From owner-freebsd-stable Mon Jun 10 9:15:45 2002 Delivered-To: freebsd-stable@freebsd.org Received: from tinker.exit.com (tinker.exit.com [206.223.0.1]) by hub.freebsd.org (Postfix) with ESMTP id 0A06037B40B for ; Mon, 10 Jun 2002 09:15:30 -0700 (PDT) Received: from realtime.exit.com (realtime [206.223.0.5]) by tinker.exit.com (8.12.3/8.12.3) with ESMTP id g5AGEH48018143; Mon, 10 Jun 2002 09:14:24 -0700 (PDT) (envelope-from frank@exit.com) Received: from realtime.exit.com (localhost [127.0.0.1]) by realtime.exit.com (8.12.3/8.12.2) with ESMTP id g5AGE6Mj088719; Mon, 10 Jun 2002 09:14:06 -0700 (PDT) (envelope-from frank@realtime.exit.com) Received: (from frank@localhost) by realtime.exit.com (8.12.3/8.12.3/Submit) id g5AGDuCb088693; Mon, 10 Jun 2002 09:13:56 -0700 (PDT) From: Frank Mayhar Message-Id: <200206101613.g5AGDuCb088693@realtime.exit.com> Subject: Re: Numerous hard hangs on TWO different ASUS P4T-E w/P4 1.6G In-Reply-To: <20020610104015.D9887@cscoms.net> To: Alain Fauconnet Date: Mon, 10 Jun 2002 09:13:56 -0700 (PDT) Cc: David Schultz , freebsd-stable@FreeBSD.ORG Reply-To: frank@exit.com X-Copyright0: Copyright 2002 Frank Mayhar. All Rights Reserved. X-Copyright1: Permission granted for electronic reproduction as Usenet News or email only. X-Mailer: ELM [version 2.4ME+ PL98b (25)] MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Alain Fauconnet wrote: > I will consider doing a CVS upgrade (first I need to learn how > to do it). I dislike the idea of a downgrade to 4.4 or whatever older > than 4.5. > > For what it's worth, the box has *not* hung again since I have > switched off DMA (forced PIO using sysctl). It's clearly not an option > for long-term production mode though, and it has already happened that > it had been stable for weeks, so that doesn't really tell much. Unfortunately, this is not an option for me, since I exclusively use SCSI. > Since the box freezes up solid and I can't enter DDB, I really wonder > what I could do to help tracking down the problem. Suggestions are > welcome. I'm in the same boat... I am completely convinced, now, that this has to do with interrupt sharing. Over the weekend I turned off the parallel port and moved cards around; I tried to turn off IDE but although the BIOS says it's off it's still being detected by probe and the IRQs aren't being freed. May be a BIOS or chipset oddity. Anyway I was able to free a couple of IRQs, 5 and 7, and though I wasn't able to force the sound card to its own interrupt, it is now only sharing IRQ 5 with half of the 39160 card. The upshot: Things are better but of course they aren't fixed. Although it's hard to quantify, the freezes seem to have been reduced in frequency and duration. No hard hangs requiring a reboot yet, but the system has only been back up for a little less than two days. -- Frank Mayhar frank@exit.com http://www.exit.com/ Exit Consulting http://www.gpsclock.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message