Re: panic in sysmon_envsys_unregister()

[mlelstv%serpens.de@localhost (Michael van Elst)]

bad%bsd.de@localhost (Christoph Badura) writes:

>Sorry, I meant the question in a more general way.
>sysmon_envsys_sensor_detach() iterates over the sensor_list with sme_mtx
>held etc.  Therefore it looks fishy to me to iterate over the sensor_list
>without sme_mtx held.

The envsys structure has been removed from the list, you are the only
user, it doesn't need to be protected anymore.


>Also it seems that sme_events_destroy() is only ever called if the
>events_list happens to be empty.  And I can't see where the sme work
>queue is destroy elsewhere.

The work queue is created for the first event handler and destroyed
after the last event handler has been deregistered.


>And while looking for places that work queues get destroyed in
>dev/sysmon/ it seems that swwdog's workqueue is created twice when
>loaded as a module.

Yep. The modularization is broken.

-- 
-- 
                                Michael van Elst
Internet: mlelstv%serpens.de@localhost
                                "A potential Snark may lurk in every tree."