summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorEric Dumazet <edumazet@google.com>2012-08-21 15:05:14 +0200
committerAl Viro <viro@zeniv.linux.org.uk>2012-08-22 10:31:05 -0400
commit88ec2789d856056344161aa20420dd37e893b0fe (patch)
tree0e944baca39ae432b11ca105fa5c8d0cdd6a47a5
parent55852635a8e2803cbc22d0e143d727813f0fcdb5 (diff)
downloadlinux-3.10-88ec2789d856056344161aa20420dd37e893b0fe.tar.gz
linux-3.10-88ec2789d856056344161aa20420dd37e893b0fe.tar.bz2
linux-3.10-88ec2789d856056344161aa20420dd37e893b0fe.zip
task_work: add a scheduling point in task_work_run()
It seems commit 4a9d4b02 (switch fput to task_work_add) reintroduced the problem addressed in commit 944be0b2 (close_files(): add scheduling point) If a server process with a lot of files (say 2 million tcp sockets) is killed, we can spend a lot of time in task_work_run() and trigger a soft lockup. Signed-off-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
-rw-r--r--kernel/task_work.c1
1 files changed, 1 insertions, 0 deletions
diff --git a/kernel/task_work.c b/kernel/task_work.c
index 91d4e1742a0..d320d44903b 100644
--- a/kernel/task_work.c
+++ b/kernel/task_work.c
@@ -75,6 +75,7 @@ void task_work_run(void)
p = q->next;
q->func(q);
q = p;
+ cond_resched();
}
}
}