call_usermodehelper && check_hung_uninterruptible_tasks

[Oleg Nesterov <oleg@redhat.com>]

(change subject, add lkml)

On 05/19, Boaz Harrosh wrote:
>
> On 05/19/2012 05:21 AM, Tetsuo Handa wrote:
>
> > By the way, when considering whether we want call_usermodehelper_timeout()
> > or not, maybe we also want to consider below hung check timer topic.
> >
> > I noticed that a task waiting for call_usermodehelper(UMH_WAIT_PROC) to return
> > is subjected to hung check timer. call_usermodehelper(UMH_WAIT_PROC) would
> > normally return within few seconds, but it can take longer than
> > hung_task_timeout_secs if the usermode helper is waiting for (e.g.) user's
> > input. Should hung check timer warn this situation?
> >
>
>
> The fix depends on if we want to enable usermodehelper to wait for user-input ?
> 	we will need to silence the hung check timer

tomoyo_load_policy() can use call_usermodehelper(UMH_WAIT_PROC | UMH_KILLABLE),
this should also silence the hung check.


But there is another problem, wait_for_completion/wait_for_completion_killable
are not freezer-friendly. Perhaps something like the patch below? Just in case,
it should silence check_hung_uninterruptible_tasks() as well.

Oleg.

--- x/kernel/kmod.c
+++ x/kernel/kmod.c
@@ -557,7 +557,9 @@ int call_usermodehelper_exec(struct subp
 		goto unlock;
 
 	if (wait & UMH_KILLABLE) {
+		freezer_do_not_count();
 		retval = wait_for_completion_killable(&done);
+		freezer_count();
 		if (!retval)
 			goto wait_done;
 
@@ -567,7 +569,9 @@ int call_usermodehelper_exec(struct subp
 		/* fallthrough, umh_complete() was already called */
 	}
 
+	freezer_do_not_count();
 	wait_for_completion(&done);
+	freezer_count();
 wait_done:
 	retval = sub_info->retval;
 out: