Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 13 Feb 2002 19:08:22 -0800
From:      "Eugene M. Kim" <gene@nttmcl.com>
To:        Mike Meyer <mwm-dated-1014085865.349159@mired.org>
Cc:        FreeBSD Multimedia Mailing List <multimedia@freebsd.org>
Subject:   Re: ich sound driver question?
Message-ID:  <20020213190822.A444@alicia.nttmcl.com>
In-Reply-To: <15467.8553.401620.914825@guru.mired.org>; from mwm-dated-1014085865.349159@mired.org on Wed, Feb 13, 2002 at 08:31:05PM -0600
References:  <15465.52738.786349.856491@guru.mired.org> <20020213160154.A20758@alicia.nttmcl.com> <15467.8553.401620.914825@guru.mired.org>

next in thread | previous in thread | raw e-mail | index | archive | help
On Wed, Feb 13, 2002 at 08:31:05PM -0600, Mike Meyer wrote:
> 
> 
> [Context recovered from top posting.]
> 
> Eugene M. Kim <gene@nttmcl.com> types:
> > On Tue, Feb 12, 2002 at 08:22:58PM -0600, Mike Meyer wrote:
> > > I've got a patched version of that driver that works with the SiS 7012
> > > built into the SiS 735 chipset found on the ECS K7S5A board - but it
> > > doesn't record on -stable at all, and MFC'ing things from -current
> > > makes it record, but it gets the data rate wrong. I don't have an ich
> > > to check this stuff on, and would like to know if this is something I
> > > have to fix in the 7012 support, or a problem with the ich in general.
> > The problem is: the ICH chipset has two distinct recording channel, one
> > for every possible input source (line-in, CD-ROM, mic, modem and so on)
> > and the other for the microphone only.  The ICH driver utilizes this
> > feature and registers two recording channels, the generic one first
> > then the microphone-only one.  But due to the LIFO characteristic of the
> > recording channel selection, the mic-only channel gets chosen by the
> > first application that opens /dev/dsp.
> 
> Note that I'm not using the ich chipset, but something that looks like
> it. Are the patches for working around the microphone problem in
> -current? If so, that would explain why it records when most of the
> changes from -current are patched into it, but not on -stable.

I haven't tried the patch on -current (had no -current box with an ICH
chipset to tset it on).  It works on -stable and solves the recording
problem (that at least I had).

The separate microphone channel is, IIRC, a design feature incorporated
in the latest revision of AC'97 codec specification, so I believe there
could be other chipset that suffers from the same problem.  But a quick
scan of the -current pcm driver shows that only the ICH driver actually
utilizes the microphone channel.  If you somehow morphed the ICH driver
to suit the SiS 7012 chipset then it could possibly be the problem.

Regards,
Eugene

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020213190822.A444>