linux-mips
[Top] [All Lists]

Re: [PATCH] Fix TCP/UDP checksums on the Broadcom SB-1

To: Matej Kupljen <matej.kupljen@ultra.si>
Subject: Re: [PATCH] Fix TCP/UDP checksums on the Broadcom SB-1
From: Ralf Baechle <ralf@linux-mips.org>
Date: Tue, 20 Sep 2005 12:06:09 +0100
Cc: "Maciej W. Rozycki" <macro@linux-mips.org>, Daniel Jacobowitz <dan@debian.org>, linux-mips@linux-mips.org
In-reply-to: <1127214005.2149.15.camel@localhost.localdomain>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <20050920032818.GA7199@nevyn.them.org> <Pine.LNX.4.61L.0509201140160.23494@blysk.ds.pg.gda.pl> <1127214005.2149.15.camel@localhost.localdomain>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Mutt/1.4.2.1i
On Tue, Sep 20, 2005 at 01:00:04PM +0200, Matej Kupljen wrote:

> > > This caused incorrect checksums in some UDP packets for NFS root.  The
> > > problem was mild when using a 10.0.1.x IP address, but severe when
> > > using 192.168.1.x.
> > 
> >  Ah!  So *that* is the reason for the absolutely abysmal NFS performance 
> > of the SWARM with 2.6!  I have had no time to track it down -- thanks a 
> > lot!
> 
> Is this for MIPS64 only?
> Because, on dbau1200 we also have poor NFS performance :-(

It's for 64-bit kernels only.  Note the difference, I didn't say MIPS64.

Also, since this bug did result in an operation that has undefined
behaviour it likely may will only have impacted some 64-bit processors -
such as the SB1 - but others may have been unaffected.

  Ralf

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