Re: what happens when the kernel stack overflows?

[Rene Herman <rene.herman@xxxxxxxxxxxx>]

On 24-04-08 17:58, Robert P. J. Day wrote:

> as a related thread to the earlier one, is there any protection against
> the kernel stack extending downwards until it overruns the thread_info
> structure?

Not really no. There's some debugging infrastructure which can warn but 
basically, you just overrun, crash, burn.

> i'm skeptical that there's *any* call chain in kernel code that would
> suck up that much kernel stack, but i'm just curious.

You needn't be skeptical. The at the moment still running "4K stacks default 
on x86" thread on LKML is all about overrrunning that 4K (generally through 
a combination of XFS and RAID/LVM).

Rene

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