diff options
Diffstat (limited to 'fs/quota')
| -rw-r--r-- | fs/quota/Kconfig | 59 | ||||
| -rw-r--r-- | fs/quota/Makefile | 14 | ||||
| -rw-r--r-- | fs/quota/dquot.c | 2564 | ||||
| -rw-r--r-- | fs/quota/quota.c | 513 | ||||
| -rw-r--r-- | fs/quota/quota_tree.c | 645 | ||||
| -rw-r--r-- | fs/quota/quota_tree.h | 25 | ||||
| -rw-r--r-- | fs/quota/quota_v1.c | 218 | ||||
| -rw-r--r-- | fs/quota/quota_v2.c | 236 | ||||
| -rw-r--r-- | fs/quota/quotaio_v1.h | 33 | ||||
| -rw-r--r-- | fs/quota/quotaio_v2.h | 60 |
10 files changed, 4367 insertions, 0 deletions
diff --git a/fs/quota/Kconfig b/fs/quota/Kconfig new file mode 100644 index 000000000000..d9750d86fde2 --- /dev/null +++ b/fs/quota/Kconfig @@ -0,0 +1,59 @@ +# +# Quota configuration +# + +config QUOTA + bool "Quota support" + help + If you say Y here, you will be able to set per user limits for disk + usage (also called disk quotas). Currently, it works for the + ext2, ext3, and reiserfs file system. ext3 also supports journalled + quotas for which you don't need to run quotacheck(8) after an unclean + shutdown. + For further details, read the Quota mini-HOWTO, available from + <http://www.tldp.org/docs.html#howto>, or the documentation provided + with the quota tools. Probably the quota support is only useful for + multi user systems. If unsure, say N. + +config QUOTA_NETLINK_INTERFACE + bool "Report quota messages through netlink interface" + depends on QUOTA && NET + help + If you say Y here, quota warnings (about exceeding softlimit, reaching + hardlimit, etc.) will be reported through netlink interface. If unsure, + say Y. + +config PRINT_QUOTA_WARNING + bool "Print quota warnings to console (OBSOLETE)" + depends on QUOTA + default y + help + If you say Y here, quota warnings (about exceeding softlimit, reaching + hardlimit, etc.) will be printed to the process' controlling terminal. + Note that this behavior is currently deprecated and may go away in + future. Please use notification via netlink socket instead. + +# Generic support for tree structured quota files. Seleted when needed. +config QUOTA_TREE + tristate + +config QFMT_V1 + tristate "Old quota format support" + depends on QUOTA + help + This quota format was (is) used by kernels earlier than 2.4.22. If + you have quota working and you don't want to convert to new quota + format say Y here. + +config QFMT_V2 + tristate "Quota format v2 support" + depends on QUOTA + select QUOTA_TREE + help + This quota format allows using quotas with 32-bit UIDs/GIDs. If you + need this functionality say Y here. + +config QUOTACTL + bool + depends on XFS_QUOTA || QUOTA + default y diff --git a/fs/quota/Makefile b/fs/quota/Makefile new file mode 100644 index 000000000000..385a0831cc99 --- /dev/null +++ b/fs/quota/Makefile @@ -0,0 +1,14 @@ +# +# Makefile for the Linux filesystems. +# +# 14 Sep 2000, Christoph Hellwig <hch@infradead.org> +# Rewritten to use lists instead of if-statements. +# + +obj-y := + +obj-$(CONFIG_QUOTA) += dquot.o +obj-$(CONFIG_QFMT_V1) += quota_v1.o +obj-$(CONFIG_QFMT_V2) += quota_v2.o +obj-$(CONFIG_QUOTA_TREE) += quota_tree.o +obj-$(CONFIG_QUOTACTL) += quota.o diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c new file mode 100644 index 000000000000..28aa14667602 --- /dev/null +++ b/fs/quota/dquot.c @@ -0,0 +1,2564 @@ +/* + * Implementation of the diskquota system for the LINUX operating system. QUOTA + * is implemented using the BSD system call interface as the means of + * communication with the user level. This file contains the generic routines + * called by the different filesystems on allocation of an inode or block. + * These routines take care of the administration needed to have a consistent + * diskquota tracking system. The ideas of both user and group quotas are based + * 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 + * + * Revised list management to avoid races + * -- Bill Hawes, <whawes@star.net>, 9/98 + * + * Fixed races in dquot_transfer(), dqget() and dquot_alloc_...(). + * As the consequence the locking was moved from dquot_decr_...(), + * dquot_incr_...() to calling functions. + * invalidate_dquots() now writes modified dquots. + * Serialized quota_off() and quota_on() for mount point. + * Fixed a few bugs in grow_dquots(). + * Fixed deadlock in write_dquot() - we no longer account quotas on + * quota files + * remove_dquot_ref() moved to inode.c - it now traverses through inodes + * add_dquot_ref() restarts after blocking + * Added check for bogus uid and fixed check for group in quotactl. + * Jan Kara, <jack@suse.cz>, sponsored by SuSE CR, 10-11/99 + * + * Used struct list_head instead of own list struct + * Invalidation of referenced dquots is no longer possible + * Improved free_dquots list management + * Quota and i_blocks are now updated in one place to avoid races + * Warnings are now delayed so we won't block in critical section + * Write updated not to require dquot lock + * Jan Kara, <jack@suse.cz>, 9/2000 + * + * Added dynamic quota structure allocation + * Jan Kara <jack@suse.cz> 12/2000 + * + * Rewritten quota interface. Implemented new quota format and + * formats registering. + * Jan Kara, <jack@suse.cz>, 2001,2002 + * + * New SMP locking. + * Jan Kara, <jack@suse.cz>, 10/2002 + * + * Added journalled quota support, fix lock inversion problems + * Jan Kara, <jack@suse.cz>, 2003,2004 + * + * (C) Copyright 1994 - 1997 Marco van Wieringen + */ + +#include <linux/errno.h> +#include <linux/kernel.h> +#include <linux/fs.h> +#include <linux/mount.h> +#include <linux/mm.h> +#include <linux/time.h> +#include <linux/types.h> +#include <linux/string.h> +#include <linux/fcntl.h> +#include <linux/stat.h> +#include <linux/tty.h> +#include <linux/file.h> +#include <linux/slab.h> +#include <linux/sysctl.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/proc_fs.h> +#include <linux/security.h> +#include <linux/kmod.h> +#include <linux/namei.h> +#include <linux/buffer_head.h> +#include <linux/capability.h> +#include <linux/quotaops.h> +#include <linux/writeback.h> /* for inode_lock, oddly enough.. */ +#ifdef CONFIG_QUOTA_NETLINK_INTERFACE +#include <net/netlink.h> +#include <net/genetlink.h> +#endif + +#include <asm/uaccess.h> + +#define __DQUOT_PARANOIA + +/* + * There are three quota SMP locks. dq_list_lock protects all lists with quotas + * and quota formats, dqstats structure containing statistics about the lists + * 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. + * + * The spinlock ordering is hence: dq_data_lock > dq_list_lock > i_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 + * (these locking rules also apply for S_NOQUOTA flag in the inode - note that + * for altering the flag i_mutex is also needed). + * + * 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. + * + * Lock ordering (including related VFS locks) is the following: + * i_mutex > dqonoff_sem > journal_lock > dqptr_sem > dquot->dq_lock > + * dqio_mutex + * 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. + * i_mutex on quota files is special (it's below dqio_mutex) + */ + +static DEFINE_SPINLOCK(dq_list_lock); +static DEFINE_SPINLOCK(dq_state_lock); +DEFINE_SPINLOCK(dq_data_lock); +EXPORT_SYMBOL(dq_data_lock); + +static char *quotatypes[] = INITQFNAMES; +static struct quota_format_type *quota_formats; /* List of registered formats */ +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) +{ + 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); + +void unregister_quota_format(struct quota_format_type *fmt) +{ + struct quota_format_type **actqf; + + spin_lock(&dq_list_lock); + for (actqf = "a_formats; *actqf && *actqf != fmt; actqf = &(*actqf)->qf_next); + if (*actqf) + *actqf = (*actqf)->qf_next; + spin_unlock(&dq_list_lock); +} +EXPORT_SYMBOL(unregister_quota_format); + +static struct quota_format_type *find_quota_format(int id) +{ + struct quota_format_type *actqf; + + spin_lock(&dq_list_lock); + for (actqf = quota_formats; actqf && actqf->qf_fmt_id != id; actqf = actqf->qf_next); + if (!actqf || !try_module_get(actqf->qf_owner)) { + int qm; + + spin_unlock(&dq_list_lock); + + for (qm = 0; module_names[qm].qm_fmt_id && module_names[qm].qm_fmt_id != id; qm++); + if (!module_names[qm].qm_fmt_id || request_module(module_names[qm].qm_mod_name)) + return NULL; + + spin_lock(&dq_list_lock); + for (actqf = quota_formats; actqf && actqf->qf_fmt_id != id; actqf = actqf->qf_next); + if (actqf && !try_module_get(actqf->qf_owner)) + actqf = NULL; + } + spin_unlock(&dq_list_lock); + return actqf; +} + +static void put_quota_format(struct quota_format_type *fmt) +{ + module_put(fmt->qf_owner); +} + +/* + * 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. + * + * 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. + * + * Dquots with a specific identity (device, type and id) are placed on + * one of the dquot_hash[] hash chains. The provides an efficient search + * mechanism to locate a specific dquot. + */ + +static LIST_HEAD(inuse_list); +static LIST_HEAD(free_dquots); +static unsigned int dq_hash_bits, dq_hash_mask; +static struct hlist_head *dquot_hash; + +struct dqstats dqstats; +EXPORT_SYMBOL(dqstats); + +static inline unsigned int +hashfn(const struct super_block *sb, unsigned int id, int type) +{ + unsigned long tmp; + + tmp = (((unsigned long)sb>>L1_CACHE_SHIFT) ^ id) * (MAXQUOTAS - type); + return (tmp + (tmp >> dq_hash_bits)) & dq_hash_mask; +} + +/* + * Following list functions expect dq_list_lock to be held + */ +static inline void insert_dquot_hash(struct dquot *dquot) +{ + struct hlist_head *head = dquot_hash + hashfn(dquot->dq_sb, dquot->dq_id, dquot->dq_type); + hlist_add_head(&dquot->dq_hash, head); +} + +static inline void remove_dquot_hash(struct dquot *dquot) +{ + hlist_del_init(&dquot->dq_hash); +} + +static inline struct dquot *find_dquot(unsigned int hashent, struct super_block *sb, unsigned int id, int type) +{ + struct hlist_node *node; + struct dquot *dquot; + + hlist_for_each (node, dquot_hash+hashent) { + dquot = hlist_entry(node, struct dquot, dq_hash); + if (dquot->dq_sb == sb && dquot->dq_id == id && dquot->dq_type == type) + return dquot; + } + return NODQUOT; +} + +/* Add a dquot to the tail of the free list */ +static inline void put_dquot_last(struct dquot *dquot) +{ + list_add_tail(&dquot->dq_free, &free_dquots); + dqstats.free_dquots++; +} + +static inline void remove_free_dquot(struct dquot *dquot) +{ + if (list_empty(&dquot->dq_free)) + return; + list_del_init(&dquot->dq_free); + dqstats.free_dquots--; +} + +static inline void put_inuse(struct dquot *dquot) +{ + /* We add to the back of inuse list so we don't have to restart + * when traversing this list and we block */ + list_add_tail(&dquot->dq_inuse, &inuse_list); + dqstats.allocated_dquots++; +} + +static inline void remove_inuse(struct dquot *dquot) +{ + dqstats.allocated_dquots--; + list_del(&dquot->dq_inuse); +} +/* + * End of list functions needing dq_list_lock + */ + +static void wait_on_dquot(struct dquot *dquot) +{ + mutex_lock(&dquot->dq_lock); + mutex_unlock(&dquot->dq_lock); +} + +static inline int dquot_dirty(struct dquot *dquot) +{ + return test_bit(DQ_MOD_B, &dquot->dq_flags); +} + +static inline int mark_dquot_dirty(struct dquot *dquot) +{ + return dquot->dq_sb->dq_op->mark_dirty(dquot); +} + +int dquot_mark_dquot_dirty(struct dquot *dquot) +{ + spin_lock(&dq_list_lock); + if (!test_and_set_bit(DQ_MOD_B, &dquot->dq_flags)) + list_add(&dquot->dq_dirty, &sb_dqopt(dquot->dq_sb)-> + info[dquot->dq_type].dqi_dirty_list); + spin_unlock(&dq_list_lock); + return 0; +} +EXPORT_SYMBOL(dquot_mark_dquot_dirty); + +/* 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)) + return 0; + list_del_init(&dquot->dq_dirty); + 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); +} +EXPORT_SYMBOL(mark_info_dirty); + +/* + * Read dquot from disk and alloc space for it + */ + +int dquot_acquire(struct dquot *dquot) +{ + int ret = 0, ret2 = 0; + 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)) + ret = dqopt->ops[dquot->dq_type]->read_dqblk(dquot); + if (ret < 0) + goto out_iolock; + set_bit(DQ_READ_B, &dquot->dq_flags); + /* Instantiate dquot if needed */ + if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && !dquot->dq_off) { + ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot); + /* Write the info if needed */ + if (info_dirty(&dqopt->info[dquot->dq_type])) + ret2 = dqopt->ops[dquot->dq_type]->write_file_info(dquot->dq_sb, dquot->dq_type); + if (ret < 0) + goto out_iolock; + if (ret2 < 0) { + ret = ret2; + goto out_iolock; + } + } + set_bit(DQ_ACTIVE_B, &dquot->dq_flags); +out_iolock: + mutex_unlock(&dqopt->dqio_mutex); + mutex_unlock(&dquot->dq_lock); + return ret; +} +EXPORT_SYMBOL(dquot_acquire); + +/* + * Write dquot to disk + */ +int dquot_commit(struct dquot *dquot) +{ + int ret = 0, ret2 = 0; + 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); + /* 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)) { + ret = dqopt->ops[dquot->dq_type]->commit_dqblk(dquot); + if (info_dirty(&dqopt->info[dquot->dq_type])) + ret2 = dqopt->ops[dquot->dq_type]->write_file_info(dquot->dq_sb, dquot->dq_type); + if (ret >= 0) + ret = ret2; + } +out_sem: + mutex_unlock(&dqopt->dqio_mutex); + return ret; +} +EXPORT_SYMBOL(dquot_commit); + +/* + * Release dquot + */ +int dquot_release(struct dquot *dquot) +{ + int ret = 0, ret2 = 0; + struct quota_info *dqopt = sb_dqopt(dquot->dq_sb); + + mutex_lock(&dquot->dq_lock); + /* Check whether we are not racing with some other dqget() */ + if (atomic_read(&dquot->dq_count) > 1) + goto out_dqlock; + mutex_lock(&dqopt->dqio_mutex); + if (dqopt->ops[dquot->dq_type]->release_dqblk) { + ret = dqopt->ops[dquot->dq_type]->release_dqblk(dquot); + /* Write the info */ + if (info_dirty(&dqopt->info[dquot->dq_type])) + ret2 = dqopt->ops[dquot->dq_type]->write_file_info(dquot->dq_sb, dquot->dq_type); + if (ret >= 0) + ret = ret2; + } + clear_bit(DQ_ACTIVE_B, &dquot->dq_flags); + mutex_unlock(&dqopt->dqio_mutex); +out_dqlock: + mutex_unlock(&dquot->dq_lock); + return ret; +} +EXPORT_SYMBOL(dquot_release); + +void dquot_destroy(struct dquot *dquot) +{ + kmem_cache_free(dquot_cachep, dquot); +} +EXPORT_SYMBOL(dquot_destroy); + +static inline void do_destroy_dquot(struct dquot *dquot) +{ + dquot->dq_sb->dq_op->destroy_dquot(dquot); +} + +/* Invalidate all dquots on the list. Note that this function is called after + * quota is disabled and pointers from inodes removed so there cannot be new + * quota users. There can still be some users of quotas due to inodes being + * just deleted or pruned by prune_icache() (those are not attached to any + * list) or parallel quotactl call. We have to wait for such users. + */ +static void invalidate_dquots(struct super_block *sb, int type) +{ + struct dquot *dquot, *tmp; + +restart: + spin_lock(&dq_list_lock); + list_for_each_entry_safe(dquot, tmp, &inuse_list, dq_inuse) { + if (dquot->dq_sb != sb) + continue; + if (dquot->dq_type != type) + 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 + * 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); + dqput(dquot); + /* At this moment dquot() need not exist (it could be + * reclaimed by prune_dqcache(). Hence we must + * restart. */ + goto restart; + } + /* + * Quota now has no users and it has been written on last + * dqput() + */ + remove_dquot_hash(dquot); + remove_free_dquot(dquot); + remove_inuse(dquot); + do_destroy_dquot(dquot); + } + spin_unlock(&dq_list_lock); +} + +/* Call callback for every active dquot on given filesystem */ +int dquot_scan_active(struct super_block *sb, + int (*fn)(struct dquot *dquot, unsigned long priv), + unsigned long priv) +{ + struct dquot *dquot, *old_dquot = NULL; + int ret = 0; + + mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); + spin_lock(&dq_list_lock); + list_for_each_entry(dquot, &inuse_list, dq_inuse) { + if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags)) + continue; + if (dquot->dq_sb != sb) + continue; + /* Now we have active dquot so we can just increase use count */ + atomic_inc(&dquot->dq_count); + dqstats.lookups++; + spin_unlock(&dq_list_lock); + dqput(old_dquot); + old_dquot = 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 */ + } + spin_unlock(&dq_list_lock); +out: + dqput(old_dquot); + mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); + return ret; +} +EXPORT_SYMBOL(dquot_scan_active); + +int vfs_quota_sync(struct super_block *sb, int type) +{ + struct list_head *dirty; + struct dquot *dquot; + struct quota_info *dqopt = sb_dqopt(sb); + int cnt; + + 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; + spin_lock(&dq_list_lock); + dirty = &dqopt->info[cnt].dqi_dirty_list; + 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); + 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); + dqstats.lookups++; + spin_unlock(&dq_list_lock); + sb->dq_op->write_dquot(dquot); + dqput(dquot); + spin_lock(&dq_list_lock); + } + spin_unlock(&dq_list_lock); + } + + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if ((cnt == type || type == -1) && sb_has_quota_active(sb, cnt) + && info_dirty(&dqopt->info[cnt])) + sb->dq_op->write_info(sb, cnt); + spin_lock(&dq_list_lock); + dqstats.syncs++; + spin_unlock(&dq_list_lock); + mutex_unlock(&dqopt->dqonoff_mutex); + + return 0; +} +EXPORT_SYMBOL(vfs_quota_sync); + +/* Free unused dquots from cache */ +static void prune_dqcache(int count) +{ + struct list_head *head; + struct dquot *dquot; + + head = free_dquots.prev; + while (head != &free_dquots && count) { + dquot = list_entry(head, struct dquot, dq_free); + remove_dquot_hash(dquot); + remove_free_dquot(dquot); + remove_inuse(dquot); + do_destroy_dquot(dquot); + count--; + head = free_dquots.prev; + } +} + +/* + * This is called from kswapd when we think we need some + * more memory + */ + +static int shrink_dqcache_memory(int nr, gfp_t gfp_mask) +{ + if (nr) { + spin_lock(&dq_list_lock); + prune_dqcache(nr); + spin_unlock(&dq_list_lock); + } + return (dqstats.free_dquots / 100) * sysctl_vfs_cache_pressure; +} + +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 __DQUOT_PARANOIA + if (!atomic_read(&dquot->dq_count)) { + printk("VFS: dqput: trying to free free dquot\n"); + printk("VFS: device %s, dquot of %s %d\n", + dquot->dq_sb->s_id, + quotatypes[dquot->dq_type], + dquot->dq_id); + BUG(); + } +#endif + + spin_lock(&dq_list_lock); + dqstats.drops++; + spin_unlock(&dq_list_lock); +we_slept: + spin_lock(&dq_list_lock); + if (atomic_read(&dquot->dq_count) > 1) { + /* We have more than one user... nothing to do */ + atomic_dec(&dquot->dq_count); + /* Releasing dquot during quotaoff phase? */ + if (!sb_has_quota_active(dquot->dq_sb, dquot->dq_type) && + atomic_read(&dquot->dq_count) == 1) + wake_up(&dquot->dq_wait_unused); + 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) { + printk(KERN_ERR "VFS: cannot write quota structure on " + "device %s (error %d). Quota may get out of " + "sync!\n", dquot->dq_sb->s_id, 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; + } + atomic_dec(&dquot->dq_count); +#ifdef __DQUOT_PARANOIA + /* sanity check */ + BUG_ON(!list_empty(&dquot->dq_free)); +#endif + put_dquot_last(dquot); + spin_unlock(&dq_list_lock); +} +EXPORT_SYMBOL(dqput); + +struct dquot *dquot_alloc(struct super_block *sb, int type) +{ + return kmem_cache_zalloc(dquot_cachep, GFP_NOFS); +} +EXPORT_SYMBOL(dquot_alloc); + +static struct dquot *get_empty_dquot(struct super_block *sb, int type) +{ + struct dquot *dquot; + + dquot = sb->dq_op->alloc_dquot(sb, type); + if(!dquot) + return NODQUOT; + + mutex_init(&dquot->dq_lock); + INIT_LIST_HEAD(&dquot->dq_free); + 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_type = type; + atomic_set(&dquot->dq_count, 1); + + return dquot; +} + +/* + * Get reference to dquot + * + * Locking is slightly tricky here. We are guarded from parallel quotaoff() + * destroying our dquot by: + * a) checking for quota flags under dq_list_lock and + * b) getting a reference to dquot before we release dq_list_lock + */ +struct dquot *dqget(struct super_block *sb, unsigned int id, int type) +{ + unsigned int hashent = hashfn(sb, id, type); + struct dquot *dquot = NODQUOT, *empty = NODQUOT; + + if (!sb_has_quota_active(sb, type)) + return NODQUOT; +we_slept: + spin_lock(&dq_list_lock); + spin_lock(&dq_state_lock); + if (!sb_has_quota_active(sb, type)) { + spin_unlock(&dq_state_lock); + spin_unlock(&dq_list_lock); + goto out; + } + spin_unlock(&dq_state_lock); + + if ((dquot = find_dquot(hashent, sb, id, type)) == NODQUOT) { + if (empty == NODQUOT) { + spin_unlock(&dq_list_lock); + if ((empty = get_empty_dquot(sb, type)) == NODQUOT) + schedule(); /* Try to wait for a moment... */ + goto we_slept; + } + dquot = empty; + empty = NODQUOT; + dquot->dq_id = id; + /* all dquots go on the inuse_list */ + put_inuse(dquot); + /* hash it first so it can be found */ + insert_dquot_hash(dquot); + dqstats.lookups++; + spin_unlock(&dq_list_lock); + } else { + if (!atomic_read(&dquot->dq_count)) + remove_free_dquot(dquot); + atomic_inc(&dquot->dq_count); + dqstats.cache_hits++; + dqstats.lookups++; + spin_unlock(&dq_list_lock); + } + /* 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 */ + wait_on_dquot(dquot); + /* Read the dquot and instantiate it (everything done only if needed) */ + if (!test_bit(DQ_ACTIVE_B, &dquot->dq_flags) && sb->dq_op->acquire_dquot(dquot) < 0) { + dqput(dquot); + dquot = NODQUOT; + goto out; + } +#ifdef __DQUOT_PARANOIA + BUG_ON(!dquot->dq_sb); /* Has somebody invalidated entry under us? */ +#endif +out: + if (empty) + do_destroy_dquot(empty); + + return dquot; +} +EXPORT_SYMBOL(dqget); + +static int dqinit_needed(struct inode *inode, int type) +{ + int cnt; + + if (IS_NOQUOTA(inode)) + return 0; + if (type != -1) + return inode->i_dquot[type] == NODQUOT; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt] == NODQUOT) + return 1; + return 0; +} + +/* This routine is guarded by dqonoff_mutex mutex */ +static void add_dquot_ref(struct super_block *sb, int type) +{ + struct inode *inode, *old_inode = NULL; + + spin_lock(&inode_lock); + list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { + if (!atomic_read(&inode->i_writecount)) + continue; + if (!dqinit_needed(inode, type)) + continue; + if (inode->i_state & (I_FREEING|I_WILL_FREE)) + continue; + + __iget(inode); + spin_unlock(&inode_lock); + + iput(old_inode); + sb->dq_op->initialize(inode, type); + /* We hold a reference to 'inode' so it couldn't have been + * removed from s_inodes list while we dropped the inode_lock. + * We cannot iput the inode now as we can be holding the last + * reference and we cannot iput it under inode_lock. So we + * keep the reference and iput it later. */ + old_inode = inode; + spin_lock(&inode_lock); + } + spin_unlock(&inode_lock); + iput(old_inode); +} + +/* 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 - add dquot to list for freeing if needed */ +/* 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] = NODQUOT; + if (dquot != NODQUOT) { + if (dqput_blocks(dquot)) { +#ifdef __DQUOT_PARANOIA + if (atomic_read(&dquot->dq_count) != 1) + printk(KERN_WARNING "VFS: Adding dquot with dq_count %d to dispose list.\n", atomic_read(&dquot->dq_count)); +#endif + spin_lock(&dq_list_lock); + list_add(&dquot->dq_free, tofree_head); /* As dquot must have currently users it can't be on the free list... */ + spin_unlock(&dq_list_lock); + return 1; + } + else + dqput(dquot); /* We have guaranteed we won't block */ + } + return 0; +} + +/* Free list of dquots - called from inode.c */ +/* dquots are removed from inodes, 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; + /* So now we have dquots on the list... Just free them */ + while (act_head != tofree_head) { + dquot = list_entry(act_head, struct dquot, dq_free); + act_head = act_head->next; + list_del_init(&dquot->dq_free); /* Remove dquot from the list so we won't have problems... */ + dqput(dquot); + } +} + +static void remove_dquot_ref(struct super_block *sb, int type, + struct list_head *tofree_head) +{ + struct inode *inode; + + spin_lock(&inode_lock); + list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { + if (!IS_NOQUOTA(inode)) + remove_inode_dquot_ref(inode, type, tofree_head); + } + spin_unlock(&inode_lock); +} + +/* 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) +{ + WARN_ON(dquot->dq_dqb.dqb_rsvspace < number); + dquot->dq_dqb.dqb_curspace += number; + dquot->dq_dqb.dqb_rsvspace -= number; +} + +static inline +void dquot_free_reserved_space(struct dquot *dquot, qsize_t number) +{ + dquot->dq_dqb.dqb_rsvspace -= number; +} + +static inline void dquot_decr_inodes(struct dquot *dquot, qsize_t number) +{ + if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NEGATIVE_USAGE || + dquot->dq_dqb.dqb_curinodes >= number) + dquot->dq_dqb.dqb_curinodes -= 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; + clear_bit(DQ_INODES_B, &dquot->dq_flags); +} + +static inline void dquot_decr_space(struct dquot *dquot, qsize_t number) +{ + if (sb_dqopt(dquot->dq_sb)->flags & DQUOT_NEGATIVE_USAGE || + dquot->dq_dqb.dqb_curspace >= 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; + clear_bit(DQ_BLKS_B, &dquot->dq_flags); +} + +static int warning_issued(struct dquot *dquot, const int warntype) +{ + int flag = (warntype == QUOTA_NL_BHARDWARN || + warntype == QUOTA_NL_BSOFTLONGWARN) ? DQ_BLKS_B : + ((warntype == QUOTA_NL_IHARDWARN || + warntype == QUOTA_NL_ISOFTLONGWARN) ? DQ_INODES_B : 0); + + if (!flag) + return 0; + return test_and_set_bit(flag, &dquot->dq_flags); +} + +#ifdef CONFIG_PRINT_QUOTA_WARNING +static int flag_print_warnings = 1; + +static inline int need_print_warning(struct dquot *dquot) +{ + if (!flag_print_warnings) + return 0; + + switch (dquot->dq_type) { + case USRQUOTA: + return current_fsuid() == dquot->dq_id; + case GRPQUOTA: + return in_group_p(dquot->dq_id); + } + return 0; +} + +/* Print warning to user which exceeded quota */ +static void print_warning(struct dquot *dquot, const int warntype) +{ + char *msg = NULL; + struct tty_struct *tty; + + if (warntype == QUOTA_NL_IHARDBELOW || + warntype == QUOTA_NL_ISOFTBELOW || + warntype == QUOTA_NL_BHARDBELOW || + warntype == QUOTA_NL_BSOFTBELOW || !need_print_warning(dquot)) + return; + + tty = get_current_tty(); + if (!tty) + return; + tty_write_message(tty, dquot->dq_sb->s_id); + if (warntype == QUOTA_NL_ISOFTWARN || warntype == QUOTA_NL_BSOFTWARN) + tty_write_message(tty, ": warning, "); + else + tty_write_message(tty, ": write failed, "); + tty_write_message(tty, quotatypes[dquot->dq_type]); + switch (warntype) { + case QUOTA_NL_IHARDWARN: + msg = " file limit reached.\r\n"; + break; + case QUOTA_NL_ISOFTLONGWARN: + msg = " file quota exceeded too long.\r\n"; + break; + case QUOTA_NL_ISOFTWARN: + msg = " file quota exceeded.\r\n"; + break; + case QUOTA_NL_BHARDWARN: + msg = " block limit reached.\r\n"; + break; + case QUOTA_NL_BSOFTLONGWARN: + msg = " block quota exceeded too long.\r\n"; + break; + case QUOTA_NL_BSOFTWARN: + msg = " block quota exceeded.\r\n"; + break; + } + tty_write_message(tty, msg); + tty_kref_put(tty); +} +#endif + +#ifdef CONFIG_QUOTA_NETLINK_INTERFACE + +/* Netlink family structure for quota */ +static struct genl_family quota_genl_family = { + .id = GENL_ID_GENERATE, + .hdrsize = 0, + .name = "VFS_DQUOT", + .version = 1, + .maxattr = QUOTA_NL_A_MAX, +}; + +/* Send warning to userspace about user which exceeded quota */ +static void send_warning(const struct dquot *dquot, const char warntype) +{ + static atomic_t seq; + struct sk_buff *skb; + void *msg_head; + int ret; + int msg_size = 4 * nla_total_size(sizeof(u32)) + + 2 * nla_total_size(sizeof(u64)); + + /* We have to allocate using GFP_NOFS as we are called from a + * filesystem performing write and thus further recursion into + * the fs to free some data could cause deadlocks. */ + skb = genlmsg_new(msg_size, GFP_NOFS); + if (!skb) { + printk(KERN_ERR + "VFS: Not enough memory to send quota warning.\n"); + return; + } + msg_head = genlmsg_put(skb, 0, atomic_add_return(1, &seq), + "a_genl_family, 0, QUOTA_NL_C_WARNING); + if (!msg_head) { + printk(KERN_ERR + "VFS: Cannot store netlink header in quota warning.\n"); + goto err_out; + } + ret = nla_put_u32(skb, QUOTA_NL_A_QTYPE, dquot->dq_type); + if (ret) + goto attr_err_out; + ret = nla_put_u64(skb, QUOTA_NL_A_EXCESS_ID, dquot->dq_id); + if (ret) + goto attr_err_out; + ret = nla_put_u32(skb, QUOTA_NL_A_WARNING, warntype); + if (ret) + goto attr_err_out; + ret = nla_put_u32(skb, QUOTA_NL_A_DEV_MAJOR, + MAJOR(dquot->dq_sb->s_dev)); + if (ret) + goto attr_err_out; + ret = nla_put_u32(skb, QUOTA_NL_A_DEV_MINOR, + MINOR(dquot->dq_sb->s_dev)); + if (ret) + goto attr_err_out; + ret = nla_put_u64(skb, QUOTA_NL_A_CAUSED_ID, current_uid()); + if (ret) + goto attr_err_out; + genlmsg_end(skb, msg_head); + + ret = genlmsg_multicast(skb, 0, quota_genl_family.id, GFP_NOFS); + if (ret < 0 && ret != -ESRCH) + printk(KERN_ERR + "VFS: Failed to send notification message: %d\n", ret); + return; +attr_err_out: + printk(KERN_ERR "VFS: Not enough space to compose quota message!\n"); +err_out: + kfree_skb(skb); +} +#endif +/* + * Write warnings to the console and send warning messages over netlink. + * + * Note that this function can sleep. + */ +static inline void flush_warnings(struct dquot * const *dquots, char *warntype) +{ + int i; + + for (i = 0; i < MAXQUOTAS; i++) + if (dquots[i] != NODQUOT && warntype[i] != QUOTA_NL_NOWARN && + !warning_issued(dquots[i], warntype[i])) { +#ifdef CONFIG_PRINT_QUOTA_WARNING + print_warning(dquots[i], warntype[i]); +#endif +#ifdef CONFIG_QUOTA_NETLINK_INTERFACE + send_warning(dquots[i], warntype[i]); +#endif + } +} + +static inline char ignore_hardlimit(struct dquot *dquot) +{ + struct mem_dqinfo *info = &sb_dqopt(dquot->dq_sb)->info[dquot->dq_type]; + + return capable(CAP_SYS_RESOURCE) && + (info->dqi_format->qf_fmt_id != QFMT_VFS_OLD || !(info->dqi_flags & V1_DQF_RSQUASH)); +} + +/* needs dq_data_lock */ +static int check_idq(struct dquot *dquot, qsize_t inodes, char *warntype) +{ + *warntype = QUOTA_NL_NOWARN; + if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type) || + test_bit(DQ_FAKE_B, &dquot->dq_flags)) + return QUOTA_OK; + + if (dquot->dq_dqb.dqb_ihardlimit && + (dquot->dq_dqb.dqb_curinodes + inodes) > dquot->dq_dqb.dqb_ihardlimit && + !ignore_hardlimit(dquot)) { + *warntype = QUOTA_NL_IHARDWARN; + return NO_QUOTA; + } + + if (dquot->dq_dqb.dqb_isoftlimit && + (dquot->dq_dqb.dqb_curinodes + inodes) > dquot->dq_dqb.dqb_isoftlimit && + dquot->dq_dqb.dqb_itime && get_seconds() >= dquot->dq_dqb.dqb_itime && + !ignore_hardlimit(dquot)) { + *warntype = QUOTA_NL_ISOFTLONGWARN; + return NO_QUOTA; + } + + if (dquot->dq_dqb.dqb_isoftlimit && + (dquot->dq_dqb.dqb_curinodes + inodes) > dquot->dq_dqb.dqb_isoftlimit && + dquot->dq_dqb.dqb_itime == 0) { + *warntype = QUOTA_NL_ISOFTWARN; + dquot->dq_dqb.dqb_itime = get_seconds() + sb_dqopt(dquot->dq_sb)->info[dquot->dq_type].dqi_igrace; + } + + return QUOTA_OK; +} + +/* needs dq_data_lock */ +static int check_bdq(struct dquot *dquot, qsize_t space, int prealloc, char *warntype) +{ + qsize_t tspace; + + *warntype = QUOTA_NL_NOWARN; + if (!sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type) || + test_bit(DQ_FAKE_B, &dquot->dq_flags)) + return QUOTA_OK; + + tspace = dquot->dq_dqb.dqb_curspace + dquot->dq_dqb.dqb_rsvspace + + space; + + if (dquot->dq_dqb.dqb_bhardlimit && + tspace > dquot->dq_dqb.dqb_bhardlimit && + !ignore_hardlimit(dquot)) { + if (!prealloc) + *warntype = QUOTA_NL_BHARDWARN; + return NO_QUOTA; + } + + if (dquot->dq_dqb.dqb_bsoftlimit && + tspace > dquot->dq_dqb.dqb_bsoftlimit && + dquot->dq_dqb.dqb_btime && get_seconds() >= dquot->dq_dqb.dqb_btime && + !ignore_hardlimit(dquot)) { + if (!prealloc) + *warntype = QUOTA_NL_BSOFTLONGWARN; + return NO_QUOTA; + } + + if (dquot->dq_dqb.dqb_bsoftlimit && + tspace > dquot->dq_dqb.dqb_bsoftlimit && + dquot->dq_dqb.dqb_btime == 0) { + if (!prealloc) { + *warntype = QUOTA_NL_BSOFTWARN; + dquot->dq_dqb.dqb_btime = get_seconds() + sb_dqopt(dquot->dq_sb)->info[dquot->dq_type].dqi_bgrace; + } + else + /* + * We don't allow preallocation to exceed softlimit so exceeding will + * be always printed + */ + return NO_QUOTA; + } + + return QUOTA_OK; +} + +static int info_idq_free(struct dquot *dquot, qsize_t inodes) +{ + if (test_bit(DQ_FAKE_B, &dquot->dq_flags) || + dquot->dq_dqb.dqb_curinodes <= dquot->dq_dqb.dqb_isoftlimit || + !sb_has_quota_limits_enabled(dquot->dq_sb, dquot->dq_type)) + return QUOTA_NL_NOWARN; + + if (dquot->dq_dqb.dqb_curinodes - inodes <= dquot->dq_dqb.dqb_isoftlimit) + return QUOTA_NL_ISOFTBELOW; + if (dquot->dq_dqb.dqb_curinodes >= dquot->dq_dqb.dqb_ihardlimit && + dquot->dq_dqb.dqb_curinodes - inodes < dquot->dq_dqb.dqb_ihardlimit) + return QUOTA_NL_IHARDBELOW; + return QUOTA_NL_NOWARN; +} + +static int info_bdq_free(struct dquot *dquot, qsize_t space) +{ + if (test_bit(DQ_FAKE_B, &dquot->dq_flags) || + dquot->dq_dqb.dqb_curspace <= dquot->dq_dqb.dqb_bsoftlimit) + return QUOTA_NL_NOWARN; + + if (dquot->dq_dqb.dqb_curspace - 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) + return QUOTA_NL_BHARDBELOW; + return QUOTA_NL_NOWARN; +} +/* + * 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. + */ +int dquot_initialize(struct inode *inode, int type) +{ + unsigned int id = 0; + int cnt, ret = 0; + struct dquot *got[MAXQUOTAS] = { NODQUOT, NODQUOT }; + struct super_block *sb = inode->i_sb; + + /* 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; + + /* First get references to structures we might need. */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (type != -1 && cnt != type) + continue; + switch (cnt) { + case USRQUOTA: + id = inode->i_uid; + break; + case GRPQUOTA: + id = inode->i_gid; + break; + } + got[cnt] = dqget(sb, id, cnt); + } + + down_write(&sb_dqopt(sb)->dqptr_sem); + /* Having dqptr_sem we know NOQUOTA flags can't be altered... */ + if (IS_NOQUOTA(inode)) + goto out_err; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (type != -1 && cnt != type) + continue; + /* Avoid races with quotaoff() */ + if (!sb_has_quota_active(sb, cnt)) + continue; + if (inode->i_dquot[cnt] == NODQUOT) { + inode->i_dquot[cnt] = got[cnt]; + got[cnt] = NODQUOT; + } + } +out_err: + up_write(&sb_dqopt(sb)->dqptr_sem); + /* Drop unused references */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + dqput(got[cnt]); + return ret; +} +EXPORT_SYMBOL(dquot_initialize); + +/* + * Release all quotas referenced by inode + */ +int dquot_drop(struct inode *inode) +{ + int cnt; + struct dquot *put[MAXQUOTAS]; + + down_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + put[cnt] = inode->i_dquot[cnt]; + inode->i_dquot[cnt] = NODQUOT; + } + up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + dqput(put[cnt]); + return 0; +} +EXPORT_SYMBOL(dquot_drop); + +/* Wrapper to remove references to quota structures from inode */ +void vfs_dq_drop(struct inode *inode) +{ + /* Here we can get arbitrary inode from clear_inode() so we have + * to be careful. OTOH we don't need locking as quota operations + * are allowed to change only at mount time */ + if (!IS_NOQUOTA(inode) && inode->i_sb && inode->i_sb->dq_op + && inode->i_sb->dq_op->drop) { + int cnt; + /* Test before calling to rule out calls from proc and such + * where we are not allowed to block. Note that this is + * actually reliable test even without the lock - the caller + * must assure that nobody can come after the DQUOT_DROP and + * add quota pointers back anyway */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt] != NODQUOT) + break; + if (cnt < MAXQUOTAS) + inode->i_sb->dq_op->drop(inode); + } +} +EXPORT_SYMBOL(vfs_dq_drop); + +/* + * Following four functions update i_blocks+i_bytes fields and + * quota information (together with appropriate checks) + * NOTE: We absolutely rely on the fact that caller dirties + * the inode (usually macros in quotaops.h care about this) and + * holds a handle for the current transaction so that dquot write and + * inode write go into the same transaction. + */ + +/* + * This operation can block, but only after everything is updated + */ +int __dquot_alloc_space(struct inode *inode, qsize_t number, + int warn, int reserve) +{ + int cnt, ret = QUOTA_OK; + char warntype[MAXQUOTAS]; + + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + warntype[cnt] = QUOTA_NL_NOWARN; + + spin_lock(&dq_data_lock); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + if (check_bdq(inode->i_dquot[cnt], number, warn, warntype+cnt) + == NO_QUOTA) { + ret = NO_QUOTA; + goto out_unlock; + } + } + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + if (reserve) + dquot_resv_space(inode->i_dquot[cnt], number); + else + dquot_incr_space(inode->i_dquot[cnt], number); + } + if (!reserve) + inode_add_bytes(inode, number); +out_unlock: + spin_unlock(&dq_data_lock); + flush_warnings(inode->i_dquot, warntype); + return ret; +} + +int dquot_alloc_space(struct inode *inode, qsize_t number, int warn) +{ + int cnt, ret = QUOTA_OK; + + /* + * First test before acquiring mutex - solves deadlocks when we + * re-enter the quota code and are already holding the mutex + */ + if (IS_NOQUOTA(inode)) { + inode_add_bytes(inode, number); + goto out; + } + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) { + inode_add_bytes(inode, number); + goto out_unlock; + } + + ret = __dquot_alloc_space(inode, number, warn, 0); + if (ret == NO_QUOTA) + goto out_unlock; + + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt]) + mark_dquot_dirty(inode->i_dquot[cnt]); +out_unlock: + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); +out: + return ret; +} +EXPORT_SYMBOL(dquot_alloc_space); + +int dquot_reserve_space(struct inode *inode, qsize_t number, int warn) +{ + int ret = QUOTA_OK; + + if (IS_NOQUOTA(inode)) + goto out; + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) + goto out_unlock; + + ret = __dquot_alloc_space(inode, number, warn, 1); +out_unlock: + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); +out: + return ret; +} +EXPORT_SYMBOL(dquot_reserve_space); + +/* + * This operation can block, but only after everything is updated + */ +int dquot_alloc_inode(const struct inode *inode, qsize_t number) +{ + int cnt, ret = NO_QUOTA; + char warntype[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 QUOTA_OK; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + warntype[cnt] = QUOTA_NL_NOWARN; + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) { + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + return QUOTA_OK; + } + spin_lock(&dq_data_lock); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + if (check_idq(inode->i_dquot[cnt], number, warntype+cnt) == NO_QUOTA) + goto warn_put_all; + } + + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + dquot_incr_inodes(inode->i_dquot[cnt], number); + } + ret = QUOTA_OK; +warn_put_all: + spin_unlock(&dq_data_lock); + if (ret == QUOTA_OK) + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt]) + mark_dquot_dirty(inode->i_dquot[cnt]); + flush_warnings(inode->i_dquot, warntype); + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + return ret; +} +EXPORT_SYMBOL(dquot_alloc_inode); + +int dquot_claim_space(struct inode *inode, qsize_t number) +{ + int cnt; + int ret = QUOTA_OK; + + if (IS_NOQUOTA(inode)) { + inode_add_bytes(inode, number); + goto out; + } + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) { + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + inode_add_bytes(inode, number); + goto out; + } + + spin_lock(&dq_data_lock); + /* Claim reserved quotas to allocated quotas */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] != NODQUOT) + dquot_claim_reserved_space(inode->i_dquot[cnt], + number); + } + /* Update inode bytes */ + inode_add_bytes(inode, number); + spin_unlock(&dq_data_lock); + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt]) + mark_dquot_dirty(inode->i_dquot[cnt]); + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); +out: + return ret; +} +EXPORT_SYMBOL(dquot_claim_space); + +/* + * Release reserved quota space + */ +void dquot_release_reserved_space(struct inode *inode, qsize_t number) +{ + int cnt; + + if (IS_NOQUOTA(inode)) + goto out; + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) + goto out_unlock; + + spin_lock(&dq_data_lock); + /* Release reserved dquots */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] != NODQUOT) + dquot_free_reserved_space(inode->i_dquot[cnt], number); + } + spin_unlock(&dq_data_lock); + +out_unlock: + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); +out: + return; +} +EXPORT_SYMBOL(dquot_release_reserved_space); + +/* + * This operation can block, but only after everything is updated + */ +int dquot_free_space(struct inode *inode, qsize_t number) +{ + unsigned int cnt; + char warntype[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)) { +out_sub: + inode_sub_bytes(inode, number); + return QUOTA_OK; + } + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + /* Now recheck reliably when holding dqptr_sem */ + if (IS_NOQUOTA(inode)) { + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + goto out_sub; + } + spin_lock(&dq_data_lock); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + warntype[cnt] = info_bdq_free(inode->i_dquot[cnt], number); + dquot_decr_space(inode->i_dquot[cnt], number); + } + inode_sub_bytes(inode, number); + spin_unlock(&dq_data_lock); + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt]) + mark_dquot_dirty(inode->i_dquot[cnt]); + flush_warnings(inode->i_dquot, warntype); + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + return QUOTA_OK; +} +EXPORT_SYMBOL(dquot_free_space); + +/* + * This operation can block, but only after everything is updated + */ +int dquot_free_inode(const struct inode *inode, qsize_t number) +{ + unsigned int cnt; + char warntype[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 QUOTA_OK; + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + /* Now recheck reliably when holding dqptr_sem */ + if (IS_NOQUOTA(inode)) { + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + return QUOTA_OK; + } + spin_lock(&dq_data_lock); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (inode->i_dquot[cnt] == NODQUOT) + continue; + warntype[cnt] = info_idq_free(inode->i_dquot[cnt], number); + dquot_decr_inodes(inode->i_dquot[cnt], number); + } + spin_unlock(&dq_data_lock); + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + if (inode->i_dquot[cnt]) + mark_dquot_dirty(inode->i_dquot[cnt]); + flush_warnings(inode->i_dquot, warntype); + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + return QUOTA_OK; +} +EXPORT_SYMBOL(dquot_free_inode); + +/* + * call back function, get reserved quota space from underlying fs + */ +qsize_t dquot_get_reserved_space(struct inode *inode) +{ + qsize_t reserved_space = 0; + + if (sb_any_quota_active(inode->i_sb) && + inode->i_sb->dq_op->get_reserved_space) + reserved_space = inode->i_sb->dq_op->get_reserved_space(inode); + return reserved_space; +} + +/* + * Transfer the number of inode and blocks from one diskquota to an other. + * + * This operation can block, but only after everything is updated + * A transaction must be started when entering this function. + */ +int dquot_transfer(struct inode *inode, struct iattr *iattr) +{ + qsize_t space, cur_space; + qsize_t rsv_space = 0; + struct dquot *transfer_from[MAXQUOTAS]; + struct dquot *transfer_to[MAXQUOTAS]; + int cnt, ret = QUOTA_OK; + int chuid = iattr->ia_valid & ATTR_UID && inode->i_uid != iattr->ia_uid, + chgid = iattr->ia_valid & ATTR_GID && inode->i_gid != iattr->ia_gid; + char warntype_to[MAXQUOTAS]; + char warntype_from_inodes[MAXQUOTAS], warntype_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 QUOTA_OK; + /* Initialize the arrays */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + transfer_from[cnt] = NODQUOT; + transfer_to[cnt] = NODQUOT; + warntype_to[cnt] = QUOTA_NL_NOWARN; + switch (cnt) { + case USRQUOTA: + if (!chuid) + continue; + transfer_to[cnt] = dqget(inode->i_sb, iattr->ia_uid, cnt); + break; + case GRPQUOTA: + if (!chgid) + continue; + transfer_to[cnt] = dqget(inode->i_sb, iattr->ia_gid, cnt); + break; + } + } + + down_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + /* Now recheck reliably when holding dqptr_sem */ + if (IS_NOQUOTA(inode)) { /* File without quota accounting? */ + up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + goto put_all; + } + spin_lock(&dq_data_lock); + cur_space = inode_get_bytes(inode); + rsv_space = dquot_get_reserved_space(inode); + space = cur_space + rsv_space; + /* Build the transfer_from list and check the limits */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (transfer_to[cnt] == NODQUOT) + continue; + transfer_from[cnt] = inode->i_dquot[cnt]; + if (check_idq(transfer_to[cnt], 1, warntype_to + cnt) == + NO_QUOTA || check_bdq(transfer_to[cnt], space, 0, + warntype_to + cnt) == NO_QUOTA) + goto over_quota; + } + + /* + * Finally perform the needed transfer from transfer_from to transfer_to + */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + /* + * Skip changes for same uid or gid or for turned off quota-type. + */ + if (transfer_to[cnt] == NODQUOT) + continue; + + /* Due to IO error we might not have transfer_from[] structure */ + if (transfer_from[cnt]) { + warntype_from_inodes[cnt] = + info_idq_free(transfer_from[cnt], 1); + warntype_from_space[cnt] = + info_bdq_free(transfer_from[cnt], space); + dquot_decr_inodes(transfer_from[cnt], 1); + dquot_decr_space(transfer_from[cnt], cur_space); + dquot_free_reserved_space(transfer_from[cnt], + rsv_space); + } + + 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]; + } + spin_unlock(&dq_data_lock); + up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + + /* Dirtify all the dquots - this can block when journalling */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (transfer_from[cnt]) + mark_dquot_dirty(transfer_from[cnt]); + if (transfer_to[cnt]) { + mark_dquot_dirty(transfer_to[cnt]); + /* The reference we got is transferred to the inode */ + transfer_to[cnt] = NODQUOT; + } + } +warn_put_all: + flush_warnings(transfer_to, warntype_to); + flush_warnings(transfer_from, warntype_from_inodes); + flush_warnings(transfer_from, warntype_from_space); +put_all: + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + dqput(transfer_from[cnt]); + dqput(transfer_to[cnt]); + } + return ret; +over_quota: + spin_unlock(&dq_data_lock); + up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); + /* Clear dquot pointers we don't want to dqput() */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + transfer_from[cnt] = NODQUOT; + ret = NO_QUOTA; + goto warn_put_all; +} +EXPORT_SYMBOL(dquot_transfer); + +/* Wrapper for transferring ownership of an inode */ +int vfs_dq_transfer(struct inode *inode, struct iattr *iattr) +{ + if (sb_any_quota_active(inode->i_sb) && !IS_NOQUOTA(inode)) { + vfs_dq_init(inode); + if (inode->i_sb->dq_op->transfer(inode, iattr) == NO_QUOTA) + return 1; + } + return 0; +} +EXPORT_SYMBOL(vfs_dq_transfer); + +/* + * Write info of quota file to disk + */ +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; +} +EXPORT_SYMBOL(dquot_commit_info); + +/* + * Definitions of diskquota operations. + */ +struct dquot_operations dquot_operations = { + .initialize = dquot_initialize, + .drop = dquot_drop, + .alloc_space = dquot_alloc_space, + .alloc_inode = dquot_alloc_inode, + .free_space = dquot_free_space, + .free_inode = dquot_free_inode, + .transfer = dquot_transfer, + .write_dquot = dquot_commit, + .acquire_dquot = dquot_acquire, + .release_dquot = dquot_release, + .mark_dirty = dquot_mark_dquot_dirty, + .write_info = dquot_commit_info, + .alloc_dquot = dquot_alloc, + .destroy_dquot = dquot_destroy, +}; + +/* + * Turn quota off on a device. type == -1 ==> quotaoff for all types (umount) + */ +int vfs_quota_disable(struct super_block *sb, int type, unsigned int flags) +{ + int cnt, ret = 0; + struct quota_info *dqopt = sb_dqopt(sb); + struct inode *toputinode[MAXQUOTAS]; + + /* Cannot turn off usage accounting without turning off limits, or + * suspend quotas and simultaneously turn quotas off. */ + if ((flags & DQUOT_USAGE_ENABLED && !(flags & DQUOT_LIMITS_ENABLED)) + || (flags & DQUOT_SUSPENDED && flags & (DQUOT_LIMITS_ENABLED | + 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); + return 0; + } + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + toputinode[cnt] = NULL; + if (type != -1 && cnt != type) + continue; + if (!sb_has_quota_loaded(sb, cnt)) + continue; + + if (flags & DQUOT_SUSPENDED) { + spin_lock(&dq_state_lock); + dqopt->flags |= + dquot_state_flag(DQUOT_SUSPENDED, cnt); + spin_unlock(&dq_state_lock); + } else { + spin_lock(&dq_state_lock); + dqopt->flags &= ~dquot_state_flag(flags, cnt); + /* Turning off suspended quotas? */ + if (!sb_has_quota_loaded(sb, cnt) && + sb_has_quota_suspended(sb, cnt)) { + dqopt->flags &= ~dquot_state_flag( + DQUOT_SUSPENDED, cnt); + spin_unlock(&dq_state_lock); + iput(dqopt->files[cnt]); + dqopt->files[cnt] = NULL; + continue; + } + spin_unlock(&dq_state_lock); + } + + /* We still have to keep quota loaded? */ + if (sb_has_quota_loaded(sb, cnt) && !(flags & DQUOT_SUSPENDED)) + continue; + + /* Note: these are blocking operations */ + drop_dquot_ref(sb, cnt); + invalidate_dquots(sb, cnt); + /* + * Now all dquots should be invalidated, all writes done so we should be only + * users of the info. No locks needed. + */ + if (info_dirty(&dqopt->info[cnt])) + sb->dq_op->write_info(sb, cnt); + 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) + goto put_inodes; + + /* Sync the superblock so that buffers with quota data are written to + * disk (and so userspace sees correct data afterwards). */ + if (sb->s_op->sync_fs) + sb->s_op->sync_fs(sb, 1); + sync_blockdev(sb->s_bdev); + /* Now the quota files are just ordinary files and we can set the + * inode flags back. Moreover we discard the pagecache so that + * userspace sees the writes we did bypassing the pagecache. We + * 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_nested(&toputinode[cnt]->i_mutex, I_MUTEX_QUOTA); + 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(toputinode[cnt]); + } + mutex_unlock(&dqopt->dqonoff_mutex); + } + if (sb->s_bdev) + invalidate_bdev(sb->s_bdev); +put_inodes: + 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; +} +EXPORT_SYMBOL(vfs_quota_disable); + +int vfs_quota_off(struct super_block *sb, int type, int remount) +{ + return vfs_quota_disable(sb, type, remount ? DQUOT_SUSPENDED : + (DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED)); +} +EXPORT_SYMBOL(vfs_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) +{ + struct quota_format_type *fmt = find_quota_format(format_id); + struct super_block *sb = inode->i_sb; + struct quota_info *dqopt = sb_dqopt(sb); + int error; + int oldflags = -1; + + if (!fmt) + return -ESRCH; + if (!S_ISREG(inode->i_mode)) { + error = -EACCES; + goto out_fmt; + } + if (IS_RDONLY(inode)) { + error = -EROFS; + goto out_fmt; + } + if (!sb->s_op->quota_write || !sb->s_op->quota_read) { + error = -EINVAL; + goto out_fmt; + } + /* Usage always has to be set... */ + if (!(flags & DQUOT_USAGE_ENABLED)) { + error = -EINVAL; + goto out_fmt; + } + + if (!(dqopt->flags & DQUOT_QUOTA_SYS_FILE)) { + /* As we bypass the pagecache we must now flush the inode so + * that we see all the changes from userspace... */ + write_inode_now(inode, 1); + /* And now flush the block cache so that kernel sees the + * changes */ + invalidate_bdev(sb->s_bdev); + } + mutex_lock(&inode->i_mutex); + 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. */ + down_write(&dqopt->dqptr_sem); + oldflags = inode->i_flags & (S_NOATIME | S_IMMUTABLE | S_NOQUOTA); + inode->i_flags |= S_NOQUOTA | S_NOATIME | S_IMMUTABLE; + up_write(&dqopt->dqptr_sem); + sb->dq_op->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; + + 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); + if ((error = dqopt->ops[type]->read_file_info(sb, type)) < 0) { + mutex_unlock(&dqopt->dqio_mutex); + goto out_file_init; + } + mutex_unlock(&dqopt->dqio_mutex); + mutex_unlock(&inode->i_mutex); + 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; + +out_file_init: + dqopt->files[type] = NULL; + iput(inode); +out_lock: + mutex_unlock(&dqopt->dqonoff_mutex); + if (oldflags != -1) { + down_write(&dqopt->dqptr_sem); + /* 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; + up_write(&dqopt->dqptr_sem); + } + mutex_unlock(&inode->i_mutex); +out_fmt: + put_quota_format(fmt); + + return error; +} + +/* Reenable quotas on remount RW */ +static int vfs_quota_on_remount(struct super_block *sb, int type) +{ + struct quota_info *dqopt = sb_dqopt(sb); + struct inode *inode; + int ret; + unsigned int flags; + + mutex_lock(&dqopt->dqonoff_mutex); + if (!sb_has_quota_suspended(sb, type)) { + mutex_unlock(&dqopt->dqonoff_mutex); + return 0; + } + inode = dqopt->files[type]; + dqopt->files[type] = NULL; + spin_lock(&dq_state_lock); + flags = dqopt->flags & dquot_state_flag(DQUOT_USAGE_ENABLED | + DQUOT_LIMITS_ENABLED, type); + dqopt->flags &= ~dquot_state_flag(DQUOT_STATE_FLAGS, type); + spin_unlock(&dq_state_lock); + mutex_unlock(&dqopt->dqonoff_mutex); + + flags = dquot_generic_flag(flags, type); + ret = vfs_load_quota_inode(inode, type, dqopt->info[type].dqi_fmt_id, + flags); + iput(inode); + + return ret; +} + +int vfs_quota_on_path(struct super_block *sb, int type, int format_id, + struct path *path) +{ + int error = security_quota_on(path->dentry); + if (error) + return error; + /* Quota file not on the same filesystem? */ + if (path->mnt->mnt_sb != sb) + error = -EXDEV; + else + error = vfs_load_quota_inode(path->dentry->d_inode, type, + format_id, DQUOT_USAGE_ENABLED | + DQUOT_LIMITS_ENABLED); + return error; +} +EXPORT_SYMBOL(vfs_quota_on_path); + +int vfs_quota_on(struct super_block *sb, int type, int format_id, char *name, + int remount) +{ + struct path path; + int error; + + if (remount) + return vfs_quota_on_remount(sb, type); + + error = kern_path(name, LOOKUP_FOLLOW, &path); + if (!error) { + error = vfs_quota_on_path(sb, type, format_id, &path); + path_put(&path); + } + return error; +} +EXPORT_SYMBOL(vfs_quota_on); + +/* + * More powerful function for turning on quotas allowing setting + * of individual quota flags + */ +int vfs_quota_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? */ + if (flags & DQUOT_SUSPENDED) + return vfs_quota_on_remount(sb, type); + 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(vfs_quota_enable); + +/* + * This function is used when filesystem needs to initialize quotas + * during mount time. + */ +int vfs_quota_on_mount(struct super_block *sb, char *qf_name, + int format_id, int type) +{ + struct dentry *dentry; + int error; + + dentry = lookup_one_len(qf_name, sb->s_root, strlen(qf_name)); + 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, + DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); + +out: + dput(dentry); + return error; +} +EXPORT_SYMBOL(vfs_quota_on_mount); + +/* Wrapper to turn on quotas when remounting rw */ +int vfs_dq_quota_on_remount(struct super_block *sb) +{ + int cnt; + int ret = 0, err; + + if (!sb->s_qcop || !sb->s_qcop->quota_on) + return -ENOSYS; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + err = sb->s_qcop->quota_on(sb, cnt, 0, NULL, 1); + if (err < 0 && !ret) + ret = err; + } + return ret; +} +EXPORT_SYMBOL(vfs_dq_quota_on_remount); + +static inline qsize_t qbtos(qsize_t blocks) +{ + return blocks << QIF_DQBLKSIZE_BITS; +} + +static inline qsize_t stoqb(qsize_t space) +{ + return (space + QIF_DQBLKSIZE - 1) >> QIF_DQBLKSIZE_BITS; +} + +/* Generic routine for getting common part of quota structure */ +static void do_get_dqblk(struct dquot *dquot, struct if_dqblk *di) +{ + struct mem_dqblk *dm = &dquot->dq_dqb; + + spin_lock(&dq_data_lock); + di->dqb_bhardlimit = stoqb(dm->dqb_bhardlimit); + di->dqb_bsoftlimit = stoqb(dm->dqb_bsoftlimit); + di->dqb_curspace = dm->dqb_curspace + dm->dqb_rsvspace; + di->dqb_ihardlimit = dm->dqb_ihardlimit; + di->dqb_isoftlimit = dm->dqb_isoftlimit; + di->dqb_curinodes = dm->dqb_curinodes; + di->dqb_btime = dm->dqb_btime; + di->dqb_itime = dm->dqb_itime; + di->dqb_valid = QIF_ALL; + spin_unlock(&dq_data_lock); +} + +int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di) +{ + struct dquot *dquot; + + dquot = dqget(sb, id, type); + if (dquot == NODQUOT) + return -ESRCH; + do_get_dqblk(dquot, di); + dqput(dquot); + + return 0; +} +EXPORT_SYMBOL(vfs_get_dqblk); + +/* Generic routine for setting common part of quota structure */ +static int do_set_dqblk(struct dquot *dquot, struct if_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_type]; + + if ((di->dqb_valid & QIF_BLIMITS && + (di->dqb_bhardlimit > dqi->dqi_maxblimit || + di->dqb_bsoftlimit > dqi->dqi_maxblimit)) || + (di->dqb_valid & QIF_ILIMITS && + (di->dqb_ihardlimit > dqi->dqi_maxilimit || + di->dqb_isoftlimit > dqi->dqi_maxilimit))) + return -ERANGE; + + spin_lock(&dq_data_lock); + if (di->dqb_valid & QIF_SPACE) { + dm->dqb_curspace = di->dqb_curspace - dm->dqb_rsvspace; + check_blim = 1; + __set_bit(DQ_LASTSET_B + QIF_SPACE_B, &dquot->dq_flags); + } + if (di->dqb_valid & QIF_BLIMITS) { + dm->dqb_bsoftlimit = qbtos(di->dqb_bsoftlimit); + dm->dqb_bhardlimit = qbtos(di->dqb_bhardlimit); + check_blim = 1; + __set_bit(DQ_LASTSET_B + QIF_BLIMITS_B, &dquot->dq_flags); + } + if (di->dqb_valid & QIF_INODES) { + dm->dqb_curinodes = di->dqb_curinodes; + check_ilim = 1; + __set_bit(DQ_LASTSET_B + QIF_INODES_B, &dquot->dq_flags); + } + if (di->dqb_valid & QIF_ILIMITS) { + dm->dqb_isoftlimit = di->dqb_isoftlimit; + dm->dqb_ihardlimit = di->dqb_ihardlimit; + check_ilim = 1; + __set_bit(DQ_LASTSET_B + QIF_ILIMITS_B, &dquot->dq_flags); + } + if (di->dqb_valid & QIF_BTIME) { + dm->dqb_btime = di->dqb_btime; + check_blim = 1; + __set_bit(DQ_LASTSET_B + QIF_BTIME_B, &dquot->dq_flags); + } + if (di->dqb_valid & QIF_ITIME) { + dm->dqb_itime = di->dqb_itime; + 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_btime = 0; + clear_bit(DQ_BLKS_B, &dquot->dq_flags); + } + else if (!(di->dqb_valid & QIF_BTIME)) /* Set grace only if user hasn't provided his own... */ + dm->dqb_btime = get_seconds() + dqi->dqi_bgrace; + } + if (check_ilim) { + if (!dm->dqb_isoftlimit || dm->dqb_curinodes < dm->dqb_isoftlimit) { + dm->dqb_itime = 0; + clear_bit(DQ_INODES_B, &dquot->dq_flags); + } + else if (!(di->dqb_valid & QIF_ITIME)) /* Set grace only if user hasn't provided his own... */ + dm->dqb_itime = get_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); + + return 0; +} + +int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *di) +{ + struct dquot *dquot; + int rc; + + dquot = dqget(sb, id, type); + if (!dquot) { + rc = -ESRCH; + goto out; + } + rc = do_set_dqblk(dquot, di); + dqput(dquot); +out: + return rc; +} +EXPORT_SYMBOL(vfs_set_dqblk); + +/* Generic routine for getting common part of quota file information */ +int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii) +{ + 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; + } + 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_MASK; + ii->dqi_valid = IIF_ALL; + spin_unlock(&dq_data_lock); + mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); + return 0; +} +EXPORT_SYMBOL(vfs_get_dqinfo); + +/* Generic routine for setting common part of quota file information */ +int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *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; + } + mi = sb_dqopt(sb)->info + type; + 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_MASK) | (ii->dqi_flags & DQF_MASK); + 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; +} +EXPORT_SYMBOL(vfs_set_dqinfo); + +struct quotactl_ops vfs_quotactl_ops = { + .quota_on = vfs_quota_on, + .quota_off = vfs_quota_off, + .quota_sync = vfs_quota_sync, + .get_info = vfs_get_dqinfo, + .set_info = vfs_set_dqinfo, + .get_dqblk = vfs_get_dqblk, + .set_dqblk = vfs_set_dqblk +}; + +static ctl_table fs_dqstats_table[] = { + { + .ctl_name = FS_DQ_LOOKUPS, + .procname = "lookups", + .data = &dqstats.lookups, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_DROPS, + .procname = "drops", + .data = &dqstats.drops, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_READS, + .procname = "reads", + .data = &dqstats.reads, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_WRITES, + .procname = "writes", + .data = &dqstats.writes, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_CACHE_HITS, + .procname = "cache_hits", + .data = &dqstats.cache_hits, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_ALLOCATED, + .procname = "allocated_dquots", + .data = &dqstats.allocated_dquots, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_FREE, + .procname = "free_dquots", + .data = &dqstats.free_dquots, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, + { + .ctl_name = FS_DQ_SYNCS, + .procname = "syncs", + .data = &dqstats.syncs, + .maxlen = sizeof(int), + .mode = 0444, + .proc_handler = &proc_dointvec, + }, +#ifdef CONFIG_PRINT_QUOTA_WARNING + { + .ctl_name = FS_DQ_WARNINGS, + .procname = "warnings", + .data = &flag_print_warnings, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = &proc_dointvec, + }, +#endif + { .ctl_name = 0 }, +}; + +static ctl_table fs_table[] = { + { + .ctl_name = FS_DQSTATS, + .procname = "quota", + .mode = 0555, + .child = fs_dqstats_table, + }, + { .ctl_name = 0 }, +}; + +static ctl_table sys_table[] = { + { + .ctl_name = CTL_FS, + .procname = "fs", + .mode = 0555, + .child = fs_table, + }, + { .ctl_name = 0 }, +}; + +static int __init dquot_init(void) +{ + int i; + unsigned long nr_hash, order; + + printk(KERN_NOTICE "VFS: Disk quotas %s\n", __DQUOT_VERSION__); + + register_sysctl_table(sys_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), + NULL); + + order = 0; + dquot_hash = (struct hlist_head *)__get_free_pages(GFP_ATOMIC, order); + if (!dquot_hash) + panic("Cannot create dquot hash table"); + + /* 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--; + + 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)); + + register_shrinker(&dqcache_shrinker); + +#ifdef CONFIG_QUOTA_NETLINK_INTERFACE + if (genl_register_family("a_genl_family) != 0) + printk(KERN_ERR "VFS: Failed to create quota netlink interface.\n"); +#endif + + return 0; +} +module_init(dquot_init); diff --git a/fs/quota/quota.c b/fs/quota/quota.c new file mode 100644 index 000000000000..d76ada914f98 --- /dev/null +++ b/fs/quota/quota.c @@ -0,0 +1,513 @@ +/* + * Quota code necessary even when VFS quota support is not compiled + * into the kernel. The interesting stuff is over in dquot.c, here + * we have symbols for initial quotactl(2) handling, the sysctl(2) + * variables, etc - things needed even when quota support disabled. + */ + +#include <linux/fs.h> +#include <linux/namei.h> +#include <linux/slab.h> +#include <asm/current.h> +#include <asm/uaccess.h> +#include <linux/compat.h> +#include <linux/kernel.h> +#include <linux/security.h> +#include <linux/syscalls.h> +#include <linux/buffer_head.h> +#include <linux/capability.h> +#include <linux/quotaops.h> +#include <linux/types.h> + +/* Check validity of generic quotactl commands */ +static int generic_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t id) +{ + if (type >= MAXQUOTAS) + return -EINVAL; + if (!sb && cmd != Q_SYNC) + return -ENODEV; + /* Is operation supported? */ + if (sb && !sb->s_qcop) + return -ENOSYS; + + switch (cmd) { + case Q_GETFMT: + break; + case Q_QUOTAON: + if (!sb->s_qcop->quota_on) + return -ENOSYS; + break; + case Q_QUOTAOFF: + if (!sb->s_qcop->quota_off) + return -ENOSYS; + break; + case Q_SETINFO: + if (!sb->s_qcop->set_info) + return -ENOSYS; + break; + case Q_GETINFO: + if (!sb->s_qcop->get_info) + return -ENOSYS; + break; + case Q_SETQUOTA: + if (!sb->s_qcop->set_dqblk) + return -ENOSYS; + break; + case Q_GETQUOTA: + if (!sb->s_qcop->get_dqblk) + return -ENOSYS; + break; + case Q_SYNC: + if (sb && !sb->s_qcop->quota_sync) + return -ENOSYS; + break; + default: + return -EINVAL; + } + + /* Is quota turned on for commands which need it? */ + switch (cmd) { + case Q_GETFMT: + case Q_GETINFO: + case Q_SETINFO: + case Q_SETQUOTA: + case Q_GETQUOTA: + /* This is just informative test so we are satisfied without a lock */ + if (!sb_has_quota_active(sb, type)) + return -ESRCH; + } + + /* Check privileges */ + if (cmd == Q_GETQUOTA) { + if (((type == USRQUOTA && current_euid() != id) || + (type == GRPQUOTA && !in_egroup_p(id))) && + !capable(CAP_SYS_ADMIN)) + return -EPERM; + } + else if (cmd != Q_GETFMT && cmd != Q_SYNC && cmd != Q_GETINFO) + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + + return 0; +} + +/* Check validity of XFS Quota Manager commands */ +static int xqm_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t id) +{ + if (type >= XQM_MAXQUOTAS) + return -EINVAL; + if (!sb) + return -ENODEV; + if (!sb->s_qcop) + return -ENOSYS; + + switch (cmd) { + case Q_XQUOTAON: + case Q_XQUOTAOFF: + case Q_XQUOTARM: + if (!sb->s_qcop->set_xstate) + return -ENOSYS; + break; + case Q_XGETQSTAT: + if (!sb->s_qcop->get_xstate) + return -ENOSYS; + break; + case Q_XSETQLIM: + if (!sb->s_qcop->set_xquota) + return -ENOSYS; + break; + case Q_XGETQUOTA: + if (!sb->s_qcop->get_xquota) + return -ENOSYS; + break; + case Q_XQUOTASYNC: + if (!sb->s_qcop->quota_sync) + return -ENOSYS; + break; + default: + return -EINVAL; + } + + /* Check privileges */ + if (cmd == Q_XGETQUOTA) { + if (((type == XQM_USRQUOTA && current_euid() != id) || + (type == XQM_GRPQUOTA && !in_egroup_p(id))) && + !capable(CAP_SYS_ADMIN)) + return -EPERM; + } else if (cmd != Q_XGETQSTAT && cmd != Q_XQUOTASYNC) { + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + } + + return 0; +} + +static int check_quotactl_valid(struct super_block *sb, int type, int cmd, qid_t id) +{ + int error; + + if (XQM_COMMAND(cmd)) + error = xqm_quotactl_valid(sb, type, cmd, id); + else + error = generic_quotactl_valid(sb, type, cmd, id); + if (!error) + error = security_quotactl(cmd, type, id, sb); + return error; +} + +static void quota_sync_sb(struct super_block *sb, int type) +{ + int cnt; + + sb->s_qcop->quota_sync(sb, type); + + if (sb_dqopt(sb)->flags & DQUOT_QUOTA_SYS_FILE) + return; + /* 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); + + /* + * Now when everything is written we can discard the pagecache so + * that userspace sees the changes. + */ + mutex_lock(&sb_dqopt(sb)->dqonoff_mutex); + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (type != -1 && cnt != type) + continue; + if (!sb_has_quota_active(sb, cnt)) + continue; + mutex_lock_nested(&sb_dqopt(sb)->files[cnt]->i_mutex, I_MUTEX_QUOTA); + truncate_inode_pages(&sb_dqopt(sb)->files[cnt]->i_data, 0); + mutex_unlock(&sb_dqopt(sb)->files[cnt]->i_mutex); + } + mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); +} + +void sync_dquots(struct super_block *sb, int type) +{ + int cnt; + + if (sb) { + if (sb->s_qcop->quota_sync) + quota_sync_sb(sb, type); + return; + } + + spin_lock(&sb_lock); +restart: + list_for_each_entry(sb, &super_blocks, s_list) { + /* This test just improves performance so it needn't be reliable... */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (type != -1 && type != cnt) + continue; + if (!sb_has_quota_active(sb, cnt)) + continue; + if (!info_dirty(&sb_dqopt(sb)->info[cnt]) && + list_empty(&sb_dqopt(sb)->info[cnt].dqi_dirty_list)) + continue; + break; + } + if (cnt == MAXQUOTAS) + continue; + sb->s_count++; + spin_unlock(&sb_lock); + down_read(&sb->s_umount); + if (sb->s_root && sb->s_qcop->quota_sync) + quota_sync_sb(sb, type); + up_read(&sb->s_umount); + spin_lock(&sb_lock); + if (__put_super_and_need_restart(sb)) + goto restart; + } + spin_unlock(&sb_lock); +} + +/* Copy parameters and call proper function */ +static int do_quotactl(struct super_block *sb, int type, int cmd, qid_t id, void __user *addr) +{ + int ret; + + switch (cmd) { + case Q_QUOTAON: { + char *pathname; + + if (IS_ERR(pathname = getname(addr))) + return PTR_ERR(pathname); + ret = sb->s_qcop->quota_on(sb, type, id, pathname, 0); + putname(pathname); + return ret; + } + case Q_QUOTAOFF: + return sb->s_qcop->quota_off(sb, type, 0); + + case Q_GETFMT: { + __u32 fmt; + + down_read(&sb_dqopt(sb)->dqptr_sem); + if (!sb_has_quota_active(sb, type)) { + up_read(&sb_dqopt(sb)->dqptr_sem); + return -ESRCH; + } + fmt = sb_dqopt(sb)->info[type].dqi_format->qf_fmt_id; + up_read(&sb_dqopt(sb)->dqptr_sem); + if (copy_to_user(addr, &fmt, sizeof(fmt))) + return -EFAULT; + return 0; + } + case Q_GETINFO: { + struct if_dqinfo info; + + if ((ret = sb->s_qcop->get_info(sb, type, &info))) + return ret; + if (copy_to_user(addr, &info, sizeof(info))) + return -EFAULT; + return 0; + } + case Q_SETINFO: { + struct if_dqinfo info; + + if (copy_from_user(&info, addr, sizeof(info))) + return -EFAULT; + return sb->s_qcop->set_info(sb, type, &info); + } + case Q_GETQUOTA: { + struct if_dqblk idq; + + if ((ret = sb->s_qcop->get_dqblk(sb, type, id, &idq))) + return ret; + if (copy_to_user(addr, &idq, sizeof(idq))) + return -EFAULT; + return 0; + } + case Q_SETQUOTA: { + struct if_dqblk idq; + + if (copy_from_user(&idq, addr, sizeof(idq))) + return -EFAULT; + return sb->s_qcop->set_dqblk(sb, type, id, &idq); + } + case Q_SYNC: + sync_dquots(sb, type); + return 0; + + case Q_XQUOTAON: + case Q_XQUOTAOFF: + case Q_XQUOTARM: { + __u32 flags; + + if (copy_from_user(&flags, addr, sizeof(flags))) + return -EFAULT; + return sb->s_qcop->set_xstate(sb, flags, cmd); + } + case Q_XGETQSTAT: { + struct fs_quota_stat fqs; + + if ((ret = sb->s_qcop->get_xstate(sb, &fqs))) + return ret; + if (copy_to_user(addr, &fqs, sizeof(fqs))) + return -EFAULT; + return 0; + } + case Q_XSETQLIM: { + struct fs_disk_quota fdq; + + if (copy_from_user(&fdq, addr, sizeof(fdq))) + return -EFAULT; + return sb->s_qcop->set_xquota(sb, type, id, &fdq); + } + case Q_XGETQUOTA: { + struct fs_disk_quota fdq; + + if ((ret = sb->s_qcop->get_xquota(sb, type, id, &fdq))) + return ret; + if (copy_to_user(addr, &fdq, sizeof(fdq))) + return -EFAULT; + return 0; + } + case Q_XQUOTASYNC: + return sb->s_qcop->quota_sync(sb, type); + /* We never reach here unless validity check is broken */ + default: + BUG(); + } + return 0; +} + +/* + * look up a superblock on which quota ops will be performed + * - use the name of a block device to find the superblock thereon + */ +static inline struct super_block *quotactl_block(const char __user *special) +{ +#ifdef CONFIG_BLOCK + struct block_device *bdev; + struct super_block *sb; + char *tmp = getname(special); + + if (IS_ERR(tmp)) + return ERR_CAST(tmp); + bdev = lookup_bdev(tmp); + putname(tmp); + if (IS_ERR(bdev)) + return ERR_CAST(bdev); + sb = get_super(bdev); + bdput(bdev); + if (!sb) + return ERR_PTR(-ENODEV); + + return sb; +#else + return ERR_PTR(-ENODEV); +#endif +} + +/* + * This is the system call interface. This communicates with + * the user-level programs. Currently this only supports diskquota + * calls. Maybe we need to add the process quotas etc. in the future, + * but we probably should use rlimits for that. + */ +SYSCALL_DEFINE4(quotactl, unsigned int, cmd, const char __user *, special, + qid_t, id, void __user *, addr) +{ + uint cmds, type; + struct super_block *sb = NULL; + int ret; + + cmds = cmd >> SUBCMDSHIFT; + type = cmd & SUBCMDMASK; + + if (cmds != Q_SYNC || special) { + sb = quotactl_block(special); + if (IS_ERR(sb)) + return PTR_ERR(sb); + } + + ret = check_quotactl_valid(sb, type, cmds, id); + if (ret >= 0) + ret = do_quotactl(sb, type, cmds, id, addr); + if (sb) + drop_super(sb); + + return ret; +} + +#if defined(CONFIG_COMPAT_FOR_U64_ALIGNMENT) +/* + * This code works only for 32 bit quota tools over 64 bit OS (x86_64, ia64) + * and is necessary due to alignment problems. + */ +struct compat_if_dqblk { + compat_u64 dqb_bhardlimit; + compat_u64 dqb_bsoftlimit; + compat_u64 dqb_curspace; + compat_u64 dqb_ihardlimit; + compat_u64 dqb_isoftlimit; + compat_u64 dqb_curinodes; + compat_u64 dqb_btime; + compat_u64 dqb_itime; + compat_uint_t dqb_valid; +}; + +/* XFS structures */ +struct compat_fs_qfilestat { + compat_u64 dqb_bhardlimit; + compat_u64 qfs_nblks; + compat_uint_t qfs_nextents; +}; + +struct compat_fs_quota_stat { + __s8 qs_version; + __u16 qs_flags; + __s8 qs_pad; + struct compat_fs_qfilestat qs_uquota; + struct compat_fs_qfilestat qs_gquota; + compat_uint_t qs_incoredqs; + compat_int_t qs_btimelimit; + compat_int_t qs_itimelimit; + compat_int_t qs_rtbtimelimit; + __u16 qs_bwarnlimit; + __u16 qs_iwarnlimit; +}; + +asmlinkage long sys32_quotactl(unsigned int cmd, const char __user *special, + qid_t id, void __user *addr) +{ + unsigned int cmds; + struct if_dqblk __user *dqblk; + struct compat_if_dqblk __user *compat_dqblk; + struct fs_quota_stat __user *fsqstat; + struct compat_fs_quota_stat __user *compat_fsqstat; + compat_uint_t data; + u16 xdata; + long ret; + + cmds = cmd >> SUBCMDSHIFT; + + switch (cmds) { + case Q_GETQUOTA: + dqblk = compat_alloc_user_space(sizeof(struct if_dqblk)); + compat_dqblk = addr; + ret = sys_quotactl(cmd, special, id, dqblk); + if (ret) + break; + if (copy_in_user(compat_dqblk, dqblk, sizeof(*compat_dqblk)) || + get_user(data, &dqblk->dqb_valid) || + put_user(data, &compat_dqblk->dqb_valid)) + ret = -EFAULT; + break; + case Q_SETQUOTA: + dqblk = compat_alloc_user_space(sizeof(struct if_dqblk)); + compat_dqblk = addr; + ret = -EFAULT; + if (copy_in_user(dqblk, compat_dqblk, sizeof(*compat_dqblk)) || + get_user(data, &compat_dqblk->dqb_valid) || + put_user(data, &dqblk->dqb_valid)) + break; + ret = sys_quotactl(cmd, special, id, dqblk); + break; + case Q_XGETQSTAT: + fsqstat = compat_alloc_user_space(sizeof(struct fs_quota_stat)); + compat_fsqstat = addr; + ret = sys_quotactl(cmd, special, id, fsqstat); + if (ret) + break; + ret = -EFAULT; + /* Copying qs_version, qs_flags, qs_pad */ + if (copy_in_user(compat_fsqstat, fsqstat, + offsetof(struct compat_fs_quota_stat, qs_uquota))) + break; + /* Copying qs_uquota */ + if (copy_in_user(&compat_fsqstat->qs_uquota, + &fsqstat->qs_uquota, + sizeof(compat_fsqstat->qs_uquota)) || + get_user(data, &fsqstat->qs_uquota.qfs_nextents) || + put_user(data, &compat_fsqstat->qs_uquota.qfs_nextents)) + break; + /* Copying qs_gquota */ + if (copy_in_user(&compat_fsqstat->qs_gquota, + &fsqstat->qs_gquota, + sizeof(compat_fsqstat->qs_gquota)) || + get_user(data, &fsqstat->qs_gquota.qfs_nextents) || + put_user(data, &compat_fsqstat->qs_gquota.qfs_nextents)) + break; + /* Copying the rest */ + if (copy_in_user(&compat_fsqstat->qs_incoredqs, + &fsqstat->qs_incoredqs, + sizeof(struct compat_fs_quota_stat) - + offsetof(struct compat_fs_quota_stat, qs_incoredqs)) || + get_user(xdata, &fsqstat->qs_iwarnlimit) || + put_user(xdata, &compat_fsqstat->qs_iwarnlimit)) + break; + ret = 0; + break; + default: + ret = sys_quotactl(cmd, special, id, addr); + } + return ret; +} +#endif diff --git a/fs/quota/quota_tree.c b/fs/quota/quota_tree.c new file mode 100644 index 000000000000..953404c95b17 --- /dev/null +++ b/fs/quota/quota_tree.c @@ -0,0 +1,645 @@ +/* + * vfsv0 quota IO operations on file + */ + +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/mount.h> +#include <linux/dqblk_v2.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/quotaops.h> + +#include <asm/byteorder.h> + +#include "quota_tree.h" + +MODULE_AUTHOR("Jan Kara"); +MODULE_DESCRIPTION("Quota trie support"); +MODULE_LICENSE("GPL"); + +#define __QUOTA_QT_PARANOIA + +typedef char *dqbuf_t; + +static int get_index(struct qtree_mem_dqinfo *info, qid_t id, int depth) +{ + unsigned int epb = info->dqi_usable_bs >> 2; + + depth = info->dqi_qtree_depth - depth - 1; + while (depth--) + id /= epb; + return id % epb; +} + +/* Number of entries in one blocks */ +static inline int qtree_dqstr_in_blk(struct qtree_mem_dqinfo *info) +{ + return (info->dqi_usable_bs - sizeof(struct qt_disk_dqdbheader)) + / info->dqi_entry_size; +} + +static dqbuf_t getdqbuf(size_t size) +{ + dqbuf_t buf = kmalloc(size, GFP_NOFS); + if (!buf) + printk(KERN_WARNING "VFS: Not enough memory for quota buffers.\n"); + return buf; +} + +static inline void freedqbuf(dqbuf_t buf) +{ + kfree(buf); +} + +static inline ssize_t read_blk(struct qtree_mem_dqinfo *info, uint blk, dqbuf_t buf) +{ + struct super_block *sb = info->dqi_sb; + + memset(buf, 0, info->dqi_usable_bs); + return sb->s_op->quota_read(sb, info->dqi_type, (char *)buf, + info->dqi_usable_bs, blk << info->dqi_blocksize_bits); +} + +static inline ssize_t write_blk(struct qtree_mem_dqinfo *info, uint blk, dqbuf_t buf) +{ + struct super_block *sb = info->dqi_sb; + + return sb->s_op->quota_write(sb, info->dqi_type, (char *)buf, + info->dqi_usable_bs, blk << info->dqi_blocksize_bits); +} + +/* Remove empty block from list and return it */ +static int get_free_dqblk(struct qtree_mem_dqinfo *info) +{ + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf; + int ret, blk; + + if (!buf) + return -ENOMEM; + if (info->dqi_free_blk) { + blk = info->dqi_free_blk; + ret = read_blk(info, blk, buf); + if (ret < 0) + goto out_buf; + info->dqi_free_blk = le32_to_cpu(dh->dqdh_next_free); + } + else { + memset(buf, 0, info->dqi_usable_bs); + /* Assure block allocation... */ + ret = write_blk(info, info->dqi_blocks, buf); + if (ret < 0) + goto out_buf; + blk = info->dqi_blocks++; + } + mark_info_dirty(info->dqi_sb, info->dqi_type); + ret = blk; +out_buf: + freedqbuf(buf); + return ret; +} + +/* Insert empty block to the list */ +static int put_free_dqblk(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk) +{ + struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf; + int err; + + dh->dqdh_next_free = cpu_to_le32(info->dqi_free_blk); + dh->dqdh_prev_free = cpu_to_le32(0); + dh->dqdh_entries = cpu_to_le16(0); + err = write_blk(info, blk, buf); + if (err < 0) + return err; + info->dqi_free_blk = blk; + mark_info_dirty(info->dqi_sb, info->dqi_type); + return 0; +} + +/* Remove given block from the list of blocks with free entries */ +static int remove_free_dqentry(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk) +{ + dqbuf_t tmpbuf = getdqbuf(info->dqi_usable_bs); + struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf; + uint nextblk = le32_to_cpu(dh->dqdh_next_free); + uint prevblk = le32_to_cpu(dh->dqdh_prev_free); + int err; + + if (!tmpbuf) + return -ENOMEM; + if (nextblk) { + err = read_blk(info, nextblk, tmpbuf); + if (err < 0) + goto out_buf; + ((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_prev_free = + dh->dqdh_prev_free; + err = write_blk(info, nextblk, tmpbuf); + if (err < 0) + goto out_buf; + } + if (prevblk) { + err = read_blk(info, prevblk, tmpbuf); + if (err < 0) + goto out_buf; + ((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_next_free = + dh->dqdh_next_free; + err = write_blk(info, prevblk, tmpbuf); + if (err < 0) + goto out_buf; + } else { + info->dqi_free_entry = nextblk; + mark_info_dirty(info->dqi_sb, info->dqi_type); + } + freedqbuf(tmpbuf); + dh->dqdh_next_free = dh->dqdh_prev_free = cpu_to_le32(0); + /* No matter whether write succeeds block is out of list */ + if (write_blk(info, blk, buf) < 0) + printk(KERN_ERR "VFS: Can't write block (%u) with free entries.\n", blk); + return 0; +out_buf: + freedqbuf(tmpbuf); + return err; +} + +/* Insert given block to the beginning of list with free entries */ +static int insert_free_dqentry(struct qtree_mem_dqinfo *info, dqbuf_t buf, uint blk) +{ + dqbuf_t tmpbuf = getdqbuf(info->dqi_usable_bs); + struct qt_disk_dqdbheader *dh = (struct qt_disk_dqdbheader *)buf; + int err; + + if (!tmpbuf) + return -ENOMEM; + dh->dqdh_next_free = cpu_to_le32(info->dqi_free_entry); + dh->dqdh_prev_free = cpu_to_le32(0); + err = write_blk(info, blk, buf); + if (err < 0) + goto out_buf; + if (info->dqi_free_entry) { + err = read_blk(info, info->dqi_free_entry, tmpbuf); + if (err < 0) + goto out_buf; + ((struct qt_disk_dqdbheader *)tmpbuf)->dqdh_prev_free = + cpu_to_le32(blk); + err = write_blk(info, info->dqi_free_entry, tmpbuf); + if (err < 0) + goto out_buf; + } + freedqbuf(tmpbuf); + info->dqi_free_entry = blk; + mark_info_dirty(info->dqi_sb, info->dqi_type); + return 0; +out_buf: + freedqbuf(tmpbuf); + return err; +} + +/* Is the entry in the block free? */ +int qtree_entry_unused(struct qtree_mem_dqinfo *info, char *disk) +{ + int i; + + for (i = 0; i < info->dqi_entry_size; i++) + if (disk[i]) + return 0; + return 1; +} +EXPORT_SYMBOL(qtree_entry_unused); + +/* Find space for dquot */ +static uint find_free_dqentry(struct qtree_mem_dqinfo *info, + struct dquot *dquot, int *err) +{ + uint blk, i; + struct qt_disk_dqdbheader *dh; + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + char *ddquot; + + *err = 0; + if (!buf) { + *err = -ENOMEM; + return 0; + } + dh = (struct qt_disk_dqdbheader *)buf; + if (info->dqi_free_entry) { + blk = info->dqi_free_entry; + *err = read_blk(info, blk, buf); + if (*err < 0) + goto out_buf; + } else { + blk = get_free_dqblk(info); + if ((int)blk < 0) { + *err = blk; + freedqbuf(buf); + return 0; + } + memset(buf, 0, info->dqi_usable_bs); + /* This is enough as block is already zeroed and entry list is empty... */ + info->dqi_free_entry = blk; + mark_info_dirty(dquot->dq_sb, dquot->dq_type); + } + /* Block will be full? */ + if (le16_to_cpu(dh->dqdh_entries) + 1 >= qtree_dqstr_in_blk(info)) { + *err = remove_free_dqentry(info, buf, blk); + if (*err < 0) { + printk(KERN_ERR "VFS: find_free_dqentry(): Can't " + "remove block (%u) from entry free list.\n", + blk); + goto out_buf; + } + } + le16_add_cpu(&dh->dqdh_entries, 1); + /* Find free structure in block */ + for (i = 0, ddquot = ((char *)buf) + sizeof(struct qt_disk_dqdbheader); + i < qtree_dqstr_in_blk(info) && !qtree_entry_unused(info, ddquot); + i++, ddquot += info->dqi_entry_size); +#ifdef __QUOTA_QT_PARANOIA + if (i == qtree_dqstr_in_blk(info)) { + printk(KERN_ERR "VFS: find_free_dqentry(): Data block full " + "but it shouldn't.\n"); + *err = -EIO; + goto out_buf; + } +#endif + *err = write_blk(info, blk, buf); + if (*err < 0) { + printk(KERN_ERR "VFS: find_free_dqentry(): Can't write quota " + "data block %u.\n", blk); + goto out_buf; + } + dquot->dq_off = (blk << info->dqi_blocksize_bits) + + sizeof(struct qt_disk_dqdbheader) + + i * info->dqi_entry_size; + freedqbuf(buf); + return blk; +out_buf: + freedqbuf(buf); + return 0; +} + +/* Insert reference to structure into the trie */ +static int do_insert_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot, + uint *treeblk, int depth) +{ + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + int ret = 0, newson = 0, newact = 0; + __le32 *ref; + uint newblk; + + if (!buf) + return -ENOMEM; + if (!*treeblk) { + ret = get_free_dqblk(info); + if (ret < 0) + goto out_buf; + *treeblk = ret; + memset(buf, 0, info->dqi_usable_bs); + newact = 1; + } else { + ret = read_blk(info, *treeblk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't read tree quota block " + "%u.\n", *treeblk); + goto out_buf; + } + } + ref = (__le32 *)buf; + newblk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]); + if (!newblk) + newson = 1; + if (depth == info->dqi_qtree_depth - 1) { +#ifdef __QUOTA_QT_PARANOIA + if (newblk) { + printk(KERN_ERR "VFS: Inserting already present quota " + "entry (block %u).\n", + le32_to_cpu(ref[get_index(info, + dquot->dq_id, depth)])); + ret = -EIO; + goto out_buf; + } +#endif + newblk = find_free_dqentry(info, dquot, &ret); + } else { + ret = do_insert_tree(info, dquot, &newblk, depth+1); + } + if (newson && ret >= 0) { + ref[get_index(info, dquot->dq_id, depth)] = + cpu_to_le32(newblk); + ret = write_blk(info, *treeblk, buf); + } else if (newact && ret < 0) { + put_free_dqblk(info, buf, *treeblk); + } +out_buf: + freedqbuf(buf); + return ret; +} + +/* Wrapper for inserting quota structure into tree */ +static inline int dq_insert_tree(struct qtree_mem_dqinfo *info, + struct dquot *dquot) +{ + int tmp = QT_TREEOFF; + return do_insert_tree(info, dquot, &tmp, 0); +} + +/* + * We don't have to be afraid of deadlocks as we never have quotas on quota files... + */ +int qtree_write_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot) +{ + int type = dquot->dq_type; + struct super_block *sb = dquot->dq_sb; + ssize_t ret; + dqbuf_t ddquot = getdqbuf(info->dqi_entry_size); + + if (!ddquot) + return -ENOMEM; + + /* dq_off is guarded by dqio_mutex */ + if (!dquot->dq_off) { + ret = dq_insert_tree(info, dquot); + if (ret < 0) { + printk(KERN_ERR "VFS: Error %zd occurred while " + "creating quota.\n", ret); + freedqbuf(ddquot); + return ret; + } + } + spin_lock(&dq_data_lock); + info->dqi_ops->mem2disk_dqblk(ddquot, dquot); + spin_unlock(&dq_data_lock); + ret = sb->s_op->quota_write(sb, type, (char *)ddquot, + info->dqi_entry_size, dquot->dq_off); + if (ret != info->dqi_entry_size) { + printk(KERN_WARNING "VFS: dquota write failed on dev %s\n", + sb->s_id); + if (ret >= 0) + ret = -ENOSPC; + } else { + ret = 0; + } + dqstats.writes++; + freedqbuf(ddquot); + + return ret; +} +EXPORT_SYMBOL(qtree_write_dquot); + +/* Free dquot entry in data block */ +static int free_dqentry(struct qtree_mem_dqinfo *info, struct dquot *dquot, + uint blk) +{ + struct qt_disk_dqdbheader *dh; + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + int ret = 0; + + if (!buf) + return -ENOMEM; + if (dquot->dq_off >> info->dqi_blocksize_bits != blk) { + printk(KERN_ERR "VFS: Quota structure has offset to other " + "block (%u) than it should (%u).\n", blk, + (uint)(dquot->dq_off >> info->dqi_blocksize_bits)); + goto out_buf; + } + ret = read_blk(info, blk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't read quota data block %u\n", blk); + goto out_buf; + } + dh = (struct qt_disk_dqdbheader *)buf; + le16_add_cpu(&dh->dqdh_entries, -1); + if (!le16_to_cpu(dh->dqdh_entries)) { /* Block got free? */ + ret = remove_free_dqentry(info, buf, blk); + if (ret >= 0) + ret = put_free_dqblk(info, buf, blk); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't move quota data block (%u) " + "to free list.\n", blk); + goto out_buf; + } + } else { + memset(buf + + (dquot->dq_off & ((1 << info->dqi_blocksize_bits) - 1)), + 0, info->dqi_entry_size); + if (le16_to_cpu(dh->dqdh_entries) == + qtree_dqstr_in_blk(info) - 1) { + /* Insert will write block itself */ + ret = insert_free_dqentry(info, buf, blk); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't insert quota data " + "block (%u) to free entry list.\n", blk); + goto out_buf; + } + } else { + ret = write_blk(info, blk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't write quota data " + "block %u\n", blk); + goto out_buf; + } + } + } + dquot->dq_off = 0; /* Quota is now unattached */ +out_buf: + freedqbuf(buf); + return ret; +} + +/* Remove reference to dquot from tree */ +static int remove_tree(struct qtree_mem_dqinfo *info, struct dquot *dquot, + uint *blk, int depth) +{ + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + int ret = 0; + uint newblk; + __le32 *ref = (__le32 *)buf; + + if (!buf) + return -ENOMEM; + ret = read_blk(info, *blk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't read quota data block %u\n", *blk); + goto out_buf; + } + newblk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]); + if (depth == info->dqi_qtree_depth - 1) { + ret = free_dqentry(info, dquot, newblk); + newblk = 0; + } else { + ret = remove_tree(info, dquot, &newblk, depth+1); + } + if (ret >= 0 && !newblk) { + int i; + ref[get_index(info, dquot->dq_id, depth)] = cpu_to_le32(0); + /* Block got empty? */ + for (i = 0; + i < (info->dqi_usable_bs >> 2) && !ref[i]; + i++); + /* Don't put the root block into the free block list */ + if (i == (info->dqi_usable_bs >> 2) + && *blk != QT_TREEOFF) { + put_free_dqblk(info, buf, *blk); + *blk = 0; + } else { + ret = write_blk(info, *blk, buf); + if (ret < 0) + printk(KERN_ERR "VFS: Can't write quota tree " + "block %u.\n", *blk); + } + } +out_buf: + freedqbuf(buf); + return ret; +} + +/* Delete dquot from tree */ +int qtree_delete_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot) +{ + uint tmp = QT_TREEOFF; + + if (!dquot->dq_off) /* Even not allocated? */ + return 0; + return remove_tree(info, dquot, &tmp, 0); +} +EXPORT_SYMBOL(qtree_delete_dquot); + +/* Find entry in block */ +static loff_t find_block_dqentry(struct qtree_mem_dqinfo *info, + struct dquot *dquot, uint blk) +{ + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + loff_t ret = 0; + int i; + char *ddquot; + + if (!buf) + return -ENOMEM; + ret = read_blk(info, blk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk); + goto out_buf; + } + for (i = 0, ddquot = ((char *)buf) + sizeof(struct qt_disk_dqdbheader); + i < qtree_dqstr_in_blk(info) && !info->dqi_ops->is_id(ddquot, dquot); + i++, ddquot += info->dqi_entry_size); + if (i == qtree_dqstr_in_blk(info)) { + printk(KERN_ERR "VFS: Quota for id %u referenced " + "but not present.\n", dquot->dq_id); + ret = -EIO; + goto out_buf; + } else { + ret = (blk << info->dqi_blocksize_bits) + sizeof(struct + qt_disk_dqdbheader) + i * info->dqi_entry_size; + } +out_buf: + freedqbuf(buf); + return ret; +} + +/* Find entry for given id in the tree */ +static loff_t find_tree_dqentry(struct qtree_mem_dqinfo *info, + struct dquot *dquot, uint blk, int depth) +{ + dqbuf_t buf = getdqbuf(info->dqi_usable_bs); + loff_t ret = 0; + __le32 *ref = (__le32 *)buf; + + if (!buf) + return -ENOMEM; + ret = read_blk(info, blk, buf); + if (ret < 0) { + printk(KERN_ERR "VFS: Can't read quota tree block %u.\n", blk); + goto out_buf; + } + ret = 0; + blk = le32_to_cpu(ref[get_index(info, dquot->dq_id, depth)]); + if (!blk) /* No reference? */ + goto out_buf; + if (depth < info->dqi_qtree_depth - 1) + ret = find_tree_dqentry(info, dquot, blk, depth+1); + else + ret = find_block_dqentry(info, dquot, blk); +out_buf: + freedqbuf(buf); + return ret; +} + +/* Find entry for given id in the tree - wrapper function */ +static inline loff_t find_dqentry(struct qtree_mem_dqinfo *info, + struct dquot *dquot) +{ + return find_tree_dqentry(info, dquot, QT_TREEOFF, 0); +} + +int qtree_read_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot) +{ + int type = dquot->dq_type; + struct super_block *sb = dquot->dq_sb; + loff_t offset; + dqbuf_t ddquot; + int ret = 0; + +#ifdef __QUOTA_QT_PARANOIA + /* Invalidated quota? */ + if (!sb_dqopt(dquot->dq_sb)->files[type]) { + printk(KERN_ERR "VFS: Quota invalidated while reading!\n"); + return -EIO; + } +#endif + /* Do we know offset of the dquot entry in the quota file? */ + if (!dquot->dq_off) { + offset = find_dqentry(info, dquot); + if (offset <= 0) { /* Entry not present? */ + if (offset < 0) + printk(KERN_ERR "VFS: Can't read quota " + "structure for id %u.\n", dquot->dq_id); + dquot->dq_off = 0; + set_bit(DQ_FAKE_B, &dquot->dq_flags); + memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk)); + ret = offset; + goto out; + } + dquot->dq_off = offset; + } + ddquot = getdqbuf(info->dqi_entry_size); + if (!ddquot) + return -ENOMEM; + ret = sb->s_op->quota_read(sb, type, (char *)ddquot, + info->dqi_entry_size, dquot->dq_off); + if (ret != info->dqi_entry_size) { + if (ret >= 0) + ret = -EIO; + printk(KERN_ERR "VFS: Error while reading quota " + "structure for id %u.\n", dquot->dq_id); + set_bit(DQ_FAKE_B, &dquot->dq_flags); + memset(&dquot->dq_dqb, 0, sizeof(struct mem_dqblk)); + freedqbuf(ddquot); + goto out; + } + spin_lock(&dq_data_lock); + info->dqi_ops->disk2mem_dqblk(dquot, ddquot); + if (!dquot->dq_dqb.dqb_bhardlimit && + !dquot->dq_dqb.dqb_bsoftlimit && + !dquot->dq_dqb.dqb_ihardlimit && + !dquot->dq_dqb.dqb_isoftlimit) + set_bit(DQ_FAKE_B, &dquot->dq_flags); + spin_unlock(&dq_data_lock); + freedqbuf(ddquot); +out: + dqstats.reads++; + return ret; +} +EXPORT_SYMBOL(qtree_read_dquot); + +/* Check whether dquot should not be deleted. We know we are + * the only one operating on dquot (thanks to dq_lock) */ +int qtree_release_dquot(struct qtree_mem_dqinfo *info, struct dquot *dquot) +{ + if (test_bit(DQ_FAKE_B, &dquot->dq_flags) && !(dquot->dq_dqb.dqb_curinodes | dquot->dq_dqb.dqb_curspace)) + return qtree_delete_dquot(info, dquot); + return 0; +} +EXPORT_SYMBOL(qtree_release_dquot); diff --git a/fs/quota/quota_tree.h b/fs/quota/quota_tree.h new file mode 100644 index 000000000000..a1ab8db81a51 --- /dev/null +++ b/fs/quota/quota_tree.h @@ -0,0 +1,25 @@ +/* + * Definitions of structures for vfsv0 quota format + */ + +#ifndef _LINUX_QUOTA_TREE_H +#define _LINUX_QUOTA_TREE_H + +#include <linux/types.h> +#include <linux/quota.h> + +/* + * Structure of header of block with quota structures. It is padded to 16 bytes so + * there will be space for exactly 21 quota-entries in a block + */ +struct qt_disk_dqdbheader { + __le32 dqdh_next_free; /* Number of next block with free entry */ + __le32 dqdh_prev_free; /* Number of previous block with free entry */ + __le16 dqdh_entries; /* Number of valid entries in block */ + __le16 dqdh_pad1; + __le32 dqdh_pad2; +}; + +#define QT_TREEOFF 1 /* Offset of tree in file in blocks */ + +#endif /* _LINUX_QUOTAIO_TREE_H */ diff --git a/fs/quota/quota_v1.c b/fs/quota/quota_v1.c new file mode 100644 index 000000000000..b4af1c69ad16 --- /dev/null +++ b/fs/quota/quota_v1.c @@ -0,0 +1,218 @@ +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/quota.h> +#include <linux/quotaops.h> +#include <linux/dqblk_v1.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> + +#include <asm/byteorder.h> + +#include "quotaio_v1.h" + +MODULE_AUTHOR("Jan Kara"); +MODULE_DESCRIPTION("Old quota format support"); +MODULE_LICENSE("GPL"); + +#define QUOTABLOCK_BITS 10 +#define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS) + +static inline qsize_t v1_stoqb(qsize_t space) +{ + return (space + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS; +} + +static inline qsize_t v1_qbtos(qsize_t blocks) +{ + return blocks << QUOTABLOCK_BITS; +} + +static void v1_disk2mem_dqblk(struct mem_dqblk *m, struct v1_disk_dqblk *d) +{ + m->dqb_ihardlimit = d->dqb_ihardlimit; + m->dqb_isoftlimit = d->dqb_isoftlimit; + m->dqb_curinodes = d->dqb_curinodes; + m->dqb_bhardlimit = v1_qbtos(d->dqb_bhardlimit); + m->dqb_bsoftlimit = v1_qbtos(d->dqb_bsoftlimit); + m->dqb_curspace = v1_qbtos(d->dqb_curblocks); + m->dqb_itime = d->dqb_itime; + m->dqb_btime = d->dqb_btime; +} + +static void v1_mem2disk_dqblk(struct v1_disk_dqblk *d, struct mem_dqblk *m) +{ + d->dqb_ihardlimit = m->dqb_ihardlimit; + d->dqb_isoftlimit = m->dqb_isoftlimit; + d->dqb_curinodes = m->dqb_curinodes; + d->dqb_bhardlimit = v1_stoqb(m->dqb_bhardlimit); + d->dqb_bsoftlimit = v1_stoqb(m->dqb_bsoftlimit); + d->dqb_curblocks = v1_stoqb(m->dqb_curspace); + d->dqb_itime = m->dqb_itime; + d->dqb_btime = m->dqb_btime; +} + +static int v1_read_dqblk(struct dquot *dquot) +{ + int type = dquot->dq_type; + struct v1_disk_dqblk dqblk; + + if (!sb_dqopt(dquot->dq_sb)->files[type]) + return -EINVAL; + + /* Set structure to 0s in case read fails/is after end of file */ + memset(&dqblk, 0, sizeof(struct v1_disk_dqblk)); + dquot->dq_sb->s_op->quota_read(dquot->dq_sb, type, (char *)&dqblk, sizeof(struct v1_disk_dqblk), v1_dqoff(dquot->dq_id)); + + v1_disk2mem_dqblk(&dquot->dq_dqb, &dqblk); + if (dquot->dq_dqb.dqb_bhardlimit == 0 && dquot->dq_dqb.dqb_bsoftlimit == 0 && + dquot->dq_dqb.dqb_ihardlimit == 0 && dquot->dq_dqb.dqb_isoftlimit == 0) + set_bit(DQ_FAKE_B, &dquot->dq_flags); + dqstats.reads++; + + return 0; +} + +static int v1_commit_dqblk(struct dquot *dquot) +{ + short type = dquot->dq_type; + ssize_t ret; + struct v1_disk_dqblk dqblk; + + v1_mem2disk_dqblk(&dqblk, &dquot->dq_dqb); + if (dquot->dq_id == 0) { + dqblk.dqb_btime = sb_dqopt(dquot->dq_sb)->info[type].dqi_bgrace; + dqblk.dqb_itime = sb_dqopt(dquot->dq_sb)->info[type].dqi_igrace; + } + ret = 0; + if (sb_dqopt(dquot->dq_sb)->files[type]) + ret = dquot->dq_sb->s_op->quota_write(dquot->dq_sb, type, (char *)&dqblk, + sizeof(struct v1_disk_dqblk), v1_dqoff(dquot->dq_id)); + if (ret != sizeof(struct v1_disk_dqblk)) { + printk(KERN_WARNING "VFS: dquota write failed on dev %s\n", + dquot->dq_sb->s_id); + if (ret >= 0) + ret = -EIO; + goto out; + } + ret = 0; + +out: + dqstats.writes++; + + return ret; +} + +/* Magics of new quota format */ +#define V2_INITQMAGICS {\ + 0xd9c01f11, /* USRQUOTA */\ + 0xd9c01927 /* GRPQUOTA */\ +} + +/* Header of new quota format */ +struct v2_disk_dqheader { + __le32 dqh_magic; /* Magic number identifying file */ + __le32 dqh_version; /* File version */ +}; + +static int v1_check_quota_file(struct super_block *sb, int type) +{ + struct inode *inode = sb_dqopt(sb)->files[type]; + ulong blocks; + size_t off; + struct v2_disk_dqheader dqhead; + ssize_t size; + loff_t isize; + static const uint quota_magics[] = V2_INITQMAGICS; + + isize = i_size_read(inode); + if (!isize) + return 0; + blocks = isize >> BLOCK_SIZE_BITS; + off = isize & (BLOCK_SIZE - 1); + if ((blocks % sizeof(struct v1_disk_dqblk) * BLOCK_SIZE + off) % sizeof(struct v1_disk_dqblk)) + return 0; + /* Doublecheck whether we didn't get file with new format - with old quotactl() this could happen */ + size = sb->s_op->quota_read(sb, type, (char *)&dqhead, sizeof(struct v2_disk_dqheader), 0); + if (size != sizeof(struct v2_disk_dqheader)) + return 1; /* Probably not new format */ + if (le32_to_cpu(dqhead.dqh_magic) != quota_magics[type]) + return 1; /* Definitely not new format */ + printk(KERN_INFO "VFS: %s: Refusing to turn on old quota format on given file. It probably contains newer quota format.\n", sb->s_id); + return 0; /* Seems like a new format file -> refuse it */ +} + +static int v1_read_file_info(struct super_block *sb, int type) +{ + struct quota_info *dqopt = sb_dqopt(sb); + struct v1_disk_dqblk dqblk; + int ret; + + if ((ret = sb->s_op->quota_read(sb, type, (char *)&dqblk, sizeof(struct v1_disk_dqblk), v1_dqoff(0))) != sizeof(struct v1_disk_dqblk)) { + if (ret >= 0) + ret = -EIO; + goto out; + } + ret = 0; + /* limits are stored as unsigned 32-bit data */ + dqopt->info[type].dqi_maxblimit = 0xffffffff; + dqopt->info[type].dqi_maxilimit = 0xffffffff; + dqopt->info[type].dqi_igrace = dqblk.dqb_itime ? dqblk.dqb_itime : MAX_IQ_TIME; + dqopt->info[type].dqi_bgrace = dqblk.dqb_btime ? dqblk.dqb_btime : MAX_DQ_TIME; +out: + return ret; +} + +static int v1_write_file_info(struct super_block *sb, int type) +{ + struct quota_info *dqopt = sb_dqopt(sb); + struct v1_disk_dqblk dqblk; + int ret; + + dqopt->info[type].dqi_flags &= ~DQF_INFO_DIRTY; + if ((ret = sb->s_op->quota_read(sb, type, (char *)&dqblk, + sizeof(struct v1_disk_dqblk), v1_dqoff(0))) != sizeof(struct v1_disk_dqblk)) { + if (ret >= 0) + ret = -EIO; + goto out; + } + dqblk.dqb_itime = dqopt->info[type].dqi_igrace; + dqblk.dqb_btime = dqopt->info[type].dqi_bgrace; + ret = sb->s_op->quota_write(sb, type, (char *)&dqblk, + sizeof(struct v1_disk_dqblk), v1_dqoff(0)); + if (ret == sizeof(struct v1_disk_dqblk)) + ret = 0; + else if (ret > 0) + ret = -EIO; +out: + return ret; +} + +static struct quota_format_ops v1_format_ops = { + .check_quota_file = v1_check_quota_file, + .read_file_info = v1_read_file_info, + .write_file_info = v1_write_file_info, + .free_file_info = NULL, + .read_dqblk = v1_read_dqblk, + .commit_dqblk = v1_commit_dqblk, +}; + +static struct quota_format_type v1_quota_format = { + .qf_fmt_id = QFMT_VFS_OLD, + .qf_ops = &v1_format_ops, + .qf_owner = THIS_MODULE +}; + +static int __init init_v1_quota_format(void) +{ + return register_quota_format(&v1_quota_format); +} + +static void __exit exit_v1_quota_format(void) +{ + unregister_quota_format(&v1_quota_format); +} + +module_init(init_v1_quota_format); +module_exit(exit_v1_quota_format); + diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c new file mode 100644 index 000000000000..b618b563635c --- /dev/null +++ b/fs/quota/quota_v2.c @@ -0,0 +1,236 @@ +/* + * vfsv0 quota IO operations on file + */ + +#include <linux/errno.h> +#include <linux/fs.h> +#include <linux/mount.h> +#include <linux/dqblk_v2.h> +#include <linux/kernel.h> +#include <linux/init.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <linux/quotaops.h> + +#include <asm/byteorder.h> + +#include "quota_tree.h" +#include "quotaio_v2.h" + +MODULE_AUTHOR("Jan Kara"); +MODULE_DESCRIPTION("Quota format v2 support"); +MODULE_LICENSE("GPL"); + +#define __QUOTA_V2_PARANOIA + +static void v2_mem2diskdqb(void *dp, struct dquot *dquot); +static void v2_disk2memdqb(struct dquot *dquot, void *dp); +static int v2_is_id(void *dp, struct dquot *dquot); + +static struct qtree_fmt_operations v2_qtree_ops = { + .mem2disk_dqblk = v2_mem2diskdqb, + .disk2mem_dqblk = v2_disk2memdqb, + .is_id = v2_is_id, +}; + +#define QUOTABLOCK_BITS 10 +#define QUOTABLOCK_SIZE (1 << QUOTABLOCK_BITS) + +static inline qsize_t v2_stoqb(qsize_t space) +{ + return (space + QUOTABLOCK_SIZE - 1) >> QUOTABLOCK_BITS; +} + +static inline qsize_t v2_qbtos(qsize_t blocks) +{ + return blocks << QUOTABLOCK_BITS; +} + +/* Check whether given file is really vfsv0 quotafile */ +static int v2_check_quota_file(struct super_block *sb, int type) +{ + struct v2_disk_dqheader dqhead; + ssize_t size; + static const uint quota_magics[] = V2_INITQMAGICS; + static const uint quota_versions[] = V2_INITQVERSIONS; + + size = sb->s_op->quota_read(sb, type, (char *)&dqhead, sizeof(struct v2_disk_dqheader), 0); + if (size != sizeof(struct v2_disk_dqheader)) { + printk("quota_v2: failed read expected=%zd got=%zd\n", + sizeof(struct v2_disk_dqheader), size); + return 0; + } + if (le32_to_cpu(dqhead.dqh_magic) != quota_magics[type] || + le32_to_cpu(dqhead.dqh_version) != quota_versions[type]) + return 0; + return 1; +} + +/* Read information header from quota file */ +static int v2_read_file_info(struct super_block *sb, int type) +{ + struct v2_disk_dqinfo dinfo; + struct mem_dqinfo *info = sb_dqinfo(sb, type); + struct qtree_mem_dqinfo *qinfo; + ssize_t size; + + size = sb->s_op->quota_read(sb, type, (char *)&dinfo, + sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF); + if (size != sizeof(struct v2_disk_dqinfo)) { + printk(KERN_WARNING "Can't read info structure on device %s.\n", + sb->s_id); + return -1; + } + info->dqi_priv = kmalloc(sizeof(struct qtree_mem_dqinfo), GFP_NOFS); + if (!info->dqi_priv) { + printk(KERN_WARNING + "Not enough memory for quota information structure.\n"); + return -1; + } + qinfo = info->dqi_priv; + /* limits are stored as unsigned 32-bit data */ + info->dqi_maxblimit = 0xffffffff; + info->dqi_maxilimit = 0xffffffff; + info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace); + info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace); + info->dqi_flags = le32_to_cpu(dinfo.dqi_flags); + qinfo->dqi_sb = sb; + qinfo->dqi_type = type; + qinfo->dqi_blocks = le32_to_cpu(dinfo.dqi_blocks); + qinfo->dqi_free_blk = le32_to_cpu(dinfo.dqi_free_blk); + qinfo->dqi_free_entry = le32_to_cpu(dinfo.dqi_free_entry); + qinfo->dqi_blocksize_bits = V2_DQBLKSIZE_BITS; + qinfo->dqi_usable_bs = 1 << V2_DQBLKSIZE_BITS; + qinfo->dqi_qtree_depth = qtree_depth(qinfo); + qinfo->dqi_entry_size = sizeof(struct v2_disk_dqblk); + qinfo->dqi_ops = &v2_qtree_ops; + return 0; +} + +/* Write information header to quota file */ +static int v2_write_file_info(struct super_block *sb, int type) +{ + struct v2_disk_dqinfo dinfo; + struct mem_dqinfo *info = sb_dqinfo(sb, type); + struct qtree_mem_dqinfo *qinfo = info->dqi_priv; + ssize_t size; + + spin_lock(&dq_data_lock); + info->dqi_flags &= ~DQF_INFO_DIRTY; + dinfo.dqi_bgrace = cpu_to_le32(info->dqi_bgrace); + dinfo.dqi_igrace = cpu_to_le32(info->dqi_igrace); + dinfo.dqi_flags = cpu_to_le32(info->dqi_flags & DQF_MASK); + spin_unlock(&dq_data_lock); + dinfo.dqi_blocks = cpu_to_le32(qinfo->dqi_blocks); + dinfo.dqi_free_blk = cpu_to_le32(qinfo->dqi_free_blk); + dinfo.dqi_free_entry = cpu_to_le32(qinfo->dqi_free_entry); + size = sb->s_op->quota_write(sb, type, (char *)&dinfo, + sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF); + if (size != sizeof(struct v2_disk_dqinfo)) { + printk(KERN_WARNING "Can't write info structure on device %s.\n", + sb->s_id); + return -1; + } + return 0; +} + +static void v2_disk2memdqb(struct dquot *dquot, void *dp) +{ + struct v2_disk_dqblk *d = dp, empty; + struct mem_dqblk *m = &dquot->dq_dqb; + + m->dqb_ihardlimit = le32_to_cpu(d->dqb_ihardlimit); + m->dqb_isoftlimit = le32_to_cpu(d->dqb_isoftlimit); + m->dqb_curinodes = le32_to_cpu(d->dqb_curinodes); + m->dqb_itime = le64_to_cpu(d->dqb_itime); + m->dqb_bhardlimit = v2_qbtos(le32_to_cpu(d->dqb_bhardlimit)); + m->dqb_bsoftlimit = v2_qbtos(le32_to_cpu(d->dqb_bsoftlimit)); + m->dqb_curspace = le64_to_cpu(d->dqb_curspace); + m->dqb_btime = le64_to_cpu(d->dqb_btime); + /* We need to escape back all-zero structure */ + memset(&empty, 0, sizeof(struct v2_disk_dqblk)); + empty.dqb_itime = cpu_to_le64(1); + if (!memcmp(&empty, dp, sizeof(struct v2_disk_dqblk))) + m->dqb_itime = 0; +} + +static void v2_mem2diskdqb(void *dp, struct dquot *dquot) +{ + struct v2_disk_dqblk *d = dp; + struct mem_dqblk *m = &dquot->dq_dqb; + struct qtree_mem_dqinfo *info = + sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv; + + d->dqb_ihardlimit = cpu_to_le32(m->dqb_ihardlimit); + d->dqb_isoftlimit = cpu_to_le32(m->dqb_isoftlimit); + d->dqb_curinodes = cpu_to_le32(m->dqb_curinodes); + d->dqb_itime = cpu_to_le64(m->dqb_itime); + d->dqb_bhardlimit = cpu_to_le32(v2_stoqb(m->dqb_bhardlimit)); + d->dqb_bsoftlimit = cpu_to_le32(v2_stoqb(m->dqb_bsoftlimit)); + d->dqb_curspace = cpu_to_le64(m->dqb_curspace); + d->dqb_btime = cpu_to_le64(m->dqb_btime); + d->dqb_id = cpu_to_le32(dquot->dq_id); + if (qtree_entry_unused(info, dp)) + d->dqb_itime = cpu_to_le64(1); +} + +static int v2_is_id(void *dp, struct dquot *dquot) +{ + struct v2_disk_dqblk *d = dp; + struct qtree_mem_dqinfo *info = + sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv; + + if (qtree_entry_unused(info, dp)) + return 0; + return le32_to_cpu(d->dqb_id) == dquot->dq_id; +} + +static int v2_read_dquot(struct dquot *dquot) +{ + return qtree_read_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot); +} + +static int v2_write_dquot(struct dquot *dquot) +{ + return qtree_write_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot); +} + +static int v2_release_dquot(struct dquot *dquot) +{ + return qtree_release_dquot(sb_dqinfo(dquot->dq_sb, dquot->dq_type)->dqi_priv, dquot); +} + +static int v2_free_file_info(struct super_block *sb, int type) +{ + kfree(sb_dqinfo(sb, type)->dqi_priv); + return 0; +} + +static struct quota_format_ops v2_format_ops = { + .check_quota_file = v2_check_quota_file, + .read_file_info = v2_read_file_info, + .write_file_info = v2_write_file_info, + .free_file_info = v2_free_file_info, + .read_dqblk = v2_read_dquot, + .commit_dqblk = v2_write_dquot, + .release_dqblk = v2_release_dquot, +}; + +static struct quota_format_type v2_quota_format = { + .qf_fmt_id = QFMT_VFS_V0, + .qf_ops = &v2_format_ops, + .qf_owner = THIS_MODULE +}; + +static int __init init_v2_quota_format(void) +{ + return register_quota_format(&v2_quota_format); +} + +static void __exit exit_v2_quota_format(void) +{ + unregister_quota_format(&v2_quota_format); +} + +module_init(init_v2_quota_format); +module_exit(exit_v2_quota_format); diff --git a/fs/quota/quotaio_v1.h b/fs/quota/quotaio_v1.h new file mode 100644 index 000000000000..746654b5de70 --- /dev/null +++ b/fs/quota/quotaio_v1.h @@ -0,0 +1,33 @@ +#ifndef _LINUX_QUOTAIO_V1_H +#define _LINUX_QUOTAIO_V1_H + +#include <linux/types.h> + +/* + * The following constants define the amount of time given a user + * before the soft limits are treated as hard limits (usually resulting + * in an allocation failure). The timer is started when the user crosses + * their soft limit, it is reset when they go below their soft limit. + */ +#define MAX_IQ_TIME 604800 /* (7*24*60*60) 1 week */ +#define MAX_DQ_TIME 604800 /* (7*24*60*60) 1 week */ + +/* + * The following structure defines the format of the disk quota file + * (as it appears on disk) - the file is an array of these structures + * indexed by user or group number. + */ +struct v1_disk_dqblk { + __u32 dqb_bhardlimit; /* absolute limit on disk blks alloc */ + __u32 dqb_bsoftlimit; /* preferred limit on disk blks */ + __u32 dqb_curblocks; /* current block count */ + __u32 dqb_ihardlimit; /* absolute limit on allocated inodes */ + __u32 dqb_isoftlimit; /* preferred inode limit */ + __u32 dqb_curinodes; /* current # allocated inodes */ + time_t dqb_btime; /* time limit for excessive disk use */ + time_t dqb_itime; /* time limit for excessive inode use */ +}; + +#define v1_dqoff(UID) ((loff_t)((UID) * sizeof (struct v1_disk_dqblk))) + +#endif /* _LINUX_QUOTAIO_V1_H */ diff --git a/fs/quota/quotaio_v2.h b/fs/quota/quotaio_v2.h new file mode 100644 index 000000000000..530fe580685c --- /dev/null +++ b/fs/quota/quotaio_v2.h @@ -0,0 +1,60 @@ +/* + * Definitions of structures for vfsv0 quota format + */ + +#ifndef _LINUX_QUOTAIO_V2_H +#define _LINUX_QUOTAIO_V2_H + +#include <linux/types.h> +#include <linux/quota.h> + +/* + * Definitions of magics and versions of current quota files + */ +#define V2_INITQMAGICS {\ + 0xd9c01f11, /* USRQUOTA */\ + 0xd9c01927 /* GRPQUOTA */\ +} + +#define V2_INITQVERSIONS {\ + 0, /* USRQUOTA */\ + 0 /* GRPQUOTA */\ +} + +/* First generic header */ +struct v2_disk_dqheader { + __le32 dqh_magic; /* Magic number identifying file */ + __le32 dqh_version; /* File version */ +}; + +/* + * The following structure defines the format of the disk quota file + * (as it appears on disk) - the file is a radix tree whose leaves point + * to blocks of these structures. + */ +struct v2_disk_dqblk { + __le32 dqb_id; /* id this quota applies to */ + __le32 dqb_ihardlimit; /* absolute limit on allocated inodes */ + __le32 dqb_isoftlimit; /* preferred inode limit */ + __le32 dqb_curinodes; /* current # allocated inodes */ + __le32 dqb_bhardlimit; /* absolute limit on disk space (in QUOTABLOCK_SIZE) */ + __le32 dqb_bsoftlimit; /* preferred limit on disk space (in QUOTABLOCK_SIZE) */ + __le64 dqb_curspace; /* current space occupied (in bytes) */ + __le64 dqb_btime; /* time limit for excessive disk use */ + __le64 dqb_itime; /* time limit for excessive inode use */ +}; + +/* Header with type and version specific information */ +struct v2_disk_dqinfo { + __le32 dqi_bgrace; /* Time before block soft limit becomes hard limit */ + __le32 dqi_igrace; /* Time before inode soft limit becomes hard limit */ + __le32 dqi_flags; /* Flags for quotafile (DQF_*) */ + __le32 dqi_blocks; /* Number of blocks in file */ + __le32 dqi_free_blk; /* Number of first free block in the list */ + __le32 dqi_free_entry; /* Number of block with at least one free entry */ +}; + +#define V2_DQINFOOFF sizeof(struct v2_disk_dqheader) /* Offset of info header in file */ +#define V2_DQBLKSIZE_BITS 10 /* Size of leaf block in tree */ + +#endif /* _LINUX_QUOTAIO_V2_H */ |