Date: Tue, 4 Mar 1997 04:09:30 -0500 From: exidor@superior.net (Christopher Masto) To: msmith@atrad.adelaide.edu.au (Michael Smith) Cc: jlemon@americantv.com (Jonathan Lemon), hackers@freebsd.org Subject: Re: xemacs crashes kernel Message-ID: <19970304040930.VV27825@@> In-Reply-To: <199703040546.QAA10965@genesis.atrad.adelaide.edu.au>; from Michael Smith on Mar 4, 1997 16:16:46 %2B1030 References: <19970303233743.01587@right.PCS> <199703040546.QAA10965@genesis.atrad.adelaide.edu.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Michael Smith writes: > Jonathan Lemon stands accused of saying: > > On Mar 03, 1997 at 11:01:57PM -0600, Jonathan Lemon wrote: > > > > > > stopped at _fsync+0x73, testb $0x40, 0x18(%eax) > > > > Just to follow up on the really obvious, this failure is in fsync(), at: > > > > error = VOP_FSYNC(vp, fp->f_cred, > > (vp->v_mount->mnt_flag & MNT_ASYNC) ? MNT_NOWAIT : MNT_WAIT, p); > > > > where 0x40 is MNT_ASYNC, and %eax is supposed to contain vp->v_mount, but > > contains a NULL pointer instead. Since this is filesystem related, I should > > mention that I have both MFS and PROCFS in the kernel, and that MFS is > > mounted on /tmp (async, local). (if this makes any difference) > > The MFS code doesn't appear to touch v_mount at all. By contrast, the UFS > code does, but it appears to only read it. If you don't mount /tmp on your > MFS, does the system still barf? I'm not using MFS at all and I can confirm this problem on my 2.2-GAMMA system. Can't get the exact uname until I get into the office tomorrow to resurrect it. :-( -- Christopher Masto . . . . chris@masto.com . . . . . Masto Consulting: info@masto.com On Permanence: It could permanently hurt a batter for a long time. - Pete Rose, Cincinnati Red, speaking about a brushback pitch.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19970304040930.VV27825>