On Tue, 2002-06-04 at 21:55, Benjamin LaHaise wrote:
> Hey folks,
>
>
>
> I had been playing with 2.5KB stacks (4KB page minus 1.5K for task_struct),
> and it is possible given a few fixes for massive (>1KB) stack allocation
> in the pci layer and a few others. So far 4KB hasn't overflowed on any
> of the tasks I normally run (checked using a stack overflow checker that
> follows).
>
Ben,
Just what are the tasks you normally run - and how many code
paths do you think there are out there which you do not run. XFS
might get a bit stack hungry in places, we try to keep it down,
but when you get into file system land things can stack up quickly:
NFS server -> file system -> block layer -> device driver
With possibly some form of volume management out there too.
I am pounding away on xfs with your code in there including the
checker, and so far it is surviving. But I only have a plain old
scsi drive underneath, and no NFS on top.
Steve
--Steve Lord voice: +1-651-683-3511 Principal Engineer, Filesystem Software email: lord@sgi.com - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Jun 07 2002 - 22:00:26 EST