Re: [LAU] 2.6.21-rt7 Oopses, More

Re: [LAU] 2.6.21-rt7 Oopses, More

[David Baron]

On Di, 29.05.07 21:37 David Baron <d_baron@012.net.il> wrote:

> >Ran for nearly two hours with none. Nada.
> 
> >Then they started up again. What triggered them?
> 
> >The only thing in the logcheck at the same time as the first outputs
> >was a pptp connect after the original connection fell. Relevant?
> 
> The offending functions are icmp_.... functions, most likely the more
> used ipv4 flavor of icmp.c.
> 
> Example:
> BUG: using smp_processor_id() in preemptible [00000000] code:
> IRQ-9/1110 caller is icmp_xmit_unlock+0xb/0x30
>  [<c01faab9>] debug_smp_processor_id+0xa9/0xb0
>  [<c02e531b>] icmp_xmit_unlock+0xb/0x30
>  [<c02e598e>] icmp_send+0x11e/0x3f0
> ....

Could you please

a) add a Re: to the subject before answering yourself, so this gets a
thread, instead of single mails.

and b) maybe CC linux-kernel@vger.kernel.org as I do not think Ingo
Molnar is subscribed to this list here, but it would be good if he also
gets your reports in order to fix it.

Thanks.