From alsa-user-owner@alsa.jcu.cz  Tue Oct 27 10:42:15 1998
Received: from entry.jcu.cz (IDENT:perex@entry.jcu.cz [160.217.1.111])
	by marvin.jcu.cz (8.9.1a/8.9.1) with SMTP id KAA16402;
	Tue, 27 Oct 1998 10:36:27 +0100
Date: Tue, 27 Oct 1998 10:36:27 +0100 (CET)
From: Jaroslav Kysela <perex@jcu.cz>
To: Paul Winkler <zarmzarm@erols.com>
cc: alsa-user@alsa.jcu.cz, alsa-devel@alsa.jcu.cz
Subject: Re: Lockups, one step closer
In-Reply-To: <36358CB9.3340@erols.com>
Message-ID: <Pine.LNX.3.96.981027102650.26753H-100000@entry.jcu.cz>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: alsa-user@alsa.jcu.cz
Sender: alsa-user-owner@alsa.jcu.cz
Precedence: list

On Tue, 27 Oct 1998, Paul Winkler wrote:

> OK, this is interesting. As long as I've got two OSes on my machine, I
> might as well use them to troubleshoot each other, right? I just
> downloaded ntrack, a shareware Win95 multitrack recorder. With no
> special configuration, I installed it and recorded three stereo tracks,
> 2 minutes long. Full duplex worked fine, no lockups, no problems.
> 
> This leads me to believe that either it's not really a hardware problem,
> or the Winblows 95 audio system knows how to work around the hardware
> and Linux doesn't.
> 
> I also just got this email from Hannu S. at 4front:
> 
> > Some PC chipsets such as VIA and Intel VX have somehow broken ISA DMA.
> > It's possible that the crashes caused by this will be fixed in the next
> > OSS version (3.9.1i).
> 
> So at least that tells me that 4front believes the problem can be worked
> around in the driver...

Maybe it is possible (check over PCI interface if broken PCI to ISA bridge
exist), but is somewhere list of broken chips with PCI vendor/device IDs
and described workarounds for them? Can someone prepare fix for ALSA?
All ISA DMA related code is in kernel/isadma.c source code...

						Jaroslav

-----
Jaroslav Kysela <perex@jcu.cz>
Academic Computer Centre, University of South Bohemia
Branisovska 31, C. Budejovice, CZ-370 05 Czech Republic


