Date: Tue, 22 Apr 1997 23:21:45 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= <ache@nagual.ru> To: Bruce Evans <bde@zeta.org.au> Cc: current@freebsd.org, dyson@freebsd.org Subject: ufs lockstatus bug again (was Re: Recent vfork kernel changes broke csh & tcsh!) Message-ID: <Pine.BSF.3.96.970422231424.205A-100000@nagual.ru> In-Reply-To: <199704221843.EAA29497@godzilla.zeta.org.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, 23 Apr 1997, Bruce Evans wrote: > >BTW, this bug easily cause another bug I report early (ufs lockstatus > >panic) when csh or tcsh is login shell. When you call a.sh from login > >shell, sometimes it cause lockstatus panic instead of exiting to getty. > > I haven't seen this. This one line .profile crash 0day-current with 90% probability: exec /usr/local/bin/ssh -c des -e none localhost (assuming you run recent sshd and _not_ explicitly allow des) This line itself not cause crash, crash always happens on the next sync (from update), so you need some little disk activity to see it: page fault in kernel mode: lockstatus+0x8: cmpw $0,0x10(%edx) stack looks like: lockstatus ufs_islocked vfs_msync sync If "exec" in .profile changed to nothing, all work right, i.e. proper error diagnostic printed from ssh and login prompt appearse again. -- Andrey A. Chernov <ache@null.net> http://www.nagual.ru/~ache/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.96.970422231424.205A-100000>