Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 24 Jun 2000 18:49:04 +0200
From:      Stefan Esser <se@freebsd.org>
To:        Steve Khoo <steve@gordian.com>
Cc:        freebsd-stable@FreeBSD.ORG, Stefan Esser <se@freebsd.org>
Subject:   Re: 4.x-stable(~ June 8) fxp driver and KAME problem
Message-ID:  <20000624184904.A47446@StefanEsser.FreeBSD.org>
In-Reply-To: <394957C4.3AAA628A@gordian.com>; from steve@gordian.com on Thu, Jun 15, 2000 at 03:25:08PM -0700
References:  <394957C4.3AAA628A@gordian.com>

next in thread | previous in thread | raw e-mail | index | archive | help
On 2000-06-15 15:25 -0700, Steve Khoo <steve@gordian.com> wrote:
> I was trying to get KAME working on 4.x-stable box and ran into some
> problems.  In searching for answers I came across this little note in
> the release notes for the latest(20000613) KAME snapshot.  It claims
> there is a problem with fxp driver.   Is this a known problem?  Is there
> a fix for it?

There was a work.-around, that was put into -current soem time 
ago. You may want to check the CVS logs of the fxp driver ...

> - FreeBSD/Openbsd/BSDI Intel EtherExpress Pro driver has some problem
> with
>   the initialization sequence KAME is using.  This is because these
>   drivers use interrupts for multicast filter setup, and KAME code calls
>   multicast initialization code in splimp() or splnet(). 
>   It is not KAME problem, it is problem in drivers.  These drivers
> should
>   be corrected not to use interropts in initialization sequence.
> 
>   KAME/FreeBSD: fxp driver, no workaround available.

I developed a workaround, which de-activated the fxp card and then
(after sleeping for a second) re-initialized it in some particular
order (don't remember the order of steps, i.e. whether the IPv6 or
IPv4 functionality was re-enabled first). You may want to play with
"ifconfig inet x.x.x.x delete" and "ifconfig inet6 x::y:z delete"
followed by ifconfig command that restore working parameters for
inet and inet6 (in either order).

I had this in a file in /usr/local/etc/rc.d, but the problem was
solved by adding some work-around to 5.0-current, which I was using
at the time. I've downgraded the system to 4.0-stable, and it still
works with 3 EtherExpress 100 Pro cards. Will have to see, whether
I have kept the script that worked for me. (Can only do that next 
wednesday.)

Regards, STefan


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20000624184904.A47446>