aboutsummaryrefslogtreecommitdiffstats
path: root/fs/notify/inotify
diff options
context:
space:
mode:
Diffstat (limited to 'fs/notify/inotify')
-rw-r--r--fs/notify/inotify/Kconfig15
-rw-r--r--fs/notify/inotify/Makefile1
-rw-r--r--fs/notify/inotify/inotify.c873
-rw-r--r--fs/notify/inotify/inotify.h7
-rw-r--r--fs/notify/inotify/inotify_fsnotify.c151
-rw-r--r--fs/notify/inotify/inotify_user.c369
6 files changed, 338 insertions, 1078 deletions
diff --git a/fs/notify/inotify/Kconfig b/fs/notify/inotify/Kconfig
index b3a159b..b981fc0 100644
--- a/fs/notify/inotify/Kconfig
+++ b/fs/notify/inotify/Kconfig
@@ -1,18 +1,3 @@
-config INOTIFY
- bool "Inotify file change notification support"
- default n
- ---help---
- Say Y here to enable legacy in kernel inotify support. Inotify is a
- file change notification system. It is a replacement for dnotify.
- This option only provides the legacy inotify in kernel API. There
- are no in tree kernel users of this interface since it is deprecated.
- You only need this if you are loading an out of tree kernel module
- that uses inotify.
-
- For more information, see <file:Documentation/filesystems/inotify.txt>
-
- If unsure, say N.
-
config INOTIFY_USER
bool "Inotify support for userspace"
select ANON_INODES
diff --git a/fs/notify/inotify/Makefile b/fs/notify/inotify/Makefile
index 9438281..a380dab 100644
--- a/fs/notify/inotify/Makefile
+++ b/fs/notify/inotify/Makefile
@@ -1,2 +1 @@
-obj-$(CONFIG_INOTIFY) += inotify.o
obj-$(CONFIG_INOTIFY_USER) += inotify_fsnotify.o inotify_user.o
diff --git a/fs/notify/inotify/inotify.c b/fs/notify/inotify/inotify.c
deleted file mode 100644
index 27b75eb..0000000
--- a/fs/notify/inotify/inotify.c
+++ /dev/null
@@ -1,873 +0,0 @@
-/*
- * fs/inotify.c - inode-based file event notifications
- *
- * Authors:
- * John McCutchan <ttb@tentacle.dhs.org>
- * Robert Love <rml@novell.com>
- *
- * Kernel API added by: Amy Griffis <amy.griffis@hp.com>
- *
- * Copyright (C) 2005 John McCutchan
- * Copyright 2006 Hewlett-Packard Development Company, L.P.
- *
- * This program is free software; you can redistribute it and/or modify it
- * under the terms of the GNU General Public License as published by the
- * Free Software Foundation; either version 2, or (at your option) any
- * later version.
- *
- * This program is distributed in the hope that it will be useful, but
- * WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
- * General Public License for more details.
- */
-
-#include <linux/module.h>
-#include <linux/kernel.h>
-#include <linux/spinlock.h>
-#include <linux/idr.h>
-#include <linux/slab.h>
-#include <linux/fs.h>
-#include <linux/sched.h>
-#include <linux/init.h>
-#include <linux/list.h>
-#include <linux/writeback.h>
-#include <linux/inotify.h>
-#include <linux/fsnotify_backend.h>
-
-static atomic_t inotify_cookie;
-
-/*
- * Lock ordering:
- *
- * dentry->d_lock (used to keep d_move() away from dentry->d_parent)
- * iprune_mutex (synchronize shrink_icache_memory())
- * inode_lock (protects the super_block->s_inodes list)
- * inode->inotify_mutex (protects inode->inotify_watches and watches->i_list)
- * inotify_handle->mutex (protects inotify_handle and watches->h_list)
- *
- * The inode->inotify_mutex and inotify_handle->mutex and held during execution
- * of a caller's event handler. Thus, the caller must not hold any locks
- * taken in their event handler while calling any of the published inotify
- * interfaces.
- */
-
-/*
- * Lifetimes of the three main data structures--inotify_handle, inode, and
- * inotify_watch--are managed by reference count.
- *
- * inotify_handle: Lifetime is from inotify_init() to inotify_destroy().
- * Additional references can bump the count via get_inotify_handle() and drop
- * the count via put_inotify_handle().
- *
- * inotify_watch: for inotify's purposes, lifetime is from inotify_add_watch()
- * to remove_watch_no_event(). Additional references can bump the count via
- * get_inotify_watch() and drop the count via put_inotify_watch(). The caller
- * is reponsible for the final put after receiving IN_IGNORED, or when using
- * IN_ONESHOT after receiving the first event. Inotify does the final put if
- * inotify_destroy() is called.
- *
- * inode: Pinned so long as the inode is associated with a watch, from
- * inotify_add_watch() to the final put_inotify_watch().
- */
-
-/*
- * struct inotify_handle - represents an inotify instance
- *
- * This structure is protected by the mutex 'mutex'.
- */
-struct inotify_handle {
- struct idr idr; /* idr mapping wd -> watch */
- struct mutex mutex; /* protects this bad boy */
- struct list_head watches; /* list of watches */
- atomic_t count; /* reference count */
- u32 last_wd; /* the last wd allocated */
- const struct inotify_operations *in_ops; /* inotify caller operations */
-};
-
-static inline void get_inotify_handle(struct inotify_handle *ih)
-{
- atomic_inc(&ih->count);
-}
-
-static inline void put_inotify_handle(struct inotify_handle *ih)
-{
- if (atomic_dec_and_test(&ih->count)) {
- idr_destroy(&ih->idr);
- kfree(ih);
- }
-}
-
-/**
- * get_inotify_watch - grab a reference to an inotify_watch
- * @watch: watch to grab
- */
-void get_inotify_watch(struct inotify_watch *watch)
-{
- atomic_inc(&watch->count);
-}
-EXPORT_SYMBOL_GPL(get_inotify_watch);
-
-int pin_inotify_watch(struct inotify_watch *watch)
-{
- struct super_block *sb = watch->inode->i_sb;
- if (atomic_inc_not_zero(&sb->s_active)) {
- atomic_inc(&watch->count);
- return 1;
- }
- return 0;
-}
-
-/**
- * put_inotify_watch - decrements the ref count on a given watch. cleans up
- * watch references if the count reaches zero. inotify_watch is freed by
- * inotify callers via the destroy_watch() op.
- * @watch: watch to release
- */
-void put_inotify_watch(struct inotify_watch *watch)
-{
- if (atomic_dec_and_test(&watch->count)) {
- struct inotify_handle *ih = watch->ih;
-
- iput(watch->inode);
- ih->in_ops->destroy_watch(watch);
- put_inotify_handle(ih);
- }
-}
-EXPORT_SYMBOL_GPL(put_inotify_watch);
-
-void unpin_inotify_watch(struct inotify_watch *watch)
-{
- struct super_block *sb = watch->inode->i_sb;
- put_inotify_watch(watch);
- deactivate_super(sb);
-}
-
-/*
- * inotify_handle_get_wd - returns the next WD for use by the given handle
- *
- * Callers must hold ih->mutex. This function can sleep.
- */
-static int inotify_handle_get_wd(struct inotify_handle *ih,
- struct inotify_watch *watch)
-{
- int ret;
-
- do {
- if (unlikely(!idr_pre_get(&ih->idr, GFP_NOFS)))
- return -ENOSPC;
- ret = idr_get_new_above(&ih->idr, watch, ih->last_wd+1, &watch->wd);
- } while (ret == -EAGAIN);
-
- if (likely(!ret))
- ih->last_wd = watch->wd;
-
- return ret;
-}
-
-/*
- * inotify_inode_watched - returns nonzero if there are watches on this inode
- * and zero otherwise. We call this lockless, we do not care if we race.
- */
-static inline int inotify_inode_watched(struct inode *inode)
-{
- return !list_empty(&inode->inotify_watches);
-}
-
-/*
- * Get child dentry flag into synch with parent inode.
- * Flag should always be clear for negative dentrys.
- */
-static void set_dentry_child_flags(struct inode *inode, int watched)
-{
- struct dentry *alias;
-
- spin_lock(&dcache_lock);
- list_for_each_entry(alias, &inode->i_dentry, d_alias) {
- struct dentry *child;
-
- list_for_each_entry(child, &alias->d_subdirs, d_u.d_child) {
- if (!child->d_inode)
- continue;
-
- spin_lock(&child->d_lock);
- if (watched)
- child->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
- else
- child->d_flags &=~DCACHE_INOTIFY_PARENT_WATCHED;
- spin_unlock(&child->d_lock);
- }
- }
- spin_unlock(&dcache_lock);
-}
-
-/*
- * inotify_find_handle - find the watch associated with the given inode and
- * handle
- *
- * Callers must hold inode->inotify_mutex.
- */
-static struct inotify_watch *inode_find_handle(struct inode *inode,
- struct inotify_handle *ih)
-{
- struct inotify_watch *watch;
-
- list_for_each_entry(watch, &inode->inotify_watches, i_list) {
- if (watch->ih == ih)
- return watch;
- }
-
- return NULL;
-}
-
-/*
- * remove_watch_no_event - remove watch without the IN_IGNORED event.
- *
- * Callers must hold both inode->inotify_mutex and ih->mutex.
- */
-static void remove_watch_no_event(struct inotify_watch *watch,
- struct inotify_handle *ih)
-{
- list_del(&watch->i_list);
- list_del(&watch->h_list);
-
- if (!inotify_inode_watched(watch->inode))
- set_dentry_child_flags(watch->inode, 0);
-
- idr_remove(&ih->idr, watch->wd);
-}
-
-/**
- * inotify_remove_watch_locked - Remove a watch from both the handle and the
- * inode. Sends the IN_IGNORED event signifying that the inode is no longer
- * watched. May be invoked from a caller's event handler.
- * @ih: inotify handle associated with watch
- * @watch: watch to remove
- *
- * Callers must hold both inode->inotify_mutex and ih->mutex.
- */
-void inotify_remove_watch_locked(struct inotify_handle *ih,
- struct inotify_watch *watch)
-{
- remove_watch_no_event(watch, ih);
- ih->in_ops->handle_event(watch, watch->wd, IN_IGNORED, 0, NULL, NULL);
-}
-EXPORT_SYMBOL_GPL(inotify_remove_watch_locked);
-
-/* Kernel API for producing events */
-
-/*
- * inotify_d_instantiate - instantiate dcache entry for inode
- */
-void inotify_d_instantiate(struct dentry *entry, struct inode *inode)
-{
- struct dentry *parent;
-
- if (!inode)
- return;
-
- spin_lock(&entry->d_lock);
- parent = entry->d_parent;
- if (parent->d_inode && inotify_inode_watched(parent->d_inode))
- entry->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
- spin_unlock(&entry->d_lock);
-}
-
-/*
- * inotify_d_move - dcache entry has been moved
- */
-void inotify_d_move(struct dentry *entry)
-{
- struct dentry *parent;
-
- parent = entry->d_parent;
- if (inotify_inode_watched(parent->d_inode))
- entry->d_flags |= DCACHE_INOTIFY_PARENT_WATCHED;
- else
- entry->d_flags &= ~DCACHE_INOTIFY_PARENT_WATCHED;
-}
-
-/**
- * inotify_inode_queue_event - queue an event to all watches on this inode
- * @inode: inode event is originating from
- * @mask: event mask describing this event
- * @cookie: cookie for synchronization, or zero
- * @name: filename, if any
- * @n_inode: inode associated with name
- */
-void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
- const char *name, struct inode *n_inode)
-{
- struct inotify_watch *watch, *next;
-
- if (!inotify_inode_watched(inode))
- return;
-
- mutex_lock(&inode->inotify_mutex);
- list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
- u32 watch_mask = watch->mask;
- if (watch_mask & mask) {
- struct inotify_handle *ih= watch->ih;
- mutex_lock(&ih->mutex);
- if (watch_mask & IN_ONESHOT)
- remove_watch_no_event(watch, ih);
- ih->in_ops->handle_event(watch, watch->wd, mask, cookie,
- name, n_inode);
- mutex_unlock(&ih->mutex);
- }
- }
- mutex_unlock(&inode->inotify_mutex);
-}
-EXPORT_SYMBOL_GPL(inotify_inode_queue_event);
-
-/**
- * inotify_dentry_parent_queue_event - queue an event to a dentry's parent
- * @dentry: the dentry in question, we queue against this dentry's parent
- * @mask: event mask describing this event
- * @cookie: cookie for synchronization, or zero
- * @name: filename, if any
- */
-void inotify_dentry_parent_queue_event(struct dentry *dentry, u32 mask,
- u32 cookie, const char *name)
-{
- struct dentry *parent;
- struct inode *inode;
-
- if (!(dentry->d_flags & DCACHE_INOTIFY_PARENT_WATCHED))
- return;
-
- spin_lock(&dentry->d_lock);
- parent = dentry->d_parent;
- inode = parent->d_inode;
-
- if (inotify_inode_watched(inode)) {
- dget(parent);
- spin_unlock(&dentry->d_lock);
- inotify_inode_queue_event(inode, mask, cookie, name,
- dentry->d_inode);
- dput(parent);
- } else
- spin_unlock(&dentry->d_lock);
-}
-EXPORT_SYMBOL_GPL(inotify_dentry_parent_queue_event);
-
-/**
- * inotify_get_cookie - return a unique cookie for use in synchronizing events.
- */
-u32 inotify_get_cookie(void)
-{
- return atomic_inc_return(&inotify_cookie);
-}
-EXPORT_SYMBOL_GPL(inotify_get_cookie);
-
-/**
- * inotify_unmount_inodes - an sb is unmounting. handle any watched inodes.
- * @list: list of inodes being unmounted (sb->s_inodes)
- *
- * Called with inode_lock held, protecting the unmounting super block's list
- * of inodes, and with iprune_mutex held, keeping shrink_icache_memory() at bay.
- * We temporarily drop inode_lock, however, and CAN block.
- */
-void inotify_unmount_inodes(struct list_head *list)
-{
- struct inode *inode, *next_i, *need_iput = NULL;
-
- list_for_each_entry_safe(inode, next_i, list, i_sb_list) {
- struct inotify_watch *watch, *next_w;
- struct inode *need_iput_tmp;
- struct list_head *watches;
-
- /*
- * We cannot __iget() an inode in state I_CLEAR, I_FREEING,
- * I_WILL_FREE, or I_NEW which is fine because by that point
- * the inode cannot have any associated watches.
- */
- if (inode->i_state & (I_CLEAR|I_FREEING|I_WILL_FREE|I_NEW))
- continue;
-
- /*
- * If i_count is zero, the inode cannot have any watches and
- * doing an __iget/iput with MS_ACTIVE clear would actually
- * evict all inodes with zero i_count from icache which is
- * unnecessarily violent and may in fact be illegal to do.
- */
- if (!atomic_read(&inode->i_count))
- continue;
-
- need_iput_tmp = need_iput;
- need_iput = NULL;
- /* In case inotify_remove_watch_locked() drops a reference. */
- if (inode != need_iput_tmp)
- __iget(inode);
- else
- need_iput_tmp = NULL;
- /* In case the dropping of a reference would nuke next_i. */
- if ((&next_i->i_sb_list != list) &&
- atomic_read(&next_i->i_count) &&
- !(next_i->i_state & (I_CLEAR | I_FREEING |
- I_WILL_FREE))) {
- __iget(next_i);
- need_iput = next_i;
- }
-
- /*
- * We can safely drop inode_lock here because we hold
- * references on both inode and next_i. Also no new inodes
- * will be added since the umount has begun. Finally,
- * iprune_mutex keeps shrink_icache_memory() away.
- */
- spin_unlock(&inode_lock);
-
- if (need_iput_tmp)
- iput(need_iput_tmp);
-
- /* for each watch, send IN_UNMOUNT and then remove it */
- mutex_lock(&inode->inotify_mutex);
- watches = &inode->inotify_watches;
- list_for_each_entry_safe(watch, next_w, watches, i_list) {
- struct inotify_handle *ih= watch->ih;
- get_inotify_watch(watch);
- mutex_lock(&ih->mutex);
- ih->in_ops->handle_event(watch, watch->wd, IN_UNMOUNT, 0,
- NULL, NULL);
- inotify_remove_watch_locked(ih, watch);
- mutex_unlock(&ih->mutex);
- put_inotify_watch(watch);
- }
- mutex_unlock(&inode->inotify_mutex);
- iput(inode);
-
- spin_lock(&inode_lock);
- }
-}
-EXPORT_SYMBOL_GPL(inotify_unmount_inodes);
-
-/**
- * inotify_inode_is_dead - an inode has been deleted, cleanup any watches
- * @inode: inode that is about to be removed
- */
-void inotify_inode_is_dead(struct inode *inode)
-{
- struct inotify_watch *watch, *next;
-
- mutex_lock(&inode->inotify_mutex);
- list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
- struct inotify_handle *ih = watch->ih;
- mutex_lock(&ih->mutex);
- inotify_remove_watch_locked(ih, watch);
- mutex_unlock(&ih->mutex);
- }
- mutex_unlock(&inode->inotify_mutex);
-}
-EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
-
-/* Kernel Consumer API */
-
-/**
- * inotify_init - allocate and initialize an inotify instance
- * @ops: caller's inotify operations
- */
-struct inotify_handle *inotify_init(const struct inotify_operations *ops)
-{
- struct inotify_handle *ih;
-
- ih = kmalloc(sizeof(struct inotify_handle), GFP_KERNEL);
- if (unlikely(!ih))
- return ERR_PTR(-ENOMEM);
-
- idr_init(&ih->idr);
- INIT_LIST_HEAD(&ih->watches);
- mutex_init(&ih->mutex);
- ih->last_wd = 0;
- ih->in_ops = ops;
- atomic_set(&ih->count, 0);
- get_inotify_handle(ih);
-
- return ih;
-}
-EXPORT_SYMBOL_GPL(inotify_init);
-
-/**
- * inotify_init_watch - initialize an inotify watch
- * @watch: watch to initialize
- */
-void inotify_init_watch(struct inotify_watch *watch)
-{
- INIT_LIST_HEAD(&watch->h_list);
- INIT_LIST_HEAD(&watch->i_list);
- atomic_set(&watch->count, 0);
- get_inotify_watch(watch); /* initial get */
-}
-EXPORT_SYMBOL_GPL(inotify_init_watch);
-
-/*
- * Watch removals suck violently. To kick the watch out we need (in this
- * order) inode->inotify_mutex and ih->mutex. That's fine if we have
- * a hold on inode; however, for all other cases we need to make damn sure
- * we don't race with umount. We can *NOT* just grab a reference to a
- * watch - inotify_unmount_inodes() will happily sail past it and we'll end
- * with reference to inode potentially outliving its superblock. Ideally
- * we just want to grab an active reference to superblock if we can; that
- * will make sure we won't go into inotify_umount_inodes() until we are
- * done. Cleanup is just deactivate_super(). However, that leaves a messy
- * case - what if we *are* racing with umount() and active references to
- * superblock can't be acquired anymore? We can bump ->s_count, grab
- * ->s_umount, which will wait until the superblock is shut down and the
- * watch in question is pining for fjords.
- *
- * And yes, this is far beyond mere "not very pretty"; so's the entire
- * concept of inotify to start with.
- */
-
-/**
- * pin_to_kill - pin the watch down for removal
- * @ih: inotify handle
- * @watch: watch to kill
- *
- * Called with ih->mutex held, drops it. Possible return values:
- * 0 - nothing to do, it has died
- * 1 - remove it, drop the reference and deactivate_super()
- */
-static int pin_to_kill(struct inotify_handle *ih, struct inotify_watch *watch)
-{
- struct super_block *sb = watch->inode->i_sb;
-
- if (atomic_inc_not_zero(&sb->s_active)) {
- get_inotify_watch(watch);
- mutex_unlock(&ih->mutex);
- return 1; /* the best outcome */
- }
- spin_lock(&sb_lock);
- sb->s_count++;
- spin_unlock(&sb_lock);
- mutex_unlock(&ih->mutex); /* can't grab ->s_umount under it */
- down_read(&sb->s_umount);
- /* fs is already shut down; the watch is dead */
- drop_super(sb);
- return 0;
-}
-
-static void unpin_and_kill(struct inotify_watch *watch)
-{
- struct super_block *sb = watch->inode->i_sb;
- put_inotify_watch(watch);
- deactivate_super(sb);
-}
-
-/**
- * inotify_destroy - clean up and destroy an inotify instance
- * @ih: inotify handle
- */
-void inotify_destroy(struct inotify_handle *ih)
-{
- /*
- * Destroy all of the watches for this handle. Unfortunately, not very
- * pretty. We cannot do a simple iteration over the list, because we
- * do not know the inode until we iterate to the watch. But we need to
- * hold inode->inotify_mutex before ih->mutex. The following works.
- *
- * AV: it had to become even uglier to start working ;-/
- */
- while (1) {
- struct inotify_watch *watch;
- struct list_head *watches;
- struct super_block *sb;
- struct inode *inode;
-
- mutex_lock(&ih->mutex);
- watches = &ih->watches;
- if (list_empty(watches)) {
- mutex_unlock(&ih->mutex);
- break;
- }
- watch = list_first_entry(watches, struct inotify_watch, h_list);
- sb = watch->inode->i_sb;
- if (!pin_to_kill(ih, watch))
- continue;
-
- inode = watch->inode;
- mutex_lock(&inode->inotify_mutex);
- mutex_lock(&ih->mutex);
-
- /* make sure we didn't race with another list removal */
- if (likely(idr_find(&ih->idr, watch->wd))) {
- remove_watch_no_event(watch, ih);
- put_inotify_watch(watch);
- }
-
- mutex_unlock(&ih->mutex);
- mutex_unlock(&inode->inotify_mutex);
- unpin_and_kill(watch);
- }
-
- /* free this handle: the put matching the get in inotify_init() */
- put_inotify_handle(ih);
-}
-EXPORT_SYMBOL_GPL(inotify_destroy);
-
-/**
- * inotify_find_watch - find an existing watch for an (ih,inode) pair
- * @ih: inotify handle
- * @inode: inode to watch
- * @watchp: pointer to existing inotify_watch
- *
- * Caller must pin given inode (via nameidata).
- */
-s32 inotify_find_watch(struct inotify_handle *ih, struct inode *inode,
- struct inotify_watch **watchp)
-{
- struct inotify_watch *old;
- int ret = -ENOENT;
-
- mutex_lock(&inode->inotify_mutex);
- mutex_lock(&ih->mutex);
-
- old = inode_find_handle(inode, ih);
- if (unlikely(old)) {
- get_inotify_watch(old); /* caller must put watch */
- *watchp = old;
- ret = old->wd;
- }
-
- mutex_unlock(&ih->mutex);
- mutex_unlock(&inode->inotify_mutex);
-
- return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_find_watch);
-
-/**
- * inotify_find_update_watch - find and update the mask of an existing watch
- * @ih: inotify handle
- * @inode: inode's watch to update
- * @mask: mask of events to watch
- *
- * Caller must pin given inode (via nameidata).
- */
-s32 inotify_find_update_watch(struct inotify_handle *ih, struct inode *inode,
- u32 mask)
-{
- struct inotify_watch *old;
- int mask_add = 0;
- int ret;
-
- if (mask & IN_MASK_ADD)
- mask_add = 1;
-
- /* don't allow invalid bits: we don't want flags set */
- mask &= IN_ALL_EVENTS | IN_ONESHOT;
- if (unlikely(!mask))
- return -EINVAL;
-
- mutex_lock(&inode->inotify_mutex);
- mutex_lock(&ih->mutex);
-
- /*
- * Handle the case of re-adding a watch on an (inode,ih) pair that we
- * are already watching. We just update the mask and return its wd.
- */
- old = inode_find_handle(inode, ih);
- if (unlikely(!old)) {
- ret = -ENOENT;
- goto out;
- }
-
- if (mask_add)
- old->mask |= mask;
- else
- old->mask = mask;
- ret = old->wd;
-out:
- mutex_unlock(&ih->mutex);
- mutex_unlock(&inode->inotify_mutex);
- return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_find_update_watch);
-
-/**
- * inotify_add_watch - add a watch to an inotify instance
- * @ih: inotify handle
- * @watch: caller allocated watch structure
- * @inode: inode to watch
- * @mask: mask of events to watch
- *
- * Caller must pin given inode (via nameidata).
- * Caller must ensure it only calls inotify_add_watch() once per watch.
- * Calls inotify_handle_get_wd() so may sleep.
- */
-s32 inotify_add_watch(struct inotify_handle *ih, struct inotify_watch *watch,
- struct inode *inode, u32 mask)
-{
- int ret = 0;
- int newly_watched;
-
- /* don't allow invalid bits: we don't want flags set */
- mask &= IN_ALL_EVENTS | IN_ONESHOT;
- if (unlikely(!mask))
- return -EINVAL;
- watch->mask = mask;
-
- mutex_lock(&inode->inotify_mutex);
- mutex_lock(&ih->mutex);
-
- /* Initialize a new watch */
- ret = inotify_handle_get_wd(ih, watch);
- if (unlikely(ret))
- goto out;
- ret = watch->wd;
-
- /* save a reference to handle and bump the count to make it official */
- get_inotify_handle(ih);
- watch->ih = ih;
-
- /*
- * Save a reference to the inode and bump the ref count to make it
- * official. We hold a reference to nameidata, which makes this safe.
- */
- watch->inode = igrab(inode);
-
- /* Add the watch to the handle's and the inode's list */
- newly_watched = !inotify_inode_watched(inode);
- list_add(&watch->h_list, &ih->watches);
- list_add(&watch->i_list, &inode->inotify_watches);
- /*
- * Set child flags _after_ adding the watch, so there is no race
- * windows where newly instantiated children could miss their parent's
- * watched flag.
- */
- if (newly_watched)
- set_dentry_child_flags(inode, 1);
-
-out:
- mutex_unlock(&ih->mutex);
- mutex_unlock(&inode->inotify_mutex);
- return ret;
-}
-EXPORT_SYMBOL_GPL(inotify_add_watch);
-
-/**
- * inotify_clone_watch - put the watch next to existing one
- * @old: already installed watch
- * @new: new watch
- *
- * Caller must hold the inotify_mutex of inode we are dealing with;
- * it is expected to remove the old watch before unlocking the inode.
- */
-s32 inotify_clone_watch(struct inotify_watch *old, struct inotify_watch *new)
-{
- struct inotify_handle *ih = old->ih;
- int ret = 0;
-
- new->mask = old->mask;
- new->ih = ih;
-
- mutex_lock(&ih->mutex);
-
- /* Initialize a new watch */
- ret = inotify_handle_get_wd(ih, new);
- if (unlikely(ret))
- goto out;
- ret = new->wd;
-
- get_inotify_handle(ih);
-
- new->inode = igrab(old->inode);
-
- list_add(&new->h_list, &ih->watches);
- list_add(&new->i_list, &old->inode->inotify_watches);
-out:
- mutex_unlock(&ih->mutex);
- return ret;
-}
-
-void inotify_evict_watch(struct inotify_watch *watch)
-{
- get_inotify_watch(watch);
- mutex_lock(&watch->ih->mutex);
- inotify_remove_watch_locked(watch->ih, watch);
- mutex_unlock(&watch->ih->mutex);
-}
-
-/**
- * inotify_rm_wd - remove a watch from an inotify instance
- * @ih: inotify handle
- * @wd: watch descriptor to remove
- *
- * Can sleep.
- */
-int inotify_rm_wd(struct inotify_handle *ih, u32 wd)
-{
- struct inotify_watch *watch;
- struct super_block *sb;
- struct inode *inode;
-
- mutex_lock(&ih->mutex);
- watch = idr_find(&ih->idr, wd);
- if (unlikely(!watch)) {
- mutex_unlock(&ih->mutex);
- return -EINVAL;
- }
- sb = watch->inode->i_sb;
- if (!pin_to_kill(ih, watch))
- return 0;
-
- inode = watch->inode;
-
- mutex_lock(&inode->inotify_mutex);
- mutex_lock(&ih->mutex);
-
- /* make sure that we did not race */
- if (likely(idr_find(&ih->idr, wd) == watch))
- inotify_remove_watch_locked(ih, watch);
-
- mutex_unlock(&ih->mutex);
- mutex_unlock(&inode->inotify_mutex);
- unpin_and_kill(watch);
-
- return 0;
-}
-EXPORT_SYMBOL_GPL(inotify_rm_wd);
-
-/**
- * inotify_rm_watch - remove a watch from an inotify instance
- * @ih: inotify handle
- * @watch: watch to remove
- *
- * Can sleep.
- */
-int inotify_rm_watch(struct inotify_handle *ih,
- struct inotify_watch *watch)
-{
- return inotify_rm_wd(ih, watch->wd);
-}
-EXPORT_SYMBOL_GPL(inotify_rm_watch);
-
-/*
- * inotify_setup - core initialization function
- */
-static int __init inotify_setup(void)
-{
- BUILD_BUG_ON(IN_ACCESS != FS_ACCESS);
- BUILD_BUG_ON(IN_MODIFY != FS_MODIFY);
- BUILD_BUG_ON(IN_ATTRIB != FS_ATTRIB);
- BUILD_BUG_ON(IN_CLOSE_WRITE != FS_CLOSE_WRITE);
- BUILD_BUG_ON(IN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
- BUILD_BUG_ON(IN_OPEN != FS_OPEN);
- BUILD_BUG_ON(IN_MOVED_FROM != FS_MOVED_FROM);
- BUILD_BUG_ON(IN_MOVED_TO != FS_MOVED_TO);
- BUILD_BUG_ON(IN_CREATE != FS_CREATE);
- BUILD_BUG_ON(IN_DELETE != FS_DELETE);
- BUILD_BUG_ON(IN_DELETE_SELF != FS_DELETE_SELF);
- BUILD_BUG_ON(IN_MOVE_SELF != FS_MOVE_SELF);
- BUILD_BUG_ON(IN_Q_OVERFLOW != FS_Q_OVERFLOW);
-
- BUILD_BUG_ON(IN_UNMOUNT != FS_UNMOUNT);
- BUILD_BUG_ON(IN_ISDIR != FS_IN_ISDIR);
- BUILD_BUG_ON(IN_IGNORED != FS_IN_IGNORED);
- BUILD_BUG_ON(IN_ONESHOT != FS_IN_ONESHOT);
-
- atomic_set(&inotify_cookie, 0);
-
- return 0;
-}
-
-module_init(inotify_setup);
diff --git a/fs/notify/inotify/inotify.h b/fs/notify/inotify/inotify.h
index f234f3a..b6642e4 100644
--- a/fs/notify/inotify/inotify.h
+++ b/fs/notify/inotify/inotify.h
@@ -9,13 +9,12 @@ struct inotify_event_private_data {
int wd;
};
-struct inotify_inode_mark_entry {
- /* fsnotify_mark_entry MUST be the first thing */
- struct fsnotify_mark_entry fsn_entry;
+struct inotify_inode_mark {
+ struct fsnotify_mark fsn_mark;
int wd;
};
-extern void inotify_ignored_and_remove_idr(struct fsnotify_mark_entry *entry,
+extern void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
struct fsnotify_group *group);
extern void inotify_free_event_priv(struct fsnotify_event_private_data *event_priv);
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index e27960c..5e73eeb 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -22,6 +22,7 @@
* General Public License for more details.
*/
+#include <linux/dcache.h> /* d_unlinked */
#include <linux/fs.h> /* struct inode */
#include <linux/fsnotify_backend.h>
#include <linux/inotify.h>
@@ -32,26 +33,84 @@
#include "inotify.h"
-static int inotify_handle_event(struct fsnotify_group *group, struct fsnotify_event *event)
+/*
+ * Check if 2 events contain the same information. We do not compare private data
+ * but at this moment that isn't a problem for any know fsnotify listeners.
+ */
+static bool event_compare(struct fsnotify_event *old, struct fsnotify_event *new)
+{
+ if ((old->mask == new->mask) &&
+ (old->to_tell == new->to_tell) &&
+ (old->data_type == new->data_type) &&
+ (old->name_len == new->name_len)) {
+ switch (old->data_type) {
+ case (FSNOTIFY_EVENT_INODE):
+ /* remember, after old was put on the wait_q we aren't
+ * allowed to look at the inode any more, only thing
+ * left to check was if the file_name is the same */
+ if (!old->name_len ||
+ !strcmp(old->file_name, new->file_name))
+ return true;
+ break;
+ case (FSNOTIFY_EVENT_FILE):
+ if ((old->file->f_path.mnt == new->file->f_path.mnt) &&
+ (old->file->f_path.dentry == new->file->f_path.dentry))
+ return true;
+ break;
+ case (FSNOTIFY_EVENT_NONE):
+ if (old->mask & FS_Q_OVERFLOW)
+ return true;
+ else if (old->mask & FS_IN_IGNORED)
+ return false;
+ return true;
+ };
+ }
+ return false;
+}
+
+static struct fsnotify_event *inotify_merge(struct list_head *list,
+ struct fsnotify_event *event)
{
- struct fsnotify_mark_entry *entry;
- struct inotify_inode_mark_entry *ientry;
+ struct fsnotify_event_holder *last_holder;
+ struct fsnotify_event *last_event;
+
+ /* and the list better be locked by something too */
+ spin_lock(&event->lock);
+
+ last_holder = list_entry(list->prev, struct fsnotify_event_holder, event_list);
+ last_event = last_holder->event;
+ if (event_compare(last_event, event))
+ fsnotify_get_event(last_event);
+ else
+ last_event = NULL;
+
+ spin_unlock(&event->lock);
+
+ return last_event;
+}
+
+static int inotify_handle_event(struct fsnotify_group *group,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmount_mark,
+ struct fsnotify_event *event)
+{
+ struct inotify_inode_mark *i_mark;
struct inode *to_tell;
struct inotify_event_private_data *event_priv;
struct fsnotify_event_private_data *fsn_event_priv;
- int wd, ret;
+ struct fsnotify_event *added_event;
+ int wd, ret = 0;
+
+ BUG_ON(vfsmount_mark);
+
+ pr_debug("%s: group=%p event=%p to_tell=%p mask=%x\n", __func__, group,
+ event, event->to_tell, event->mask);
to_tell = event->to_tell;
- spin_lock(&to_tell->i_lock);
- entry = fsnotify_find_mark_entry(group, to_tell);
- spin_unlock(&to_tell->i_lock);
- /* race with watch removal? We already passes should_send */
- if (unlikely(!entry))
- return 0;
- ientry = container_of(entry, struct inotify_inode_mark_entry,
- fsn_entry);
- wd = ientry->wd;
+ i_mark = container_of(inode_mark, struct inotify_inode_mark,
+ fsn_mark);
+ wd = i_mark->wd;
event_priv = kmem_cache_alloc(event_priv_cachep, GFP_KERNEL);
if (unlikely(!event_priv))
@@ -62,48 +121,40 @@ static int inotify_handle_event(struct fsnotify_group *group, struct fsnotify_ev
fsn_event_priv->group = group;
event_priv->wd = wd;
- ret = fsnotify_add_notify_event(group, event, fsn_event_priv);
- if (ret) {
+ added_event = fsnotify_add_notify_event(group, event, fsn_event_priv, inotify_merge);
+ if (added_event) {
inotify_free_event_priv(fsn_event_priv);
- /* EEXIST says we tail matched, EOVERFLOW isn't something
- * to report up the stack. */
- if ((ret == -EEXIST) ||
- (ret == -EOVERFLOW))
- ret = 0;
+ if (!IS_ERR(added_event))
+ fsnotify_put_event(added_event);
+ else
+ ret = PTR_ERR(added_event);
}
- /*
- * If we hold the entry until after the event is on the queue
- * IN_IGNORED won't be able to pass this event in the queue
- */
- fsnotify_put_mark(entry);
+ if (inode_mark->mask & IN_ONESHOT)
+ fsnotify_destroy_mark(inode_mark);
return ret;
}
-static void inotify_freeing_mark(struct fsnotify_mark_entry *entry, struct fsnotify_group *group)
+static void inotify_freeing_mark(struct fsnotify_mark *fsn_mark, struct fsnotify_group *group)
{
- inotify_ignored_and_remove_idr(entry, group);
+ inotify_ignored_and_remove_idr(fsn_mark, group);
}
-static bool inotify_should_send_event(struct fsnotify_group *group, struct inode *inode, __u32 mask)
+static bool inotify_should_send_event(struct fsnotify_group *group, struct inode *inode,
+ struct fsnotify_mark *inode_mark,
+ struct fsnotify_mark *vfsmount_mark,
+ __u32 mask, void *data, int data_type)
{
- struct fsnotify_mark_entry *entry;
- bool send;
-
- spin_lock(&inode->i_lock);
- entry = fsnotify_find_mark_entry(group, inode);
- spin_unlock(&inode->i_lock);
- if (!entry)
- return false;
+ if ((inode_mark->mask & FS_EXCL_UNLINK) &&
+ (data_type == FSNOTIFY_EVENT_FILE)) {
+ struct file *file = data;
- mask = (mask & ~FS_EVENT_ON_CHILD);
- send = (entry->mask & mask);
-
- /* find took a reference */
- fsnotify_put_mark(entry);
+ if (d_unlinked(file->f_path.dentry))
+ return false;
+ }
- return send;
+ return true;
}
/*
@@ -115,18 +166,18 @@ static bool inotify_should_send_event(struct fsnotify_group *group, struct inode
*/
static int idr_callback(int id, void *p, void *data)
{
- struct fsnotify_mark_entry *entry;
- struct inotify_inode_mark_entry *ientry;
+ struct fsnotify_mark *fsn_mark;
+ struct inotify_inode_mark *i_mark;
static bool warned = false;
if (warned)
return 0;
warned = true;
- entry = p;
- ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
+ fsn_mark = p;
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
- WARN(1, "inotify closing but id=%d for entry=%p in group=%p still in "
+ WARN(1, "inotify closing but id=%d for fsn_mark=%p in group=%p still in "
"idr. Probably leaking memory\n", id, p, data);
/*
@@ -135,9 +186,9 @@ static int idr_callback(int id, void *p, void *data)
* out why we got here and the panic is no worse than the original
* BUG() that was here.
*/
- if (entry)
- printk(KERN_WARNING "entry->group=%p inode=%p wd=%d\n",
- entry->group, entry->inode, ientry->wd);
+ if (fsn_mark)
+ printk(KERN_WARNING "fsn_mark->group=%p inode=%p wd=%d\n",
+ fsn_mark->group, fsn_mark->i.inode, i_mark->wd);
return 0;
}
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index e46ca68..bf7f6d7 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -46,17 +46,11 @@
/* these are configurable via /proc/sys/fs/inotify/ */
static int inotify_max_user_instances __read_mostly;
static int inotify_max_queued_events __read_mostly;
-int inotify_max_user_watches __read_mostly;
+static int inotify_max_user_watches __read_mostly;
static struct kmem_cache *inotify_inode_mark_cachep __read_mostly;
struct kmem_cache *event_priv_cachep __read_mostly;
-/*
- * When inotify registers a new group it increments this and uses that
- * value as an offset to set the fsnotify group "name" and priority.
- */
-static atomic_t inotify_grp_num;
-
#ifdef CONFIG_SYSCTL
#include <linux/sysctl.h>
@@ -96,11 +90,14 @@ static inline __u32 inotify_arg_to_mask(u32 arg)
{
__u32 mask;
- /* everything should accept their own ignored and cares about children */
- mask = (FS_IN_IGNORED | FS_EVENT_ON_CHILD);
+ /*
+ * everything should accept their own ignored, cares about children,
+ * and should receive events when the inode is unmounted
+ */
+ mask = (FS_IN_IGNORED | FS_EVENT_ON_CHILD | FS_UNMOUNT);
/* mask off the flags used to open the fd */
- mask |= (arg & (IN_ALL_EVENTS | IN_ONESHOT));
+ mask |= (arg & (IN_ALL_EVENTS | IN_ONESHOT | IN_EXCL_UNLINK));
return mask;
}
@@ -144,6 +141,8 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
event = fsnotify_peek_notify_event(group);
+ pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
if (event->name_len)
event_size += roundup(event->name_len + 1, event_size);
@@ -173,6 +172,8 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
size_t event_size = sizeof(struct inotify_event);
size_t name_len = 0;
+ pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+
/* we get the inotify watch descriptor from the event private data */
spin_lock(&event->lock);
fsn_priv = fsnotify_remove_priv_from_event(group, event);
@@ -245,6 +246,8 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
kevent = get_one_event(group, count);
mutex_unlock(&group->notification_mutex);
+ pr_debug("%s: group=%p kevent=%p\n", __func__, group, kevent);
+
if (kevent) {
ret = PTR_ERR(kevent);
if (IS_ERR(kevent))
@@ -289,6 +292,8 @@ static int inotify_release(struct inode *ignored, struct file *file)
struct fsnotify_group *group = file->private_data;
struct user_struct *user = group->inotify_data.user;
+ pr_debug("%s: group=%p\n", __func__, group);
+
fsnotify_clear_marks_by_group(group);
/* free this group, matching get was inotify_init->fsnotify_obtain_group */
@@ -312,6 +317,8 @@ static long inotify_ioctl(struct file *file, unsigned int cmd,
group = file->private_data;
p = (void __user *) arg;
+ pr_debug("%s: group=%p cmd=%u\n", __func__, group, cmd);
+
switch (cmd) {
case FIONREAD:
mutex_lock(&group->notification_mutex);
@@ -357,59 +364,159 @@ static int inotify_find_inode(const char __user *dirname, struct path *path, uns
return error;
}
+static int inotify_add_to_idr(struct idr *idr, spinlock_t *idr_lock,
+ int *last_wd,
+ struct inotify_inode_mark *i_mark)
+{
+ int ret;
+
+ do {
+ if (unlikely(!idr_pre_get(idr, GFP_KERNEL)))
+ return -ENOMEM;
+
+ spin_lock(idr_lock);
+ ret = idr_get_new_above(idr, i_mark, *last_wd + 1,
+ &i_mark->wd);
+ /* we added the mark to the idr, take a reference */
+ if (!ret) {
+ *last_wd = i_mark->wd;
+ fsnotify_get_mark(&i_mark->fsn_mark);
+ }
+ spin_unlock(idr_lock);
+ } while (ret == -EAGAIN);
+
+ return ret;
+}
+
+static struct inotify_inode_mark *inotify_idr_find_locked(struct fsnotify_group *group,
+ int wd)
+{
+ struct idr *idr = &group->inotify_data.idr;
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+ struct inotify_inode_mark *i_mark;
+
+ assert_spin_locked(idr_lock);
+
+ i_mark = idr_find(idr, wd);
+ if (i_mark) {
+ struct fsnotify_mark *fsn_mark = &i_mark->fsn_mark;
+
+ fsnotify_get_mark(fsn_mark);
+ /* One ref for being in the idr, one ref we just took */
+ BUG_ON(atomic_read(&fsn_mark->refcnt) < 2);
+ }
+
+ return i_mark;
+}
+
+static struct inotify_inode_mark *inotify_idr_find(struct fsnotify_group *group,
+ int wd)
+{
+ struct inotify_inode_mark *i_mark;
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+
+ spin_lock(idr_lock);
+ i_mark = inotify_idr_find_locked(group, wd);
+ spin_unlock(idr_lock);
+
+ return i_mark;
+}
+
+static void do_inotify_remove_from_idr(struct fsnotify_group *group,
+ struct inotify_inode_mark *i_mark)
+{
+ struct idr *idr = &group->inotify_data.idr;
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+ int wd = i_mark->wd;
+
+ assert_spin_locked(idr_lock);
+
+ idr_remove(idr, wd);
+
+ /* removed from the idr, drop that ref */
+ fsnotify_put_mark(&i_mark->fsn_mark);
+}
+
/*
* Remove the mark from the idr (if present) and drop the reference
* on the mark because it was in the idr.
*/
static void inotify_remove_from_idr(struct fsnotify_group *group,
- struct inotify_inode_mark_entry *ientry)
+ struct inotify_inode_mark *i_mark)
{
- struct idr *idr;
- struct fsnotify_mark_entry *entry;
- struct inotify_inode_mark_entry *found_ientry;
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+ struct inotify_inode_mark *found_i_mark = NULL;
int wd;
- spin_lock(&group->inotify_data.idr_lock);
- idr = &group->inotify_data.idr;
- wd = ientry->wd;
+ spin_lock(idr_lock);
+ wd = i_mark->wd;
- if (wd == -1)
+ /*
+ * does this i_mark think it is in the idr? we shouldn't get called
+ * if it wasn't....
+ */
+ if (wd == -1) {
+ WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+ " i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+ i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
goto out;
+ }
- entry = idr_find(&group->inotify_data.idr, wd);
- if (unlikely(!entry))
+ /* Lets look in the idr to see if we find it */
+ found_i_mark = inotify_idr_find_locked(group, wd);
+ if (unlikely(!found_i_mark)) {
+ WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+ " i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+ i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
goto out;
+ }
- found_ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
- if (unlikely(found_ientry != ientry)) {
- /* We found an entry in the idr with the right wd, but it's
- * not the entry we were told to remove. eparis seriously
- * fucked up somewhere. */
- WARN_ON(1);
- ientry->wd = -1;
+ /*
+ * We found an mark in the idr at the right wd, but it's
+ * not the mark we were told to remove. eparis seriously
+ * fucked up somewhere.
+ */
+ if (unlikely(found_i_mark != i_mark)) {
+ WARN_ONCE(1, "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p "
+ "mark->inode=%p found_i_mark=%p found_i_mark->wd=%d "
+ "found_i_mark->group=%p found_i_mark->inode=%p\n",
+ __func__, i_mark, i_mark->wd, i_mark->fsn_mark.group,
+ i_mark->fsn_mark.i.inode, found_i_mark, found_i_mark->wd,
+ found_i_mark->fsn_mark.group,
+ found_i_mark->fsn_mark.i.inode);
goto out;
}
- /* One ref for being in the idr, one ref held by the caller */
- BUG_ON(atomic_read(&entry->refcnt) < 2);
-
- idr_remove(idr, wd);
- ientry->wd = -1;
+ /*
+ * One ref for being in the idr
+ * one ref held by the caller trying to kill us
+ * one ref grabbed by inotify_idr_find
+ */
+ if (unlikely(atomic_read(&i_mark->fsn_mark.refcnt) < 3)) {
+ printk(KERN_ERR "%s: i_mark=%p i_mark->wd=%d i_mark->group=%p"
+ " i_mark->inode=%p\n", __func__, i_mark, i_mark->wd,
+ i_mark->fsn_mark.group, i_mark->fsn_mark.i.inode);
+ /* we can't really recover with bad ref cnting.. */
+ BUG();
+ }
- /* removed from the idr, drop that ref */
- fsnotify_put_mark(entry);
+ do_inotify_remove_from_idr(group, i_mark);
out:
- spin_unlock(&group->inotify_data.idr_lock);
+ /* match the ref taken by inotify_idr_find_locked() */
+ if (found_i_mark)
+ fsnotify_put_mark(&found_i_mark->fsn_mark);
+ i_mark->wd = -1;
+ spin_unlock(idr_lock);
}
/*
* Send IN_IGNORED for this wd, remove this wd from the idr.
*/
-void inotify_ignored_and_remove_idr(struct fsnotify_mark_entry *entry,
+void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
struct fsnotify_group *group)
{
- struct inotify_inode_mark_entry *ientry;
- struct fsnotify_event *ignored_event;
+ struct inotify_inode_mark *i_mark;
+ struct fsnotify_event *ignored_event, *notify_event;
struct inotify_event_private_data *event_priv;
struct fsnotify_event_private_data *fsn_event_priv;
int ret;
@@ -420,7 +527,7 @@ void inotify_ignored_and_remove_idr(struct fsnotify_mark_entry *entry,
if (!ignored_event)
return;
- ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
event_priv = kmem_cache_alloc(event_priv_cachep, GFP_NOFS);
if (unlikely(!event_priv))
@@ -429,37 +536,44 @@ void inotify_ignored_and_remove_idr(struct fsnotify_mark_entry *entry,
fsn_event_priv = &event_priv->fsnotify_event_priv_data;
fsn_event_priv->group = group;
- event_priv->wd = ientry->wd;
-
- ret = fsnotify_add_notify_event(group, ignored_event, fsn_event_priv);
- if (ret)
+ event_priv->wd = i_mark->wd;
+
+ notify_event = fsnotify_add_notify_event(group, ignored_event, fsn_event_priv, NULL);
+ if (notify_event) {
+ if (IS_ERR(notify_event))
+ ret = PTR_ERR(notify_event);
+ else
+ fsnotify_put_event(notify_event);
inotify_free_event_priv(fsn_event_priv);
+ }
skip_send_ignore:
/* matches the reference taken when the event was created */
fsnotify_put_event(ignored_event);
- /* remove this entry from the idr */
- inotify_remove_from_idr(group, ientry);
+ /* remove this mark from the idr */
+ inotify_remove_from_idr(group, i_mark);
atomic_dec(&group->inotify_data.user->inotify_watches);
}
/* ding dong the mark is dead */
-static void inotify_free_mark(struct fsnotify_mark_entry *entry)
+static void inotify_free_mark(struct fsnotify_mark *fsn_mark)
{
- struct inotify_inode_mark_entry *ientry = (struct inotify_inode_mark_entry *)entry;
+ struct inotify_inode_mark *i_mark;
+
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
- kmem_cache_free(inotify_inode_mark_cachep, ientry);
+ kmem_cache_free(inotify_inode_mark_cachep, i_mark);
}
static int inotify_update_existing_watch(struct fsnotify_group *group,
struct inode *inode,
u32 arg)
{
- struct fsnotify_mark_entry *entry;
- struct inotify_inode_mark_entry *ientry;
+ struct fsnotify_mark *fsn_mark;
+ struct inotify_inode_mark *i_mark;
__u32 old_mask, new_mask;
__u32 mask;
int add = (arg & IN_MASK_ADD);
@@ -467,52 +581,43 @@ static int inotify_update_existing_watch(struct fsnotify_group *group,
/* don't allow invalid bits: we don't want flags set */
mask = inotify_arg_to_mask(arg);
- if (unlikely(!mask))
+ if (unlikely(!(mask & IN_ALL_EVENTS)))
return -EINVAL;
- spin_lock(&inode->i_lock);
- entry = fsnotify_find_mark_entry(group, inode);
- spin_unlock(&inode->i_lock);
- if (!entry)
+ fsn_mark = fsnotify_find_inode_mark(group, inode);
+ if (!fsn_mark)
return -ENOENT;
- ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
- spin_lock(&entry->lock);
+ spin_lock(&fsn_mark->lock);
- old_mask = entry->mask;
- if (add) {
- entry->mask |= mask;
- new_mask = entry->mask;
- } else {
- entry->mask = mask;
- new_mask = entry->mask;
- }
+ old_mask = fsn_mark->mask;
+ if (add)
+ fsnotify_set_mark_mask_locked(fsn_mark, (fsn_mark->mask | mask));
+ else
+ fsnotify_set_mark_mask_locked(fsn_mark, mask);
+ new_mask = fsn_mark->mask;
- spin_unlock(&entry->lock);
+ spin_unlock(&fsn_mark->lock);
if (old_mask != new_mask) {
/* more bits in old than in new? */
int dropped = (old_mask & ~new_mask);
- /* more bits in this entry than the inode's mask? */
+ /* more bits in this fsn_mark than the inode's mask? */
int do_inode = (new_mask & ~inode->i_fsnotify_mask);
- /* more bits in this entry than the group? */
- int do_group = (new_mask & ~group->mask);
- /* update the inode with this new entry */
+ /* update the inode with this new fsn_mark */
if (dropped || do_inode)
fsnotify_recalc_inode_mask(inode);
- /* update the group mask with the new mask */
- if (dropped || do_group)
- fsnotify_recalc_group_mask(group);
}
/* return the wd */
- ret = ientry->wd;
+ ret = i_mark->wd;
- /* match the get from fsnotify_find_mark_entry() */
- fsnotify_put_mark(entry);
+ /* match the get from fsnotify_find_mark() */
+ fsnotify_put_mark(fsn_mark);
return ret;
}
@@ -521,73 +626,51 @@ static int inotify_new_watch(struct fsnotify_group *group,
struct inode *inode,
u32 arg)
{
- struct inotify_inode_mark_entry *tmp_ientry;
+ struct inotify_inode_mark *tmp_i_mark;
__u32 mask;
int ret;
+ struct idr *idr = &group->inotify_data.idr;
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
/* don't allow invalid bits: we don't want flags set */
mask = inotify_arg_to_mask(arg);
- if (unlikely(!mask))
+ if (unlikely(!(mask & IN_ALL_EVENTS)))
return -EINVAL;
- tmp_ientry = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
- if (unlikely(!tmp_ientry))
+ tmp_i_mark = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
+ if (unlikely(!tmp_i_mark))
return -ENOMEM;
- fsnotify_init_mark(&tmp_ientry->fsn_entry, inotify_free_mark);
- tmp_ientry->fsn_entry.mask = mask;
- tmp_ientry->wd = -1;
+ fsnotify_init_mark(&tmp_i_mark->fsn_mark, inotify_free_mark);
+ tmp_i_mark->fsn_mark.mask = mask;
+ tmp_i_mark->wd = -1;
ret = -ENOSPC;
if (atomic_read(&group->inotify_data.user->inotify_watches) >= inotify_max_user_watches)
goto out_err;
-retry:
- ret = -ENOMEM;
- if (unlikely(!idr_pre_get(&group->inotify_data.idr, GFP_KERNEL)))
- goto out_err;
- /* we are putting the mark on the idr, take a reference */
- fsnotify_get_mark(&tmp_ientry->fsn_entry);
-
- spin_lock(&group->inotify_data.idr_lock);
- ret = idr_get_new_above(&group->inotify_data.idr, &tmp_ientry->fsn_entry,
- group->inotify_data.last_wd+1,
- &tmp_ientry->wd);
- spin_unlock(&group->inotify_data.idr_lock);
- if (ret) {
- /* we didn't get on the idr, drop the idr reference */
- fsnotify_put_mark(&tmp_ientry->fsn_entry);
-
- /* idr was out of memory allocate and try again */
- if (ret == -EAGAIN)
- goto retry;
+ ret = inotify_add_to_idr(idr, idr_lock, &group->inotify_data.last_wd,
+ tmp_i_mark);
+ if (ret)
goto out_err;
- }
/* we are on the idr, now get on the inode */
- ret = fsnotify_add_mark(&tmp_ientry->fsn_entry, group, inode);
+ ret = fsnotify_add_mark(&tmp_i_mark->fsn_mark, group, inode, NULL, 0);
if (ret) {
/* we failed to get on the inode, get off the idr */
- inotify_remove_from_idr(group, tmp_ientry);
+ inotify_remove_from_idr(group, tmp_i_mark);
goto out_err;
}
- /* update the idr hint, who cares about races, it's just a hint */
- group->inotify_data.last_wd = tmp_ientry->wd;
-
/* increment the number of watches the user has */
atomic_inc(&group->inotify_data.user->inotify_watches);
- /* return the watch descriptor for this new entry */
- ret = tmp_ientry->wd;
-
- /* if this mark added a new event update the group mask */
- if (mask & ~group->mask)
- fsnotify_recalc_group_mask(group);
+ /* return the watch descriptor for this new mark */
+ ret = tmp_i_mark->wd;
out_err:
- /* match the ref from fsnotify_init_markentry() */
- fsnotify_put_mark(&tmp_ientry->fsn_entry);
+ /* match the ref from fsnotify_init_mark() */
+ fsnotify_put_mark(&tmp_i_mark->fsn_mark);
return ret;
}
@@ -616,11 +699,8 @@ retry:
static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsigned int max_events)
{
struct fsnotify_group *group;
- unsigned int grp_num;
- /* fsnotify_obtain_group took a reference to group, we put this when we kill the file in the end */
- grp_num = (INOTIFY_GROUP_NUM - atomic_inc_return(&inotify_grp_num));
- group = fsnotify_obtain_group(grp_num, 0, &inotify_fsnotify_ops);
+ group = fsnotify_alloc_group(&inotify_fsnotify_ops);
if (IS_ERR(group))
return group;
@@ -726,7 +806,7 @@ fput_and_out:
SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd)
{
struct fsnotify_group *group;
- struct fsnotify_mark_entry *entry;
+ struct inotify_inode_mark *i_mark;
struct file *filp;
int ret = 0, fput_needed;
@@ -735,25 +815,23 @@ SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd)
return -EBADF;
/* verify that this is indeed an inotify instance */
- if (unlikely(filp->f_op != &inotify_fops)) {
- ret = -EINVAL;
+ ret = -EINVAL;
+ if (unlikely(filp->f_op != &inotify_fops))
goto out;
- }
group = filp->private_data;
- spin_lock(&group->inotify_data.idr_lock);
- entry = idr_find(&group->inotify_data.idr, wd);
- if (unlikely(!entry)) {
- spin_unlock(&group->inotify_data.idr_lock);
- ret = -EINVAL;
+ ret = -EINVAL;
+ i_mark = inotify_idr_find(group, wd);
+ if (unlikely(!i_mark))
goto out;
- }
- fsnotify_get_mark(entry);
- spin_unlock(&group->inotify_data.idr_lock);
- fsnotify_destroy_mark_by_entry(entry);
- fsnotify_put_mark(entry);
+ ret = 0;
+
+ fsnotify_destroy_mark(&i_mark->fsn_mark);
+
+ /* match ref taken by inotify_idr_find */
+ fsnotify_put_mark(&i_mark->fsn_mark);
out:
fput_light(filp, fput_needed);
@@ -767,7 +845,28 @@ out:
*/
static int __init inotify_user_setup(void)
{
- inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark_entry, SLAB_PANIC);
+ BUILD_BUG_ON(IN_ACCESS != FS_ACCESS);
+ BUILD_BUG_ON(IN_MODIFY != FS_MODIFY);
+ BUILD_BUG_ON(IN_ATTRIB != FS_ATTRIB);
+ BUILD_BUG_ON(IN_CLOSE_WRITE != FS_CLOSE_WRITE);
+ BUILD_BUG_ON(IN_CLOSE_NOWRITE != FS_CLOSE_NOWRITE);
+ BUILD_BUG_ON(IN_OPEN != FS_OPEN);
+ BUILD_BUG_ON(IN_MOVED_FROM != FS_MOVED_FROM);
+ BUILD_BUG_ON(IN_MOVED_TO != FS_MOVED_TO);
+ BUILD_BUG_ON(IN_CREATE != FS_CREATE);
+ BUILD_BUG_ON(IN_DELETE != FS_DELETE);
+ BUILD_BUG_ON(IN_DELETE_SELF != FS_DELETE_SELF);
+ BUILD_BUG_ON(IN_MOVE_SELF != FS_MOVE_SELF);
+ BUILD_BUG_ON(IN_UNMOUNT != FS_UNMOUNT);
+ BUILD_BUG_ON(IN_Q_OVERFLOW != FS_Q_OVERFLOW);
+ BUILD_BUG_ON(IN_IGNORED != FS_IN_IGNORED);
+ BUILD_BUG_ON(IN_EXCL_UNLINK != FS_EXCL_UNLINK);
+ BUILD_BUG_ON(IN_ISDIR != FS_IN_ISDIR);
+ BUILD_BUG_ON(IN_ONESHOT != FS_IN_ONESHOT);
+
+ BUG_ON(hweight32(ALL_INOTIFY_BITS) != 21);
+
+ inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark, SLAB_PANIC);
event_priv_cachep = KMEM_CACHE(inotify_event_private_data, SLAB_PANIC);
inotify_max_queued_events = 16384;