Re: [PATCH] 2.6.17-rt1 : fix x86_64 oops

[Dipankar Sarma]

On Wed, Jun 28, 2006 at 09:32:56PM +0200, Ingo Molnar wrote:
> 
> * Dipankar Sarma <dipankar@xxxxxxxxxx> wrote:
> 
> > Turns out that kprobe_flush_task() is called from finish_task_switch() 
> > with preemption disabled and it uses a converted spin lock. The 
> > following patch fixed the problem for me and I was able to boot my 
> > x86_64 box.
> 
> ah, subtle problem and nice fix! We are using an RCU trick to delay task 
> freeing in finish_task_switch(), but kprobe_flush_task() isnt done in 

Yes, otherwise it would have been hell to do __put_task_struct()
with preemption disabled.

> put_task_struct(). [neither would it be right to flush kprobes there.] 
> I've released -rt4 with this included.

OK, I need to catch up, but I see a lot of oops while running rcutorture
in my box (rt1). I am investigating this atm.

Thanks
Dipankar
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/