epoll: make epoll_wait() use the hrtimer range feature
This make epoll use hrtimers for the timeout value which prevents
epoll_wait() from timing out up to a millisecond early.
This mirrors the behavior of select() and poll().
Signed-off-by: Shawn Bohrer <shawn.bohrer@gmail.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Acked-by: Davide Libenzi <davidel@xmailserver.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 256bb7b..8cf0724 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -77,9 +77,6 @@
/* Maximum number of nesting allowed inside epoll sets */
#define EP_MAX_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)
-
#define EP_MAX_EVENTS (INT_MAX / sizeof(struct epoll_event))
#define EP_UNACTIVE_PTR ((void *) -1L)
@@ -1117,18 +1114,22 @@
static int ep_poll(struct eventpoll *ep, struct epoll_event __user *events,
int maxevents, long timeout)
{
- int res, eavail;
+ int res, eavail, timed_out = 0;
unsigned long flags;
- long jtimeout;
+ long slack;
wait_queue_t wait;
+ struct timespec end_time;
+ ktime_t expires, *to = NULL;
- /*
- * 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;
+ if (timeout > 0) {
+ ktime_get_ts(&end_time);
+ timespec_add_ns(&end_time, (u64)timeout * NSEC_PER_MSEC);
+ slack = select_estimate_accuracy(&end_time);
+ to = &expires;
+ *to = timespec_to_ktime(end_time);
+ } else if (timeout == 0) {
+ timed_out = 1;
+ }
retry:
spin_lock_irqsave(&ep->lock, flags);
@@ -1150,7 +1151,7 @@
* to TASK_INTERRUPTIBLE before doing the checks.
*/
set_current_state(TASK_INTERRUPTIBLE);
- if (!list_empty(&ep->rdllist) || !jtimeout)
+ if (!list_empty(&ep->rdllist) || timed_out)
break;
if (signal_pending(current)) {
res = -EINTR;
@@ -1158,7 +1159,9 @@
}
spin_unlock_irqrestore(&ep->lock, flags);
- jtimeout = schedule_timeout(jtimeout);
+ if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS))
+ timed_out = 1;
+
spin_lock_irqsave(&ep->lock, flags);
}
__remove_wait_queue(&ep->wq, &wait);
@@ -1176,7 +1179,7 @@
* more luck.
*/
if (!res && eavail &&
- !(res = ep_send_events(ep, events, maxevents)) && jtimeout)
+ !(res = ep_send_events(ep, events, maxevents)) && !timed_out)
goto retry;
return res;
diff --git a/fs/select.c b/fs/select.c
index 5f023f9..b7b10aa 100644
--- a/fs/select.c
+++ b/fs/select.c
@@ -67,7 +67,7 @@
return slack;
}
-static long select_estimate_accuracy(struct timespec *tv)
+long select_estimate_accuracy(struct timespec *tv)
{
unsigned long ret;
struct timespec now;
diff --git a/include/linux/poll.h b/include/linux/poll.h
index 600cc1f..56e76af 100644
--- a/include/linux/poll.h
+++ b/include/linux/poll.h
@@ -73,6 +73,8 @@
extern void poll_freewait(struct poll_wqueues *pwq);
extern int poll_schedule_timeout(struct poll_wqueues *pwq, int state,
ktime_t *expires, unsigned long slack);
+extern long select_estimate_accuracy(struct timespec *tv);
+
static inline int poll_schedule(struct poll_wqueues *pwq, int state)
{
