Re: del_timer_sync(), function epilog and rmmod puzzle

[Ender.Dai <ender.dai@xxxxxxxxx>]

Hi,

On Jan 8, 2008 1:49 PM, Michael Blizek
<michi1@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote:
>
> The timer function is executed in interrupt context. Calling
> wake_up_interruptable in the timer function may cause problems.

wake_up_interruptable() was called in a task queue task handler, and
the task was put into the scheduler queue by schedule_task(). This
queue was running in process context (keventd),  so I guess there are
no problem here, but this is not the point. The key problem is, will
the epilog of the task handler cause kernel oops?

> On non-smp
> systems you do not need any synchronization. On smp machines you can try calling
> sleep ??ms (do not know if you can sleep in the module unload function) after
> stopping the timer and then unload. I know, this is dirty,  but if the timer
> function is running, it will not be interrupted and rescheduled.
>

And thanks for your kindly reply.

-- 
Invent and fit; have fits and reinvent!
We toast the Lisp programmer who pens his thoughts within nests of parentheses.

--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ