Date: Wed, 2 Jun 2004 15:51:21 -0400 (EDT) From: Robert Watson <rwatson@FreeBSD.org> To: Bosko Milekic <bmilekic@FreeBSD.org> Cc: current@FreeBSD.org Subject: Re: uipc_socket2.c panic Message-ID: <Pine.NEB.3.96L.1040602154936.17221K-100000@fledge.watson.org> In-Reply-To: <20040602170800.GA56275@freefall.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 2 Jun 2004, Bosko Milekic wrote:
> You wrote:
> >Additional information (I forgot to mention it - I apology)
> >I am running with debug.mpsafenet=1
>
> This is highly relevant. It means that there is probably a problem
> in network locking somewhere. I have CC'd Robert to make sure he
> is in on the loop for this.
Actually, what it means is that someone didn't read the UPDATING entry
relating to debug.mpsafenet or the heads up message saying "Do not use
debug.mpsafenet=1". Not all of the netperf work is merged yet, and
specifically, the socket locking isn't merged. This could easily explain
socket buffer corruption. I'll be merging a lot of that locking this
week, but until that merging is done, you don't want to run with
debug.mpsafenet=1 unless you're also running with the netperf patch set
at:
http://www.watson.org/~robert/freebsd/netperf/
I'm in the throes of updating that patch for recent merges of accept
locking yesterday.
Robert N M Watson FreeBSD Core Team, TrustedBSD Projects
robert@fledge.watson.org Senior Research Scientist, McAfee Research
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.3.96L.1040602154936.17221K-100000>