/*
- * fs/eventpoll.c ( Efficent event polling implementation )
- * Copyright (C) 2001,...,2003 Davide Libenzi
+ * fs/eventpoll.c (Efficient event retrieval implementation)
+ * Copyright (C) 2001,...,2009 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/module.h>
#include <linux/init.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/poll.h>
-#include <linux/smp_lock.h>
#include <linux/string.h>
#include <linux/list.h>
#include <linux/hash.h>
#include <linux/spinlock.h>
#include <linux/syscalls.h>
-#include <linux/rwsem.h>
#include <linux/rbtree.h>
#include <linux/wait.h>
#include <linux/eventpoll.h>
#include <linux/mount.h>
#include <linux/bitops.h>
#include <linux/mutex.h>
+#include <linux/anon_inodes.h>
#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/io.h>
#include <asm/mman.h>
-#include <asm/atomic.h>
-#include <asm/semaphore.h>
-
+#include <linux/atomic.h>
/*
* LOCKING:
* There are three level of locking required by epoll :
*
* 1) epmutex (mutex)
- * 2) ep->sem (rw_semaphore)
- * 3) ep->lock (rw_lock)
+ * 2) ep->mtx (mutex)
+ * 3) ep->lock (spinlock)
*
* The acquire order is the one listed above, from 1 to 3.
* We need a spinlock (ep->lock) because we manipulate objects
* a spinlock. During the event transfer loop (from kernel to
* user space) we could end up sleeping due a copy_to_user(), so
* we need a lock that will allow us to sleep. This lock is a
- * read-write semaphore (ep->sem). It is acquired on read during
- * the event transfer loop and in write during epoll_ctl(EPOLL_CTL_DEL)
- * and during eventpoll_release_file(). Then we also need a global
- * semaphore to serialize eventpoll_release_file() and ep_free().
- * This semaphore is acquired by ep_free() during the epoll file
+ * mutex (ep->mtx). It is acquired during the event transfer loop,
+ * during epoll_ctl(EPOLL_CTL_DEL) and during eventpoll_release_file().
+ * Then we also need a global mutex to serialize eventpoll_release_file()
+ * and ep_free().
+ * This mutex is acquired by ep_free() during the epoll file
* cleanup path and it is also acquired by eventpoll_release_file()
* 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 "epmutex" (together with "ep->lock") to have it working,
- * but having "ep->sem" will make the interface more scalable.
+ * close()d without a previous call to epoll_ctl(EPOLL_CTL_DEL).
+ * It is also acquired when inserting an epoll fd onto another epoll
+ * fd. We do this so that we walk the epoll tree and ensure that this
+ * insertion does not create a cycle of epoll file descriptors, which
+ * could lead to deadlock. We need a global mutex to prevent two
+ * simultaneous inserts (A into B and B into A) from racing and
+ * constructing a cycle without either insert observing that it is
+ * going to.
+ * It is possible to drop the "ep->mtx" and to use the global
+ * mutex "epmutex" (together with "ep->lock") to have it working,
+ * but having "ep->mtx" will make the interface more scalable.
* Events that require holding "epmutex" are very rare, while for
- * normal operations the epoll private "ep->sem" will guarantee
- * a greater scalability.
+ * normal operations the epoll private "ep->mtx" will guarantee
+ * a better scalability.
*/
-
-#define EVENTPOLLFS_MAGIC 0x03111965 /* My birthday should work for this :) */
-
-#define DEBUG_EPOLL 0
-
-#if DEBUG_EPOLL > 0
-#define DPRINTK(x) printk x
-#define DNPRINTK(n, x) do { if ((n) <= DEBUG_EPOLL) printk x; } while (0)
-#else /* #if DEBUG_EPOLL > 0 */
-#define DPRINTK(x) (void) 0
-#define DNPRINTK(n, x) (void) 0
-#endif /* #if DEBUG_EPOLL > 0 */
-
-#define DEBUG_EPI 0
-
-#if DEBUG_EPI != 0
-#define EPI_SLAB_DEBUG (SLAB_DEBUG_FREE | SLAB_RED_ZONE /* | SLAB_POISON */)
-#else /* #if DEBUG_EPI != 0 */
-#define EPI_SLAB_DEBUG 0
-#endif /* #if DEBUG_EPI != 0 */
-
/* Epoll private bits inside the event mask */
#define EP_PRIVATE_BITS (EPOLLONESHOT | EPOLLET)
-/* Maximum number of poll wake up nests we are allowing */
-#define EP_MAX_POLLWAKE_NESTS 4
+/* Maximum number of nesting allowed inside epoll sets */
+#define EP_MAX_NESTS 4
+
+#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
-/* Maximum msec timeout value storeable in a long int */
-#define EP_MAX_MSTIMEO min(1000ULL * MAX_SCHEDULE_TIMEOUT / HZ, (LONG_MAX - 999ULL) / HZ)
+#define EP_UNACTIVE_PTR ((void *) -1L)
+#define EP_ITEM_COST (sizeof(struct epitem) + sizeof(struct eppoll_entry))
struct epoll_filefd {
struct file *file;
};
/*
- * Node that is linked into the "wake_task_list" member of the "struct poll_safewake".
- * It is used to keep track on all tasks that are currently inside the wake_up() code
- * to 1) short-circuit the one coming from the same task and same wait queue head
- * ( loop ) 2) allow a maximum number of epoll descriptors inclusion nesting
- * 3) let go the ones coming from other tasks.
+ * Structure used to track possible nested calls, for too deep recursions
+ * and loop cycles.
*/
-struct wake_task_node {
+struct nested_call_node {
struct list_head llink;
- task_t *task;
- wait_queue_head_t *wq;
+ void *cookie;
+ void *ctx;
};
/*
- * This is used to implement the safe poll wake up avoiding to reenter
- * the poll callback from inside wake_up().
+ * This structure is used as collector for nested calls, to check for
+ * maximum recursion dept and loop cycles.
*/
-struct poll_safewake {
- struct list_head wake_task_list;
+struct nested_calls {
+ struct list_head tasks_call_list;
spinlock_t lock;
};
+/*
+ * Each file descriptor added to the eventpoll interface will
+ * have an entry of this type linked to the "rbr" RB tree.
+ */
+struct epitem {
+ /* RB tree node used to link this structure to the eventpoll RB tree */
+ struct rb_node rbn;
+
+ /* List header used to link this structure to the eventpoll ready list */
+ struct list_head rdllink;
+
+ /*
+ * Works together "struct eventpoll"->ovflist in keeping the
+ * single linked chain of items.
+ */
+ struct epitem *next;
+
+ /* The file descriptor information this item refers to */
+ struct epoll_filefd ffd;
+
+ /* Number of active wait queue attached to poll operations */
+ int nwait;
+
+ /* List containing poll wait queues */
+ struct list_head pwqlist;
+
+ /* The "container" of this item */
+ struct eventpoll *ep;
+
+ /* List header used to link this item to the "struct file" items list */
+ struct list_head fllink;
+
+ /* The structure that describe the interested events and the source fd */
+ struct epoll_event event;
+};
+
/*
* This structure is stored inside the "private_data" member of the file
- * structure and rapresent the main data sructure for the eventpoll
+ * structure and represents the main data structure for the eventpoll
* interface.
*/
struct eventpoll {
- /* Protect the this structure access */
- rwlock_t lock;
+ /* Protect the access to this structure */
+ spinlock_t lock;
/*
- * This semaphore is used to ensure that files are not removed
- * while epoll is using them. This is read-held during the event
- * collection loop and it is write-held during the file cleanup
- * path, the epoll file exit code and the ctl operations.
+ * This mutex is used to ensure that files are not removed
+ * while epoll is using them. This is held during the event
+ * collection loop, the file cleanup path, the epoll file exit
+ * code and the ctl operations.
*/
- struct rw_semaphore sem;
+ struct mutex mtx;
/* Wait queue used by sys_epoll_wait() */
wait_queue_head_t wq;
/* List of ready file descriptors */
struct list_head rdllist;
- /* RB-Tree root used to store monitored fd structs */
+ /* RB tree root used to store monitored fd structs */
struct rb_root rbr;
+
+ /*
+ * This is a single linked list that chains all the "struct epitem" that
+ * happened while transferring ready events to userspace w/out
+ * holding ->lock.
+ */
+ struct epitem *ovflist;
+
+ /* The user that created the eventpoll descriptor */
+ struct user_struct *user;
};
/* Wait structure used by the poll hooks */
struct list_head llink;
/* The "base" pointer is set to the container "struct epitem" */
- void *base;
+ struct epitem *base;
/*
* Wait queue item that will be linked to the target file wait
wait_queue_head_t *whead;
};
-/*
- * Each file descriptor added to the eventpoll interface will
- * have an entry of this type linked to the hash.
- */
-struct epitem {
- /* RB-Tree node used to link this structure to the eventpoll rb-tree */
- struct rb_node rbn;
-
- /* List header used to link this structure to the eventpoll ready list */
- struct list_head rdllink;
-
- /* The file descriptor information this item refers to */
- struct epoll_filefd ffd;
-
- /* Number of active wait queue attached to poll operations */
- int nwait;
-
- /* List containing poll wait queues */
- struct list_head pwqlist;
-
- /* The "container" of this item */
- struct eventpoll *ep;
-
- /* The structure that describe the interested events and the source fd */
- struct epoll_event event;
-
- /*
- * Used to keep track of the usage count of the structure. This avoids
- * that the structure will desappear from underneath our processing.
- */
- atomic_t usecnt;
-
- /* List header used to link this item to the "struct file" items list */
- struct list_head fllink;
-
- /* List header used to link the item to the transfer list */
- struct list_head txlink;
-
- /*
- * This is used during the collection/transfer of events to userspace
- * to pin items empty events set.
- */
- unsigned int revents;
-};
-
/* Wrapper struct used by poll queueing */
struct ep_pqueue {
poll_table pt;
struct epitem *epi;
};
-
-
-static void ep_poll_safewake_init(struct poll_safewake *psw);
-static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq);
-static int ep_getfd(int *efd, struct inode **einode, struct file **efile,
- struct eventpoll *ep);
-static int ep_alloc(struct eventpoll **pep);
-static void ep_free(struct eventpoll *ep);
-static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd);
-static void ep_use_epitem(struct epitem *epi);
-static void ep_release_epitem(struct epitem *epi);
-static void ep_ptable_queue_proc(struct file *file, wait_queue_head_t *whead,
- poll_table *pt);
-static void ep_rbtree_insert(struct eventpoll *ep, struct epitem *epi);
-static int ep_insert(struct eventpoll *ep, struct epoll_event *event,
- struct file *tfile, int fd);
-static int ep_modify(struct eventpoll *ep, struct epitem *epi,
- struct epoll_event *event);
-static void ep_unregister_pollwait(struct eventpoll *ep, struct epitem *epi);
-static int ep_unlink(struct eventpoll *ep, struct epitem *epi);
-static int ep_remove(struct eventpoll *ep, struct epitem *epi);
-static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *key);
-static int ep_eventpoll_close(struct inode *inode, struct file *file);
-static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait);
-static int ep_collect_ready_items(struct eventpoll *ep,
- struct list_head *txlist, int maxevents);
-static int ep_send_events(struct eventpoll *ep, struct list_head *txlist,
- struct epoll_event __user *events);
-static void ep_reinject_items(struct eventpoll *ep, struct list_head *txlist);
-static int ep_events_transfer(struct eventpoll *ep,
- struct epoll_event __user *events,
- int maxevents);
-static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
- 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);
+/* Used by the ep_send_events() function as callback private data */
+struct ep_send_events_data {
+ int maxevents;
+ struct epoll_event __user *events;
+};
/*
- * This semaphore is used to serialize ep_free() and eventpoll_release_file().
+ * Configuration options available inside /proc/sys/fs/epoll/
*/
-static struct mutex epmutex;
+/* Maximum number of epoll watched descriptors, per user */
+static long max_user_watches __read_mostly;
-/* Safe wake up implementation */
-static struct poll_safewake psw;
-
-/* Slab cache used to allocate "struct epitem" */
-static kmem_cache_t *epi_cache __read_mostly;
+/*
+ * This mutex is used to serialize ep_free() and eventpoll_release_file().
+ */
+static DEFINE_MUTEX(epmutex);
-/* Slab cache used to allocate "struct eppoll_entry" */
-static kmem_cache_t *pwq_cache __read_mostly;
+/* Used to check for epoll file descriptor inclusion loops */
+static struct nested_calls poll_loop_ncalls;
-/* Virtual fs used to allocate inodes for eventpoll files */
-static struct vfsmount *eventpoll_mnt __read_mostly;
+/* Used for safe wake up implementation */
+static struct nested_calls poll_safewake_ncalls;
-/* File callbacks that implement the eventpoll file behaviour */
-static const struct file_operations eventpoll_fops = {
- .release = ep_eventpoll_close,
- .poll = ep_eventpoll_poll
-};
+/* Used to call file's f_op->poll() under the nested calls boundaries */
+static struct nested_calls poll_readywalk_ncalls;
-/*
- * This is used to register the virtual file system from where
- * eventpoll inodes are allocated.
- */
-static struct file_system_type eventpoll_fs_type = {
- .name = "eventpollfs",
- .get_sb = eventpollfs_get_sb,
- .kill_sb = kill_anon_super,
-};
+/* Slab cache used to allocate "struct epitem" */
+static struct kmem_cache *epi_cache __read_mostly;
-/* Very basic directory entry operations for the eventpoll virtual file system */
-static struct dentry_operations eventpollfs_dentry_operations = {
- .d_delete = eventpollfs_delete_dentry,
+/* Slab cache used to allocate "struct eppoll_entry" */
+static struct kmem_cache *pwq_cache __read_mostly;
+
+#ifdef CONFIG_SYSCTL
+
+#include <linux/sysctl.h>
+
+static long zero;
+static long long_max = LONG_MAX;
+
+ctl_table epoll_table[] = {
+ {
+ .procname = "max_user_watches",
+ .data = &max_user_watches,
+ .maxlen = sizeof(max_user_watches),
+ .mode = 0644,
+ .proc_handler = proc_doulongvec_minmax,
+ .extra1 = &zero,
+ .extra2 = &long_max,
+ },
+ { }
};
+#endif /* CONFIG_SYSCTL */
-
-/* Fast test to see if the file is an evenpoll file */
-static inline int is_file_epoll(struct file *f)
-{
- return f->f_op == &eventpoll_fops;
-}
-
-/* Setup the structure that is used as key for the rb-tree */
+/* Setup the structure that is used as key for the RB tree */
static inline void ep_set_ffd(struct epoll_filefd *ffd,
struct file *file, int fd)
{
ffd->fd = fd;
}
-/* Compare rb-tree keys */
+/* Compare RB tree keys */
static inline int ep_cmp_ffd(struct epoll_filefd *p1,
struct epoll_filefd *p2)
{
(p1->file < p2->file ? -1 : p1->fd - p2->fd));
}
-/* Special initialization for the rb-tree node to detect linkage */
-static inline void ep_rb_initnode(struct rb_node *n)
-{
- n->rb_parent = 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;
-}
-
-/* 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;
-}
-
-/*
- * Remove the item from the list and perform its initialization.
- * This is useful for us because we can test if the item is linked
- * using "ep_is_linked(p)".
- */
-static inline void ep_list_del(struct list_head *p)
-{
- list_del(p);
- INIT_LIST_HEAD(p);
-}
-
/* Tells us if the item is currently linked */
static inline int ep_is_linked(struct list_head *p)
{
}
/* Get the "struct epitem" from a wait queue pointer */
-static inline struct epitem * ep_item_from_wait(wait_queue_t *p)
+static inline struct epitem *ep_item_from_wait(wait_queue_t *p)
{
return container_of(p, struct eppoll_entry, wait)->base;
}
/* Get the "struct epitem" from an epoll queue wrapper */
-static inline struct epitem * ep_item_from_epqueue(poll_table *p)
+static inline struct epitem *ep_item_from_epqueue(poll_table *p)
{
return container_of(p, struct ep_pqueue, pt)->epi;
}
/* Tells if the epoll_ctl(2) operation needs an event copy from userspace */
-static inline int ep_op_hash_event(int op)
+static inline int ep_op_has_event(int op)
{
return op != EPOLL_CTL_DEL;
}
/* Initialize the poll safe wake up structure */
-static void ep_poll_safewake_init(struct poll_safewake *psw)
+static void ep_nested_calls_init(struct nested_calls *ncalls)
{
-
- INIT_LIST_HEAD(&psw->wake_task_list);
- spin_lock_init(&psw->lock);
+ INIT_LIST_HEAD(&ncalls->tasks_call_list);
+ spin_lock_init(&ncalls->lock);
}
+/**
+ * ep_events_available - Checks if ready events might be available.
+ *
+ * @ep: Pointer to the eventpoll context.
+ *
+ * Returns: Returns a value different than zero if ready events are available,
+ * or zero otherwise.
+ */
+static inline int ep_events_available(struct eventpoll *ep)
+{
+ return !list_empty(&ep->rdllist) || ep->ovflist != EP_UNACTIVE_PTR;
+}
-/*
- * Perform a safe wake up of the poll wait list. The problem is that
- * with the new callback'd wake up system, it is possible that the
- * poll callback is reentered from inside the call to wake_up() done
- * on the poll wait queue head. The rule is that we cannot reenter the
- * wake up code from the same task more than EP_MAX_POLLWAKE_NESTS times,
- * and we cannot reenter the same wait queue head at all. This will
- * enable to have a hierarchy of epoll file descriptor of no more than
- * EP_MAX_POLLWAKE_NESTS deep. We need the irq version of the spin lock
- * because this one gets called by the poll callback, that in turn is called
- * from inside a wake_up(), that might be called from irq context.
+/**
+ * ep_call_nested - Perform a bound (possibly) nested call, by checking
+ * that the recursion limit is not exceeded, and that
+ * the same nested call (by the meaning of same cookie) is
+ * no re-entered.
+ *
+ * @ncalls: Pointer to the nested_calls structure to be used for this call.
+ * @max_nests: Maximum number of allowed nesting calls.
+ * @nproc: Nested call core function pointer.
+ * @priv: Opaque data to be passed to the @nproc callback.
+ * @cookie: Cookie to be used to identify this nested call.
+ * @ctx: This instance context.
+ *
+ * Returns: Returns the code returned by the @nproc callback, or -1 if
+ * the maximum recursion limit has been exceeded.
*/
-static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq)
+static int ep_call_nested(struct nested_calls *ncalls, int max_nests,
+ int (*nproc)(void *, void *, int), void *priv,
+ void *cookie, void *ctx)
{
- int wake_nests = 0;
+ int error, call_nests = 0;
unsigned long flags;
- task_t *this_task = current;
- struct list_head *lsthead = &psw->wake_task_list, *lnk;
- struct wake_task_node *tncur;
- struct wake_task_node tnode;
-
- spin_lock_irqsave(&psw->lock, flags);
+ struct list_head *lsthead = &ncalls->tasks_call_list;
+ struct nested_call_node *tncur;
+ struct nested_call_node tnode;
- /* Try to see if the current task is already inside this wakeup call */
- list_for_each(lnk, lsthead) {
- tncur = list_entry(lnk, struct wake_task_node, llink);
+ spin_lock_irqsave(&ncalls->lock, flags);
- if (tncur->wq == wq ||
- (tncur->task == this_task && ++wake_nests > EP_MAX_POLLWAKE_NESTS)) {
+ /*
+ * Try to see if the current task is already inside this wakeup call.
+ * We use a list here, since the population inside this set is always
+ * very much limited.
+ */
+ list_for_each_entry(tncur, lsthead, llink) {
+ if (tncur->ctx == ctx &&
+ (tncur->cookie == cookie || ++call_nests > max_nests)) {
/*
* Ops ... loop detected or maximum nest level reached.
* We abort this wake by breaking the cycle itself.
*/
- spin_unlock_irqrestore(&psw->lock, flags);
- return;
+ error = -1;
+ goto out_unlock;
}
}
- /* Add the current task to the list */
- tnode.task = this_task;
- tnode.wq = wq;
+ /* Add the current task and cookie to the list */
+ tnode.ctx = ctx;
+ tnode.cookie = cookie;
list_add(&tnode.llink, lsthead);
- spin_unlock_irqrestore(&psw->lock, flags);
+ spin_unlock_irqrestore(&ncalls->lock, flags);
- /* Do really wake up now */
- wake_up(wq);
+ /* Call the nested function */
+ error = (*nproc)(priv, cookie, call_nests);
/* Remove the current task from the list */
- spin_lock_irqsave(&psw->lock, flags);
+ spin_lock_irqsave(&ncalls->lock, flags);
list_del(&tnode.llink);
- spin_unlock_irqrestore(&psw->lock, flags);
+out_unlock:
+ spin_unlock_irqrestore(&ncalls->lock, flags);
+
+ return error;
+}
+
+#ifdef CONFIG_DEBUG_LOCK_ALLOC
+static inline void ep_wake_up_nested(wait_queue_head_t *wqueue,
+ unsigned long events, int subclass)
+{
+ unsigned long flags;
+
+ spin_lock_irqsave_nested(&wqueue->lock, flags, subclass);
+ wake_up_locked_poll(wqueue, events);
+ spin_unlock_irqrestore(&wqueue->lock, flags);
+}
+#else
+static inline void ep_wake_up_nested(wait_queue_head_t *wqueue,
+ unsigned long events, int subclass)
+{
+ wake_up_poll(wqueue, events);
}
+#endif
+static int ep_poll_wakeup_proc(void *priv, void *cookie, int call_nests)
+{
+ ep_wake_up_nested((wait_queue_head_t *) cookie, POLLIN,
+ 1 + call_nests);
+ return 0;
+}
/*
- * 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
- * closed without being removed from the eventpoll interface.
+ * Perform a safe wake up of the poll wait list. The problem is that
+ * with the new callback'd wake up system, it is possible that the
+ * poll callback is reentered from inside the call to wake_up() done
+ * on the poll wait queue head. The rule is that we cannot reenter the
+ * wake up code from the same task more than EP_MAX_NESTS times,
+ * and we cannot reenter the same wait queue head at all. This will
+ * enable to have a hierarchy of epoll file descriptor of no more than
+ * EP_MAX_NESTS deep.
*/
-void eventpoll_release_file(struct file *file)
+static void ep_poll_safewake(wait_queue_head_t *wq)
{
- struct list_head *lsthead = &file->f_ep_links;
- struct eventpoll *ep;
- struct epitem *epi;
+ int this_cpu = get_cpu();
- /*
- * We don't want to get "file->f_ep_lock" because it is not
- * necessary. It is not necessary because we're in the "struct file"
- * 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 "epmutex" because ep_remove() requires it when called
- * from anywhere but ep_free().
- */
- mutex_lock(&epmutex);
+ ep_call_nested(&poll_safewake_ncalls, EP_MAX_NESTS,
+ ep_poll_wakeup_proc, NULL, wq, (void *) (long) this_cpu);
+
+ put_cpu();
+}
+
+/*
+ * This function unregisters poll callbacks from the associated file
+ * descriptor. Must be called with "mtx" held (or "epmutex" if called from
+ * ep_free).
+ */
+static void ep_unregister_pollwait(struct eventpoll *ep, struct epitem *epi)
+{
+ struct list_head *lsthead = &epi->pwqlist;
+ struct eppoll_entry *pwq;
while (!list_empty(lsthead)) {
- epi = list_entry(lsthead->next, struct epitem, fllink);
+ pwq = list_first_entry(lsthead, struct eppoll_entry, llink);
- ep = epi->ep;
- ep_list_del(&epi->fllink);
- down_write(&ep->sem);
- ep_remove(ep, epi);
- up_write(&ep->sem);
+ list_del(&pwq->llink);
+ remove_wait_queue(pwq->whead, &pwq->wait);
+ kmem_cache_free(pwq_cache, pwq);
}
-
- mutex_unlock(&epmutex);
}
-
-/*
- * It opens an eventpoll file descriptor by suggesting a storage of "size"
- * file descriptors. The size parameter is just an hint about how to size
- * data structures. It won't prevent the user to store more than "size"
- * file descriptors inside the epoll interface. It is the kernel part of
- * the userspace epoll_create(2).
+/**
+ * ep_scan_ready_list - Scans the ready list in a way that makes possible for
+ * the scan code, to call f_op->poll(). Also allows for
+ * O(NumReady) performance.
+ *
+ * @ep: Pointer to the epoll private data structure.
+ * @sproc: Pointer to the scan callback.
+ * @priv: Private opaque data passed to the @sproc callback.
+ *
+ * Returns: The same integer error code returned by the @sproc callback.
*/
-asmlinkage long sys_epoll_create(int size)
+static int ep_scan_ready_list(struct eventpoll *ep,
+ int (*sproc)(struct eventpoll *,
+ struct list_head *, void *),
+ void *priv)
{
- int error, fd;
- struct eventpoll *ep;
- struct inode *inode;
- struct file *file;
+ int error, pwake = 0;
+ unsigned long flags;
+ struct epitem *epi, *nepi;
+ LIST_HEAD(txlist);
+
+ /*
+ * We need to lock this because we could be hit by
+ * eventpoll_release_file() and epoll_ctl().
+ */
+ mutex_lock(&ep->mtx);
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d)\n",
- current, size));
+ /*
+ * Steal the ready list, and re-init the original one to the
+ * empty list. Also, set ep->ovflist to NULL so that events
+ * happening while looping w/out locks, are not lost. We cannot
+ * have the poll callback to queue directly on ep->rdllist,
+ * because we want the "sproc" callback to be able to do it
+ * in a lockless way.
+ */
+ spin_lock_irqsave(&ep->lock, flags);
+ list_splice_init(&ep->rdllist, &txlist);
+ ep->ovflist = NULL;
+ spin_unlock_irqrestore(&ep->lock, flags);
/*
- * Sanity check on the size parameter, and create the internal data
- * structure ( "struct eventpoll" ).
+ * Now call the callback function.
*/
- error = -EINVAL;
- if (size <= 0 || (error = ep_alloc(&ep)) != 0)
- goto eexit_1;
+ error = (*sproc)(ep, &txlist, priv);
+ spin_lock_irqsave(&ep->lock, flags);
/*
- * Creates all the items needed to setup an eventpoll file. That is,
- * a file structure, and inode and a free file descriptor.
+ * During the time we spent inside the "sproc" callback, some
+ * other events might have been queued by the poll callback.
+ * We re-insert them inside the main ready-list here.
+ */
+ for (nepi = ep->ovflist; (epi = nepi) != NULL;
+ nepi = epi->next, epi->next = EP_UNACTIVE_PTR) {
+ /*
+ * We need to check if the item is already in the list.
+ * During the "sproc" callback execution time, items are
+ * queued into ->ovflist but the "txlist" might already
+ * contain them, and the list_splice() below takes care of them.
+ */
+ if (!ep_is_linked(&epi->rdllink))
+ list_add_tail(&epi->rdllink, &ep->rdllist);
+ }
+ /*
+ * We need to set back ep->ovflist to EP_UNACTIVE_PTR, so that after
+ * releasing the lock, events will be queued in the normal way inside
+ * ep->rdllist.
+ */
+ ep->ovflist = EP_UNACTIVE_PTR;
+
+ /*
+ * Quickly re-inject items left on "txlist".
*/
- error = ep_getfd(&fd, &inode, &file, ep);
- if (error)
- goto eexit_2;
+ list_splice(&txlist, &ep->rdllist);
+
+ if (!list_empty(&ep->rdllist)) {
+ /*
+ * Wake up (if active) both the eventpoll wait list and
+ * the ->poll() wait list (delayed after we release the lock).
+ */
+ if (waitqueue_active(&ep->wq))
+ wake_up_locked(&ep->wq);
+ if (waitqueue_active(&ep->poll_wait))
+ pwake++;
+ }
+ spin_unlock_irqrestore(&ep->lock, flags);
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
- current, size, fd));
+ mutex_unlock(&ep->mtx);
- return fd;
+ /* We have to call this outside the lock */
+ if (pwake)
+ ep_poll_safewake(&ep->poll_wait);
-eexit_2:
- ep_free(ep);
- kfree(ep);
-eexit_1:
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_create(%d) = %d\n",
- current, size, error));
return error;
}
-
/*
- * The following function implements the controller interface for
- * the eventpoll file that enables the insertion/removal/change of
- * file descriptors inside the interest set. It represents
- * the kernel part of the user space epoll_ctl(2).
+ * Removes a "struct epitem" from the eventpoll RB tree and deallocates
+ * all the associated resources. Must be called with "mtx" held.
*/
-asmlinkage long
-sys_epoll_ctl(int epfd, int op, int fd, struct epoll_event __user *event)
+static int ep_remove(struct eventpoll *ep, struct epitem *epi)
{
- int error;
- struct file *file, *tfile;
- struct eventpoll *ep;
- struct epitem *epi;
- struct epoll_event epds;
+ unsigned long flags;
+ struct file *file = epi->ffd.file;
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_ctl(%d, %d, %d, %p)\n",
- current, epfd, op, fd, event));
+ /*
+ * Removes poll wait queue hooks. We _have_ to do this without holding
+ * the "ep->lock" otherwise a deadlock might occur. This because of the
+ * sequence of the lock acquisition. Here we do "ep->lock" then the wait
+ * queue head lock when unregistering the wait queue. The wakeup callback
+ * will run by holding the wait queue head lock and will call our callback
+ * that will try to get "ep->lock".
+ */
+ ep_unregister_pollwait(ep, epi);
- error = -EFAULT;
- if (ep_op_hash_event(op) &&
- copy_from_user(&epds, event, sizeof(struct epoll_event)))
- goto eexit_1;
+ /* Remove the current item from the list of epoll hooks */
+ spin_lock(&file->f_lock);
+ if (ep_is_linked(&epi->fllink))
+ list_del_init(&epi->fllink);
+ spin_unlock(&file->f_lock);
- /* Get the "struct file *" for the eventpoll file */
- error = -EBADF;
- file = fget(epfd);
- if (!file)
- goto eexit_1;
+ rb_erase(&epi->rbn, &ep->rbr);
- /* Get the "struct file *" for the target file */
- tfile = fget(fd);
- if (!tfile)
- goto eexit_2;
+ spin_lock_irqsave(&ep->lock, flags);
+ if (ep_is_linked(&epi->rdllink))
+ list_del_init(&epi->rdllink);
+ spin_unlock_irqrestore(&ep->lock, flags);
- /* The target file descriptor must support poll */
- error = -EPERM;
- if (!tfile->f_op || !tfile->f_op->poll)
- goto eexit_3;
+ /* At this point it is safe to free the eventpoll item */
+ kmem_cache_free(epi_cache, epi);
- /*
- * We have to check that the file structure underneath the file descriptor
- * the user passed to us _is_ an eventpoll file. And also we do not permit
- * adding an epoll file descriptor inside itself.
- */
- error = -EINVAL;
- if (file == tfile || !is_file_epoll(file))
- goto eexit_3;
+ atomic_long_dec(&ep->user->epoll_watches);
- /*
- * At this point it is safe to assume that the "private_data" contains
- * our own data structure.
- */
- ep = file->private_data;
+ return 0;
+}
- down_write(&ep->sem);
+static void ep_free(struct eventpoll *ep)
+{
+ struct rb_node *rbp;
+ struct epitem *epi;
- /* Try to lookup the file inside our hash table */
- epi = ep_find(ep, tfile, fd);
+ /* We need to release all tasks waiting for these file */
+ if (waitqueue_active(&ep->poll_wait))
+ ep_poll_safewake(&ep->poll_wait);
- error = -EINVAL;
- switch (op) {
- case EPOLL_CTL_ADD:
- if (!epi) {
- epds.events |= POLLERR | POLLHUP | POLLRDHUP;
+ /*
+ * We need to lock this because we could be hit by
+ * eventpoll_release_file() while we're freeing the "struct eventpoll".
+ * We do not need to hold "ep->mtx" 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 "epmutex" is sufficient here.
+ */
+ mutex_lock(&epmutex);
- error = ep_insert(ep, &epds, tfile, fd);
- } else
- error = -EEXIST;
- break;
- case EPOLL_CTL_DEL:
- if (epi)
- error = ep_remove(ep, epi);
- else
- error = -ENOENT;
- break;
- case EPOLL_CTL_MOD:
- if (epi) {
- epds.events |= POLLERR | POLLHUP | POLLRDHUP;
- error = ep_modify(ep, epi, &epds);
- } else
- error = -ENOENT;
- break;
+ /*
+ * Walks through the whole tree by unregistering poll callbacks.
+ */
+ for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
+ epi = rb_entry(rbp, struct epitem, rbn);
+
+ ep_unregister_pollwait(ep, epi);
}
/*
- * The function ep_find() increments the usage count of the structure
- * so, if this is not NULL, we need to release it.
+ * Walks through the whole tree by freeing each "struct epitem". At this
+ * point we are sure no poll callbacks will be lingering around, and also by
+ * holding "epmutex" we can be sure that no file cleanup code will hit
+ * us during this operation. So we can avoid the lock on "ep->lock".
*/
- if (epi)
- ep_release_epitem(epi);
+ while ((rbp = rb_first(&ep->rbr)) != NULL) {
+ epi = rb_entry(rbp, struct epitem, rbn);
+ ep_remove(ep, epi);
+ }
- up_write(&ep->sem);
+ mutex_unlock(&epmutex);
+ mutex_destroy(&ep->mtx);
+ free_uid(ep->user);
+ kfree(ep);
+}
-eexit_3:
- fput(tfile);
-eexit_2:
- fput(file);
-eexit_1:
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_ctl(%d, %d, %d, %p) = %d\n",
- current, epfd, op, fd, event, error));
+static int ep_eventpoll_release(struct inode *inode, struct file *file)
+{
+ struct eventpoll *ep = file->private_data;
- return error;
-}
+ if (ep)
+ ep_free(ep);
-#define MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
+ return 0;
+}
-/*
- * Implement the event wait interface for the eventpoll file. It is the kernel
- * part of the user space epoll_wait(2).
- */
-asmlinkage long sys_epoll_wait(int epfd, struct epoll_event __user *events,
- int maxevents, int timeout)
+static int ep_read_events_proc(struct eventpoll *ep, struct list_head *head,
+ void *priv)
{
- int error;
- struct file *file;
- struct eventpoll *ep;
+ struct epitem *epi, *tmp;
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_wait(%d, %p, %d, %d)\n",
- current, epfd, events, maxevents, timeout));
+ list_for_each_entry_safe(epi, tmp, head, rdllink) {
+ if (epi->ffd.file->f_op->poll(epi->ffd.file, NULL) &
+ epi->event.events)
+ return POLLIN | POLLRDNORM;
+ else {
+ /*
+ * Item has been dropped into the ready list by the poll
+ * callback, but it's not actually ready, as far as
+ * caller requested events goes. We can remove it here.
+ */
+ list_del_init(&epi->rdllink);
+ }
+ }
- /* The maximum number of event must be greater than zero */
- if (maxevents <= 0 || maxevents > MAX_EVENTS)
- return -EINVAL;
+ return 0;
+}
- /* Verify that the area passed by the user is writeable */
- if (!access_ok(VERIFY_WRITE, events, maxevents * sizeof(struct epoll_event))) {
- error = -EFAULT;
- goto eexit_1;
- }
+static int ep_poll_readyevents_proc(void *priv, void *cookie, int call_nests)
+{
+ return ep_scan_ready_list(priv, ep_read_events_proc, NULL);
+}
- /* Get the "struct file *" for the eventpoll file */
- error = -EBADF;
- file = fget(epfd);
- if (!file)
- goto eexit_1;
+static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait)
+{
+ int pollflags;
+ struct eventpoll *ep = file->private_data;
- /*
- * We have to check that the file structure underneath the fd
- * the user passed to us _is_ an eventpoll file.
- */
- error = -EINVAL;
- if (!is_file_epoll(file))
- goto eexit_2;
+ /* Insert inside our poll wait queue */
+ poll_wait(file, &ep->poll_wait, wait);
/*
- * At this point it is safe to assume that the "private_data" contains
- * our own data structure.
+ * Proceed to find out if wanted events are really available inside
+ * the ready list. This need to be done under ep_call_nested()
+ * supervision, since the call to f_op->poll() done on listed files
+ * could re-enter here.
*/
- ep = file->private_data;
+ pollflags = ep_call_nested(&poll_readywalk_ncalls, EP_MAX_NESTS,
+ ep_poll_readyevents_proc, ep, ep, current);
- /* Time to fish for events ... */
- error = ep_poll(ep, events, maxevents, timeout);
+ return pollflags != -1 ? pollflags : 0;
+}
-eexit_2:
- fput(file);
-eexit_1:
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: sys_epoll_wait(%d, %p, %d, %d) = %d\n",
- current, epfd, events, maxevents, timeout, error));
+/* File callbacks that implement the eventpoll file behaviour */
+static const struct file_operations eventpoll_fops = {
+ .release = ep_eventpoll_release,
+ .poll = ep_eventpoll_poll,
+ .llseek = noop_llseek,
+};
- return error;
+/* Fast test to see if the file is an evenpoll file */
+static inline int is_file_epoll(struct file *f)
+{
+ return f->f_op == &eventpoll_fops;
}
-
/*
- * Creates the file descriptor to be used by the epoll interface.
+ * 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
+ * closed without being removed from the eventpoll interface.
*/
-static int ep_getfd(int *efd, struct inode **einode, struct file **efile,
- struct eventpoll *ep)
+void eventpoll_release_file(struct file *file)
{
- struct qstr this;
- char name[32];
- struct dentry *dentry;
- struct inode *inode;
- struct file *file;
- int error, fd;
-
- /* Get an ready to use file */
- error = -ENFILE;
- file = get_empty_filp();
- if (!file)
- goto eexit_1;
-
- /* Allocates an inode from the eventpoll file system */
- inode = ep_eventpoll_inode();
- error = PTR_ERR(inode);
- if (IS_ERR(inode))
- goto eexit_2;
-
- /* Allocates a free descriptor to plug the file onto */
- error = get_unused_fd();
- if (error < 0)
- goto eexit_3;
- fd = error;
+ struct list_head *lsthead = &file->f_ep_links;
+ struct eventpoll *ep;
+ struct epitem *epi;
/*
- * Link the inode to a directory entry by creating a unique name
- * using the inode number.
+ * We don't want to get "file->f_lock" because it is not
+ * necessary. It is not necessary because we're in the "struct file"
+ * cleanup path, and this means that no one is using this file anymore.
+ * So, for example, epoll_ctl() cannot hit here since if we reach this
+ * point, the file counter already went to zero and fget() would fail.
+ * The only hit might come from ep_free() but by holding the mutex
+ * will correctly serialize the operation. We do need to acquire
+ * "ep->mtx" after "epmutex" because ep_remove() requires it when called
+ * from anywhere but ep_free().
+ *
+ * Besides, ep_remove() acquires the lock, so we can't hold it here.
*/
- error = -ENOMEM;
- sprintf(name, "[%lu]", inode->i_ino);
- this.name = name;
- this.len = strlen(name);
- this.hash = inode->i_ino;
- dentry = d_alloc(eventpoll_mnt->mnt_sb->s_root, &this);
- if (!dentry)
- goto eexit_4;
- dentry->d_op = &eventpollfs_dentry_operations;
- d_add(dentry, inode);
- file->f_vfsmnt = mntget(eventpoll_mnt);
- file->f_dentry = dentry;
- file->f_mapping = inode->i_mapping;
-
- file->f_pos = 0;
- file->f_flags = O_RDONLY;
- file->f_op = &eventpoll_fops;
- file->f_mode = FMODE_READ;
- file->f_version = 0;
- file->private_data = ep;
-
- /* Install the new setup file into the allocated fd. */
- fd_install(fd, file);
-
- *efd = fd;
- *einode = inode;
- *efile = file;
- return 0;
+ mutex_lock(&epmutex);
-eexit_4:
- put_unused_fd(fd);
-eexit_3:
- iput(inode);
-eexit_2:
- put_filp(file);
-eexit_1:
- return error;
-}
+ while (!list_empty(lsthead)) {
+ epi = list_first_entry(lsthead, struct epitem, fllink);
+ ep = epi->ep;
+ list_del_init(&epi->fllink);
+ mutex_lock(&ep->mtx);
+ ep_remove(ep, epi);
+ mutex_unlock(&ep->mtx);
+ }
+
+ mutex_unlock(&epmutex);
+}
static int ep_alloc(struct eventpoll **pep)
{
- struct eventpoll *ep = kzalloc(sizeof(*ep), GFP_KERNEL);
+ int error;
+ struct user_struct *user;
+ struct eventpoll *ep;
- if (!ep)
- return -ENOMEM;
+ user = get_current_user();
+ error = -ENOMEM;
+ ep = kzalloc(sizeof(*ep), GFP_KERNEL);
+ if (unlikely(!ep))
+ goto free_uid;
- rwlock_init(&ep->lock);
- init_rwsem(&ep->sem);
+ spin_lock_init(&ep->lock);
+ mutex_init(&ep->mtx);
init_waitqueue_head(&ep->wq);
init_waitqueue_head(&ep->poll_wait);
INIT_LIST_HEAD(&ep->rdllist);
ep->rbr = RB_ROOT;
+ ep->ovflist = EP_UNACTIVE_PTR;
+ ep->user = user;
*pep = ep;
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_alloc() ep=%p\n",
- current, ep));
return 0;
-}
-
-
-static void ep_free(struct eventpoll *ep)
-{
- struct rb_node *rbp;
- struct epitem *epi;
-
- /* We need to release all tasks waiting for these file */
- if (waitqueue_active(&ep->poll_wait))
- ep_poll_safewake(&psw, &ep->poll_wait);
-
- /*
- * We need to lock this because we could be hit by
- * eventpoll_release_file() while we're freeing the "struct eventpoll".
- * 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 "epmutex" is sufficent here.
- */
- mutex_lock(&epmutex);
-
- /*
- * Walks through the whole tree by unregistering poll callbacks.
- */
- for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
- epi = rb_entry(rbp, struct epitem, rbn);
-
- ep_unregister_pollwait(ep, epi);
- }
-
- /*
- * Walks through the whole hash by freeing each "struct epitem". At this
- * point we are sure no poll callbacks will be lingering around, and also by
- * write-holding "sem" we can be sure that no file cleanup code will hit
- * us during this operation. So we can avoid the lock on "ep->lock".
- */
- while ((rbp = rb_first(&ep->rbr)) != 0) {
- epi = rb_entry(rbp, struct epitem, rbn);
- ep_remove(ep, epi);
- }
- mutex_unlock(&epmutex);
+free_uid:
+ free_uid(user);
+ return error;
}
-
/*
- * Search the file inside the eventpoll hash. It add usage count to
- * the returned item, so the caller must call ep_release_epitem()
- * after finished using the "struct epitem".
+ * Search the file inside the eventpoll tree. The RB tree operations
+ * are protected by the "mtx" mutex, and ep_find() must be called with
+ * "mtx" held.
*/
static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd)
{
int kcmp;
- unsigned long flags;
struct rb_node *rbp;
struct epitem *epi, *epir = NULL;
struct epoll_filefd ffd;
ep_set_ffd(&ffd, file, fd);
- read_lock_irqsave(&ep->lock, flags);
for (rbp = ep->rbr.rb_node; rbp; ) {
epi = rb_entry(rbp, struct epitem, rbn);
kcmp = ep_cmp_ffd(&ffd, &epi->ffd);
else if (kcmp < 0)
rbp = rbp->rb_left;
else {
- ep_use_epitem(epi);
epir = epi;
break;
}
}
- read_unlock_irqrestore(&ep->lock, flags);
-
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_find(%p) -> %p\n",
- current, file, epir));
return epir;
}
-
/*
- * Increment the usage count of the "struct epitem" making it sure
- * that the user will have a valid pointer to reference.
+ * This is the callback that is passed to the wait queue wakeup
+ * mechanism. It is called by the stored file descriptors when they
+ * have events to report.
*/
-static void ep_use_epitem(struct epitem *epi)
+static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *key)
{
+ int pwake = 0;
+ unsigned long flags;
+ struct epitem *epi = ep_item_from_wait(wait);
+ struct eventpoll *ep = epi->ep;
- atomic_inc(&epi->usecnt);
-}
+ spin_lock_irqsave(&ep->lock, flags);
+
+ /*
+ * If the event mask does not contain any poll(2) event, we consider the
+ * descriptor to be disabled. This condition is likely the effect of the
+ * EPOLLONESHOT bit that disables the descriptor when an event is received,
+ * until the next EPOLL_CTL_MOD will be issued.
+ */
+ if (!(epi->event.events & ~EP_PRIVATE_BITS))
+ goto out_unlock;
+ /*
+ * Check the events coming with the callback. At this stage, not
+ * every device reports the events in the "key" parameter of the
+ * callback. We need to be able to handle both cases here, hence the
+ * test for "key" != NULL before the event match test.
+ */
+ if (key && !((unsigned long) key & epi->event.events))
+ goto out_unlock;
-/*
- * Decrement ( release ) the usage count by signaling that the user
- * has finished using the structure. It might lead to freeing the
- * structure itself if the count goes to zero.
- */
-static void ep_release_epitem(struct epitem *epi)
-{
+ /*
+ * If we are transferring events to userspace, we can hold no locks
+ * (because we're accessing user memory, and because of linux f_op->poll()
+ * semantics). All the events that happen during that period of time are
+ * chained in ep->ovflist and requeued later on.
+ */
+ if (unlikely(ep->ovflist != EP_UNACTIVE_PTR)) {
+ if (epi->next == EP_UNACTIVE_PTR) {
+ epi->next = ep->ovflist;
+ ep->ovflist = epi;
+ }
+ goto out_unlock;
+ }
- if (atomic_dec_and_test(&epi->usecnt))
- kmem_cache_free(epi_cache, epi);
-}
+ /* If this file is already in the ready list we exit soon */
+ if (!ep_is_linked(&epi->rdllink))
+ list_add_tail(&epi->rdllink, &ep->rdllist);
+
+ /*
+ * Wake up ( if active ) both the eventpoll wait list and the ->poll()
+ * wait list.
+ */
+ if (waitqueue_active(&ep->wq))
+ wake_up_locked(&ep->wq);
+ if (waitqueue_active(&ep->poll_wait))
+ pwake++;
+
+out_unlock:
+ spin_unlock_irqrestore(&ep->lock, flags);
+ /* We have to call this outside the lock */
+ if (pwake)
+ ep_poll_safewake(&ep->poll_wait);
+
+ return 1;
+}
/*
* This is the callback that is used to add our wait queue to the
struct epitem *epi = ep_item_from_epqueue(pt);
struct eppoll_entry *pwq;
- if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, SLAB_KERNEL))) {
+ if (epi->nwait >= 0 && (pwq = kmem_cache_alloc(pwq_cache, GFP_KERNEL))) {
init_waitqueue_func_entry(&pwq->wait, ep_poll_callback);
pwq->whead = whead;
pwq->base = epi;
}
}
-
static void ep_rbtree_insert(struct eventpoll *ep, struct epitem *epi)
{
int kcmp;
rb_insert_color(&epi->rbn, &ep->rbr);
}
-
+/*
+ * Must be called with "mtx" held.
+ */
static int ep_insert(struct eventpoll *ep, struct epoll_event *event,
struct file *tfile, int fd)
{
int error, revents, pwake = 0;
unsigned long flags;
+ long user_watches;
struct epitem *epi;
struct ep_pqueue epq;
- error = -ENOMEM;
- if (!(epi = kmem_cache_alloc(epi_cache, SLAB_KERNEL)))
- goto eexit_1;
+ user_watches = atomic_long_read(&ep->user->epoll_watches);
+ if (unlikely(user_watches >= max_user_watches))
+ return -ENOSPC;
+ if (!(epi = kmem_cache_alloc(epi_cache, GFP_KERNEL)))
+ return -ENOMEM;
/* Item initialization follow here ... */
- ep_rb_initnode(&epi->rbn);
INIT_LIST_HEAD(&epi->rdllink);
INIT_LIST_HEAD(&epi->fllink);
- INIT_LIST_HEAD(&epi->txlink);
INIT_LIST_HEAD(&epi->pwqlist);
epi->ep = ep;
ep_set_ffd(&epi->ffd, tfile, fd);
epi->event = *event;
- atomic_set(&epi->usecnt, 1);
epi->nwait = 0;
+ epi->next = EP_UNACTIVE_PTR;
/* Initialize the poll table using the queue callback */
epq.epi = epi;
/*
* Attach the item to the poll hooks and get current event bits.
* We can safely use the file* here because its usage count has
- * been increased by the caller of this function.
+ * been increased by the caller of this function. Note that after
+ * this operation completes, the poll callback can start hitting
+ * the new item.
*/
revents = tfile->f_op->poll(tfile, &epq.pt);
* install process. Namely an allocation for a wait queue failed due
* high memory pressure.
*/
+ error = -ENOMEM;
if (epi->nwait < 0)
- goto eexit_2;
+ goto error_unregister;
/* Add the current item to the list of active epoll hook for this file */
- spin_lock(&tfile->f_ep_lock);
+ spin_lock(&tfile->f_lock);
list_add_tail(&epi->fllink, &tfile->f_ep_links);
- spin_unlock(&tfile->f_ep_lock);
+ spin_unlock(&tfile->f_lock);
- /* We have to drop the new item inside our item list to keep track of it */
- write_lock_irqsave(&ep->lock, flags);
-
- /* Add the current item to the rb-tree */
+ /*
+ * Add the current item to the RB tree. All RB tree operations are
+ * protected by "mtx", and ep_insert() is called with "mtx" held.
+ */
ep_rbtree_insert(ep, epi);
+ /* We have to drop the new item inside our item list to keep track of it */
+ spin_lock_irqsave(&ep->lock, flags);
+
/* If the file is already "ready" we drop it inside the ready list */
if ((revents & event->events) && !ep_is_linked(&epi->rdllink)) {
list_add_tail(&epi->rdllink, &ep->rdllist);
/* Notify waiting tasks that events are available */
if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
+ wake_up_locked(&ep->wq);
if (waitqueue_active(&ep->poll_wait))
pwake++;
}
- write_unlock_irqrestore(&ep->lock, flags);
+ spin_unlock_irqrestore(&ep->lock, flags);
+
+ atomic_long_inc(&ep->user->epoll_watches);
/* We have to call this outside the lock */
if (pwake)
- ep_poll_safewake(&psw, &ep->poll_wait);
-
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_insert(%p, %p, %d)\n",
- current, ep, tfile, fd));
+ ep_poll_safewake(&ep->poll_wait);
return 0;
-eexit_2:
+error_unregister:
ep_unregister_pollwait(ep, epi);
/*
* We need to do this because an event could have been arrived on some
- * allocated wait queue.
+ * allocated wait queue. Note that we don't care about the ep->ovflist
+ * list, since that is used/cleaned only inside a section bound by "mtx".
+ * And ep_insert() is called with "mtx" held.
*/
- write_lock_irqsave(&ep->lock, flags);
+ spin_lock_irqsave(&ep->lock, flags);
if (ep_is_linked(&epi->rdllink))
- ep_list_del(&epi->rdllink);
- write_unlock_irqrestore(&ep->lock, flags);
+ list_del_init(&epi->rdllink);
+ spin_unlock_irqrestore(&ep->lock, flags);
kmem_cache_free(epi_cache, epi);
-eexit_1:
+
return error;
}
-
/*
* Modify the interest event mask by dropping an event if the new mask
- * has a match in the current file status.
+ * has a match in the current file status. Must be called with "mtx" held.
*/
static int ep_modify(struct eventpoll *ep, struct epitem *epi, struct epoll_event *event)
{
int pwake = 0;
unsigned int revents;
- unsigned long flags;
/*
- * Set the new event interest mask before calling f_op->poll(), otherwise
- * a potential race might occur. In fact if we do this operation inside
- * the lock, an event might happen between the f_op->poll() call and the
- * new event set registering.
+ * Set the new event interest mask before calling f_op->poll();
+ * otherwise we might miss an event that happens between the
+ * f_op->poll() call and the new event set registering.
*/
epi->event.events = event->events;
+ epi->event.data = event->data; /* protected by mtx */
/*
* Get current event bits. We can safely use the file* here because
*/
revents = epi->ffd.file->f_op->poll(epi->ffd.file, NULL);
- write_lock_irqsave(&ep->lock, flags);
-
- /* Copy the data member from inside the lock */
- epi->event.data = event->data;
-
/*
- * If the item is not linked to the hash it means that it's on its
- * way toward the removal. Do nothing in this case.
+ * If the item is "hot" and it is not registered inside the ready
+ * list, push it inside.
*/
- if (ep_rb_linked(&epi->rbn)) {
- /*
- * If the item is "hot" and it is not registered inside the ready
- * list, push it inside. If the item is not "hot" and it is currently
- * registered inside the ready list, unlink it.
- */
- if (revents & event->events) {
- if (!ep_is_linked(&epi->rdllink)) {
- list_add_tail(&epi->rdllink, &ep->rdllist);
+ if (revents & event->events) {
+ spin_lock_irq(&ep->lock);
+ if (!ep_is_linked(&epi->rdllink)) {
+ list_add_tail(&epi->rdllink, &ep->rdllist);
- /* Notify waiting tasks that events are available */
- if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
- if (waitqueue_active(&ep->poll_wait))
- pwake++;
- }
+ /* Notify waiting tasks that events are available */
+ if (waitqueue_active(&ep->wq))
+ wake_up_locked(&ep->wq);
+ if (waitqueue_active(&ep->poll_wait))
+ pwake++;
}
+ spin_unlock_irq(&ep->lock);
}
- write_unlock_irqrestore(&ep->lock, flags);
-
/* We have to call this outside the lock */
if (pwake)
- ep_poll_safewake(&psw, &ep->poll_wait);
+ ep_poll_safewake(&ep->poll_wait);
return 0;
}
-
-/*
- * This function unregister poll callbacks from the associated file descriptor.
- * Since this must be called without holding "ep->lock" the atomic exchange trick
- * will protect us from multiple unregister.
- */
-static void ep_unregister_pollwait(struct eventpoll *ep, struct epitem *epi)
+static int ep_send_events_proc(struct eventpoll *ep, struct list_head *head,
+ void *priv)
{
- int nwait;
- struct list_head *lsthead = &epi->pwqlist;
- struct eppoll_entry *pwq;
+ struct ep_send_events_data *esed = priv;
+ int eventcnt;
+ unsigned int revents;
+ struct epitem *epi;
+ struct epoll_event __user *uevent;
+
+ /*
+ * We can loop without lock because we are passed a task private list.
+ * Items cannot vanish during the loop because ep_scan_ready_list() is
+ * holding "mtx" during this call.
+ */
+ for (eventcnt = 0, uevent = esed->events;
+ !list_empty(head) && eventcnt < esed->maxevents;) {
+ epi = list_first_entry(head, struct epitem, rdllink);
- /* This is called without locks, so we need the atomic exchange */
- nwait = xchg(&epi->nwait, 0);
+ list_del_init(&epi->rdllink);
- if (nwait) {
- while (!list_empty(lsthead)) {
- pwq = list_entry(lsthead->next, struct eppoll_entry, llink);
+ revents = epi->ffd.file->f_op->poll(epi->ffd.file, NULL) &
+ epi->event.events;
- ep_list_del(&pwq->llink);
- remove_wait_queue(pwq->whead, &pwq->wait);
- kmem_cache_free(pwq_cache, pwq);
+ /*
+ * If the event mask intersect the caller-requested one,
+ * deliver the event to userspace. Again, ep_scan_ready_list()
+ * is holding "mtx", so no operations coming from userspace
+ * can change the item.
+ */
+ if (revents) {
+ if (__put_user(revents, &uevent->events) ||
+ __put_user(epi->event.data, &uevent->data)) {
+ list_add(&epi->rdllink, head);
+ return eventcnt ? eventcnt : -EFAULT;
+ }
+ eventcnt++;
+ uevent++;
+ if (epi->event.events & EPOLLONESHOT)
+ epi->event.events &= EP_PRIVATE_BITS;
+ else if (!(epi->event.events & EPOLLET)) {
+ /*
+ * If this file has been added with Level
+ * Trigger mode, we need to insert back inside
+ * the ready list, so that the next call to
+ * epoll_wait() will check again the events
+ * availability. At this point, no one can insert
+ * into ep->rdllist besides us. The epoll_ctl()
+ * callers are locked out by
+ * ep_scan_ready_list() holding "mtx" and the
+ * poll callback will queue them in ep->ovflist.
+ */
+ list_add_tail(&epi->rdllink, &ep->rdllist);
+ }
}
}
-}
+ return eventcnt;
+}
-/*
- * Unlink the "struct epitem" from all places it might have been hooked up.
- * This function must be called with write IRQ lock on "ep->lock".
- */
-static int ep_unlink(struct eventpoll *ep, struct epitem *epi)
+static int ep_send_events(struct eventpoll *ep,
+ struct epoll_event __user *events, int maxevents)
{
- int error;
-
- /*
- * It can happen that this one is called for an item already unlinked.
- * The check protect us from doing a double unlink ( crash ).
- */
- error = -ENOENT;
- if (!ep_rb_linked(&epi->rbn))
- goto eexit_1;
-
- /*
- * Clear the event mask for the unlinked item. This will avoid item
- * notifications to be sent after the unlink operation from inside
- * the kernel->userspace event transfer loop.
- */
- epi->event.events = 0;
-
- /*
- * At this point is safe to do the job, unlink the item from our rb-tree.
- * This operation togheter with the above check closes the door to
- * double unlinks.
- */
- ep_rb_erase(&epi->rbn, &ep->rbr);
+ struct ep_send_events_data esed;
- /*
- * If the item we are going to remove is inside the ready file descriptors
- * we want to remove it from this list to avoid stale events.
- */
- if (ep_is_linked(&epi->rdllink))
- ep_list_del(&epi->rdllink);
+ esed.maxevents = maxevents;
+ esed.events = events;
- error = 0;
-eexit_1:
+ return ep_scan_ready_list(ep, ep_send_events_proc, &esed);
+}
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_unlink(%p, %p) = %d\n",
- current, ep, epi->file, error));
+static inline struct timespec ep_set_mstimeout(long ms)
+{
+ struct timespec now, ts = {
+ .tv_sec = ms / MSEC_PER_SEC,
+ .tv_nsec = NSEC_PER_MSEC * (ms % MSEC_PER_SEC),
+ };
- return error;
+ ktime_get_ts(&now);
+ return timespec_add_safe(now, ts);
}
-
-/*
- * Removes a "struct epitem" from the eventpoll hash and deallocates
- * all the associated resources.
+/**
+ * ep_poll - Retrieves ready events, and delivers them to the caller supplied
+ * event buffer.
+ *
+ * @ep: Pointer to the eventpoll context.
+ * @events: Pointer to the userspace buffer where the ready events should be
+ * stored.
+ * @maxevents: Size (in terms of number of events) of the caller event buffer.
+ * @timeout: Maximum timeout for the ready events fetch operation, in
+ * milliseconds. If the @timeout is zero, the function will not block,
+ * while if the @timeout is less than zero, the function will block
+ * until at least one event has been retrieved (or an error
+ * occurred).
+ *
+ * Returns: Returns the number of ready events which have been fetched, or an
+ * error code, in case of error.
*/
-static int ep_remove(struct eventpoll *ep, struct epitem *epi)
+static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
+ int maxevents, long timeout)
{
- int error;
+ int res = 0, eavail, timed_out = 0;
unsigned long flags;
- struct file *file = epi->ffd.file;
-
- /*
- * Removes poll wait queue hooks. We _have_ to do this without holding
- * the "ep->lock" otherwise a deadlock might occur. This because of the
- * sequence of the lock acquisition. Here we do "ep->lock" then the wait
- * queue head lock when unregistering the wait queue. The wakeup callback
- * will run by holding the wait queue head lock and will call our callback
- * that will try to get "ep->lock".
- */
- ep_unregister_pollwait(ep, epi);
-
- /* Remove the current item from the list of epoll hooks */
- spin_lock(&file->f_ep_lock);
- if (ep_is_linked(&epi->fllink))
- ep_list_del(&epi->fllink);
- spin_unlock(&file->f_ep_lock);
-
- /* We need to acquire the write IRQ lock before calling ep_unlink() */
- write_lock_irqsave(&ep->lock, flags);
-
- /* Really unlink the item from the hash */
- error = ep_unlink(ep, epi);
-
- write_unlock_irqrestore(&ep->lock, flags);
-
- if (error)
- goto eexit_1;
-
- /* At this point it is safe to free the eventpoll item */
- ep_release_epitem(epi);
+ long slack = 0;
+ wait_queue_t wait;
+ ktime_t expires, *to = NULL;
- error = 0;
-eexit_1:
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: ep_remove(%p, %p) = %d\n",
- current, ep, file, error));
+ if (timeout > 0) {
+ struct timespec end_time = ep_set_mstimeout(timeout);
- return error;
-}
+ slack = select_estimate_accuracy(&end_time);
+ to = &expires;
+ *to = timespec_to_ktime(end_time);
+ } else if (timeout == 0) {
+ /*
+ * Avoid the unnecessary trip to the wait queue loop, if the
+ * caller specified a non blocking operation.
+ */
+ timed_out = 1;
+ spin_lock_irqsave(&ep->lock, flags);
+ goto check_events;
+ }
+fetch_events:
+ spin_lock_irqsave(&ep->lock, flags);
-/*
- * This is the callback that is passed to the wait queue wakeup
- * machanism. It is called by the stored file descriptors when they
- * have events to report.
- */
-static int ep_poll_callback(wait_queue_t *wait, unsigned mode, int sync, void *key)
-{
- int pwake = 0;
- unsigned long flags;
- struct epitem *epi = ep_item_from_wait(wait);
- struct eventpoll *ep = epi->ep;
+ if (!ep_events_available(ep)) {
+ /*
+ * We don't have any available event to return to the caller.
+ * We need to sleep here, and we will be wake up by
+ * ep_poll_callback() when events will become available.
+ */
+ init_waitqueue_entry(&wait, current);
+ __add_wait_queue_exclusive(&ep->wq, &wait);
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: poll_callback(%p) epi=%p ep=%p\n",
- current, epi->file, epi, ep));
+ for (;;) {
+ /*
+ * We don't want to sleep if the ep_poll_callback() sends us
+ * a wakeup in between. That's why we set the task state
+ * to TASK_INTERRUPTIBLE before doing the checks.
+ */
+ set_current_state(TASK_INTERRUPTIBLE);
+ if (ep_events_available(ep) || timed_out)
+ break;
+ if (signal_pending(current)) {
+ res = -EINTR;
+ break;
+ }
- write_lock_irqsave(&ep->lock, flags);
+ spin_unlock_irqrestore(&ep->lock, flags);
+ if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS))
+ timed_out = 1;
- /*
- * If the event mask does not contain any poll(2) event, we consider the
- * descriptor to be disabled. This condition is likely the effect of the
- * EPOLLONESHOT bit that disables the descriptor when an event is received,
- * until the next EPOLL_CTL_MOD will be issued.
- */
- if (!(epi->event.events & ~EP_PRIVATE_BITS))
- goto is_disabled;
+ spin_lock_irqsave(&ep->lock, flags);
+ }
+ __remove_wait_queue(&ep->wq, &wait);
- /* If this file is already in the ready list we exit soon */
- if (ep_is_linked(&epi->rdllink))
- goto is_linked;
+ set_current_state(TASK_RUNNING);
+ }
+check_events:
+ /* Is it worth to try to dig for events ? */
+ eavail = ep_events_available(ep);
- list_add_tail(&epi->rdllink, &ep->rdllist);
+ spin_unlock_irqrestore(&ep->lock, flags);
-is_linked:
/*
- * Wake up ( if active ) both the eventpoll wait list and the ->poll()
- * wait list.
+ * Try to transfer events to user space. In case we get 0 events and
+ * there's still timeout left over, we go trying again in search of
+ * more luck.
*/
- if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
- if (waitqueue_active(&ep->poll_wait))
- pwake++;
-
-is_disabled:
- write_unlock_irqrestore(&ep->lock, flags);
-
- /* We have to call this outside the lock */
- if (pwake)
- ep_poll_safewake(&psw, &ep->poll_wait);
+ if (!res && eavail &&
+ !(res = ep_send_events(ep, events, maxevents)) && !timed_out)
+ goto fetch_events;
- return 1;
+ return res;
}
-
-static int ep_eventpoll_close(struct inode *inode, struct file *file)
+/**
+ * ep_loop_check_proc - Callback function to be passed to the @ep_call_nested()
+ * API, to verify that adding an epoll file inside another
+ * epoll structure, does not violate the constraints, in
+ * terms of closed loops, or too deep chains (which can
+ * result in excessive stack usage).
+ *
+ * @priv: Pointer to the epoll file to be currently checked.
+ * @cookie: Original cookie for this call. This is the top-of-the-chain epoll
+ * data structure pointer.
+ * @call_nests: Current dept of the @ep_call_nested() call stack.
+ *
+ * Returns: Returns zero if adding the epoll @file inside current epoll
+ * structure @ep does not violate the constraints, or -1 otherwise.
+ */
+static int ep_loop_check_proc(void *priv, void *cookie, int call_nests)
{
+ int error = 0;
+ struct file *file = priv;
struct eventpoll *ep = file->private_data;
+ struct rb_node *rbp;
+ struct epitem *epi;
- if (ep) {
- ep_free(ep);
- kfree(ep);
+ mutex_lock(&ep->mtx);
+ for (rbp = rb_first(&ep->rbr); rbp; rbp = rb_next(rbp)) {
+ epi = rb_entry(rbp, struct epitem, rbn);
+ if (unlikely(is_file_epoll(epi->ffd.file))) {
+ error = ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS,
+ ep_loop_check_proc, epi->ffd.file,
+ epi->ffd.file->private_data, current);
+ if (error != 0)
+ break;
+ }
}
+ mutex_unlock(&ep->mtx);
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: close() ep=%p\n", current, ep));
- return 0;
+ return error;
}
-
-static unsigned int ep_eventpoll_poll(struct file *file, poll_table *wait)
+/**
+ * ep_loop_check - Performs a check to verify that adding an epoll file (@file)
+ * another epoll file (represented by @ep) does not create
+ * closed loops or too deep chains.
+ *
+ * @ep: Pointer to the epoll private data structure.
+ * @file: Pointer to the epoll file to be checked.
+ *
+ * Returns: Returns zero if adding the epoll @file inside current epoll
+ * structure @ep does not violate the constraints, or -1 otherwise.
+ */
+static int ep_loop_check(struct eventpoll *ep, struct file *file)
{
- unsigned int pollflags = 0;
- unsigned long flags;
- struct eventpoll *ep = file->private_data;
-
- /* Insert inside our poll wait queue */
- poll_wait(file, &ep->poll_wait, wait);
-
- /* Check our condition */
- read_lock_irqsave(&ep->lock, flags);
- if (!list_empty(&ep->rdllist))
- pollflags = POLLIN | POLLRDNORM;
- read_unlock_irqrestore(&ep->lock, flags);
-
- return pollflags;
+ return ep_call_nested(&poll_loop_ncalls, EP_MAX_NESTS,
+ ep_loop_check_proc, file, ep, current);
}
-
/*
- * Since we have to release the lock during the __copy_to_user() operation and
- * during the f_op->poll() call, we try to collect the maximum number of items
- * by reducing the irqlock/irqunlock switching rate.
+ * Open an eventpoll file descriptor.
*/
-static int ep_collect_ready_items(struct eventpoll *ep, struct list_head *txlist, int maxevents)
+SYSCALL_DEFINE1(epoll_create1, int, flags)
{
- int nepi;
- unsigned long flags;
- struct list_head *lsthead = &ep->rdllist, *lnk;
- struct epitem *epi;
-
- write_lock_irqsave(&ep->lock, flags);
-
- for (nepi = 0, lnk = lsthead->next; lnk != lsthead && nepi < maxevents;) {
- epi = list_entry(lnk, struct epitem, rdllink);
-
- lnk = lnk->next;
-
- /* If this file is already in the ready list we exit soon */
- if (!ep_is_linked(&epi->txlink)) {
- /*
- * This is initialized in this way so that the default
- * behaviour of the reinjecting code will be to push back
- * the item inside the ready list.
- */
- epi->revents = epi->event.events;
-
- /* Link the ready item into the transfer list */
- list_add(&epi->txlink, txlist);
- nepi++;
+ int error;
+ struct eventpoll *ep = NULL;
- /*
- * Unlink the item from the ready list.
- */
- ep_list_del(&epi->rdllink);
- }
- }
+ /* Check the EPOLL_* constant for consistency. */
+ BUILD_BUG_ON(EPOLL_CLOEXEC != O_CLOEXEC);
- write_unlock_irqrestore(&ep->lock, flags);
+ if (flags & ~EPOLL_CLOEXEC)
+ return -EINVAL;
+ /*
+ * Create the internal data structure ("struct eventpoll").
+ */
+ error = ep_alloc(&ep);
+ if (error < 0)
+ return error;
+ /*
+ * Creates all the items needed to setup an eventpoll file. That is,
+ * a file structure and a free file descriptor.
+ */
+ error = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep,
+ O_RDWR | (flags & O_CLOEXEC));
+ if (error < 0)
+ ep_free(ep);
- return nepi;
+ return error;
}
-
-/*
- * This function is called without holding the "ep->lock" since the call to
- * __copy_to_user() might sleep, and also f_op->poll() might reenable the IRQ
- * because of the way poll() is traditionally implemented in Linux.
- */
-static int ep_send_events(struct eventpoll *ep, struct list_head *txlist,
- struct epoll_event __user *events)
+SYSCALL_DEFINE1(epoll_create, int, size)
{
- int eventcnt = 0;
- unsigned int revents;
- struct list_head *lnk;
- struct epitem *epi;
-
- /*
- * We can loop without lock because this is a task private list.
- * The test done during the collection loop will guarantee us that
- * another task will not try to collect this file. Also, items
- * cannot vanish during the loop because we are holding "sem".
- */
- list_for_each(lnk, txlist) {
- epi = list_entry(lnk, struct epitem, txlink);
-
- /*
- * Get the ready file event set. We can safely use the file
- * because we are holding the "sem" in read and this will
- * guarantee that both the file and the item will not vanish.
- */
- revents = epi->ffd.file->f_op->poll(epi->ffd.file, NULL);
+ if (size <= 0)
+ return -EINVAL;
- /*
- * Set the return event set for the current file descriptor.
- * Note that only the task task was successfully able to link
- * the item to its "txlist" will write this field.
- */
- epi->revents = revents & epi->event.events;
-
- if (epi->revents) {
- if (__put_user(epi->revents,
- &events[eventcnt].events) ||
- __put_user(epi->event.data,
- &events[eventcnt].data))
- return -EFAULT;
- if (epi->event.events & EPOLLONESHOT)
- epi->event.events &= EP_PRIVATE_BITS;
- eventcnt++;
- }
- }
- return eventcnt;
+ return sys_epoll_create1(0);
}
-
/*
- * Walk through the transfer list we collected with ep_collect_ready_items()
- * and, if 1) the item is still "alive" 2) its event set is not empty 3) it's
- * not already linked, links it to the ready list. Same as above, we are holding
- * "sem" so items cannot vanish underneath our nose.
+ * The following function implements the controller interface for
+ * the eventpoll file that enables the insertion/removal/change of
+ * file descriptors inside the interest set.
*/
-static void ep_reinject_items(struct eventpoll *ep, struct list_head *txlist)
+SYSCALL_DEFINE4(epoll_ctl, int, epfd, int, op, int, fd,
+ struct epoll_event __user *, event)
{
- int ricnt = 0, pwake = 0;
- unsigned long flags;
+ int error;
+ int did_lock_epmutex = 0;
+ struct file *file, *tfile;
+ struct eventpoll *ep;
struct epitem *epi;
+ struct epoll_event epds;
- write_lock_irqsave(&ep->lock, flags);
-
- while (!list_empty(txlist)) {
- epi = list_entry(txlist->next, struct epitem, txlink);
+ error = -EFAULT;
+ if (ep_op_has_event(op) &&
+ copy_from_user(&epds, event, sizeof(struct epoll_event)))
+ goto error_return;
- /* Unlink the current item from the transfer list */
- ep_list_del(&epi->txlink);
+ /* Get the "struct file *" for the eventpoll file */
+ error = -EBADF;
+ file = fget(epfd);
+ if (!file)
+ goto error_return;
- /*
- * If the item is no more linked to the interest set, we don't
- * have to push it inside the ready list because the following
- * ep_release_epitem() is going to drop it. Also, if the current
- * item is set to have an Edge Triggered behaviour, we don't have
- * to push it back either.
- */
- if (ep_rb_linked(&epi->rbn) && !(epi->event.events & EPOLLET) &&
- (epi->revents & epi->event.events) && !ep_is_linked(&epi->rdllink)) {
- list_add_tail(&epi->rdllink, &ep->rdllist);
- ricnt++;
- }
- }
+ /* Get the "struct file *" for the target file */
+ tfile = fget(fd);
+ if (!tfile)
+ goto error_fput;
- if (ricnt) {
- /*
- * Wake up ( if active ) both the eventpoll wait list and the ->poll()
- * wait list.
- */
- if (waitqueue_active(&ep->wq))
- wake_up(&ep->wq);
- if (waitqueue_active(&ep->poll_wait))
- pwake++;
- }
+ /* The target file descriptor must support poll */
+ error = -EPERM;
+ if (!tfile->f_op || !tfile->f_op->poll)
+ goto error_tgt_fput;
- write_unlock_irqrestore(&ep->lock, flags);
+ /*
+ * We have to check that the file structure underneath the file descriptor
+ * the user passed to us _is_ an eventpoll file. And also we do not permit
+ * adding an epoll file descriptor inside itself.
+ */
+ error = -EINVAL;
+ if (file == tfile || !is_file_epoll(file))
+ goto error_tgt_fput;
- /* We have to call this outside the lock */
- if (pwake)
- ep_poll_safewake(&psw, &ep->poll_wait);
-}
+ /*
+ * At this point it is safe to assume that the "private_data" contains
+ * our own data structure.
+ */
+ ep = file->private_data;
+ /*
+ * When we insert an epoll file descriptor, inside another epoll file
+ * descriptor, there is the change of creating closed loops, which are
+ * better be handled here, than in more critical paths.
+ *
+ * We hold epmutex across the loop check and the insert in this case, in
+ * order to prevent two separate inserts from racing and each doing the
+ * insert "at the same time" such that ep_loop_check passes on both
+ * before either one does the insert, thereby creating a cycle.
+ */
+ if (unlikely(is_file_epoll(tfile) && op == EPOLL_CTL_ADD)) {
+ mutex_lock(&epmutex);
+ did_lock_epmutex = 1;
+ error = -ELOOP;
+ if (ep_loop_check(ep, tfile) != 0)
+ goto error_tgt_fput;
+ }
-/*
- * Perform the transfer of events to user space.
- */
-static int ep_events_transfer(struct eventpoll *ep,
- struct epoll_event __user *events, int maxevents)
-{
- int eventcnt = 0;
- struct list_head txlist;
- INIT_LIST_HEAD(&txlist);
+ mutex_lock(&ep->mtx);
/*
- * We need to lock this because we could be hit by
- * eventpoll_release_file() and epoll_ctl(EPOLL_CTL_DEL).
+ * Try to lookup the file inside our RB tree, Since we grabbed "mtx"
+ * above, we can be sure to be able to use the item looked up by
+ * ep_find() till we release the mutex.
*/
- down_read(&ep->sem);
-
- /* Collect/extract ready items */
- if (ep_collect_ready_items(ep, &txlist, maxevents) > 0) {
- /* Build result set in userspace */
- eventcnt = ep_send_events(ep, &txlist, events);
+ epi = ep_find(ep, tfile, fd);
- /* Reinject ready items into the ready list */
- ep_reinject_items(ep, &txlist);
+ error = -EINVAL;
+ switch (op) {
+ case EPOLL_CTL_ADD:
+ if (!epi) {
+ epds.events |= POLLERR | POLLHUP;
+ error = ep_insert(ep, &epds, tfile, fd);
+ } else
+ error = -EEXIST;
+ break;
+ case EPOLL_CTL_DEL:
+ if (epi)
+ error = ep_remove(ep, epi);
+ else
+ error = -ENOENT;
+ break;
+ case EPOLL_CTL_MOD:
+ if (epi) {
+ epds.events |= POLLERR | POLLHUP;
+ error = ep_modify(ep, epi, &epds);
+ } else
+ error = -ENOENT;
+ break;
}
+ mutex_unlock(&ep->mtx);
- up_read(&ep->sem);
+error_tgt_fput:
+ if (unlikely(did_lock_epmutex))
+ mutex_unlock(&epmutex);
- return eventcnt;
-}
+ fput(tfile);
+error_fput:
+ fput(file);
+error_return:
+ return error;
+}
-static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
- int maxevents, long timeout)
+/*
+ * Implement the event wait interface for the eventpoll file. It is the kernel
+ * part of the user space epoll_wait(2).
+ */
+SYSCALL_DEFINE4(epoll_wait, int, epfd, struct epoll_event __user *, events,
+ int, maxevents, int, timeout)
{
- int res, eavail;
- unsigned long flags;
- long jtimeout;
- wait_queue_t wait;
-
- /*
- * Calculate the timeout by checking for the "infinite" value ( -1 )
- * and the overflow condition. The passed timeout is in milliseconds,
- * that why (t * HZ) / 1000.
- */
- jtimeout = (timeout < 0 || timeout >= EP_MAX_MSTIMEO) ?
- MAX_SCHEDULE_TIMEOUT : (timeout * HZ + 999) / 1000;
-
-retry:
- write_lock_irqsave(&ep->lock, flags);
-
- res = 0;
- if (list_empty(&ep->rdllist)) {
- /*
- * We don't have any available event to return to the caller.
- * We need to sleep here, and we will be wake up by
- * ep_poll_callback() when events will become available.
- */
- init_waitqueue_entry(&wait, current);
- add_wait_queue(&ep->wq, &wait);
-
- for (;;) {
- /*
- * We don't want to sleep if the ep_poll_callback() sends us
- * a wakeup in between. That's why we set the task state
- * to TASK_INTERRUPTIBLE before doing the checks.
- */
- set_current_state(TASK_INTERRUPTIBLE);
- if (!list_empty(&ep->rdllist) || !jtimeout)
- break;
- if (signal_pending(current)) {
- res = -EINTR;
- break;
- }
+ int error;
+ struct file *file;
+ struct eventpoll *ep;
- write_unlock_irqrestore(&ep->lock, flags);
- jtimeout = schedule_timeout(jtimeout);
- write_lock_irqsave(&ep->lock, flags);
- }
- remove_wait_queue(&ep->wq, &wait);
+ /* The maximum number of event must be greater than zero */
+ if (maxevents <= 0 || maxevents > EP_MAX_EVENTS)
+ return -EINVAL;
- set_current_state(TASK_RUNNING);
+ /* Verify that the area passed by the user is writeable */
+ if (!access_ok(VERIFY_WRITE, events, maxevents * sizeof(struct epoll_event))) {
+ error = -EFAULT;
+ goto error_return;
}
- /* Is it worth to try to dig for events ? */
- eavail = !list_empty(&ep->rdllist);
-
- write_unlock_irqrestore(&ep->lock, flags);
+ /* Get the "struct file *" for the eventpoll file */
+ error = -EBADF;
+ file = fget(epfd);
+ if (!file)
+ goto error_return;
/*
- * Try to transfer events to user space. In case we get 0 events and
- * there's still timeout left over, we go trying again in search of
- * more luck.
+ * We have to check that the file structure underneath the fd
+ * the user passed to us _is_ an eventpoll file.
*/
- if (!res && eavail &&
- !(res = ep_events_transfer(ep, events, maxevents)) && jtimeout)
- goto retry;
+ error = -EINVAL;
+ if (!is_file_epoll(file))
+ goto error_fput;
- return res;
-}
+ /*
+ * At this point it is safe to assume that the "private_data" contains
+ * our own data structure.
+ */
+ ep = file->private_data;
+ /* Time to fish for events ... */
+ error = ep_poll(ep, events, maxevents, timeout);
-static int eventpollfs_delete_dentry(struct dentry *dentry)
-{
+error_fput:
+ fput(file);
+error_return:
- return 1;
+ return error;
}
+#ifdef HAVE_SET_RESTORE_SIGMASK
-static struct inode *ep_eventpoll_inode(void)
+/*
+ * Implement the event wait interface for the eventpoll file. It is the kernel
+ * part of the user space epoll_pwait(2).
+ */
+SYSCALL_DEFINE6(epoll_pwait, int, epfd, struct epoll_event __user *, events,
+ int, maxevents, int, timeout, const sigset_t __user *, sigmask,
+ size_t, sigsetsize)
{
- int error = -ENOMEM;
- struct inode *inode = new_inode(eventpoll_mnt->mnt_sb);
+ int error;
+ sigset_t ksigmask, sigsaved;
- if (!inode)
- goto eexit_1;
+ /*
+ * If the caller wants a certain signal mask to be set during the wait,
+ * we apply it here.
+ */
+ if (sigmask) {
+ if (sigsetsize != sizeof(sigset_t))
+ return -EINVAL;
+ if (copy_from_user(&ksigmask, sigmask, sizeof(ksigmask)))
+ return -EFAULT;
+ sigdelsetmask(&ksigmask, sigmask(SIGKILL) | sigmask(SIGSTOP));
+ sigprocmask(SIG_SETMASK, &ksigmask, &sigsaved);
+ }
- inode->i_fop = &eventpoll_fops;
+ error = sys_epoll_wait(epfd, events, maxevents, timeout);
/*
- * Mark the inode dirty from the very beginning,
- * that way it will never be moved to the dirty
- * list because mark_inode_dirty() will think
- * that it already _is_ on the dirty list.
+ * If we changed the signal mask, we need to restore the original one.
+ * In case we've got a signal while waiting, we do not restore the
+ * signal mask yet, and we allow do_signal() to deliver the signal on
+ * the way back to userspace, before the signal mask is restored.
*/
- inode->i_state = I_DIRTY;
- inode->i_mode = S_IRUSR | S_IWUSR;
- inode->i_uid = current->fsuid;
- inode->i_gid = current->fsgid;
- inode->i_atime = inode->i_mtime = inode->i_ctime = CURRENT_TIME;
- inode->i_blksize = PAGE_SIZE;
- return inode;
-
-eexit_1:
- return ERR_PTR(error);
-}
-
+ if (sigmask) {
+ if (error == -EINTR) {
+ memcpy(¤t->saved_sigmask, &sigsaved,
+ sizeof(sigsaved));
+ set_restore_sigmask();
+ } else
+ sigprocmask(SIG_SETMASK, &sigsaved, NULL);
+ }
-static struct super_block *
-eventpollfs_get_sb(struct file_system_type *fs_type, int flags,
- const char *dev_name, void *data)
-{
- return get_sb_pseudo(fs_type, "eventpoll:", NULL, EVENTPOLLFS_MAGIC);
+ return error;
}
+#endif /* HAVE_SET_RESTORE_SIGMASK */
static int __init eventpoll_init(void)
{
- int error;
+ struct sysinfo si;
+
+ si_meminfo(&si);
+ /*
+ * Allows top 4% of lomem to be allocated for epoll watches (per user).
+ */
+ max_user_watches = (((si.totalram - si.totalhigh) / 25) << PAGE_SHIFT) /
+ EP_ITEM_COST;
+ BUG_ON(max_user_watches < 0);
- mutex_init(&epmutex);
+ /*
+ * Initialize the structure used to perform epoll file descriptor
+ * inclusion loops checks.
+ */
+ ep_nested_calls_init(&poll_loop_ncalls);
/* Initialize the structure used to perform safe poll wait head wake ups */
- ep_poll_safewake_init(&psw);
+ ep_nested_calls_init(&poll_safewake_ncalls);
+
+ /* Initialize the structure used to perform file's f_op->poll() calls */
+ ep_nested_calls_init(&poll_readywalk_ncalls);
/* Allocates slab cache used to allocate "struct epitem" items */
epi_cache = kmem_cache_create("eventpoll_epi", sizeof(struct epitem),
- 0, SLAB_HWCACHE_ALIGN|EPI_SLAB_DEBUG|SLAB_PANIC,
- NULL, NULL);
+ 0, SLAB_HWCACHE_ALIGN | SLAB_PANIC, NULL);
/* Allocates slab cache used to allocate "struct eppoll_entry" */
pwq_cache = kmem_cache_create("eventpoll_pwq",
- sizeof(struct eppoll_entry), 0,
- EPI_SLAB_DEBUG|SLAB_PANIC, NULL, NULL);
+ sizeof(struct eppoll_entry), 0, SLAB_PANIC, NULL);
- /*
- * Register the virtual file system that will be the source of inodes
- * for the eventpoll files
- */
- error = register_filesystem(&eventpoll_fs_type);
- if (error)
- goto epanic;
-
- /* Mount the above commented virtual file system */
- eventpoll_mnt = kern_mount(&eventpoll_fs_type);
- error = PTR_ERR(eventpoll_mnt);
- if (IS_ERR(eventpoll_mnt))
- goto epanic;
-
- DNPRINTK(3, (KERN_INFO "[%p] eventpoll: successfully initialized.\n",
- current));
return 0;
-
-epanic:
- panic("eventpoll_init() failed\n");
}
-
-
-static void __exit eventpoll_exit(void)
-{
- /* Undo all operations done inside eventpoll_init() */
- unregister_filesystem(&eventpoll_fs_type);
- mntput(eventpoll_mnt);
- kmem_cache_destroy(pwq_cache);
- kmem_cache_destroy(epi_cache);
-}
-
-module_init(eventpoll_init);
-module_exit(eventpoll_exit);
-
-MODULE_LICENSE("GPL");
+fs_initcall(eventpoll_init);
Code Review / linux-2.6.git / blobdiff