Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 12 Nov 2004 22:18:25 -0800
From:      Clint Olsen <clint@0lsen.net>
To:        Peter Wemm <peter@wemm.org>
Cc:        FreeBSD amd64 mailing list <freebsd-amd64@freebsd.org>
Subject:   Re: [PATCH] if_sk(4) rx/tx "hangs"
Message-ID:  <20041113061825.GS25188@0lsen.net>
In-Reply-To: <200411121739.40804.peter@wemm.org>
References:  <Pine.BSF.4.53.0411130037360.85716@e0-0.zab2.int.zabbadoz.net> <200411131134.21189.doconnor@gsoft.com.au> <200411121739.40804.peter@wemm.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Nov 12, Peter Wemm wrote:
> I think he was saying he created a 1GB file from the contents of
> /dev/urandom and then timed the transfer of this over 100Mbit/sec link.
> 
> Anyway, the point was that it worked!  Could the problem with the K8V SE
> really be as simple as we've been hardcoding 128K of ram for a device
> that only has 64K?

So, are you suggesting that it's worth testing this patch?  I'm on
5.3-STABLE, and I got so pissed off last night with my sk0 interface that I
brought the machine down and disabled the onboard network and installed a
Linksys 100Mbit card.  My board is the ASUS P4P800-SE, but I believe the
interfaces are the same/similar.

My favorite way to hang it was to use gtk-gnutella.  Massive amounts of
small-sized inbound traffic seemed to bring on the hang (the tell-tale
'watchdog timeout' kernel message).

-Clint



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20041113061825.GS25188>