339ddb53d373 (fs/epoll: remove unnecessary wakeups of nested epoll) changed
the userspace visible behavior of exclusive waiters blocked on a common
epoll descriptor upon a single event becoming ready. Previously, all tasks
doing epoll_wait would awake, and now only one is awoken, potentially causing
missed wakeups on applications that rely on this behavior, such as Apache Qpid.
While the aforementioned commit aims at having only a wakeup single path in
ep_poll_callback (with the exceptions of epoll_ctl cases), we need to restore
the wakeup in what was the old ep_scan_ready_list() such that the next thread
can be awoken, in a cascading style, after the waker's corresponding ep_send_events().
Signed-off-by: Davidlohr Bueso <dbueso@xxxxxxx>
---
fs/eventpoll.c | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 73138ea68342..1e596e1d0bba 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -657,6 +657,12 @@ static void ep_done_scan(struct eventpoll *ep,
*/
list_splice(txlist, &ep->rdllist);
__pm_relax(ep->ws);
+
+ if (!list_empty(&ep->rdllist)) {
+ if (waitqueue_active(&ep->wq))
+ wake_up(&ep->wq);
+ }
+
write_unlock_irq(&ep->lock);
}
--
2.26.2
