/*
* fs/eventpoll.c ( Efficent event polling implementation )
- * Copyright (C) 2001,...,2003 Davide Libenzi
+ * Copyright (C) 2001,...,2006 Davide Libenzi
*
* 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
#include <linux/eventpoll.h>
#include <linux/mount.h>
#include <linux/bitops.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/io.h>
* LOCKING:
* There are three level of locking required by epoll :
*
- * 1) epsem (semaphore)
+ * 1) epmutex (mutex)
* 2) ep->sem (rw_semaphore)
* 3) ep->lock (rw_lock)
*
* if a file has been pushed inside an epoll set and it is then
* close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
* It is possible to drop the "ep->sem" and to use the global
- * semaphore "epsem" (together with "ep->lock") to have it working,
+ * semaphore "epmutex" (together with "ep->lock") to have it working,
* but having "ep->sem" will make the interface more scalable.
- * Events that require holding "epsem" are very rare, while for
+ * Events that require holding "epmutex" are very rare, while for
* normal operations the epoll private "ep->sem" will guarantee
* a greater scalability.
*/
/* Maximum number of poll wake up nests we are allowing */
#define EP_MAX_POLLWAKE_NESTS 4
+/* Maximum msec timeout value storeable in a long int */
+#define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ)
+
+
struct epoll_filefd {
struct file *file;
int fd;
int maxevents, long timeout);
static int eventpollfs_delete_dentry(struct dentry *dentry);
static struct inode *ep_eventpoll_inode(void);
-static struct super_block *eventpollfs_get_sb(struct file_system_type *fs_type,
- int flags, const char *dev_name,
- void *data);
+static int eventpollfs_get_sb(struct file_system_type *fs_type,
+ int flags, const char *dev_name,
+ void *data, struct vfsmount *mnt);
/*
* This semaphore is used to serialize ep_free() and eventpoll_release_file().
*/
-static struct semaphore epsem;
+static struct mutex epmutex;
/* Safe wake up implementation */
static struct poll_safewake psw;
/* Slab cache used to allocate "struct epitem" */
-static kmem_cache_t *epi_cache;
+static kmem_cache_t *epi_cache __read_mostly;
/* Slab cache used to allocate "struct eppoll_entry" */
-static kmem_cache_t *pwq_cache;
+static kmem_cache_t *pwq_cache __read_mostly;
/* Virtual fs used to allocate inodes for eventpoll files */
-static struct vfsmount *eventpoll_mnt;
+static struct vfsmount *eventpoll_mnt __read_mostly;
/* File callbacks that implement the eventpoll file behaviour */
-static struct file_operations eventpoll_fops = {
+static const struct file_operations eventpoll_fops = {
.release = ep_eventpoll_close,
.poll = ep_eventpoll_poll
};
/* Special initialization for the rb-tree node to detect linkage */
static inline void ep_rb_initnode(struct rb_node *n)
{
- n->rb_parent = n;
+ rb_set_parent(n, n);
}
/* Removes a node from the rb-tree and marks it for a fast is-linked check */
static inline void ep_rb_erase(struct rb_node *n, struct rb_root *r)
{
rb_erase(n, r);
- n->rb_parent = n;
+ rb_set_parent(n, n);
}
/* Fast check to verify that the item is linked to the main rb-tree */
static inline int ep_rb_linked(struct rb_node *n)
{
- return n->rb_parent != n;
+ return rb_parent(n) != n;
}
/*
}
-/* Used to initialize the epoll bits inside the "struct file" */
-void eventpoll_init_file(struct file *file)
-{
-
- INIT_LIST_HEAD(&file->f_ep_links);
- spin_lock_init(&file->f_ep_lock);
-}
-
-
/*
* This is called from eventpoll_release() to unlink files from the eventpoll
* interface. We need to have this facility to cleanup correctly files that are
* cleanup path, and this means that noone is using this file anymore.
* The only hit might come from ep_free() but by holding the semaphore
* will correctly serialize the operation. We do need to acquire
- * "ep->sem" after "epsem" because ep_remove() requires it when called
+ * "ep->sem" after "epmutex" because ep_remove() requires it when called
* from anywhere but ep_free().
*/
- down(&epsem);
+ mutex_lock(&epmutex);
while (!list_empty(lsthead)) {
epi = list_entry(lsthead->next, struct epitem, fllink);
up_write(&ep->sem);
}
- up(&epsem);
+ mutex_unlock(&epmutex);
}
* We do not need to hold "ep->sem" here because the epoll file
* is on the way to be removed and no one has references to it
* anymore. The only hit might come from eventpoll_release_file() but
- * holding "epsem" is sufficent here.
+ * holding "epmutex" is sufficent here.
*/
- down(&epsem);
+ mutex_lock(&epmutex);
/*
* Walks through the whole tree by unregistering poll callbacks.
ep_remove(ep, epi);
}
- up(&epsem);
+ mutex_unlock(&epmutex);
}
/* Notify waiting tasks that events are available */
if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
+ __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE | TASK_INTERRUPTIBLE);
if (waitqueue_active(&ep->poll_wait))
pwake++;
}
/* Notify waiting tasks that events are available */
if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
+ __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE |
+ TASK_INTERRUPTIBLE);
if (waitqueue_active(&ep->poll_wait))
pwake++;
}
* wait list.
*/
if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
+ __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE |
+ TASK_INTERRUPTIBLE);
if (waitqueue_active(&ep->poll_wait))
pwake++;
* wait list.
*/
if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
+ __wake_up_locked(&ep->wq, TASK_UNINTERRUPTIBLE |
+ TASK_INTERRUPTIBLE);
if (waitqueue_active(&ep->poll_wait))
pwake++;
}
* and the overflow condition. The passed timeout is in milliseconds,
* that why (t * HZ) / 1000.
*/
- jtimeout = timeout == -1 || timeout > (MAX_SCHEDULE_TIMEOUT - 1000) / HZ ?
- MAX_SCHEDULE_TIMEOUT: (timeout * HZ + 999) / 1000;
+ jtimeout = (timeout < 0 || timeout >= EP_MAX_MSTIMEO) ?
+ MAX_SCHEDULE_TIMEOUT : (timeout * HZ + 999) / 1000;
retry:
write_lock_irqsave(&ep->lock, flags);
* ep_poll_callback() when events will become available.
*/
init_waitqueue_entry(&wait, current);
- add_wait_queue(&ep->wq, &wait);
+ __add_wait_queue(&ep->wq, &wait);
for (;;) {
/*
jtimeout = schedule_timeout(jtimeout);
write_lock_irqsave(&ep->lock, flags);
}
- remove_wait_queue(&ep->wq, &wait);
+ __remove_wait_queue(&ep->wq, &wait);
set_current_state(TASK_RUNNING);
}
}
-static struct super_block *
+static int
eventpollfs_get_sb(struct file_system_type *fs_type, int flags,
- const char *dev_name, void *data)
+ const char *dev_name, void *data, struct vfsmount *mnt)
{
- return get_sb_pseudo(fs_type, "eventpoll:", NULL, EVENTPOLLFS_MAGIC);
+ return get_sb_pseudo(fs_type, "eventpoll:", NULL, EVENTPOLLFS_MAGIC,
+ mnt);
}
{
int error;
- init_MUTEX(&epsem);
+ mutex_init(&epmutex);
/* Initialize the structure used to perform safe poll wait head wake ups */
ep_poll_safewake_init(&psw);
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff