netdev
[Top] [All Lists]

Re: Adding just a pinch of icache/dcache pressure...

To: Jan Harkes <jaharkes@xxxxxxxxxx>
Subject: Re: Adding just a pinch of icache/dcache pressure...
From: Andrew Morton <andrewm@xxxxxxxxxx>
Date: Sun, 25 Mar 2001 03:29:23 +1000
Cc: Andi Kleen <ak@xxxxxx>, "netdev@xxxxxxxxxxx" <netdev@xxxxxxxxxxx>
References: <20010323015358Z129164-406+3041@vger.kernel.org> <Pine.LNX.4.21.0103230403370.29682-100000@imladris.rielhome.conectiva> <20010323122815.A6428@win.tue.nl> <m1hf0k1qvi.fsf@frodo.biederman.org> <3ABB6833.183E9188@mandrakesoft.com> <20010323111056.A9332@cs.cmu.edu> <20010323171716.28420@colin.muc.de>, <20010323171716.28420@colin.muc.de>; <20010323115123.A12720@cs.cmu.edu> <3ABC4B33.F1833B5A@uow.edu.au>, <3ABC4B33.F1833B5A@uow.edu.au>; from andrewm@uow.edu.au on Sat, Mar 24, 2001 at 06:22:27PM +1100 <20010324110822.A23436@cs.cmu.edu>
Sender: owner-netdev@xxxxxxxxxxx
Jan Harkes wrote:
> 
> I do have a SMP system, so it could be an obscure race. The thing that
> might have been triggering it were the backups that consisted of about
> 121 tcp connection with a total datatransfer of about 1GB. But it could
> also be UDP fragments, Coda tends to send 4KB UDP packets when
> revalidating it's caches.
> 

OK, thanks.  I'll test UDP with large packets and random corruption.
I guess testing with NFS client didn't cover that path adequately.

<Prev in Thread] Current Thread [Next in Thread>