Messages in this thread Patch in this message |  | | From | Spencer Baugh <> | | Subject | [PATCH] soft lockup: kill realtime threads before panic | | Date | Tue, 21 Jul 2015 15:07:57 -0700 |
| |
From: Joern Engel <joern@logfs.org>
We have observed cases where the soft lockup detector triggered, but no kernel bug existed. Instead we had a buggy realtime thread that monopolized a cpu. So let's kill the responsible party and not panic the entire system.
Signed-off-by: Joern Engel <joern@logfs.org> Signed-off-by: Spencer Baugh <sbaugh@catern.com> --- kernel/watchdog.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/kernel/watchdog.c b/kernel/watchdog.c index a6ffa43..2355bd5 100644 --- a/kernel/watchdog.c +++ b/kernel/watchdog.c @@ -428,7 +428,10 @@ static enum hrtimer_restart watchdog_timer_fn(struct hrtimer *hrtimer) } add_taint(TAINT_SOFTLOCKUP, LOCKDEP_STILL_OK); - if (softlockup_panic) + if (rt_prio(current->prio)) { + pr_emerg("killing realtime thread\n"); + send_sig(SIGILL, current, 0); + } else if (softlockup_panic) panic("softlockup: hung tasks"); __this_cpu_write(soft_watchdog_warn, true); } else -- 2.4.3
|  |