Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 23 Jan 2002 03:37:58 -0600
From:      Erik Greenwald <erik@smluc.org>
To:        freebsd-current@freebsd.org
Subject:   Re: pcm: cracking noise for CMedia driver
Message-ID:  <200201230937.DAA02818@smluc.org>

next in thread | raw e-mail | index | archive | help
> Hi!
> 
> I own an ASUS A7V266-E motherboard with onboard-sound.
> 
> With this kernel:
> FreeBSD zerogravity.kawo2.rwth-aachen.de 5.0-CURRENT FreeBSD 5.0-CURRENT #1: Thu Dec 20 00:34:04 CET 2001     alex@zerogravity.kawo2.rwth-aachen.de:/storage/zerogravity/obj/usr/src/sys/CICHLIDS  i386
> 
> Sound works fine, as it should (ok, I can't get line-in/mic to work, but
> that's another issue).
> 
> alex@zerogravity ~dir $ cat /dev/sndstat 
> FreeBSD Audio Driver (newpcm)
> Installed devices:
> pcm0: <CMedia CMI8738> at io 0xd800 irq 10 (1p/1r/0v channels duplex default)

I've about the same setup, an iwill kk266 mb with a cmi 8738 on the mb
pcm0: <CMedia CMI8738> at io 0xe000 irq 9 (1p/1r/0v channels duplex default)
FreeBSD fenris 5.0-CURRENT FreeBSD 5.0-CURRENT #6: Sun Jan 20 17:14:31 CST 2002     root@fenris:/usr/src/sys/i386/compile/FENRIS  i386

> 
> With a kernel starting January 18th or so I always get a cracking noise
> on tones with high loudness, e.g. on MP3s with much bass.

hrm, I started getting it a lot earlier, but the same problem... I did a lot of
experimenting, including de-evolving the driver back to the 4.x level where
it didn't have any problems (just meant removing the locking), and trying to
interface the audio system in different ways. 

I usually use 'esd' to do audio, and the popping is only roughly a couple 
times a second with a 512 byte audio buffer, but with straight 'oss', it puts 
almost a buzz over the sound, that's a 4k buffer I think

It still seems to be there when the volume isn't very high, but much less
noticable... some songs are smoove enough to really show it off. Certain
operations cause it to really static out so it sounds like random noise, um,
a long sequence of stat() calls one right after another, scrolling with the
arrow keys in a mozilla widget (the mozilla widged also seems a whole lot
slower on these kernels, so I have a feeling that the flaw is a system call
or kernel component that suddenly has more latency due to new locking
or something? I d'no, I'm guessing). I've also noticed that the pop is
inserted into the sound stream, not overlaid, so if a clip takes 10 seconds
and there's a lot of popping, it pushes it up to 12 or whatever. 

> 
> I can't see any changes to the cmi-driver in that timespan, though, so
> I wonder what might have changed.
> 
> Any idea?
> 
> Thanks
> 
> Alex
> 
> PS:
> lock order reversal
>  1st 0xc66e7b00 pcm0 @ /usr/src/sys/dev/sound/pcm/sound.c:132
>  2nd 0xc66e7940 pcm0:play:0 @ /usr/src/sys/dev/sound/pcm/sound.c:189

        -Erik <erik@smluc.org> [http://math.smsu.edu/~erik]

The opinions expressed by me are not necessarily opinions. In all probability,
they are random rambling, and to be ignored. Failure to ignore may result in
severe boredom or confusion. Shake well before opening. Keep Refrigerated.

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




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200201230937.DAA02818>