From owner-freebsd-current  Sun Oct 29 12:17:57 1995
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id MAA16029
          for current-outgoing; Sun, 29 Oct 1995 12:17:57 -0800
Received: from haywire.DIALix.COM (news@haywire.DIALix.COM [192.203.228.65])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id MAA16010
          for <freebsd-current@freebsd.org>; Sun, 29 Oct 1995 12:17:49 -0800
Received: (from news@localhost) by haywire.DIALix.COM (sendmail) id EAA08588 for freebsd-current@freebsd.org; Mon, 30 Oct 1995 04:17:40 +0800 (WST)
Received: from GATEWAY by haywire.DIALix.COM with netnews
	for freebsd-current@freebsd.org (problems to: usenet@haywire.dialix.com)
To: freebsd-current@freebsd.org
Date: 30 Oct 1995 04:17:33 +0800
From: peter@haywire.dialix.com (Peter Wemm)
Message-ID: <470ngt$8c6$1@haywire.DIALix.COM>
Organization: DIALix Services, Perth, Australia.
References: <199510291438.BAA09021@asstdc.scgt.oz.au>, <199510291545.RAA09345@shadows.cs.hut.fi>
Subject: Re: load related problem or my compilation ?
Sender: owner-current@freebsd.org
Precedence: bulk

hsu@cs.hut.fi (Heikki Suonsivu) writes:
>michael butler writes:
> > Watching the mbuf clusters, I see no more than ~120k allocated in ~40-50
> > clusters so there doesn't seem to be any shortage .. I did another compile
> > with NMBCLUSTERS=2048 but, as expected, nothing different happened. It still
> > dies without logging anything at all about the failure event.
> > 
> > I have a feeling (more than actual observation as the machine's 4km from
> > here) that it may be dying when one of the modems drops the link whilst in
> > full flight .. ~16k of data in the PPP send queue with more rapidly arriving
> > .. but I haven't been able to catch it doing it whilst I'm actually there
> > and watching it.

>The panics I get are almost always in the same place, and it could be
>something like this, but also it could be just getting a freed mbuf
>mid-fligth.  I have had similar panic problems with a slow (38.4k) leased
>line which certainly does not loose carrier, and it panics sometimes,
>usually when a load peak suddenly arrives after idling (start up a heavy X
>program or like).

Hmm.. Are you using Bruce's cy driver?

I have this vague recollection that his cypoll and siopoll hooks are
launching into the tty sybsystem code at splsofttty().  I am not sure,
but I dont think that splsofttty() would mask the networking code,
which could mean that it might be possible for the networking code to
be started up on the return from some hardware interrupt that had
preempted the tty code.

I wonder if what you are seeing is the cypoll() call to pppstart() being
interrupted by networking code, which is manipulating the mbuf
pointers?

I've not looked at the old ppp-2.1.2 code for quite some time now, as
I've been working on getting ppp-2.2 up and running.

I remember having this discussion with Bruce a few weeks ago, but I
can't find his answer and I can't for the life of me remember exactly
what he said.  Just thinking about the lengths that had been gone to
in getting ppp-2.2 up without totally interlocking spltty() and
splimp() makes my brain turn back to mush.

-Peter