On 11-06-2008 07:40, David Miller wrote:
...

cassini: Does flush_scheduled_work() from ->change_mtu() method, which
	 also holds RTNL semaphore.  Seems tricky to fix as it's trying
	 to schedule work and then immediately wait for it to
	 complete synchronously before returning from the ->change_mtu
	 method.  Maybe calling cas_reset_task() directly would work,
	 but this could conflict with reset tasks scheduled in other
	 contexts.  NOT FIXED

Could you give an example of such a conflict? (I wonder how it's
avoided with current solution.)

Jarek P.