Re: Crashes in perf_event_ctx_lock_nested

[Thomas Gleixner]

On Wed, 1 Nov 2017, Peter Zijlstra wrote:

> On Tue, Oct 31, 2017 at 10:32:00PM +0100, Thomas Gleixner wrote:
> > That means we can have the following situation:
> > 
> >     lock(watchdog_mutex);
> >     lockup_detector_reconfigure();
> >         cpus_read_lock();
> > 	stop();
> > 	   park()
> > 	update();
> > 	start();
> > 	   unpark()
> > 	cpus_read_unlock();		thread runs()
> > 	cleanup();						
> >     unlock(watchdog_mutex);
> > 
> 
> Isn't there also a where hardlockup_detector_perf_init() creates an
> event to 'probe' stuff, and then hardlockup_detector_perf_enable()
> _again_ creates the event?

probe() releases the event.