From owner-freebsd-hackers  Tue Dec 11  1:13: 1 2001
Delivered-To: freebsd-hackers@freebsd.org
Received: from falcon.prod.itd.earthlink.net (falcon.mail.pas.earthlink.net [207.217.120.74])
	by hub.freebsd.org (Postfix) with ESMTP
	id 3252237B416; Tue, 11 Dec 2001 01:12:52 -0800 (PST)
Received: from pool0063.cvx21-bradley.dialup.earthlink.net ([209.179.192.63] helo=mindspring.com)
	by falcon.prod.itd.earthlink.net with esmtp (Exim 3.33 #1)
	id 16Dixo-0003sf-00; Tue, 11 Dec 2001 01:12:36 -0800
Message-ID: <3C15CE0B.FA53281B@mindspring.com>
Date: Tue, 11 Dec 2001 01:12:43 -0800
From: Terry Lambert <tlambert2@mindspring.com>
X-Mailer: Mozilla 4.7 [en]C-CCK-MCD {Sony}  (Win98; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Danny Braniss <danny@cs.huji.ac.il>
Cc: Bernd Walter <ticso@cicely8.cicely.de>,
	Mike Smith <msmith@FreeBSD.ORG>, freebsd-hackers@FreeBSD.ORG
Subject: Re: irq
References: <E16DiiV-000AEf-00@pampa.cs.huji.ac.il>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-freebsd-hackers@FreeBSD.ORG
Precedence: bulk
List-ID: <freebsd-hackers.FreeBSD.ORG>
List-Archive: <http://docs.freebsd.org/mail/> (Web Archive)
List-Help: <mailto:majordomo@FreeBSD.ORG?subject=help> (List Instructions)
List-Subscribe: <mailto:majordomo@FreeBSD.ORG?subject=subscribe%20freebsd-hackers>
List-Unsubscribe: <mailto:majordomo@FreeBSD.ORG?subject=unsubscribe%20freebsd-hackers>
X-Loop: FreeBSD.ORG

Danny Braniss wrote:
> well, if it's not software it must be hardware problem (true or false ?)
> the test:
>         video capture (using a modified meteor driver)
> doing full size 24bit colour, the meteor would complain about FIFO errors
> (which probably mean that the dma did not finish in time - correct?)
> and sometimes, the adaptec would also complain.
> 
> so i moved cards around.
> 
> 1st try: disaster, the adaptec realy complained, had to reboot.
> 2nd try: Success, fifo errors are down to less than .1% (before it
>         used to be 90%).

It looks like IRQ sharing is the only issue here.

I don't see this as a hardware problem, unless tou are complaining
about the Meteor design (which I'd probably agree with).  Sharing
of IRQs is supposed to be possible, but if you have incredible bus
on times and incredible "I caused this interrupt" register access
latency, and are stacking captures into writes ont disk serviced at
the same interrupt, well, then, you should expect problems.

But I would still blame software (in particular, the firmware on
your Meteor).

-- Terry

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message