2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* linux/fs/file_table.c
|
|
|
|
*
|
|
|
|
* Copyright (C) 1991, 1992 Linus Torvalds
|
|
|
|
* Copyright (C) 1997 David S. Miller (davem@caip.rutgers.edu)
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/string.h>
|
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/file.h>
|
|
|
|
#include <linux/init.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/smp_lock.h>
|
|
|
|
#include <linux/fs.h>
|
|
|
|
#include <linux/security.h>
|
|
|
|
#include <linux/eventpoll.h>
|
2005-09-09 14:04:13 -06:00
|
|
|
#include <linux/rcupdate.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/mount.h>
|
2006-01-11 13:17:46 -07:00
|
|
|
#include <linux/capability.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
#include <linux/cdev.h>
|
[PATCH] inotify
inotify is intended to correct the deficiencies of dnotify, particularly
its inability to scale and its terrible user interface:
* dnotify requires the opening of one fd per each directory
that you intend to watch. This quickly results in too many
open files and pins removable media, preventing unmount.
* dnotify is directory-based. You only learn about changes to
directories. Sure, a change to a file in a directory affects
the directory, but you are then forced to keep a cache of
stat structures.
* dnotify's interface to user-space is awful. Signals?
inotify provides a more usable, simple, powerful solution to file change
notification:
* inotify's interface is a system call that returns a fd, not SIGIO.
You get a single fd, which is select()-able.
* inotify has an event that says "the filesystem that the item
you were watching is on was unmounted."
* inotify can watch directories or files.
Inotify is currently used by Beagle (a desktop search infrastructure),
Gamin (a FAM replacement), and other projects.
See Documentation/filesystems/inotify.txt.
Signed-off-by: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-12 15:06:03 -06:00
|
|
|
#include <linux/fsnotify.h>
|
2006-03-07 22:55:35 -07:00
|
|
|
#include <linux/sysctl.h>
|
|
|
|
#include <linux/percpu_counter.h>
|
|
|
|
|
|
|
|
#include <asm/atomic.h>
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/* sysctl tunables... */
|
|
|
|
struct files_stat_struct files_stat = {
|
|
|
|
.max_files = NR_FILE
|
|
|
|
};
|
|
|
|
|
|
|
|
/* public. Not pretty! */
|
2006-03-07 22:55:35 -07:00
|
|
|
__cacheline_aligned_in_smp DEFINE_SPINLOCK(files_lock);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
static struct percpu_counter nr_files __cacheline_aligned_in_smp;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
static inline void file_free_rcu(struct rcu_head *head)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-03-07 22:55:35 -07:00
|
|
|
struct file *f = container_of(head, struct file, f_u.fu_rcuhead);
|
|
|
|
kmem_cache_free(filp_cachep, f);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
static inline void file_free(struct file *f)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-03-07 22:55:35 -07:00
|
|
|
percpu_counter_dec(&nr_files);
|
|
|
|
call_rcu(&f->f_u.fu_rcuhead, file_free_rcu);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
/*
|
|
|
|
* Return the total number of open files in the system
|
|
|
|
*/
|
|
|
|
static int get_nr_files(void)
|
2005-04-16 16:20:36 -06:00
|
|
|
{
|
2006-03-07 22:55:35 -07:00
|
|
|
return percpu_counter_read_positive(&nr_files);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
/*
|
|
|
|
* Return the maximum number of open files in the system
|
|
|
|
*/
|
|
|
|
int get_max_files(void)
|
2005-09-09 14:04:13 -06:00
|
|
|
{
|
2006-03-07 22:55:35 -07:00
|
|
|
return files_stat.max_files;
|
2005-09-09 14:04:13 -06:00
|
|
|
}
|
2006-03-07 22:55:35 -07:00
|
|
|
EXPORT_SYMBOL_GPL(get_max_files);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Handle nr_files sysctl
|
|
|
|
*/
|
|
|
|
#if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS)
|
|
|
|
int proc_nr_files(ctl_table *table, int write, struct file *filp,
|
|
|
|
void __user *buffer, size_t *lenp, loff_t *ppos)
|
|
|
|
{
|
|
|
|
files_stat.nr_files = get_nr_files();
|
|
|
|
return proc_dointvec(table, write, filp, buffer, lenp, ppos);
|
|
|
|
}
|
|
|
|
#else
|
|
|
|
int proc_nr_files(ctl_table *table, int write, struct file *filp,
|
|
|
|
void __user *buffer, size_t *lenp, loff_t *ppos)
|
|
|
|
{
|
|
|
|
return -ENOSYS;
|
|
|
|
}
|
|
|
|
#endif
|
2005-09-09 14:04:13 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Find an unused file structure and return a pointer to it.
|
|
|
|
* Returns NULL, if there are no more free file structures or
|
|
|
|
* we run out of memory.
|
|
|
|
*/
|
|
|
|
struct file *get_empty_filp(void)
|
|
|
|
{
|
2006-03-23 04:01:03 -07:00
|
|
|
struct task_struct *tsk;
|
2005-06-23 01:09:50 -06:00
|
|
|
static int old_max;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct file * f;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Privileged users can go above max_files
|
|
|
|
*/
|
2006-03-07 22:55:35 -07:00
|
|
|
if (get_nr_files() >= files_stat.max_files && !capable(CAP_SYS_ADMIN)) {
|
|
|
|
/*
|
|
|
|
* percpu_counters are inaccurate. Do an expensive check before
|
|
|
|
* we go and fail.
|
|
|
|
*/
|
|
|
|
if (percpu_counter_sum(&nr_files) >= files_stat.max_files)
|
|
|
|
goto over;
|
|
|
|
}
|
2005-06-23 01:09:50 -06:00
|
|
|
|
|
|
|
f = kmem_cache_alloc(filp_cachep, GFP_KERNEL);
|
|
|
|
if (f == NULL)
|
|
|
|
goto fail;
|
|
|
|
|
2006-03-07 22:55:35 -07:00
|
|
|
percpu_counter_inc(&nr_files);
|
2005-06-23 01:09:50 -06:00
|
|
|
memset(f, 0, sizeof(*f));
|
|
|
|
if (security_file_alloc(f))
|
|
|
|
goto fail_sec;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-03-23 04:01:03 -07:00
|
|
|
tsk = current;
|
|
|
|
INIT_LIST_HEAD(&f->f_u.fu_list);
|
2005-06-23 01:09:50 -06:00
|
|
|
atomic_set(&f->f_count, 1);
|
|
|
|
rwlock_init(&f->f_owner.lock);
|
2006-03-23 04:01:03 -07:00
|
|
|
f->f_uid = tsk->fsuid;
|
|
|
|
f->f_gid = tsk->fsgid;
|
|
|
|
eventpoll_init_file(f);
|
2005-06-23 01:09:50 -06:00
|
|
|
/* f->f_version: 0 */
|
|
|
|
return f;
|
|
|
|
|
|
|
|
over:
|
2005-04-16 16:20:36 -06:00
|
|
|
/* Ran out of filps - report that */
|
2006-03-07 22:55:35 -07:00
|
|
|
if (get_nr_files() > old_max) {
|
2005-04-16 16:20:36 -06:00
|
|
|
printk(KERN_INFO "VFS: file-max limit %d reached\n",
|
2006-03-07 22:55:35 -07:00
|
|
|
get_max_files());
|
|
|
|
old_max = get_nr_files();
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2005-06-23 01:09:50 -06:00
|
|
|
goto fail;
|
|
|
|
|
|
|
|
fail_sec:
|
|
|
|
file_free(f);
|
2005-04-16 16:20:36 -06:00
|
|
|
fail:
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL(get_empty_filp);
|
|
|
|
|
|
|
|
void fastcall fput(struct file *file)
|
|
|
|
{
|
2006-01-08 02:02:19 -07:00
|
|
|
if (atomic_dec_and_test(&file->f_count))
|
2005-04-16 16:20:36 -06:00
|
|
|
__fput(file);
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL(fput);
|
|
|
|
|
|
|
|
/* __fput is called from task context when aio completion releases the last
|
|
|
|
* last use of a struct file *. Do not use otherwise.
|
|
|
|
*/
|
|
|
|
void fastcall __fput(struct file *file)
|
|
|
|
{
|
|
|
|
struct dentry *dentry = file->f_dentry;
|
|
|
|
struct vfsmount *mnt = file->f_vfsmnt;
|
|
|
|
struct inode *inode = dentry->d_inode;
|
|
|
|
|
|
|
|
might_sleep();
|
[PATCH] inotify
inotify is intended to correct the deficiencies of dnotify, particularly
its inability to scale and its terrible user interface:
* dnotify requires the opening of one fd per each directory
that you intend to watch. This quickly results in too many
open files and pins removable media, preventing unmount.
* dnotify is directory-based. You only learn about changes to
directories. Sure, a change to a file in a directory affects
the directory, but you are then forced to keep a cache of
stat structures.
* dnotify's interface to user-space is awful. Signals?
inotify provides a more usable, simple, powerful solution to file change
notification:
* inotify's interface is a system call that returns a fd, not SIGIO.
You get a single fd, which is select()-able.
* inotify has an event that says "the filesystem that the item
you were watching is on was unmounted."
* inotify can watch directories or files.
Inotify is currently used by Beagle (a desktop search infrastructure),
Gamin (a FAM replacement), and other projects.
See Documentation/filesystems/inotify.txt.
Signed-off-by: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Cc: Christoph Hellwig <hch@lst.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-07-12 15:06:03 -06:00
|
|
|
|
|
|
|
fsnotify_close(file);
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* The function eventpoll_release() should be the first called
|
|
|
|
* in the file cleanup chain.
|
|
|
|
*/
|
|
|
|
eventpoll_release(file);
|
|
|
|
locks_remove_flock(file);
|
|
|
|
|
|
|
|
if (file->f_op && file->f_op->release)
|
|
|
|
file->f_op->release(inode, file);
|
|
|
|
security_file_free(file);
|
|
|
|
if (unlikely(inode->i_cdev != NULL))
|
|
|
|
cdev_put(inode->i_cdev);
|
|
|
|
fops_put(file->f_op);
|
|
|
|
if (file->f_mode & FMODE_WRITE)
|
|
|
|
put_write_access(inode);
|
|
|
|
file_kill(file);
|
|
|
|
file->f_dentry = NULL;
|
|
|
|
file->f_vfsmnt = NULL;
|
|
|
|
file_free(file);
|
|
|
|
dput(dentry);
|
|
|
|
mntput(mnt);
|
|
|
|
}
|
|
|
|
|
|
|
|
struct file fastcall *fget(unsigned int fd)
|
|
|
|
{
|
|
|
|
struct file *file;
|
|
|
|
struct files_struct *files = current->files;
|
|
|
|
|
2005-09-09 14:04:13 -06:00
|
|
|
rcu_read_lock();
|
2005-04-16 16:20:36 -06:00
|
|
|
file = fcheck_files(files, fd);
|
2005-09-09 14:04:13 -06:00
|
|
|
if (file) {
|
2006-01-08 02:02:19 -07:00
|
|
|
if (!atomic_inc_not_zero(&file->f_count)) {
|
2005-09-09 14:04:13 -06:00
|
|
|
/* File object ref couldn't be taken */
|
|
|
|
rcu_read_unlock();
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
rcu_read_unlock();
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return file;
|
|
|
|
}
|
|
|
|
|
|
|
|
EXPORT_SYMBOL(fget);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Lightweight file lookup - no refcnt increment if fd table isn't shared.
|
|
|
|
* You can use this only if it is guranteed that the current task already
|
|
|
|
* holds a refcnt to that file. That check has to be done at fget() only
|
|
|
|
* and a flag is returned to be passed to the corresponding fput_light().
|
|
|
|
* There must not be a cloning between an fget_light/fput_light pair.
|
|
|
|
*/
|
|
|
|
struct file fastcall *fget_light(unsigned int fd, int *fput_needed)
|
|
|
|
{
|
|
|
|
struct file *file;
|
|
|
|
struct files_struct *files = current->files;
|
|
|
|
|
|
|
|
*fput_needed = 0;
|
|
|
|
if (likely((atomic_read(&files->count) == 1))) {
|
|
|
|
file = fcheck_files(files, fd);
|
|
|
|
} else {
|
2005-09-09 14:04:13 -06:00
|
|
|
rcu_read_lock();
|
2005-04-16 16:20:36 -06:00
|
|
|
file = fcheck_files(files, fd);
|
|
|
|
if (file) {
|
2006-01-08 02:02:19 -07:00
|
|
|
if (atomic_inc_not_zero(&file->f_count))
|
2005-09-09 14:04:13 -06:00
|
|
|
*fput_needed = 1;
|
|
|
|
else
|
|
|
|
/* Didn't get the reference, someone's freed */
|
|
|
|
file = NULL;
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2005-09-09 14:04:13 -06:00
|
|
|
rcu_read_unlock();
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|
2005-09-09 14:04:13 -06:00
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
return file;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
void put_filp(struct file *file)
|
|
|
|
{
|
2006-01-08 02:02:19 -07:00
|
|
|
if (atomic_dec_and_test(&file->f_count)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
security_file_free(file);
|
|
|
|
file_kill(file);
|
|
|
|
file_free(file);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
void file_move(struct file *file, struct list_head *list)
|
|
|
|
{
|
|
|
|
if (!list)
|
|
|
|
return;
|
|
|
|
file_list_lock();
|
2005-10-30 16:02:16 -07:00
|
|
|
list_move(&file->f_u.fu_list, list);
|
2005-04-16 16:20:36 -06:00
|
|
|
file_list_unlock();
|
|
|
|
}
|
|
|
|
|
|
|
|
void file_kill(struct file *file)
|
|
|
|
{
|
2005-10-30 16:02:16 -07:00
|
|
|
if (!list_empty(&file->f_u.fu_list)) {
|
2005-04-16 16:20:36 -06:00
|
|
|
file_list_lock();
|
2005-10-30 16:02:16 -07:00
|
|
|
list_del_init(&file->f_u.fu_list);
|
2005-04-16 16:20:36 -06:00
|
|
|
file_list_unlock();
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
int fs_may_remount_ro(struct super_block *sb)
|
|
|
|
{
|
|
|
|
struct list_head *p;
|
|
|
|
|
|
|
|
/* Check that no files are currently opened for writing. */
|
|
|
|
file_list_lock();
|
|
|
|
list_for_each(p, &sb->s_files) {
|
2005-10-30 16:02:16 -07:00
|
|
|
struct file *file = list_entry(p, struct file, f_u.fu_list);
|
2005-04-16 16:20:36 -06:00
|
|
|
struct inode *inode = file->f_dentry->d_inode;
|
|
|
|
|
|
|
|
/* File with pending delete? */
|
|
|
|
if (inode->i_nlink == 0)
|
|
|
|
goto too_bad;
|
|
|
|
|
|
|
|
/* Writeable file? */
|
|
|
|
if (S_ISREG(inode->i_mode) && (file->f_mode & FMODE_WRITE))
|
|
|
|
goto too_bad;
|
|
|
|
}
|
|
|
|
file_list_unlock();
|
|
|
|
return 1; /* Tis' cool bro. */
|
|
|
|
too_bad:
|
|
|
|
file_list_unlock();
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
void __init files_init(unsigned long mempages)
|
|
|
|
{
|
|
|
|
int n;
|
|
|
|
/* One file with associated inode and dcache is very roughly 1K.
|
|
|
|
* Per default don't use more than 10% of our memory for files.
|
|
|
|
*/
|
|
|
|
|
|
|
|
n = (mempages * (PAGE_SIZE / 1024)) / 10;
|
|
|
|
files_stat.max_files = n;
|
|
|
|
if (files_stat.max_files < NR_FILE)
|
|
|
|
files_stat.max_files = NR_FILE;
|
2005-09-09 14:04:13 -06:00
|
|
|
files_defer_init();
|
2006-06-23 03:05:41 -06:00
|
|
|
percpu_counter_init(&nr_files, 0);
|
2005-04-16 16:20:36 -06:00
|
|
|
}
|