->dq_dqb.dqb_ihardlimit && !ignore_hardlimit(dquot)) { prepare_warning(warn, dquot, QUOTA_NL_IHARDWARN); - return -EDQUOT; + ret = -EDQUOT; + goto out; } if (dquot->dq_dqb.dqb_isoftlimit && newinodes > dquot->dq_dqb.dqb_isoftlimit && dquot->dq_dqb.dqb_itime && - get_seconds() >= dquot->dq_dqb.dqb_itime && + ktime_get_real_seconds() >= dquot->dq_dqb.dqb_itime && !ignore_hardlimit(dquot)) { prepare_warning(warn, dquot, QUOTA_NL_ISOFTLONGWARN); - return -EDQUOT; + ret = -EDQUOT; + goto out; } if (dquot->dq_dqb.dqb_isoftlimit && newinodes > dquot->dq_dqb.dqb_isoftlimit && dquot->dq_dqb.dqb_itime == 0) { prepare_warning(warn, dquot, QUOTA_NL_ISOFTWARN); - dquot->dq_dqb.dqb_itime = get_seconds() + + dquot->dq_dqb.dqb_itime = ktime_get_real_seconds() + sb_dqopt(dquot->dq_sb)->info[dquot->dq_id.type].dqi_igrace; } +add: + dquot->dq_dqb.dqb_curinodes = newinodes; - return 0; +out: + spin_unlock(&dquot->dq_dqb_lock); + return ret; } -/* needs dq_data_lock */ -static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, - struct dquot_warn *warn) +static int dquot_add_space(struct dquot *dquot, qsize_t space, + qsize_t rsv_space, unsigned int flags, + struct dquot_warn *warn) { qsize_t tspace; struct super_block *sb = dquot->dq_sb; + int ret = 0; + spin_lock(&dquot->dq_dqb_lock); if (!sb_has_quota_limits_enabled(sb, dquot->dq_id.type) || test_bit(DQ_FAKE_B, &dquot->dq_flags)) - return 0; + goto finish; tspace = dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace - + space; + + space + rsv_space; if (dquot->dq_dqb.dqb_bhardlimit && tspace > dquot->dq_dqb.dqb_bhardlimit && !ignore_hardlimit(dquot)) { - if (!prealloc) + if (flags & DQUOT_SPACE_WARN) prepare_warning(warn, dquot, QUOTA_NL_BHARDWARN); - return -EDQUOT; + ret = -EDQUOT; + goto finish; } if (dquot->dq_dqb.dqb_bsoftlimit && tspace > dquot->dq_dqb.dqb_bsoftlimit && dquot->dq_dqb.dqb_btime && - get_seconds() >= dquot->dq_dqb.dqb_btime && + ktime_get_real_seconds() >= dquot->dq_dqb.dqb_btime && !ignore_hardlimit(dquot)) { - if (!prealloc) + if (flags & DQUOT_SPACE_WARN) prepare_warning(warn, dquot, QUOTA_NL_BSOFTLONGWARN); - return -EDQUOT; + ret = -EDQUOT; + goto finish; } if (dquot->dq_dqb.dqb_bsoftlimit && tspace > dquot->dq_dqb.dqb_bsoftlimit && dquot->dq_dqb.dqb_btime == 0) { - if (!prealloc) { + if (flags & DQUOT_SPACE_WARN) { prepare_warning(warn, dquot, QUOTA_NL_BSOFTWARN); - dquot->dq_dqb.dqb_btime = get_seconds() + + dquot->dq_dqb.dqb_btime = ktime_get_real_seconds() + sb_dqopt(sb)->info[dquot->dq_id.type].dqi_bgrace; - } - else + } else { /* * We don't allow preallocation to exceed softlimit so exceeding will * be always printed */ - return -EDQUOT; + ret = -EDQUOT; + goto finish; + } } - - return 0; +finish: + /* + * We have to be careful and go through warning generation & grace time + * setting even if DQUOT_SPACE_NOFAIL is set. That's why we check it + * only here... + */ + if (flags & DQUOT_SPACE_NOFAIL) + ret = 0; + if (!ret) { + dquot->dq_dqb.dqb_rsvspace += rsv_space; + dquot->dq_dqb.dqb_curspace += space; + } + spin_unlock(&dquot->dq_dqb_lock); + return ret; } static int info_idq_free(struct dquot *dquot, qsize_t inodes) @@ -1360,19 +1424,23 @@ static int info_idq_free(struct dquot *dquot, qsize_t inodes) static int info_bdq_free(struct dquot *dquot, qsize_t space) { + qsize_t tspace; + + tspace = dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace; + if (test_bit(DQ_FAKE_B, &dquot->dq_flags) || - dquot->dq_dqb.dqb_curspace <= dquot->dq_dqb.dqb_bsoftlimit) + tspace <= dquot->dq_dqb.dqb_bsoftlimit) return QUOTA_NL_NOWARN; - if (dquot->dq_dqb.dqb_curspace - space <= dquot->dq_dqb.dqb_bsoftlimit) + if (tspace - space <= dquot->dq_dqb.dqb_bsoftlimit) return QUOTA_NL_BSOFTBELOW; - if (dquot->dq_dqb.dqb_curspace >= dquot->dq_dqb.dqb_bhardlimit && - dquot->dq_dqb.dqb_curspace - space < dquot->dq_dqb.dqb_bhardlimit) + if (tspace >= dquot->dq_dqb.dqb_bhardlimit && + tspace - space < dquot->dq_dqb.dqb_bhardlimit) return QUOTA_NL_BHARDBELOW; return QUOTA_NL_NOWARN; } -static int dquot_active(const struct inode *inode) +static int inode_quota_active(const struct inode *inode) { struct super_block *sb = inode->i_sb; @@ -1384,30 +1452,45 @@ static int dquot_active(const struct inode *inode) /* * Initialize quota pointers in inode * - * We do things in a bit complicated way but by that we avoid calling - * dqget() and thus filesystem callbacks under dqptr_sem. - * * It is better to call this function outside of any transaction as it * might need a lot of space in journal for dquot structure allocation. */ -static void __dquot_initialize(struct inode *inode, int type) +static int __dquot_initialize(struct inode *inode, int type) { - int cnt; - struct dquot *got[MAXQUOTAS]; + int cnt, init_needed = 0; + struct dquot __rcu **dquots; + struct dquot *got[MAXQUOTAS] = {}; struct super_block *sb = inode->i_sb; qsize_t rsv; + int ret = 0; - /* First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex */ - if (!dquot_active(inode)) - return; + if (!inode_quota_active(inode)) + return 0; + + dquots = i_dquot(inode); /* First get references to structures we might need. */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { struct kqid qid; - got[cnt] = NULL; + kprojid_t projid; + int rc; + struct dquot *dquot; + if (type != -1 && cnt != type) continue; + /* + * The i_dquot should have been initialized in most cases, + * we check it without locking here to avoid unnecessary + * dqget()/dqput() calls. + */ + if (dquots[cnt]) + continue; + + if (!sb_has_quota_active(sb, cnt)) + continue; + + init_needed = 1; + switch (cnt) { case USRQUOTA: qid = make_kqid_uid(inode->i_uid); @@ -1415,13 +1498,32 @@ static void __dquot_initialize(struct inode *inode, int type) case GRPQUOTA: qid = make_kqid_gid(inode->i_gid); break; + case PRJQUOTA: + rc = inode->i_sb->dq_op->get_projid(inode, &projid); + if (rc) + continue; + qid = make_kqid_projid(projid); + break; } - got[cnt] = dqget(sb, qid); + dquot = dqget(sb, qid); + if (IS_ERR(dquot)) { + /* We raced with somebody turning quotas off... */ + if (PTR_ERR(dquot) != -ESRCH) { + ret = PTR_ERR(dquot); + goto out_put; + } + dquot = NULL; + } + got[cnt] = dquot; } - down_write(&sb_dqopt(sb)->dqptr_sem); + /* All required i_dquot has been initialized */ + if (!init_needed) + return 0; + + spin_lock(&dq_data_lock); if (IS_NOQUOTA(inode)) - goto out_err; + goto out_lock; for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (type != -1 && cnt != type) continue; @@ -1431,8 +1533,8 @@ static void __dquot_initialize(struct inode *inode, int type) /* We could race with quotaon or dqget() could have failed */ if (!got[cnt]) continue; - if (!inode->i_dquot[cnt]) { - inode->i_dquot[cnt] = got[cnt]; + if (!dquots[cnt]) { + rcu_assign_pointer(dquots[cnt], got[cnt]); got[cnt] = NULL; /* * Make quota reservation system happy if someone @@ -1440,43 +1542,78 @@ static void __dquot_initialize(struct inode *inode, int type) */ rsv = inode_get_rsv_space(inode); if (unlikely(rsv)) { - spin_lock(&dq_data_lock); - dquot_resv_space(inode->i_dquot[cnt], rsv); - spin_unlock(&dq_data_lock); + struct dquot *dquot = srcu_dereference_check( + dquots[cnt], &dquot_srcu, + lockdep_is_held(&dq_data_lock)); + + spin_lock(&inode->i_lock); + /* Get reservation again under proper lock */ + rsv = __inode_get_rsv_space(inode); + spin_lock(&dquot->dq_dqb_lock); + dquot->dq_dqb.dqb_rsvspace += rsv; + spin_unlock(&dquot->dq_dqb_lock); + spin_unlock(&inode->i_lock); } } } -out_err: - up_write(&sb_dqopt(sb)->dqptr_sem); +out_lock: + spin_unlock(&dq_data_lock); +out_put: /* Drop unused references */ dqput_all(got); + + return ret; } -void dquot_initialize(struct inode *inode) +int dquot_initialize(struct inode *inode) { - __dquot_initialize(inode, -1); + return __dquot_initialize(inode, -1); } EXPORT_SYMBOL(dquot_initialize); +bool dquot_initialize_needed(struct inode *inode) +{ + struct dquot __rcu **dquots; + int i; + + if (!inode_quota_active(inode)) + return false; + + dquots = i_dquot(inode); + for (i = 0; i < MAXQUOTAS; i++) + if (!dquots[i] && sb_has_quota_active(inode->i_sb, i)) + return true; + return false; +} +EXPORT_SYMBOL(dquot_initialize_needed); + /* - * Release all quotas referenced by inode + * Release all quotas referenced by inode. + * + * This function only be called on inode free or converting + * a file to quota file, no other users for the i_dquot in + * both cases, so we needn't call synchronize_srcu() after + * clearing i_dquot. */ static void __dquot_drop(struct inode *inode) { int cnt; + struct dquot __rcu **dquots = i_dquot(inode); struct dquot *put[MAXQUOTAS]; - down_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + spin_lock(&dq_data_lock); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - put[cnt] = inode->i_dquot[cnt]; - inode->i_dquot[cnt] = NULL; + put[cnt] = srcu_dereference_check(dquots[cnt], &dquot_srcu, + lockdep_is_held(&dq_data_lock)); + rcu_assign_pointer(dquots[cnt], NULL); } - up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + spin_unlock(&dq_data_lock); dqput_all(put); } void dquot_drop(struct inode *inode) { + struct dquot __rcu * const *dquots; int cnt; if (IS_NOQUOTA(inode)) @@ -1489,8 +1626,9 @@ void dquot_drop(struct inode *inode) * must assure that nobody can come after the DQUOT_DROP and * add quota pointers back anyway. */ + dquots = i_dquot(inode); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (inode->i_dquot[cnt]) + if (dquots[cnt]) break; } @@ -1511,30 +1649,12 @@ static qsize_t *inode_reserved_space(struct inode * inode) return inode->i_sb->dq_op->get_reserved_space(inode); } -void inode_add_rsv_space(struct inode *inode, qsize_t number) -{ - spin_lock(&inode->i_lock); - *inode_reserved_space(inode) += number; - spin_unlock(&inode->i_lock); -} -EXPORT_SYMBOL(inode_add_rsv_space); - -void inode_claim_rsv_space(struct inode *inode, qsize_t number) -{ - spin_lock(&inode->i_lock); - *inode_reserved_space(inode) -= number; - __inode_add_bytes(inode, number); - spin_unlock(&inode->i_lock); -} -EXPORT_SYMBOL(inode_claim_rsv_space); - -void inode_sub_rsv_space(struct inode *inode, qsize_t number) +static qsize_t __inode_get_rsv_space(struct inode *inode) { - spin_lock(&inode->i_lock); - *inode_reserved_space(inode) -= number; - spin_unlock(&inode->i_lock); + if (!inode->i_sb->dq_op->get_reserved_space) + return 0; + return *inode_reserved_space(inode); } -EXPORT_SYMBOL(inode_sub_rsv_space); static qsize_t inode_get_rsv_space(struct inode *inode) { @@ -1543,28 +1663,11 @@ static qsize_t inode_get_rsv_space(struct inode *inode) if (!inode->i_sb->dq_op->get_reserved_space) return 0; spin_lock(&inode->i_lock); - ret = *inode_reserved_space(inode); + ret = __inode_get_rsv_space(inode); spin_unlock(&inode->i_lock); return ret; } -static void inode_incr_space(struct inode *inode, qsize_t number, - int reserve) -{ - if (reserve) - inode_add_rsv_space(inode, number); - else - inode_add_bytes(inode, number); -} - -static void inode_decr_space(struct inode *inode, qsize_t number, int reserve) -{ - if (reserve) - inode_sub_rsv_space(inode, number); - else - inode_sub_bytes(inode, number); -} - /* * This functions updates i_blocks+i_bytes fields and quota information * (together with appropriate checks). @@ -1580,51 +1683,66 @@ static void inode_decr_space(struct inode *inode, qsize_t number, int reserve) */ int __dquot_alloc_space(struct inode *inode, qsize_t number, int flags) { - int cnt, ret = 0; + int cnt, ret = 0, index; struct dquot_warn warn[MAXQUOTAS]; - struct dquot **dquots = inode->i_dquot; int reserve = flags & DQUOT_SPACE_RESERVE; + struct dquot __rcu **dquots; + struct dquot *dquot; - /* - * First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex - */ - if (!dquot_active(inode)) { - inode_incr_space(inode, number, reserve); + if (!inode_quota_active(inode)) { + if (reserve) { + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) += number; + spin_unlock(&inode->i_lock); + } else { + inode_add_bytes(inode, number); + } goto out; } for (cnt = 0; cnt < MAXQUOTAS; cnt++) warn[cnt].w_type = QUOTA_NL_NOWARN; - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - spin_lock(&dq_data_lock); + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (!dquots[cnt]) + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) continue; - ret = check_bdq(dquots[cnt], number, - !(flags & DQUOT_SPACE_WARN), &warn[cnt]); - if (ret && !(flags & DQUOT_SPACE_NOFAIL)) { - spin_unlock(&dq_data_lock); + if (reserve) { + ret = dquot_add_space(dquot, 0, number, flags, &warn[cnt]); + } else { + ret = dquot_add_space(dquot, number, 0, flags, &warn[cnt]); + } + if (ret) { + /* Back out changes we already did */ + for (cnt--; cnt >= 0; cnt--) { + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) + continue; + spin_lock(&dquot->dq_dqb_lock); + if (reserve) + dquot_free_reserved_space(dquot, number); + else + dquot_decr_space(dquot, number); + spin_unlock(&dquot->dq_dqb_lock); + } + spin_unlock(&inode->i_lock); goto out_flush_warn; } } - for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (!dquots[cnt]) - continue; - if (reserve) - dquot_resv_space(dquots[cnt], number); - else - dquot_incr_space(dquots[cnt], number); - } - inode_incr_space(inode, number, reserve); - spin_unlock(&dq_data_lock); + if (reserve) + *inode_reserved_space(inode) += number; + else + __inode_add_bytes(inode, number); + spin_unlock(&inode->i_lock); if (reserve) goto out_flush_warn; - mark_all_dquot_dirty(dquots); + ret = mark_all_dquot_dirty(dquots); out_flush_warn: - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); out: return ret; @@ -1634,39 +1752,45 @@ EXPORT_SYMBOL(__dquot_alloc_space); /* * This operation can block, but only after everything is updated */ -int dquot_alloc_inode(const struct inode *inode) +int dquot_alloc_inode(struct inode *inode) { - int cnt, ret = 0; + int cnt, ret = 0, index; struct dquot_warn warn[MAXQUOTAS]; - struct dquot * const *dquots = inode->i_dquot; + struct dquot __rcu * const *dquots; + struct dquot *dquot; - /* First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex */ - if (!dquot_active(inode)) + if (!inode_quota_active(inode)) return 0; for (cnt = 0; cnt < MAXQUOTAS; cnt++) warn[cnt].w_type = QUOTA_NL_NOWARN; - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - spin_lock(&dq_data_lock); - for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (!dquots[cnt]) - continue; - ret = check_idq(dquots[cnt], 1, &warn[cnt]); - if (ret) - goto warn_put_all; - } + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (!dquots[cnt]) + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) continue; - dquot_incr_inodes(dquots[cnt], 1); + ret = dquot_add_inodes(dquot, 1, &warn[cnt]); + if (ret) { + for (cnt--; cnt >= 0; cnt--) { + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) + continue; + /* Back out changes we already did */ + spin_lock(&dquot->dq_dqb_lock); + dquot_decr_inodes(dquot, 1); + spin_unlock(&dquot->dq_dqb_lock); + } + goto warn_put_all; + } } warn_put_all: - spin_unlock(&dq_data_lock); + spin_unlock(&inode->i_lock); if (ret == 0) - mark_all_dquot_dirty(dquots); - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + ret = mark_all_dquot_dirty(dquots); + srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); return ret; } @@ -1675,73 +1799,138 @@ EXPORT_SYMBOL(dquot_alloc_inode); /* * Convert in-memory reserved quotas to real consumed quotas */ -int dquot_claim_space_nodirty(struct inode *inode, qsize_t number) +void dquot_claim_space_nodirty(struct inode *inode, qsize_t number) { - int cnt; + struct dquot __rcu **dquots; + struct dquot *dquot; + int cnt, index; - if (!dquot_active(inode)) { - inode_claim_rsv_space(inode, number); - return 0; + if (!inode_quota_active(inode)) { + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) -= number; + __inode_add_bytes(inode, number); + spin_unlock(&inode->i_lock); + return; } - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - spin_lock(&dq_data_lock); + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); /* Claim reserved quotas to allocated quotas */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (inode->i_dquot[cnt]) - dquot_claim_reserved_space(inode->i_dquot[cnt], - number); + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (dquot) { + spin_lock(&dquot->dq_dqb_lock); + if (WARN_ON_ONCE(dquot->dq_dqb.dqb_rsvspace < number)) + number = dquot->dq_dqb.dqb_rsvspace; + dquot->dq_dqb.dqb_curspace += number; + dquot->dq_dqb.dqb_rsvspace -= number; + spin_unlock(&dquot->dq_dqb_lock); + } } /* Update inode bytes */ - inode_claim_rsv_space(inode, number); - spin_unlock(&dq_data_lock); - mark_all_dquot_dirty(inode->i_dquot); - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - return 0; + *inode_reserved_space(inode) -= number; + __inode_add_bytes(inode, number); + spin_unlock(&inode->i_lock); + mark_all_dquot_dirty(dquots); + srcu_read_unlock(&dquot_srcu, index); } EXPORT_SYMBOL(dquot_claim_space_nodirty); /* + * Convert allocated space back to in-memory reserved quotas + */ +void dquot_reclaim_space_nodirty(struct inode *inode, qsize_t number) +{ + struct dquot __rcu **dquots; + struct dquot *dquot; + int cnt, index; + + if (!inode_quota_active(inode)) { + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) += number; + __inode_sub_bytes(inode, number); + spin_unlock(&inode->i_lock); + return; + } + + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); + /* Claim reserved quotas to allocated quotas */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (dquot) { + spin_lock(&dquot->dq_dqb_lock); + if (WARN_ON_ONCE(dquot->dq_dqb.dqb_curspace < number)) + number = dquot->dq_dqb.dqb_curspace; + dquot->dq_dqb.dqb_rsvspace += number; + dquot->dq_dqb.dqb_curspace -= number; + spin_unlock(&dquot->dq_dqb_lock); + } + } + /* Update inode bytes */ + *inode_reserved_space(inode) += number; + __inode_sub_bytes(inode, number); + spin_unlock(&inode->i_lock); + mark_all_dquot_dirty(dquots); + srcu_read_unlock(&dquot_srcu, index); +} +EXPORT_SYMBOL(dquot_reclaim_space_nodirty); + +/* * This operation can block, but only after everything is updated */ void __dquot_free_space(struct inode *inode, qsize_t number, int flags) { unsigned int cnt; struct dquot_warn warn[MAXQUOTAS]; - struct dquot **dquots = inode->i_dquot; - int reserve = flags & DQUOT_SPACE_RESERVE; + struct dquot __rcu **dquots; + struct dquot *dquot; + int reserve = flags & DQUOT_SPACE_RESERVE, index; - /* First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex */ - if (!dquot_active(inode)) { - inode_decr_space(inode, number, reserve); + if (!inode_quota_active(inode)) { + if (reserve) { + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) -= number; + spin_unlock(&inode->i_lock); + } else { + inode_sub_bytes(inode, number); + } return; } - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - spin_lock(&dq_data_lock); + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { int wtype; warn[cnt].w_type = QUOTA_NL_NOWARN; - if (!dquots[cnt]) + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) continue; - wtype = info_bdq_free(dquots[cnt], number); + spin_lock(&dquot->dq_dqb_lock); + wtype = info_bdq_free(dquot, number); if (wtype != QUOTA_NL_NOWARN) - prepare_warning(&warn[cnt], dquots[cnt], wtype); + prepare_warning(&warn[cnt], dquot, wtype); if (reserve) - dquot_free_reserved_space(dquots[cnt], number); + dquot_free_reserved_space(dquot, number); else - dquot_decr_space(dquots[cnt], number); + dquot_decr_space(dquot, number); + spin_unlock(&dquot->dq_dqb_lock); } - inode_decr_space(inode, number, reserve); - spin_unlock(&dq_data_lock); + if (reserve) + *inode_reserved_space(inode) -= number; + else + __inode_sub_bytes(inode, number); + spin_unlock(&inode->i_lock); if (reserve) goto out_unlock; mark_all_dquot_dirty(dquots); out_unlock: - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); } EXPORT_SYMBOL(__dquot_free_space); @@ -1749,33 +1938,36 @@ EXPORT_SYMBOL(__dquot_free_space); /* * This operation can block, but only after everything is updated */ -void dquot_free_inode(const struct inode *inode) +void dquot_free_inode(struct inode *inode) { unsigned int cnt; struct dquot_warn warn[MAXQUOTAS]; - struct dquot * const *dquots = inode->i_dquot; + struct dquot __rcu * const *dquots; + struct dquot *dquot; + int index; - /* First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex */ - if (!dquot_active(inode)) + if (!inode_quota_active(inode)) return; - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - spin_lock(&dq_data_lock); + dquots = i_dquot(inode); + index = srcu_read_lock(&dquot_srcu); + spin_lock(&inode->i_lock); for (cnt = 0; cnt < MAXQUOTAS; cnt++) { int wtype; - warn[cnt].w_type = QUOTA_NL_NOWARN; - if (!dquots[cnt]) + dquot = srcu_dereference(dquots[cnt], &dquot_srcu); + if (!dquot) continue; - wtype = info_idq_free(dquots[cnt], 1); + spin_lock(&dquot->dq_dqb_lock); + wtype = info_idq_free(dquot, 1); if (wtype != QUOTA_NL_NOWARN) - prepare_warning(&warn[cnt], dquots[cnt], wtype); - dquot_decr_inodes(dquots[cnt], 1); + prepare_warning(&warn[cnt], dquot, wtype); + dquot_decr_inodes(dquot, 1); + spin_unlock(&dquot->dq_dqb_lock); } - spin_unlock(&dq_data_lock); + spin_unlock(&inode->i_lock); mark_all_dquot_dirty(dquots); - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + srcu_read_unlock(&dquot_srcu, index); flush_warnings(warn); } EXPORT_SYMBOL(dquot_free_inode); @@ -1789,38 +1981,52 @@ EXPORT_SYMBOL(dquot_free_inode); * This operation can block, but only after everything is updated * A transaction must be started when entering this function. * + * We are holding reference on transfer_from & transfer_to, no need to + * protect them by srcu_read_lock(). */ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) { - qsize_t space, cur_space; + qsize_t cur_space; qsize_t rsv_space = 0; + qsize_t inode_usage = 1; + struct dquot __rcu **dquots; struct dquot *transfer_from[MAXQUOTAS] = {}; - int cnt, ret = 0; + int cnt, index, ret = 0, err; char is_valid[MAXQUOTAS] = {}; struct dquot_warn warn_to[MAXQUOTAS]; struct dquot_warn warn_from_inodes[MAXQUOTAS]; struct dquot_warn warn_from_space[MAXQUOTAS]; - /* First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex */ if (IS_NOQUOTA(inode)) return 0; + + if (inode->i_sb->dq_op->get_inode_usage) { + ret = inode->i_sb->dq_op->get_inode_usage(inode, &inode_usage); + if (ret) + return ret; + } + /* Initialize the arrays */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { warn_to[cnt].w_type = QUOTA_NL_NOWARN; warn_from_inodes[cnt].w_type = QUOTA_NL_NOWARN; warn_from_space[cnt].w_type = QUOTA_NL_NOWARN; } - down_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + + spin_lock(&dq_data_lock); + spin_lock(&inode->i_lock); if (IS_NOQUOTA(inode)) { /* File without quota accounting? */ - up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + spin_unlock(&inode->i_lock); + spin_unlock(&dq_data_lock); return 0; } - spin_lock(&dq_data_lock); - cur_space = inode_get_bytes(inode); - rsv_space = inode_get_rsv_space(inode); - space = cur_space + rsv_space; - /* Build the transfer_from list and check the limits */ + cur_space = __inode_get_bytes(inode); + rsv_space = __inode_get_rsv_space(inode); + dquots = i_dquot(inode); + /* + * Build the transfer_from list, check limits, and update usage in + * the target structures. + */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { /* * Skip changes for same uid or gid or for turned off quota-type. @@ -1831,49 +2037,65 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) if (!sb_has_quota_active(inode->i_sb, cnt)) continue; is_valid[cnt] = 1; - transfer_from[cnt] = inode->i_dquot[cnt]; - ret = check_idq(transfer_to[cnt], 1, &warn_to[cnt]); + transfer_from[cnt] = srcu_dereference_check(dquots[cnt], + &dquot_srcu, lockdep_is_held(&dq_data_lock)); + ret = dquot_add_inodes(transfer_to[cnt], inode_usage, + &warn_to[cnt]); if (ret) goto over_quota; - ret = check_bdq(transfer_to[cnt], space, 0, &warn_to[cnt]); - if (ret) + ret = dquot_add_space(transfer_to[cnt], cur_space, rsv_space, + DQUOT_SPACE_WARN, &warn_to[cnt]); + if (ret) { + spin_lock(&transfer_to[cnt]->dq_dqb_lock); + dquot_decr_inodes(transfer_to[cnt], inode_usage); + spin_unlock(&transfer_to[cnt]->dq_dqb_lock); goto over_quota; + } } - /* - * Finally perform the needed transfer from transfer_from to transfer_to - */ + /* Decrease usage for source structures and update quota pointers */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (!is_valid[cnt]) continue; /* Due to IO error we might not have transfer_from[] structure */ if (transfer_from[cnt]) { int wtype; - wtype = info_idq_free(transfer_from[cnt], 1); + + spin_lock(&transfer_from[cnt]->dq_dqb_lock); + wtype = info_idq_free(transfer_from[cnt], inode_usage); if (wtype != QUOTA_NL_NOWARN) prepare_warning(&warn_from_inodes[cnt], transfer_from[cnt], wtype); - wtype = info_bdq_free(transfer_from[cnt], space); + wtype = info_bdq_free(transfer_from[cnt], + cur_space + rsv_space); if (wtype != QUOTA_NL_NOWARN) prepare_warning(&warn_from_space[cnt], transfer_from[cnt], wtype); - dquot_decr_inodes(transfer_from[cnt], 1); + dquot_decr_inodes(transfer_from[cnt], inode_usage); dquot_decr_space(transfer_from[cnt], cur_space); dquot_free_reserved_space(transfer_from[cnt], rsv_space); + spin_unlock(&transfer_from[cnt]->dq_dqb_lock); } - - dquot_incr_inodes(transfer_to[cnt], 1); - dquot_incr_space(transfer_to[cnt], cur_space); - dquot_resv_space(transfer_to[cnt], rsv_space); - - inode->i_dquot[cnt] = transfer_to[cnt]; + rcu_assign_pointer(dquots[cnt], transfer_to[cnt]); } + spin_unlock(&inode->i_lock); spin_unlock(&dq_data_lock); - up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); - mark_all_dquot_dirty(transfer_from); - mark_all_dquot_dirty(transfer_to); + /* + * These arrays are local and we hold dquot references so we don't need + * the srcu protection but still take dquot_srcu to avoid warning in + * mark_all_dquot_dirty(). + */ + index = srcu_read_lock(&dquot_srcu); + err = mark_all_dquot_dirty((struct dquot __rcu **)transfer_from); + if (err < 0) + ret = err; + err = mark_all_dquot_dirty((struct dquot __rcu **)transfer_to); + if (err < 0) + ret = err; + srcu_read_unlock(&dquot_srcu, index); + flush_warnings(warn_to); flush_warnings(warn_from_inodes); flush_warnings(warn_from_space); @@ -1881,10 +2103,20 @@ int __dquot_transfer(struct inode *inode, struct dquot **transfer_to) for (cnt = 0; cnt < MAXQUOTAS; cnt++) if (is_valid[cnt]) transfer_to[cnt] = transfer_from[cnt]; - return 0; + return ret; over_quota: + /* Back out changes we already did */ + for (cnt--; cnt >= 0; cnt--) { + if (!is_valid[cnt]) + continue; + spin_lock(&transfer_to[cnt]->dq_dqb_lock); + dquot_decr_inodes(transfer_to[cnt], inode_usage); + dquot_decr_space(transfer_to[cnt], cur_space); + dquot_free_reserved_space(transfer_to[cnt], rsv_space); + spin_unlock(&transfer_to[cnt]->dq_dqb_lock); + } + spin_unlock(&inode->i_lock); spin_unlock(&dq_data_lock); - up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); flush_warnings(warn_to); return ret; } @@ -1893,21 +2125,47 @@ EXPORT_SYMBOL(__dquot_transfer); /* Wrapper for transferring ownership of an inode for uid/gid only * Called from FSXXX_setattr() */ -int dquot_transfer(struct inode *inode, struct iattr *iattr) +int dquot_transfer(struct mnt_idmap *idmap, struct inode *inode, + struct iattr *iattr) { struct dquot *transfer_to[MAXQUOTAS] = {}; + struct dquot *dquot; struct super_block *sb = inode->i_sb; int ret; - if (!dquot_active(inode)) + if (!inode_quota_active(inode)) return 0; - if (iattr->ia_valid & ATTR_UID && !uid_eq(iattr->ia_uid, inode->i_uid)) - transfer_to[USRQUOTA] = dqget(sb, make_kqid_uid(iattr->ia_uid)); - if (iattr->ia_valid & ATTR_GID && !gid_eq(iattr->ia_gid, inode->i_gid)) - transfer_to[GRPQUOTA] = dqget(sb, make_kqid_gid(iattr->ia_gid)); + if (i_uid_needs_update(idmap, iattr, inode)) { + kuid_t kuid = from_vfsuid(idmap, i_user_ns(inode), + iattr->ia_vfsuid); + dquot = dqget(sb, make_kqid_uid(kuid)); + if (IS_ERR(dquot)) { + if (PTR_ERR(dquot) != -ESRCH) { + ret = PTR_ERR(dquot); + goto out_put; + } + dquot = NULL; + } + transfer_to[USRQUOTA] = dquot; + } + if (i_gid_needs_update(idmap, iattr, inode)) { + kgid_t kgid = from_vfsgid(idmap, i_user_ns(inode), + iattr->ia_vfsgid); + + dquot = dqget(sb, make_kqid_gid(kgid)); + if (IS_ERR(dquot)) { + if (PTR_ERR(dquot) != -ESRCH) { + ret = PTR_ERR(dquot); + goto out_put; + } + dquot = NULL; + } + transfer_to[GRPQUOTA] = dquot; + } ret = __dquot_transfer(inode, transfer_to); +out_put: dqput_all(transfer_to); return ret; } @@ -1918,16 +2176,24 @@ EXPORT_SYMBOL(dquot_transfer); */ int dquot_commit_info(struct super_block *sb, int type) { - int ret; struct quota_info *dqopt = sb_dqopt(sb); - mutex_lock(&dqopt->dqio_mutex); - ret = dqopt->ops[type]->write_file_info(sb, type); - mutex_unlock(&dqopt->dqio_mutex); - return ret; + return dqopt->ops[type]->write_file_info(sb, type); } EXPORT_SYMBOL(dquot_commit_info); +int dquot_get_next_id(struct super_block *sb, struct kqid *qid) +{ + struct quota_info *dqopt = sb_dqopt(sb); + + if (!sb_has_quota_active(sb, qid->type)) + return -ESRCH; + if (!dqopt->ops[qid->type]->get_next_id) + return -ENOSYS; + return dqopt->ops[qid->type]->get_next_id(sb, qid); +} +EXPORT_SYMBOL(dquot_get_next_id); + /* * Definitions of diskquota operations. */ @@ -1939,6 +2205,7 @@ const struct dquot_operations dquot_operations = { .write_info = dquot_commit_info, .alloc_dquot = dquot_alloc, .destroy_dquot = dquot_destroy, + .get_next_id = dquot_get_next_id, }; EXPORT_SYMBOL(dquot_operations); @@ -1951,19 +2218,36 @@ int dquot_file_open(struct inode *inode, struct file *file) error = generic_file_open(inode, file); if (!error && (file->f_mode & FMODE_WRITE)) - dquot_initialize(inode); + error = dquot_initialize(inode); return error; } EXPORT_SYMBOL(dquot_file_open); +static void vfs_cleanup_quota_inode(struct super_block *sb, int type) +{ + struct quota_info *dqopt = sb_dqopt(sb); + struct inode *inode = dqopt->files[type]; + + if (!inode) + return; + if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) { + inode_lock(inode); + inode->i_flags &= ~S_NOQUOTA; + inode_unlock(inode); + } + dqopt->files[type] = NULL; + iput(inode); +} + /* * Turn quota off on a device. type == -1 ==> quotaoff for all types (umount) */ int dquot_disable(struct super_block *sb, int type, unsigned int flags) { - int cnt, ret = 0; + int cnt; struct quota_info *dqopt = sb_dqopt(sb); - struct inode *toputinode[MAXQUOTAS]; + + rwsem_assert_held_write(&sb->s_umount); /* Cannot turn off usage accounting without turning off limits, or * suspend quotas and simultaneously turn quotas off. */ @@ -1972,20 +2256,15 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags) DQUOT_USAGE_ENABLED))) return -EINVAL; - /* We need to serialize quota_off() for device */ - mutex_lock(&dqopt->dqonoff_mutex); - /* * Skip everything if there's nothing to do. We have to do this because * sometimes we are called when fill_super() failed and calling * sync_fs() in such cases does no good. */ - if (!sb_any_quota_loaded(sb)) { - mutex_unlock(&dqopt->dqonoff_mutex); + if (!sb_any_quota_loaded(sb)) return 0; - } + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - toputinode[cnt] = NULL; if (type != -1 && cnt != type) continue; if (!sb_has_quota_loaded(sb, cnt)) @@ -2005,8 +2284,7 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags) dqopt->flags &= ~dquot_state_flag( DQUOT_SUSPENDED, cnt); spin_unlock(&dq_state_lock); - iput(dqopt->files[cnt]); - dqopt->files[cnt] = NULL; + vfs_cleanup_quota_inode(sb, cnt); continue; } spin_unlock(&dq_state_lock); @@ -2028,16 +2306,11 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags) if (dqopt->ops[cnt]->free_file_info) dqopt->ops[cnt]->free_file_info(sb, cnt); put_quota_format(dqopt->info[cnt].dqi_format); - - toputinode[cnt] = dqopt->files[cnt]; - if (!sb_has_quota_loaded(sb, cnt)) - dqopt->files[cnt] = NULL; dqopt->info[cnt].dqi_flags = 0; dqopt->info[cnt].dqi_igrace = 0; dqopt->info[cnt].dqi_bgrace = 0; dqopt->ops[cnt] = NULL; } - mutex_unlock(&dqopt->dqonoff_mutex); /* Skip syncing and setting flags if quota files are hidden */ if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) @@ -2054,39 +2327,22 @@ int dquot_disable(struct super_block *sb, int type, unsigned int flags) * must also discard the blockdev buffers so that we see the * changes done by userspace on the next quotaon() */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (toputinode[cnt]) { - mutex_lock(&dqopt->dqonoff_mutex); - /* If quota was reenabled in the meantime, we have - * nothing to do */ - if (!sb_has_quota_loaded(sb, cnt)) { - mutex_lock(&toputinode[cnt]->i_mutex); - toputinode[cnt]->i_flags &= ~(S_IMMUTABLE | - S_NOATIME | S_NOQUOTA); - truncate_inode_pages(&toputinode[cnt]->i_data, - 0); - mutex_unlock(&toputinode[cnt]->i_mutex); - mark_inode_dirty_sync(toputinode[cnt]); - } - mutex_unlock(&dqopt->dqonoff_mutex); + if (!sb_has_quota_loaded(sb, cnt) && dqopt->files[cnt]) { + inode_lock(dqopt->files[cnt]); + truncate_inode_pages(&dqopt->files[cnt]->i_data, 0); + inode_unlock(dqopt->files[cnt]); } if (sb->s_bdev) invalidate_bdev(sb->s_bdev); put_inodes: + /* We are done when suspending quotas */ + if (flags & DQUOT_SUSPENDED) + return 0; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (toputinode[cnt]) { - /* On remount RO, we keep the inode pointer so that we - * can reenable quota on the subsequent remount RW. We - * have to check 'flags' variable and not use sb_has_ - * function because another quotaon / quotaoff could - * change global state before we got here. We refuse - * to suspend quotas when there is pending delete on - * the quota file... */ - if (!(flags & DQUOT_SUSPENDED)) - iput(toputinode[cnt]); - else if (!toputinode[cnt]->i_nlink) - ret = -EBUSY; - } - return ret; + if (!sb_has_quota_loaded(sb, cnt)) + vfs_cleanup_quota_inode(sb, cnt); + return 0; } EXPORT_SYMBOL(dquot_disable); @@ -2101,30 +2357,76 @@ EXPORT_SYMBOL(dquot_quota_off); * Turn quotas on on a device */ -/* - * Helper function to turn quotas on when we already have the inode of - * quota file and no quota information is loaded. - */ -static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, - unsigned int flags) +static int vfs_setup_quota_inode(struct inode *inode, int type) { - struct quota_format_type *fmt = find_quota_format(format_id); struct super_block *sb = inode->i_sb; struct quota_info *dqopt = sb_dqopt(sb); + + if (is_bad_inode(inode)) + return -EUCLEAN; + if (!S_ISREG(inode->i_mode)) + return -EACCES; + if (IS_RDONLY(inode)) + return -EROFS; + if (sb_has_quota_loaded(sb, type)) + return -EBUSY; + + /* + * Quota files should never be encrypted. They should be thought of as + * filesystem metadata, not user data. New-style internal quota files + * cannot be encrypted by users anyway, but old-style external quota + * files could potentially be incorrectly created in an encrypted + * directory, hence this explicit check. Some reasons why encrypted + * quota files don't work include: (1) some filesystems that support + * encryption don't handle it in their quota_read and quota_write, and + * (2) cleaning up encrypted quota files at unmount would need special + * consideration, as quota files are cleaned up later than user files. + */ + if (IS_ENCRYPTED(inode)) + return -EINVAL; + + dqopt->files[type] = igrab(inode); + if (!dqopt->files[type]) + return -EIO; + if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) { + /* We don't want quota and atime on quota files (deadlocks + * possible) Also nobody should write to the file - we use + * special IO operations which ignore the immutable bit. */ + inode_lock(inode); + inode->i_flags |= S_NOQUOTA; + inode_unlock(inode); + /* + * When S_NOQUOTA is set, remove dquot references as no more + * references can be added + */ + __dquot_drop(inode); + } + return 0; +} + +int dquot_load_quota_sb(struct super_block *sb, int type, int format_id, + unsigned int flags) +{ + struct quota_format_type *fmt; + struct quota_info *dqopt = sb_dqopt(sb); int error; - int oldflags = -1; + lockdep_assert_held_write(&sb->s_umount); + + /* Just unsuspend quotas? */ + if (WARN_ON_ONCE(flags & DQUOT_SUSPENDED)) + return -EINVAL; + + fmt = find_quota_format(format_id); if (!fmt) return -ESRCH; - if (!S_ISREG(inode->i_mode)) { - error = -EACCES; - goto out_fmt; - } - if (IS_RDONLY(inode)) { - error = -EROFS; + if (!sb->dq_op || !sb->s_qcop || + (type == PRJQUOTA && sb->dq_op->get_projid == NULL)) { + error = -EINVAL; goto out_fmt; } - if (!sb->s_op->quota_write || !sb->s_op->quota_read) { + /* Filesystems outside of init_user_ns not yet supported */ + if (sb->s_user_ns != &init_user_ns) { error = -EINVAL; goto out_fmt; } @@ -2133,6 +2435,10 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, error = -EINVAL; goto out_fmt; } + if (sb_has_quota_loaded(sb, type)) { + error = -EBUSY; + goto out_fmt; + } if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) { /* As we bypass the pagecache we must now flush all the @@ -2144,108 +2450,86 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, sync_filesystem(sb); invalidate_bdev(sb->s_bdev); } - mutex_lock(&dqopt->dqonoff_mutex); - if (sb_has_quota_loaded(sb, type)) { - error = -EBUSY; - goto out_lock; - } - if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) { - /* We don't want quota and atime on quota files (deadlocks - * possible) Also nobody should write to the file - we use - * special IO operations which ignore the immutable bit. */ - mutex_lock(&inode->i_mutex); - oldflags = inode->i_flags & (S_NOATIME | S_IMMUTABLE | - S_NOQUOTA); - inode->i_flags |= S_NOQUOTA | S_NOATIME | S_IMMUTABLE; - mutex_unlock(&inode->i_mutex); - /* - * When S_NOQUOTA is set, remove dquot references as no more - * references can be added - */ - __dquot_drop(inode); - } - - error = -EIO; - dqopt->files[type] = igrab(inode); - if (!dqopt->files[type]) - goto out_lock; error = -EINVAL; if (!fmt->qf_ops->check_quota_file(sb, type)) - goto out_file_init; + goto out_fmt; dqopt->ops[type] = fmt->qf_ops; dqopt->info[type].dqi_format = fmt; dqopt->info[type].dqi_fmt_id = format_id; INIT_LIST_HEAD(&dqopt->info[type].dqi_dirty_list); - mutex_lock(&dqopt->dqio_mutex); error = dqopt->ops[type]->read_file_info(sb, type); - if (error < 0) { - mutex_unlock(&dqopt->dqio_mutex); - goto out_file_init; - } - if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) + if (error < 0) + goto out_fmt; + if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) { + spin_lock(&dq_data_lock); dqopt->info[type].dqi_flags |= DQF_SYS_FILE; - mutex_unlock(&dqopt->dqio_mutex); + spin_unlock(&dq_data_lock); + } spin_lock(&dq_state_lock); dqopt->flags |= dquot_state_flag(flags, type); spin_unlock(&dq_state_lock); - add_dquot_ref(sb, type); - mutex_unlock(&dqopt->dqonoff_mutex); - - return 0; + error = add_dquot_ref(sb, type); + if (error) + dquot_disable(sb, type, + DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); -out_file_init: - dqopt->files[type] = NULL; - iput(inode); -out_lock: - if (oldflags != -1) { - mutex_lock(&inode->i_mutex); - /* Set the flags back (in the case of accidental quotaon() - * on a wrong file we don't want to mess up the flags) */ - inode->i_flags &= ~(S_NOATIME | S_NOQUOTA | S_IMMUTABLE); - inode->i_flags |= oldflags; - mutex_unlock(&inode->i_mutex); - } - mutex_unlock(&dqopt->dqonoff_mutex); + return error; out_fmt: put_quota_format(fmt); - return error; + return error; +} +EXPORT_SYMBOL(dquot_load_quota_sb); + +/* + * More powerful function for turning on quotas on given quota inode allowing + * setting of individual quota flags + */ +int dquot_load_quota_inode(struct inode *inode, int type, int format_id, + unsigned int flags) +{ + int err; + + err = vfs_setup_quota_inode(inode, type); + if (err < 0) + return err; + err = dquot_load_quota_sb(inode->i_sb, type, format_id, flags); + if (err < 0) + vfs_cleanup_quota_inode(inode->i_sb, type); + return err; } +EXPORT_SYMBOL(dquot_load_quota_inode); /* Reenable quotas on remount RW */ int dquot_resume(struct super_block *sb, int type) { struct quota_info *dqopt = sb_dqopt(sb); - struct inode *inode; int ret = 0, cnt; unsigned int flags; + rwsem_assert_held_write(&sb->s_umount); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { if (type != -1 && cnt != type) continue; - - mutex_lock(&dqopt->dqonoff_mutex); - if (!sb_has_quota_suspended(sb, cnt)) { - mutex_unlock(&dqopt->dqonoff_mutex); + if (!sb_has_quota_suspended(sb, cnt)) continue; - } - inode = dqopt->files[cnt]; - dqopt->files[cnt] = NULL; + spin_lock(&dq_state_lock); flags = dqopt->flags & dquot_state_flag(DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED, cnt); dqopt->flags &= ~dquot_state_flag(DQUOT_STATE_FLAGS, cnt); spin_unlock(&dq_state_lock); - mutex_unlock(&dqopt->dqonoff_mutex); flags = dquot_generic_flag(flags, cnt); - ret = vfs_load_quota_inode(inode, cnt, - dqopt->info[cnt].dqi_fmt_id, flags); - iput(inode); + ret = dquot_load_quota_sb(sb, cnt, dqopt->info[cnt].dqi_fmt_id, + flags); + if (ret < 0) + vfs_cleanup_quota_inode(sb, cnt); } return ret; @@ -2253,7 +2537,7 @@ int dquot_resume(struct super_block *sb, int type) EXPORT_SYMBOL(dquot_resume); int dquot_quota_on(struct super_block *sb, int type, int format_id, - struct path *path) + const struct path *path) { int error = security_quota_on(path->dentry); if (error) @@ -2262,7 +2546,7 @@ int dquot_quota_on(struct super_block *sb, int type, int format_id, if (path->dentry->d_sb != sb) error = -EXDEV; else - error = vfs_load_quota_inode(path->dentry->d_inode, type, + error = dquot_load_quota_inode(d_inode(path->dentry), type, format_id, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); return error; @@ -2270,53 +2554,6 @@ int dquot_quota_on(struct super_block *sb, int type, int format_id, EXPORT_SYMBOL(dquot_quota_on); /* - * More powerful function for turning on quotas allowing setting - * of individual quota flags - */ -int dquot_enable(struct inode *inode, int type, int format_id, - unsigned int flags) -{ - int ret = 0; - struct super_block *sb = inode->i_sb; - struct quota_info *dqopt = sb_dqopt(sb); - - /* Just unsuspend quotas? */ - BUG_ON(flags & DQUOT_SUSPENDED); - - if (!flags) - return 0; - /* Just updating flags needed? */ - if (sb_has_quota_loaded(sb, type)) { - mutex_lock(&dqopt->dqonoff_mutex); - /* Now do a reliable test... */ - if (!sb_has_quota_loaded(sb, type)) { - mutex_unlock(&dqopt->dqonoff_mutex); - goto load_quota; - } - if (flags & DQUOT_USAGE_ENABLED && - sb_has_quota_usage_enabled(sb, type)) { - ret = -EBUSY; - goto out_lock; - } - if (flags & DQUOT_LIMITS_ENABLED && - sb_has_quota_limits_enabled(sb, type)) { - ret = -EBUSY; - goto out_lock; - } - spin_lock(&dq_state_lock); - sb_dqopt(sb)->flags |= dquot_state_flag(flags, type); - spin_unlock(&dq_state_lock); -out_lock: - mutex_unlock(&dqopt->dqonoff_mutex); - return ret; - } - -load_quota: - return vfs_load_quota_inode(inode, type, format_id, flags); -} -EXPORT_SYMBOL(dquot_enable); - -/* * This function is used when filesystem needs to initialize quotas * during mount time. */ @@ -2326,69 +2563,130 @@ int dquot_quota_on_mount(struct super_block *sb, char *qf_name, struct dentry *dentry; int error; - mutex_lock(&sb->s_root->d_inode->i_mutex); - dentry = lookup_one_len(qf_name, sb->s_root, strlen(qf_name)); - mutex_unlock(&sb->s_root->d_inode->i_mutex); + dentry = lookup_noperm_positive_unlocked(&QSTR(qf_name), sb->s_root); if (IS_ERR(dentry)) return PTR_ERR(dentry); - if (!dentry->d_inode) { - error = -ENOENT; - goto out; - } - error = security_quota_on(dentry); if (!error) - error = vfs_load_quota_inode(dentry->d_inode, type, format_id, + error = dquot_load_quota_inode(d_inode(dentry), type, format_id, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); -out: dput(dentry); return error; } EXPORT_SYMBOL(dquot_quota_on_mount); -static inline qsize_t qbtos(qsize_t blocks) +static int dquot_quota_enable(struct super_block *sb, unsigned int flags) { - return blocks << QIF_DQBLKSIZE_BITS; + int ret; + int type; + struct quota_info *dqopt = sb_dqopt(sb); + + if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) + return -ENOSYS; + /* Accounting cannot be turned on while fs is mounted */ + flags &= ~(FS_QUOTA_UDQ_ACCT | FS_QUOTA_GDQ_ACCT | FS_QUOTA_PDQ_ACCT); + if (!flags) + return -EINVAL; + for (type = 0; type < MAXQUOTAS; type++) { + if (!(flags & qtype_enforce_flag(type))) + continue; + /* Can't enforce without accounting */ + if (!sb_has_quota_usage_enabled(sb, type)) { + ret = -EINVAL; + goto out_err; + } + if (sb_has_quota_limits_enabled(sb, type)) { + /* compatible with XFS */ + ret = -EEXIST; + goto out_err; + } + spin_lock(&dq_state_lock); + dqopt->flags |= dquot_state_flag(DQUOT_LIMITS_ENABLED, type); + spin_unlock(&dq_state_lock); + } + return 0; +out_err: + /* Backout enforcement enablement we already did */ + for (type--; type >= 0; type--) { + if (flags & qtype_enforce_flag(type)) + dquot_disable(sb, type, DQUOT_LIMITS_ENABLED); + } + return ret; } -static inline qsize_t stoqb(qsize_t space) +static int dquot_quota_disable(struct super_block *sb, unsigned int flags) { - return (space + QIF_DQBLKSIZE - 1) >> QIF_DQBLKSIZE_BITS; + int ret; + int type; + struct quota_info *dqopt = sb_dqopt(sb); + + if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) + return -ENOSYS; + /* + * We don't support turning off accounting via quotactl. In principle + * quota infrastructure can do this but filesystems don't expect + * userspace to be able to do it. + */ + if (flags & + (FS_QUOTA_UDQ_ACCT | FS_QUOTA_GDQ_ACCT | FS_QUOTA_PDQ_ACCT)) + return -EOPNOTSUPP; + + /* Filter out limits not enabled */ + for (type = 0; type < MAXQUOTAS; type++) + if (!sb_has_quota_limits_enabled(sb, type)) + flags &= ~qtype_enforce_flag(type); + /* Nothing left? */ + if (!flags) + return -EEXIST; + for (type = 0; type < MAXQUOTAS; type++) { + if (flags & qtype_enforce_flag(type)) { + ret = dquot_disable(sb, type, DQUOT_LIMITS_ENABLED); + if (ret < 0) + goto out_err; + } + } + return 0; +out_err: + /* Backout enforcement disabling we already did */ + for (type--; type >= 0; type--) { + if (flags & qtype_enforce_flag(type)) { + spin_lock(&dq_state_lock); + dqopt->flags |= + dquot_state_flag(DQUOT_LIMITS_ENABLED, type); + spin_unlock(&dq_state_lock); + } + } + return ret; } /* Generic routine for getting common part of quota structure */ -static void do_get_dqblk(struct dquot *dquot, struct fs_disk_quota *di) +static void do_get_dqblk(struct dquot *dquot, struct qc_dqblk *di) { struct mem_dqblk *dm = &dquot->dq_dqb; memset(di, 0, sizeof(*di)); - di->d_version = FS_DQUOT_VERSION; - di->d_flags = dquot->dq_id.type == USRQUOTA ? - FS_USER_QUOTA : FS_GROUP_QUOTA; - di->d_id = from_kqid_munged(current_user_ns(), dquot->dq_id); - - spin_lock(&dq_data_lock); - di->d_blk_hardlimit = stoqb(dm->dqb_bhardlimit); - di->d_blk_softlimit = stoqb(dm->dqb_bsoftlimit); + spin_lock(&dquot->dq_dqb_lock); + di->d_spc_hardlimit = dm->dqb_bhardlimit; + di->d_spc_softlimit = dm->dqb_bsoftlimit; di->d_ino_hardlimit = dm->dqb_ihardlimit; di->d_ino_softlimit = dm->dqb_isoftlimit; - di->d_bcount = dm->dqb_curspace + dm->dqb_rsvspace; - di->d_icount = dm->dqb_curinodes; - di->d_btimer = dm->dqb_btime; - di->d_itimer = dm->dqb_itime; - spin_unlock(&dq_data_lock); + di->d_space = dm->dqb_curspace + dm->dqb_rsvspace; + di->d_ino_count = dm->dqb_curinodes; + di->d_spc_timer = dm->dqb_btime; + di->d_ino_timer = dm->dqb_itime; + spin_unlock(&dquot->dq_dqb_lock); } int dquot_get_dqblk(struct super_block *sb, struct kqid qid, - struct fs_disk_quota *di) + struct qc_dqblk *di) { struct dquot *dquot; dquot = dqget(sb, qid); - if (!dquot) - return -ESRCH; + if (IS_ERR(dquot)) + return PTR_ERR(dquot); do_get_dqblk(dquot, di); dqput(dquot); @@ -2396,112 +2694,135 @@ int dquot_get_dqblk(struct super_block *sb, struct kqid qid, } EXPORT_SYMBOL(dquot_get_dqblk); -#define VFS_FS_DQ_MASK \ - (FS_DQ_BCOUNT | FS_DQ_BSOFT | FS_DQ_BHARD | \ - FS_DQ_ICOUNT | FS_DQ_ISOFT | FS_DQ_IHARD | \ - FS_DQ_BTIMER | FS_DQ_ITIMER) +int dquot_get_next_dqblk(struct super_block *sb, struct kqid *qid, + struct qc_dqblk *di) +{ + struct dquot *dquot; + int err; + + if (!sb->dq_op->get_next_id) + return -ENOSYS; + err = sb->dq_op->get_next_id(sb, qid); + if (err < 0) + return err; + dquot = dqget(sb, *qid); + if (IS_ERR(dquot)) + return PTR_ERR(dquot); + do_get_dqblk(dquot, di); + dqput(dquot); + + return 0; +} +EXPORT_SYMBOL(dquot_get_next_dqblk); + +#define VFS_QC_MASK \ + (QC_SPACE | QC_SPC_SOFT | QC_SPC_HARD | \ + QC_INO_COUNT | QC_INO_SOFT | QC_INO_HARD | \ + QC_SPC_TIMER | QC_INO_TIMER) /* Generic routine for setting common part of quota structure */ -static int do_set_dqblk(struct dquot *dquot, struct fs_disk_quota *di) +static int do_set_dqblk(struct dquot *dquot, struct qc_dqblk *di) { struct mem_dqblk *dm = &dquot->dq_dqb; int check_blim = 0, check_ilim = 0; struct mem_dqinfo *dqi = &sb_dqopt(dquot->dq_sb)->info[dquot->dq_id.type]; + int ret; - if (di->d_fieldmask & ~VFS_FS_DQ_MASK) + if (di->d_fieldmask & ~VFS_QC_MASK) return -EINVAL; - if (((di->d_fieldmask & FS_DQ_BSOFT) && - (di->d_blk_softlimit > dqi->dqi_maxblimit)) || - ((di->d_fieldmask & FS_DQ_BHARD) && - (di->d_blk_hardlimit > dqi->dqi_maxblimit)) || - ((di->d_fieldmask & FS_DQ_ISOFT) && - (di->d_ino_softlimit > dqi->dqi_maxilimit)) || - ((di->d_fieldmask & FS_DQ_IHARD) && - (di->d_ino_hardlimit > dqi->dqi_maxilimit))) + if (((di->d_fieldmask & QC_SPC_SOFT) && + di->d_spc_softlimit > dqi->dqi_max_spc_limit) || + ((di->d_fieldmask & QC_SPC_HARD) && + di->d_spc_hardlimit > dqi->dqi_max_spc_limit) || + ((di->d_fieldmask & QC_INO_SOFT) && + (di->d_ino_softlimit > dqi->dqi_max_ino_limit)) || + ((di->d_fieldmask & QC_INO_HARD) && + (di->d_ino_hardlimit > dqi->dqi_max_ino_limit))) return -ERANGE; - spin_lock(&dq_data_lock); - if (di->d_fieldmask & FS_DQ_BCOUNT) { - dm->dqb_curspace = di->d_bcount - dm->dqb_rsvspace; + spin_lock(&dquot->dq_dqb_lock); + if (di->d_fieldmask & QC_SPACE) { + dm->dqb_curspace = di->d_space - dm->dqb_rsvspace; check_blim = 1; set_bit(DQ_LASTSET_B + QIF_SPACE_B, &dquot->dq_flags); } - if (di->d_fieldmask & FS_DQ_BSOFT) - dm->dqb_bsoftlimit = qbtos(di->d_blk_softlimit); - if (di->d_fieldmask & FS_DQ_BHARD) - dm->dqb_bhardlimit = qbtos(di->d_blk_hardlimit); - if (di->d_fieldmask & (FS_DQ_BSOFT | FS_DQ_BHARD)) { + if (di->d_fieldmask & QC_SPC_SOFT) + dm->dqb_bsoftlimit = di->d_spc_softlimit; + if (di->d_fieldmask & QC_SPC_HARD) + dm->dqb_bhardlimit = di->d_spc_hardlimit; + if (di->d_fieldmask & (QC_SPC_SOFT | QC_SPC_HARD)) { check_blim = 1; set_bit(DQ_LASTSET_B + QIF_BLIMITS_B, &dquot->dq_flags); } - if (di->d_fieldmask & FS_DQ_ICOUNT) { - dm->dqb_curinodes = di->d_icount; + if (di->d_fieldmask & QC_INO_COUNT) { + dm->dqb_curinodes = di->d_ino_count; check_ilim = 1; set_bit(DQ_LASTSET_B + QIF_INODES_B, &dquot->dq_flags); } - if (di->d_fieldmask & FS_DQ_ISOFT) + if (di->d_fieldmask & QC_INO_SOFT) dm->dqb_isoftlimit = di->d_ino_softlimit; - if (di->d_fieldmask & FS_DQ_IHARD) + if (di->d_fieldmask & QC_INO_HARD) dm->dqb_ihardlimit = di->d_ino_hardlimit; - if (di->d_fieldmask & (FS_DQ_ISOFT | FS_DQ_IHARD)) { + if (di->d_fieldmask & (QC_INO_SOFT | QC_INO_HARD)) { check_ilim = 1; set_bit(DQ_LASTSET_B + QIF_ILIMITS_B, &dquot->dq_flags); } - if (di->d_fieldmask & FS_DQ_BTIMER) { - dm->dqb_btime = di->d_btimer; + if (di->d_fieldmask & QC_SPC_TIMER) { + dm->dqb_btime = di->d_spc_timer; check_blim = 1; set_bit(DQ_LASTSET_B + QIF_BTIME_B, &dquot->dq_flags); } - if (di->d_fieldmask & FS_DQ_ITIMER) { - dm->dqb_itime = di->d_itimer; + if (di->d_fieldmask & QC_INO_TIMER) { + dm->dqb_itime = di->d_ino_timer; check_ilim = 1; set_bit(DQ_LASTSET_B + QIF_ITIME_B, &dquot->dq_flags); } if (check_blim) { if (!dm->dqb_bsoftlimit || - dm->dqb_curspace < dm->dqb_bsoftlimit) { + dm->dqb_curspace + dm->dqb_rsvspace <= dm->dqb_bsoftlimit) { dm->dqb_btime = 0; clear_bit(DQ_BLKS_B, &dquot->dq_flags); - } else if (!(di->d_fieldmask & FS_DQ_BTIMER)) + } else if (!(di->d_fieldmask & QC_SPC_TIMER)) /* Set grace only if user hasn't provided his own... */ - dm->dqb_btime = get_seconds() + dqi->dqi_bgrace; + dm->dqb_btime = ktime_get_real_seconds() + dqi->dqi_bgrace; } if (check_ilim) { if (!dm->dqb_isoftlimit || - dm->dqb_curinodes < dm->dqb_isoftlimit) { + dm->dqb_curinodes <= dm->dqb_isoftlimit) { dm->dqb_itime = 0; clear_bit(DQ_INODES_B, &dquot->dq_flags); - } else if (!(di->d_fieldmask & FS_DQ_ITIMER)) + } else if (!(di->d_fieldmask & QC_INO_TIMER)) /* Set grace only if user hasn't provided his own... */ - dm->dqb_itime = get_seconds() + dqi->dqi_igrace; + dm->dqb_itime = ktime_get_real_seconds() + dqi->dqi_igrace; } if (dm->dqb_bhardlimit || dm->dqb_bsoftlimit || dm->dqb_ihardlimit || dm->dqb_isoftlimit) clear_bit(DQ_FAKE_B, &dquot->dq_flags); else set_bit(DQ_FAKE_B, &dquot->dq_flags); - spin_unlock(&dq_data_lock); - mark_dquot_dirty(dquot); - + spin_unlock(&dquot->dq_dqb_lock); + ret = mark_dquot_dirty(dquot); + if (ret < 0) + return ret; return 0; } int dquot_set_dqblk(struct super_block *sb, struct kqid qid, - struct fs_disk_quota *di) + struct qc_dqblk *di) { struct dquot *dquot; int rc; dquot = dqget(sb, qid); - if (!dquot) { - rc = -ESRCH; + if (IS_ERR(dquot)) { + rc = PTR_ERR(dquot); goto out; } rc = do_set_dqblk(dquot, di); @@ -2512,133 +2833,156 @@ out: EXPORT_SYMBOL(dquot_set_dqblk); /* Generic routine for getting common part of quota file information */ -int dquot_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii) +int dquot_get_state(struct super_block *sb, struct qc_state *state) { struct mem_dqinfo *mi; - - mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); - if (!sb_has_quota_active(sb, type)) { - mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); - return -ESRCH; + struct qc_type_state *tstate; + struct quota_info *dqopt = sb_dqopt(sb); + int type; + + memset(state, 0, sizeof(*state)); + for (type = 0; type < MAXQUOTAS; type++) { + if (!sb_has_quota_active(sb, type)) + continue; + tstate = state->s_state + type; + mi = sb_dqopt(sb)->info + type; + tstate->flags = QCI_ACCT_ENABLED; + spin_lock(&dq_data_lock); + if (mi->dqi_flags & DQF_SYS_FILE) + tstate->flags |= QCI_SYSFILE; + if (mi->dqi_flags & DQF_ROOT_SQUASH) + tstate->flags |= QCI_ROOT_SQUASH; + if (sb_has_quota_limits_enabled(sb, type)) + tstate->flags |= QCI_LIMITS_ENFORCED; + tstate->spc_timelimit = mi->dqi_bgrace; + tstate->ino_timelimit = mi->dqi_igrace; + if (dqopt->files[type]) { + tstate->ino = dqopt->files[type]->i_ino; + tstate->blocks = dqopt->files[type]->i_blocks; + } + tstate->nextents = 1; /* We don't know... */ + spin_unlock(&dq_data_lock); } - mi = sb_dqopt(sb)->info + type; - spin_lock(&dq_data_lock); - ii->dqi_bgrace = mi->dqi_bgrace; - ii->dqi_igrace = mi->dqi_igrace; - ii->dqi_flags = mi->dqi_flags & DQF_GETINFO_MASK; - ii->dqi_valid = IIF_ALL; - spin_unlock(&dq_data_lock); - mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); return 0; } -EXPORT_SYMBOL(dquot_get_dqinfo); +EXPORT_SYMBOL(dquot_get_state); /* Generic routine for setting common part of quota file information */ -int dquot_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii) +int dquot_set_dqinfo(struct super_block *sb, int type, struct qc_info *ii) { struct mem_dqinfo *mi; - int err = 0; - mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); - if (!sb_has_quota_active(sb, type)) { - err = -ESRCH; - goto out; - } + if ((ii->i_fieldmask & QC_WARNS_MASK) || + (ii->i_fieldmask & QC_RT_SPC_TIMER)) + return -EINVAL; + if (!sb_has_quota_active(sb, type)) + return -ESRCH; mi = sb_dqopt(sb)->info + type; + if (ii->i_fieldmask & QC_FLAGS) { + if ((ii->i_flags & QCI_ROOT_SQUASH && + mi->dqi_format->qf_fmt_id != QFMT_VFS_OLD)) + return -EINVAL; + } spin_lock(&dq_data_lock); - if (ii->dqi_valid & IIF_BGRACE) - mi->dqi_bgrace = ii->dqi_bgrace; - if (ii->dqi_valid & IIF_IGRACE) - mi->dqi_igrace = ii->dqi_igrace; - if (ii->dqi_valid & IIF_FLAGS) - mi->dqi_flags = (mi->dqi_flags & ~DQF_SETINFO_MASK) | - (ii->dqi_flags & DQF_SETINFO_MASK); + if (ii->i_fieldmask & QC_SPC_TIMER) + mi->dqi_bgrace = ii->i_spc_timelimit; + if (ii->i_fieldmask & QC_INO_TIMER) + mi->dqi_igrace = ii->i_ino_timelimit; + if (ii->i_fieldmask & QC_FLAGS) { + if (ii->i_flags & QCI_ROOT_SQUASH) + mi->dqi_flags |= DQF_ROOT_SQUASH; + else + mi->dqi_flags &= ~DQF_ROOT_SQUASH; + } spin_unlock(&dq_data_lock); mark_info_dirty(sb, type); /* Force write to disk */ - sb->dq_op->write_info(sb, type); -out: - mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); - return err; + return sb->dq_op->write_info(sb, type); } EXPORT_SYMBOL(dquot_set_dqinfo); -const struct quotactl_ops dquot_quotactl_ops = { - .quota_on = dquot_quota_on, - .quota_off = dquot_quota_off, +const struct quotactl_ops dquot_quotactl_sysfile_ops = { + .quota_enable = dquot_quota_enable, + .quota_disable = dquot_quota_disable, .quota_sync = dquot_quota_sync, - .get_info = dquot_get_dqinfo, + .get_state = dquot_get_state, .set_info = dquot_set_dqinfo, .get_dqblk = dquot_get_dqblk, + .get_nextdqblk = dquot_get_next_dqblk, .set_dqblk = dquot_set_dqblk }; -EXPORT_SYMBOL(dquot_quotactl_ops); +EXPORT_SYMBOL(dquot_quotactl_sysfile_ops); -static int do_proc_dqstats(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, loff_t *ppos) +static int do_proc_dqstats(const struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) { - unsigned int type = (int *)table->data - dqstats.stat; + unsigned int type = (unsigned long *)table->data - dqstats.stat; + s64 value = percpu_counter_sum(&dqstats.counter[type]); + + /* Filter negative values for non-monotonic counters */ + if (value < 0 && (type == DQST_ALLOC_DQUOTS || + type == DQST_FREE_DQUOTS)) + value = 0; /* Update global table */ - dqstats.stat[type] = - percpu_counter_sum_positive(&dqstats.counter[type]); - return proc_dointvec(table, write, buffer, lenp, ppos); + dqstats.stat[type] = value; + return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); } -static struct ctl_table fs_dqstats_table[] = { +static const struct ctl_table fs_dqstats_table[] = { { .procname = "lookups", .data = &dqstats.stat[DQST_LOOKUPS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "drops", .data = &dqstats.stat[DQST_DROPS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "reads", .data = &dqstats.stat[DQST_READS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "writes", .data = &dqstats.stat[DQST_WRITES], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "cache_hits", .data = &dqstats.stat[DQST_CACHE_HITS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "allocated_dquots", .data = &dqstats.stat[DQST_ALLOC_DQUOTS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "free_dquots", .data = &dqstats.stat[DQST_FREE_DQUOTS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, { .procname = "syncs", .data = &dqstats.stat[DQST_SYNCS], - .maxlen = sizeof(int), + .maxlen = sizeof(unsigned long), .mode = 0444, .proc_handler = do_proc_dqstats, }, @@ -2651,71 +2995,60 @@ static struct ctl_table fs_dqstats_table[] = { .proc_handler = proc_dointvec, }, #endif - { }, -}; - -static struct ctl_table fs_table[] = { - { - .procname = "quota", - .mode = 0555, - .child = fs_dqstats_table, - }, - { }, -}; - -static struct ctl_table sys_table[] = { - { - .procname = "fs", - .mode = 0555, - .child = fs_table, - }, - { }, }; static int __init dquot_init(void) { int i, ret; unsigned long nr_hash, order; + struct shrinker *dqcache_shrinker; printk(KERN_NOTICE "VFS: Disk quotas %s\n", __DQUOT_VERSION__); - register_sysctl_table(sys_table); + register_sysctl_init("fs/quota", fs_dqstats_table); dquot_cachep = kmem_cache_create("dquot", sizeof(struct dquot), sizeof(unsigned long) * 4, (SLAB_HWCACHE_ALIGN|SLAB_RECLAIM_ACCOUNT| - SLAB_MEM_SPREAD|SLAB_PANIC), + SLAB_PANIC), NULL); order = 0; - dquot_hash = (struct hlist_head *)__get_free_pages(GFP_ATOMIC, order); + dquot_hash = (struct hlist_head *)__get_free_pages(GFP_KERNEL, order); if (!dquot_hash) panic("Cannot create dquot hash table"); - for (i = 0; i < _DQST_DQSTAT_LAST; i++) { - ret = percpu_counter_init(&dqstats.counter[i], 0); - if (ret) - panic("Cannot create dquot stat counters"); - } + ret = percpu_counter_init_many(dqstats.counter, 0, GFP_KERNEL, + _DQST_DQSTAT_LAST); + if (ret) + panic("Cannot create dquot stat counters"); /* Find power-of-two hlist_heads which can fit into allocation */ nr_hash = (1UL << order) * PAGE_SIZE / sizeof(struct hlist_head); - dq_hash_bits = 0; - do { - dq_hash_bits++; - } while (nr_hash >> dq_hash_bits); - dq_hash_bits--; + dq_hash_bits = ilog2(nr_hash); nr_hash = 1UL << dq_hash_bits; dq_hash_mask = nr_hash - 1; for (i = 0; i < nr_hash; i++) INIT_HLIST_HEAD(dquot_hash + i); - printk("Dquot-cache hash table entries: %ld (order %ld, %ld bytes)\n", - nr_hash, order, (PAGE_SIZE << order)); + pr_info("VFS: Dquot-cache hash table entries: %ld (order %ld," + " %ld bytes)\n", nr_hash, order, (PAGE_SIZE << order)); + + dqcache_shrinker = shrinker_alloc(0, "dquota-cache"); + if (!dqcache_shrinker) + panic("Cannot allocate dquot shrinker"); + + dqcache_shrinker->count_objects = dqcache_shrink_count; + dqcache_shrinker->scan_objects = dqcache_shrink_scan; + + shrinker_register(dqcache_shrinker); - register_shrinker(&dqcache_shrinker); + quota_unbound_wq = alloc_workqueue("quota_events_unbound", + WQ_UNBOUND | WQ_MEM_RECLAIM, WQ_MAX_ACTIVE); + if (!quota_unbound_wq) + panic("Cannot create quota_unbound_wq\n"); return 0; } -module_init(dquot_init); +fs_initcall(dquot_init); |