From owner-freebsd-hackers Wed May 7 22:55:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id WAA10789 for hackers-outgoing; Wed, 7 May 1997 22:55:58 -0700 (PDT) Received: from sendero.i-connect.net (sendero-ppp.i-Connect.Net [206.190.143.100]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id WAA10778 for ; Wed, 7 May 1997 22:55:52 -0700 (PDT) Received: (qmail 25576 invoked by uid 1000); 8 May 1997 05:44:56 -0000 Message-ID: X-Mailer: XFMail 1.1-alpha [p0] on FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <199705080300.MAA27647@genesis.atrad.adelaide.edu.au> Date: Wed, 07 May 1997 22:32:27 -0700 (PDT) Organization: iConnect Corp. From: Simon Shapiro To: Michael Smith Subject: Re: Privileged Instruction Fault... Cc: FreeBSD-Hackers@FreeBSD.ORG, FreeBSD-SCSI@FreeBSD.ORG Sender: owner-hackers@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi Michael Smith; On 08-May-97 you wrote: ... > Note that the kernel stack is _very_ small; you > should not us use automatics of any substantial size. This is an excellent advise. Thanx! ... > Printf() uses more stack. I am _guessing_ that you are either running > off the end of a local, or using too much stack; it's hard to be > sure with so little data and no indication of where the fault IP > actually lies within your kernel. Actually, your ideas prompted a day of hunting. The game is in the bag. It appears to have been stack problem. The strange thing is that adding printf() calls actually improves things (uses a separate stack?). There are not too many automatics in the code and the depth of recursion is minimal. The (ugly) code is actually posted on sendero-ppp.i-connect.net (206.190.143.100), in /crash. The version there compiles but already too old. Thanx for the great ideas, now I have to chase this queue leak I just found. Simon