aboutsummaryrefslogtreecommitdiff
path: root/fs/notify/inotify
diff options
context:
space:
mode:
Diffstat (limited to 'fs/notify/inotify')
-rw-r--r--fs/notify/inotify/Kconfig18
-rw-r--r--fs/notify/inotify/Makefile1
-rw-r--r--fs/notify/inotify/inotify.c933
-rw-r--r--fs/notify/inotify/inotify.h29
-rw-r--r--fs/notify/inotify/inotify_fsnotify.c185
-rw-r--r--fs/notify/inotify/inotify_user.c707
6 files changed, 522 insertions, 1351 deletions
diff --git a/fs/notify/inotify/Kconfig b/fs/notify/inotify/Kconfig
index 5356884289a..b981fc0c837 100644
--- a/fs/notify/inotify/Kconfig
+++ b/fs/notify/inotify/Kconfig
@@ -1,21 +1,7 @@
-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"
- depends on FSNOTIFY
+ select ANON_INODES
+ select FSNOTIFY
default y
---help---
Say Y here to enable inotify support for userspace, including the
diff --git a/fs/notify/inotify/Makefile b/fs/notify/inotify/Makefile
index 94382817136..a380dabe09d 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 40b1cf914cc..00000000000
--- a/fs/notify/inotify/inotify.c
+++ /dev/null
@@ -1,933 +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;
- spin_lock(&sb_lock);
- if (sb->s_count >= S_BIAS) {
- atomic_inc(&sb->s_active);
- spin_unlock(&sb_lock);
- atomic_inc(&watch->count);
- return 1;
- }
- spin_unlock(&sb_lock);
- 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 almost certainly wait until the superblock is shut
- * down and the watch in question is pining for fjords. That's fine, but
- * there is a problem - we might have hit the window between ->s_active
- * getting to 0 / ->s_count - below S_BIAS (i.e. the moment when superblock
- * is past the point of no return and is heading for shutdown) and the
- * moment when deactivate_super() acquires ->s_umount. We could just do
- * drop_super() yield() and retry, but that's rather antisocial and this
- * stuff is luser-triggerable. OTOH, having grabbed ->s_umount and having
- * found that we'd got there first (i.e. that ->s_root is non-NULL) we know
- * that we won't race with inotify_umount_inodes(). So we could grab a
- * reference to watch and do the rest as above, just with drop_super() instead
- * of deactivate_super(), right? Wrong. We had to drop ih->mutex before we
- * could grab ->s_umount. So the watch could've been gone already.
- *
- * That still can be dealt with - we need to save watch->wd, do idr_find()
- * and compare its result with our pointer. If they match, we either have
- * the damn thing still alive or we'd lost not one but two races at once,
- * the watch had been killed and a new one got created with the same ->wd
- * at the same address. That couldn't have happened in inotify_destroy(),
- * but inotify_rm_wd() could run into that. Still, "new one got created"
- * is not a problem - we have every right to kill it or leave it alone,
- * whatever's more convenient.
- *
- * So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
- * "grab it and kill it" check. If it's been our original watch, we are
- * fine, if it's a newcomer - nevermind, just pretend that we'd won the
- * race and kill the fscker anyway; we are safe since we know that its
- * superblock won't be going away.
- *
- * 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()
- * 2 - remove it, drop the reference and drop_super(); we tried hard to avoid
- * that variant, since it involved a lot of PITA, but that's the best that
- * could've been done.
- */
-static int pin_to_kill(struct inotify_handle *ih, struct inotify_watch *watch)
-{
- struct super_block *sb = watch->inode->i_sb;
- s32 wd = watch->wd;
-
- spin_lock(&sb_lock);
- if (sb->s_count >= S_BIAS) {
- atomic_inc(&sb->s_active);
- spin_unlock(&sb_lock);
- get_inotify_watch(watch);
- mutex_unlock(&ih->mutex);
- return 1; /* the best outcome */
- }
- sb->s_count++;
- spin_unlock(&sb_lock);
- mutex_unlock(&ih->mutex); /* can't grab ->s_umount under it */
- down_read(&sb->s_umount);
- if (likely(!sb->s_root)) {
- /* fs is already shut down; the watch is dead */
- drop_super(sb);
- return 0;
- }
- /* raced with the final deactivate_super() */
- mutex_lock(&ih->mutex);
- if (idr_find(&ih->idr, wd) != watch || watch->inode->i_sb != sb) {
- /* the watch is dead */
- mutex_unlock(&ih->mutex);
- drop_super(sb);
- return 0;
- }
- /* still alive or freed and reused with the same sb and wd; kill */
- get_inotify_watch(watch);
- mutex_unlock(&ih->mutex);
- return 2;
-}
-
-static void unpin_and_kill(struct inotify_watch *watch, int how)
-{
- struct super_block *sb = watch->inode->i_sb;
- put_inotify_watch(watch);
- switch (how) {
- case 1:
- deactivate_super(sb);
- break;
- case 2:
- drop_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;
- int how;
-
- 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;
- how = pin_to_kill(ih, watch);
- if (!how)
- 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, how);
- }
-
- /* 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;
- int how;
-
- mutex_lock(&ih->mutex);
- watch = idr_find(&ih->idr, wd);
- if (unlikely(!watch)) {
- mutex_unlock(&ih->mutex);
- return -EINVAL;
- }
- sb = watch->inode->i_sb;
- how = pin_to_kill(ih, watch);
- if (!how)
- 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, how);
-
- 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 ea2605a58b8..ed855ef6f07 100644
--- a/fs/notify/inotify/inotify.h
+++ b/fs/notify/inotify/inotify.h
@@ -2,20 +2,31 @@
#include <linux/inotify.h>
#include <linux/slab.h> /* struct kmem_cache */
-extern struct kmem_cache *event_priv_cachep;
-
-struct inotify_event_private_data {
- struct fsnotify_event_private_data fsnotify_event_priv_data;
+struct inotify_event_info {
+ struct fsnotify_event fse;
int wd;
+ u32 sync_cookie;
+ int name_len;
+ char name[];
};
-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_destroy_mark_entry(struct fsnotify_mark_entry *entry, struct fsnotify_group *group);
-extern void inotify_free_event_priv(struct fsnotify_event_private_data *event_priv);
+static inline struct inotify_event_info *INOTIFY_E(struct fsnotify_event *fse)
+{
+ return container_of(fse, struct inotify_event_info, fse);
+}
+
+extern void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
+ struct fsnotify_group *group);
+extern int inotify_handle_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,
+ const unsigned char *file_name, u32 cookie);
extern const struct fsnotify_ops inotify_fsnotify_ops;
diff --git a/fs/notify/inotify/inotify_fsnotify.c b/fs/notify/inotify/inotify_fsnotify.c
index 160da548683..43ab1e1a07a 100644
--- a/fs/notify/inotify/inotify_fsnotify.c
+++ b/fs/notify/inotify/inotify_fsnotify.c
@@ -22,116 +22,161 @@
* 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>
#include <linux/path.h> /* struct path */
#include <linux/slab.h> /* kmem_* */
#include <linux/types.h>
+#include <linux/sched.h>
#include "inotify.h"
-static int inotify_handle_event(struct fsnotify_group *group, struct fsnotify_event *event)
+/*
+ * Check if 2 events contain the same information.
+ */
+static bool event_compare(struct fsnotify_event *old_fsn,
+ struct fsnotify_event *new_fsn)
{
- struct fsnotify_mark_entry *entry;
- struct inotify_inode_mark_entry *ientry;
- struct inode *to_tell;
- struct inotify_event_private_data *event_priv;
- struct fsnotify_event_private_data *fsn_event_priv;
- int wd, ret;
-
- 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;
+ struct inotify_event_info *old, *new;
- event_priv = kmem_cache_alloc(event_priv_cachep, GFP_KERNEL);
- if (unlikely(!event_priv))
- return -ENOMEM;
+ if (old_fsn->mask & FS_IN_IGNORED)
+ return false;
+ old = INOTIFY_E(old_fsn);
+ new = INOTIFY_E(new_fsn);
+ if ((old_fsn->mask == new_fsn->mask) &&
+ (old_fsn->inode == new_fsn->inode) &&
+ (old->name_len == new->name_len) &&
+ (!old->name_len || !strcmp(old->name, new->name)))
+ return true;
+ return false;
+}
+
+static int inotify_merge(struct list_head *list,
+ struct fsnotify_event *event)
+{
+ struct fsnotify_event *last_event;
- fsn_event_priv = &event_priv->fsnotify_event_priv_data;
+ last_event = list_entry(list->prev, struct fsnotify_event, list);
+ return event_compare(last_event, event);
+}
- fsn_event_priv->group = group;
- event_priv->wd = wd;
+int inotify_handle_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,
+ const unsigned char *file_name, u32 cookie)
+{
+ struct inotify_inode_mark *i_mark;
+ struct inotify_event_info *event;
+ struct fsnotify_event *fsn_event;
+ int ret;
+ int len = 0;
+ int alloc_len = sizeof(struct inotify_event_info);
+
+ BUG_ON(vfsmount_mark);
+
+ if ((inode_mark->mask & FS_EXCL_UNLINK) &&
+ (data_type == FSNOTIFY_EVENT_PATH)) {
+ struct path *path = data;
+
+ if (d_unlinked(path->dentry))
+ return 0;
+ }
+ if (file_name) {
+ len = strlen(file_name);
+ alloc_len += len + 1;
+ }
+
+ pr_debug("%s: group=%p inode=%p mask=%x\n", __func__, group, inode,
+ mask);
+
+ i_mark = container_of(inode_mark, struct inotify_inode_mark,
+ fsn_mark);
+
+ event = kmalloc(alloc_len, GFP_KERNEL);
+ if (unlikely(!event))
+ return -ENOMEM;
- ret = fsnotify_add_notify_event(group, event, fsn_event_priv);
- /* EEXIST is not an error */
- if (ret == -EEXIST)
- ret = 0;
+ fsn_event = &event->fse;
+ fsnotify_init_event(fsn_event, inode, mask);
+ event->wd = i_mark->wd;
+ event->sync_cookie = cookie;
+ event->name_len = len;
+ if (len)
+ strcpy(event->name, file_name);
- /* did event_priv get attached? */
- if (list_empty(&fsn_event_priv->event_list))
- inotify_free_event_priv(fsn_event_priv);
+ ret = fsnotify_add_notify_event(group, fsn_event, inotify_merge);
+ if (ret) {
+ /* Our event wasn't used in the end. Free it. */
+ fsnotify_destroy_event(group, fsn_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, group);
- return ret;
+ return 0;
}
-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_destroy_mark_entry(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)
+/*
+ * This is NEVER supposed to be called. Inotify marks should either have been
+ * removed from the idr when the watch was removed or in the
+ * fsnotify_destroy_mark_by_group() call when the inotify instance was being
+ * torn down. This is only called if the idr is about to be freed but there
+ * are still marks in it.
+ */
+static int idr_callback(int id, void *p, void *data)
{
- struct fsnotify_mark_entry *entry;
- bool send;
+ struct fsnotify_mark *fsn_mark;
+ struct inotify_inode_mark *i_mark;
+ static bool warned = false;
- spin_lock(&inode->i_lock);
- entry = fsnotify_find_mark_entry(group, inode);
- spin_unlock(&inode->i_lock);
- if (!entry)
- return false;
-
- send = (entry->mask & mask);
+ if (warned)
+ return 0;
- /* find took a reference */
- fsnotify_put_mark(entry);
+ warned = true;
+ fsn_mark = p;
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
- return send;
-}
+ WARN(1, "inotify closing but id=%d for fsn_mark=%p in group=%p still in "
+ "idr. Probably leaking memory\n", id, p, data);
-static int idr_callback(int id, void *p, void *data)
-{
- BUG();
+ /*
+ * I'm taking the liberty of assuming that the mark in question is a
+ * valid address and I'm dereferencing it. This might help to figure
+ * out why we got here and the panic is no worse than the original
+ * BUG() that was here.
+ */
+ 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;
}
static void inotify_free_group_priv(struct fsnotify_group *group)
{
- /* ideally the idr is empty and we won't hit the BUG in teh callback */
- idr_for_each(&group->inotify_data.idr, idr_callback, NULL);
- idr_remove_all(&group->inotify_data.idr);
+ /* ideally the idr is empty and we won't hit the BUG in the callback */
+ idr_for_each(&group->inotify_data.idr, idr_callback, group);
idr_destroy(&group->inotify_data.idr);
+ atomic_dec(&group->inotify_data.user->inotify_devs);
+ free_uid(group->inotify_data.user);
}
-void inotify_free_event_priv(struct fsnotify_event_private_data *fsn_event_priv)
+static void inotify_free_event(struct fsnotify_event *fsn_event)
{
- struct inotify_event_private_data *event_priv;
-
-
- event_priv = container_of(fsn_event_priv, struct inotify_event_private_data,
- fsnotify_event_priv_data);
-
- kmem_cache_free(event_priv_cachep, event_priv);
+ kfree(INOTIFY_E(fsn_event));
}
const struct fsnotify_ops inotify_fsnotify_ops = {
.handle_event = inotify_handle_event,
- .should_send_event = inotify_should_send_event,
.free_group_priv = inotify_free_group_priv,
- .free_event_priv = inotify_free_event_priv,
+ .free_event = inotify_free_event,
.freeing_mark = inotify_freeing_mark,
};
diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c
index 982a412ac5b..cc423a30a0c 100644
--- a/fs/notify/inotify/inotify_user.c
+++ b/fs/notify/inotify/inotify_user.c
@@ -29,41 +29,27 @@
#include <linux/init.h> /* module_init */
#include <linux/inotify.h>
#include <linux/kernel.h> /* roundup() */
-#include <linux/magic.h> /* superblock magic number */
-#include <linux/mount.h> /* mntget */
#include <linux/namei.h> /* LOOKUP_FOLLOW */
-#include <linux/path.h> /* struct path */
#include <linux/sched.h> /* struct user */
#include <linux/slab.h> /* struct kmem_cache */
#include <linux/syscalls.h>
#include <linux/types.h>
+#include <linux/anon_inodes.h>
#include <linux/uaccess.h>
#include <linux/poll.h>
#include <linux/wait.h>
#include "inotify.h"
+#include "../fdinfo.h"
#include <asm/ioctls.h>
-static struct vfsmount *inotify_mnt __read_mostly;
-
-/* this just sits here and wastes global memory. used to just pad userspace messages with zeros */
-static struct inotify_event nul_inotify_event;
-
/* 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;
-static struct fsnotify_event *inotify_ignored_event;
-
-/*
- * 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
@@ -71,38 +57,32 @@ static atomic_t inotify_grp_num;
static int zero;
-ctl_table inotify_table[] = {
+struct ctl_table inotify_table[] = {
{
- .ctl_name = INOTIFY_MAX_USER_INSTANCES,
.procname = "max_user_instances",
.data = &inotify_max_user_instances,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = &proc_dointvec_minmax,
- .strategy = &sysctl_intvec,
+ .proc_handler = proc_dointvec_minmax,
.extra1 = &zero,
},
{
- .ctl_name = INOTIFY_MAX_USER_WATCHES,
.procname = "max_user_watches",
.data = &inotify_max_user_watches,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = &proc_dointvec_minmax,
- .strategy = &sysctl_intvec,
+ .proc_handler = proc_dointvec_minmax,
.extra1 = &zero,
},
{
- .ctl_name = INOTIFY_MAX_QUEUED_EVENTS,
.procname = "max_queued_events",
.data = &inotify_max_queued_events,
.maxlen = sizeof(int),
.mode = 0644,
- .proc_handler = &proc_dointvec_minmax,
- .strategy = &sysctl_intvec,
+ .proc_handler = proc_dointvec_minmax,
.extra1 = &zero
},
- { .ctl_name = 0 }
+ { }
};
#endif /* CONFIG_SYSCTL */
@@ -110,11 +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;
}
@@ -140,6 +123,16 @@ static unsigned int inotify_poll(struct file *file, poll_table *wait)
return ret;
}
+static int round_event_name_len(struct fsnotify_event *fsn_event)
+{
+ struct inotify_event_info *event;
+
+ event = INOTIFY_E(fsn_event);
+ if (!event->name_len)
+ return 0;
+ return roundup(event->name_len + 1, sizeof(struct inotify_event));
+}
+
/*
* Get an inotify_kernel_event if one exists and is small
* enough to fit in "count". Return an error pointer if
@@ -158,8 +151,9 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
event = fsnotify_peek_notify_event(group);
- event_size += roundup(event->name_len, event_size);
+ pr_debug("%s: group=%p event=%p\n", __func__, group, event);
+ event_size += round_event_name_len(event);
if (event_size > count)
return ERR_PTR(-EINVAL);
@@ -177,34 +171,27 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group,
* buffer we had in "get_one_event()" above.
*/
static ssize_t copy_event_to_user(struct fsnotify_group *group,
- struct fsnotify_event *event,
+ struct fsnotify_event *fsn_event,
char __user *buf)
{
struct inotify_event inotify_event;
- struct fsnotify_event_private_data *fsn_priv;
- struct inotify_event_private_data *priv;
+ struct inotify_event_info *event;
size_t event_size = sizeof(struct inotify_event);
size_t name_len;
+ size_t pad_name_len;
- /* we get the inotify watch descriptor from the event private data */
- spin_lock(&event->lock);
- fsn_priv = fsnotify_remove_priv_from_event(group, event);
- spin_unlock(&event->lock);
-
- if (!fsn_priv)
- inotify_event.wd = -1;
- else {
- priv = container_of(fsn_priv, struct inotify_event_private_data,
- fsnotify_event_priv_data);
- inotify_event.wd = priv->wd;
- inotify_free_event_priv(fsn_priv);
- }
-
- /* round up event->name_len so it is a multiple of event_size */
- name_len = roundup(event->name_len, event_size);
- inotify_event.len = name_len;
+ pr_debug("%s: group=%p event=%p\n", __func__, group, fsn_event);
- inotify_event.mask = inotify_mask_to_arg(event->mask);
+ event = INOTIFY_E(fsn_event);
+ name_len = event->name_len;
+ /*
+ * round up name length so it is a multiple of event_size
+ * plus an extra byte for the terminating '\0'.
+ */
+ pad_name_len = round_event_name_len(fsn_event);
+ inotify_event.len = pad_name_len;
+ inotify_event.mask = inotify_mask_to_arg(fsn_event->mask);
+ inotify_event.wd = event->wd;
inotify_event.cookie = event->sync_cookie;
/* send the main event */
@@ -216,20 +203,18 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,
/*
* fsnotify only stores the pathname, so here we have to send the pathname
* and then pad that pathname out to a multiple of sizeof(inotify_event)
- * with zeros. I get my zeros from the nul_inotify_event.
+ * with zeros.
*/
- if (name_len) {
- unsigned int len_to_zero = name_len - event->name_len;
+ if (pad_name_len) {
/* copy the path name */
- if (copy_to_user(buf, event->file_name, event->name_len))
+ if (copy_to_user(buf, event->name, name_len))
return -EFAULT;
- buf += event->name_len;
+ buf += name_len;
- /* fill userspace with 0's from nul_inotify_event */
- if (copy_to_user(buf, &nul_inotify_event, len_to_zero))
+ /* fill userspace with 0's */
+ if (clear_user(buf, pad_name_len - name_len))
return -EFAULT;
- buf += len_to_zero;
- event_size += name_len;
+ event_size += pad_name_len;
}
return event_size;
@@ -254,12 +239,14 @@ 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))
break;
ret = copy_event_to_user(group, kevent, buf);
- fsnotify_put_event(kevent);
+ fsnotify_destroy_event(group, kevent);
if (ret < 0)
break;
buf += ret;
@@ -270,7 +257,7 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
ret = -EAGAIN;
if (file->f_flags & O_NONBLOCK)
break;
- ret = -EINTR;
+ ret = -ERESTARTSYS;
if (signal_pending(current))
break;
@@ -286,21 +273,14 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
return ret;
}
-static int inotify_fasync(int fd, struct file *file, int on)
-{
- struct fsnotify_group *group = file->private_data;
-
- return fasync_helper(fd, file, on, &group->inotify_data.fa) >= 0 ? 0 : -EIO;
-}
-
static int inotify_release(struct inode *ignored, struct file *file)
{
struct fsnotify_group *group = file->private_data;
- fsnotify_clear_marks_by_group(group);
+ pr_debug("%s: group=%p\n", __func__, group);
/* free this group, matching get was inotify_init->fsnotify_obtain_group */
- fsnotify_put_group(group);
+ fsnotify_destroy_group(group);
return 0;
}
@@ -309,8 +289,7 @@ static long inotify_ioctl(struct file *file, unsigned int cmd,
unsigned long arg)
{
struct fsnotify_group *group;
- struct fsnotify_event_holder *holder;
- struct fsnotify_event *event;
+ struct fsnotify_event *fsn_event;
void __user *p;
int ret = -ENOTTY;
size_t send_len = 0;
@@ -318,14 +297,15 @@ 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);
- list_for_each_entry(holder, &group->notification_list, event_list) {
- event = holder->event;
+ list_for_each_entry(fsn_event, &group->notification_list,
+ list) {
send_len += sizeof(struct inotify_event);
- send_len += roundup(event->name_len,
- sizeof(struct inotify_event));
+ send_len += round_event_name_len(fsn_event);
}
mutex_unlock(&group->notification_mutex);
ret = put_user(send_len, (int __user *) p);
@@ -336,12 +316,14 @@ static long inotify_ioctl(struct file *file, unsigned int cmd,
}
static const struct file_operations inotify_fops = {
+ .show_fdinfo = inotify_show_fdinfo,
.poll = inotify_poll,
.read = inotify_read,
- .fasync = inotify_fasync,
+ .fasync = fsnotify_fasync,
.release = inotify_release,
.unlocked_ioctl = inotify_ioctl,
.compat_ioctl = inotify_ioctl,
+ .llseek = noop_llseek,
};
@@ -362,198 +344,323 @@ 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,
+ struct inotify_inode_mark *i_mark)
+{
+ int ret;
+
+ idr_preload(GFP_KERNEL);
+ spin_lock(idr_lock);
+
+ ret = idr_alloc_cyclic(idr, i_mark, 1, 0, GFP_NOWAIT);
+ if (ret >= 0) {
+ /* we added the mark to the idr, take a reference */
+ i_mark->wd = ret;
+ fsnotify_get_mark(&i_mark->fsn_mark);
+ }
+
+ spin_unlock(idr_lock);
+ idr_preload_end();
+ return ret < 0 ? ret : 0;
+}
+
+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);
+}
+
/*
- * When, for whatever reason, inotify is done with a mark (or what used to be a
- * watch) we need to remove that watch from the idr and we need to send IN_IGNORED
- * for the given wd.
- *
- * There is a bit of recursion here. The loop looks like:
- * inotify_destroy_mark_entry -> fsnotify_destroy_mark_by_entry ->
- * inotify_freeing_mark -> inotify_destory_mark_entry -> restart
- * But the loop is broken in 2 places. fsnotify_destroy_mark_by_entry sets
- * entry->group = NULL before the call to inotify_freeing_mark, so the if (egroup)
- * test below will not call back to fsnotify again. But even if that test wasn't
- * there this would still be safe since fsnotify_destroy_mark_by_entry() is
- * safe from recursion.
+ * Remove the mark from the idr (if present) and drop the reference
+ * on the mark because it was in the idr.
*/
-void inotify_destroy_mark_entry(struct fsnotify_mark_entry *entry, struct fsnotify_group *group)
+static void inotify_remove_from_idr(struct fsnotify_group *group,
+ struct inotify_inode_mark *i_mark)
{
- struct inotify_inode_mark_entry *ientry;
- struct inotify_event_private_data *event_priv;
- struct fsnotify_event_private_data *fsn_event_priv;
- struct fsnotify_group *egroup;
- struct idr *idr;
-
- spin_lock(&entry->lock);
- egroup = entry->group;
-
- /* if egroup we aren't really done and something might still send events
- * for this inode, on the callback we'll send the IN_IGNORED */
- if (egroup) {
- spin_unlock(&entry->lock);
- fsnotify_destroy_mark_by_entry(entry);
- return;
- }
- spin_unlock(&entry->lock);
+ spinlock_t *idr_lock = &group->inotify_data.idr_lock;
+ struct inotify_inode_mark *found_i_mark = NULL;
+ int wd;
+
+ spin_lock(idr_lock);
+ wd = i_mark->wd;
- ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
+ /*
+ * 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;
+ }
- event_priv = kmem_cache_alloc(event_priv_cachep, GFP_KERNEL);
- if (unlikely(!event_priv))
- goto skip_send_ignore;
+ /* 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;
+ }
- fsn_event_priv = &event_priv->fsnotify_event_priv_data;
+ /*
+ * 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;
+ }
- fsn_event_priv->group = group;
- event_priv->wd = ientry->wd;
+ /*
+ * 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();
+ }
- fsnotify_add_notify_event(group, inotify_ignored_event, fsn_event_priv);
+ do_inotify_remove_from_idr(group, i_mark);
+out:
+ /* 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);
+}
- /* did the private data get added? */
- if (list_empty(&fsn_event_priv->event_list))
- inotify_free_event_priv(fsn_event_priv);
+/*
+ * Send IN_IGNORED for this wd, remove this wd from the idr.
+ */
+void inotify_ignored_and_remove_idr(struct fsnotify_mark *fsn_mark,
+ struct fsnotify_group *group)
+{
+ struct inotify_inode_mark *i_mark;
-skip_send_ignore:
+ /* Queue ignore event for the watch */
+ inotify_handle_event(group, NULL, fsn_mark, NULL, FS_IN_IGNORED,
+ NULL, FSNOTIFY_EVENT_NONE, NULL, 0);
- /* remove this entry from the idr */
- spin_lock(&group->inotify_data.idr_lock);
- idr = &group->inotify_data.idr;
- idr_remove(idr, ientry->wd);
- spin_unlock(&group->inotify_data.idr_lock);
+ i_mark = container_of(fsn_mark, struct inotify_inode_mark, fsn_mark);
+ /* remove this mark from the idr */
+ inotify_remove_from_idr(group, i_mark);
- /* removed from idr, drop that reference */
- fsnotify_put_mark(entry);
+ 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_watch(struct fsnotify_group *group, struct inode *inode, u32 arg)
+static int inotify_update_existing_watch(struct fsnotify_group *group,
+ struct inode *inode,
+ u32 arg)
{
- struct fsnotify_mark_entry *entry = NULL;
- struct inotify_inode_mark_entry *ientry;
- int ret = 0;
- int add = (arg & IN_MASK_ADD);
- __u32 mask;
+ struct fsnotify_mark *fsn_mark;
+ struct inotify_inode_mark *i_mark;
__u32 old_mask, new_mask;
+ __u32 mask;
+ int add = (arg & IN_MASK_ADD);
+ int ret;
- /* don't allow invalid bits: we don't want flags set */
mask = inotify_arg_to_mask(arg);
- if (unlikely(!mask))
- return -EINVAL;
- ientry = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
- if (unlikely(!ientry))
- return -ENOMEM;
- /* we set the mask at the end after attaching it */
- fsnotify_init_mark(&ientry->fsn_entry, inotify_free_mark);
- ientry->wd = 0;
-
-find_entry:
- spin_lock(&inode->i_lock);
- entry = fsnotify_find_mark_entry(group, inode);
- spin_unlock(&inode->i_lock);
- if (entry) {
- kmem_cache_free(inotify_inode_mark_cachep, ientry);
- ientry = container_of(entry, struct inotify_inode_mark_entry, fsn_entry);
- } else {
- if (atomic_read(&group->inotify_data.user->inotify_watches) >= inotify_max_user_watches) {
- ret = -ENOSPC;
- goto out_err;
- }
+ fsn_mark = fsnotify_find_inode_mark(group, inode);
+ if (!fsn_mark)
+ return -ENOENT;
- ret = fsnotify_add_mark(&ientry->fsn_entry, group, inode);
- if (ret == -EEXIST)
- goto find_entry;
- else if (ret)
- goto out_err;
-
- entry = &ientry->fsn_entry;
-retry:
- ret = -ENOMEM;
- if (unlikely(!idr_pre_get(&group->inotify_data.idr, GFP_KERNEL)))
- goto out_err;
-
- spin_lock(&group->inotify_data.idr_lock);
- /* if entry is added to the idr we keep the reference obtained
- * through fsnotify_mark_add. remember to drop this reference
- * when entry is removed from idr */
- ret = idr_get_new_above(&group->inotify_data.idr, entry,
- ++group->inotify_data.last_wd,
- &ientry->wd);
- spin_unlock(&group->inotify_data.idr_lock);
- if (ret) {
- if (ret == -EAGAIN)
- goto retry;
- goto out_err;
- }
- atomic_inc(&group->inotify_data.user->inotify_watches);
- }
+ 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 ientry->wd;
+ /* return the wd */
+ ret = i_mark->wd;
-out_err:
- /* see this isn't supposed to happen, just kill the watch */
- if (entry) {
- fsnotify_destroy_mark_by_entry(entry);
- fsnotify_put_mark(entry);
+ /* match the get from fsnotify_find_mark() */
+ fsnotify_put_mark(fsn_mark);
+
+ return ret;
+}
+
+static int inotify_new_watch(struct fsnotify_group *group,
+ struct inode *inode,
+ u32 arg)
+{
+ 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;
+
+ mask = inotify_arg_to_mask(arg);
+
+ tmp_i_mark = kmem_cache_alloc(inotify_inode_mark_cachep, GFP_KERNEL);
+ if (unlikely(!tmp_i_mark))
+ return -ENOMEM;
+
+ 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;
+
+ ret = inotify_add_to_idr(idr, idr_lock, tmp_i_mark);
+ if (ret)
+ goto out_err;
+
+ /* we are on the idr, now get on the inode */
+ ret = fsnotify_add_mark_locked(&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_i_mark);
+ goto out_err;
}
+
+ /* increment the number of watches the user has */
+ atomic_inc(&group->inotify_data.user->inotify_watches);
+
+ /* return the watch descriptor for this new mark */
+ ret = tmp_i_mark->wd;
+
+out_err:
+ /* match the ref from fsnotify_init_mark() */
+ fsnotify_put_mark(&tmp_i_mark->fsn_mark);
+
+ return ret;
+}
+
+static int inotify_update_watch(struct fsnotify_group *group, struct inode *inode, u32 arg)
+{
+ int ret = 0;
+
+ mutex_lock(&group->mark_mutex);
+ /* try to update and existing watch with the new arg */
+ ret = inotify_update_existing_watch(group, inode, arg);
+ /* no mark present, try to add a new one */
+ if (ret == -ENOENT)
+ ret = inotify_new_watch(group, inode, arg);
+ mutex_unlock(&group->mark_mutex);
+
return ret;
}
-static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsigned int max_events)
+static struct fsnotify_group *inotify_new_group(unsigned int max_events)
{
struct fsnotify_group *group;
- unsigned int grp_num;
+ struct inotify_event_info *oevent;
- /* 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;
+ oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL);
+ if (unlikely(!oevent)) {
+ fsnotify_destroy_group(group);
+ return ERR_PTR(-ENOMEM);
+ }
+ group->overflow_event = &oevent->fse;
+ fsnotify_init_event(group->overflow_event, NULL, FS_Q_OVERFLOW);
+ oevent->wd = -1;
+ oevent->sync_cookie = 0;
+ oevent->name_len = 0;
+
group->max_events = max_events;
spin_lock_init(&group->inotify_data.idr_lock);
idr_init(&group->inotify_data.idr);
- group->inotify_data.last_wd = 0;
- group->inotify_data.user = user;
- group->inotify_data.fa = NULL;
+ group->inotify_data.user = get_current_user();
+
+ if (atomic_inc_return(&group->inotify_data.user->inotify_devs) >
+ inotify_max_user_instances) {
+ fsnotify_destroy_group(group);
+ return ERR_PTR(-EMFILE);
+ }
return group;
}
@@ -563,9 +670,7 @@ static struct fsnotify_group *inotify_new_group(struct user_struct *user, unsign
SYSCALL_DEFINE1(inotify_init1, int, flags)
{
struct fsnotify_group *group;
- struct user_struct *user;
- struct file *filp;
- int fd, ret;
+ int ret;
/* Check the IN_* constants for consistency. */
BUILD_BUG_ON(IN_CLOEXEC != O_CLOEXEC);
@@ -574,49 +679,16 @@ SYSCALL_DEFINE1(inotify_init1, int, flags)
if (flags & ~(IN_CLOEXEC | IN_NONBLOCK))
return -EINVAL;
- fd = get_unused_fd_flags(flags & O_CLOEXEC);
- if (fd < 0)
- return fd;
-
- filp = get_empty_filp();
- if (!filp) {
- ret = -ENFILE;
- goto out_put_fd;
- }
-
- user = get_current_user();
- if (unlikely(atomic_read(&user->inotify_devs) >=
- inotify_max_user_instances)) {
- ret = -EMFILE;
- goto out_free_uid;
- }
-
/* fsnotify_obtain_group took a reference to group, we put this when we kill the file in the end */
- group = inotify_new_group(user, inotify_max_queued_events);
- if (IS_ERR(group)) {
- ret = PTR_ERR(group);
- goto out_free_uid;
- }
-
- filp->f_op = &inotify_fops;
- filp->f_path.mnt = mntget(inotify_mnt);
- filp->f_path.dentry = dget(inotify_mnt->mnt_root);
- filp->f_mapping = filp->f_path.dentry->d_inode->i_mapping;
- filp->f_mode = FMODE_READ;
- filp->f_flags = O_RDONLY | (flags & O_NONBLOCK);
- filp->private_data = group;
-
- atomic_inc(&user->inotify_devs);
-
- fd_install(fd, filp);
+ group = inotify_new_group(inotify_max_queued_events);
+ if (IS_ERR(group))
+ return PTR_ERR(group);
- return fd;
+ ret = anon_inode_getfd("inotify", &inotify_fops, group,
+ O_RDONLY | flags);
+ if (ret < 0)
+ fsnotify_destroy_group(group);
-out_free_uid:
- free_uid(user);
- put_filp(filp);
-out_put_fd:
- put_unused_fd(fd);
return ret;
}
@@ -631,16 +703,20 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
struct fsnotify_group *group;
struct inode *inode;
struct path path;
- struct file *filp;
- int ret, fput_needed;
+ struct fd f;
+ int ret;
unsigned flags = 0;
- filp = fget_light(fd, &fput_needed);
- if (unlikely(!filp))
+ /* don't allow invalid bits: we don't want flags set */
+ if (unlikely(!(mask & ALL_INOTIFY_BITS)))
+ return -EINVAL;
+
+ f = fdget(fd);
+ if (unlikely(!f.file))
return -EBADF;
/* verify that this is indeed an inotify instance */
- if (unlikely(filp->f_op != &inotify_fops)) {
+ if (unlikely(f.file->f_op != &inotify_fops)) {
ret = -EINVAL;
goto fput_and_out;
}
@@ -656,93 +732,80 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname,
/* inode held in place by reference to path; group by fget on fd */
inode = path.dentry->d_inode;
- group = filp->private_data;
+ group = f.file->private_data;
/* create/update an inode mark */
ret = inotify_update_watch(group, inode, mask);
- if (unlikely(ret))
- goto path_put_and_out;
-
-path_put_and_out:
path_put(&path);
fput_and_out:
- fput_light(filp, fput_needed);
+ fdput(f);
return ret;
}
SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd)
{
struct fsnotify_group *group;
- struct fsnotify_mark_entry *entry;
- struct file *filp;
- int ret = 0, fput_needed;
+ struct inotify_inode_mark *i_mark;
+ struct fd f;
+ int ret = 0;
- filp = fget_light(fd, &fput_needed);
- if (unlikely(!filp))
+ f = fdget(fd);
+ if (unlikely(!f.file))
return -EBADF;
/* verify that this is indeed an inotify instance */
- if (unlikely(filp->f_op != &inotify_fops)) {
- ret = -EINVAL;
+ ret = -EINVAL;
+ if (unlikely(f.file->f_op != &inotify_fops))
goto out;
- }
- group = filp->private_data;
+ group = f.file->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);
- inotify_destroy_mark_entry(entry, group);
- fsnotify_put_mark(entry);
+ ret = 0;
+
+ fsnotify_destroy_mark(&i_mark->fsn_mark, group);
+
+ /* match ref taken by inotify_idr_find */
+ fsnotify_put_mark(&i_mark->fsn_mark);
out:
- fput_light(filp, fput_needed);
+ fdput(f);
return ret;
}
-static int
-inotify_get_sb(struct file_system_type *fs_type, int flags,
- const char *dev_name, void *data, struct vfsmount *mnt)
-{
- return get_sb_pseudo(fs_type, "inotify", NULL,
- INOTIFYFS_SUPER_MAGIC, mnt);
-}
-
-static struct file_system_type inotify_fs_type = {
- .name = "inotifyfs",
- .get_sb = inotify_get_sb,
- .kill_sb = kill_anon_super,
-};
-
/*
- * inotify_user_setup - Our initialization function. Note that we cannnot return
+ * inotify_user_setup - Our initialization function. Note that we cannot return
* error because we have compiled-in VFS hooks. So an (unlikely) failure here
* must result in panic().
*/
static int __init inotify_user_setup(void)
{
- int ret;
-
- ret = register_filesystem(&inotify_fs_type);
- if (unlikely(ret))
- panic("inotify: register_filesystem returned %d!\n", ret);
-
- inotify_mnt = kern_mount(&inotify_fs_type);
- if (IS_ERR(inotify_mnt))
- panic("inotify: kern_mount ret %ld!\n", PTR_ERR(inotify_mnt));
-
- inotify_inode_mark_cachep = KMEM_CACHE(inotify_inode_mark_entry, SLAB_PANIC);
- event_priv_cachep = KMEM_CACHE(inotify_event_private_data, SLAB_PANIC);
- inotify_ignored_event = fsnotify_create_event(NULL, FS_IN_IGNORED, NULL, FSNOTIFY_EVENT_NONE, NULL, 0);
- if (!inotify_ignored_event)
- panic("unable to allocate the inotify ignored event\n");
+ 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_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);
inotify_max_queued_events = 16384;
inotify_max_user_instances = 128;