Messages in this thread | | | Date | Thu, 1 Dec 2011 10:09:45 +0800 | From | Yong Zhang <> | Subject | Re: [3.2-rc3] 100% CPU usage while in del_timer_sync from iwl3945_rs_free_sta |
| |
On Tue, Nov 29, 2011 at 11:07:27AM +0100, Michal Hocko wrote: > Looks we are stuck in lock_timer_base loop. I am not familiar with the > code but either base is NULL or we are racing with timer->base changes. > I guess that the first one sounds more probable. The timer in question > is &rs_sta->rate_scale_flush.
FYI
If you could apply below commits in -tip/core/debugobjects dc4218b timer: Use debugobjects to catch deletion of uninitialized timers fb16b8c timer: Setup uninitialized timer with a stub callback b84d435 debugobjects: Extend to assert that an object is initialized feac18d debugobjects: Be smarter about static objects
and enable DEBUG_OBJECTS_TIMERS, maybe you can catch something easily.
Thanks, Yong
| |