[snd_hda] Choppiness under high cpu - STABLE

Norberto Meijome freebsd at meijome.net
Mon Jun 4 13:08:00 UTC 2007


On Mon, 4 Jun 2007 16:44:20 +0800
Ariff Abdullah <ariff at FreeBSD.org> wrote:

> > 
> > Have you tried with 4096 ? Note: this should go into
> > hint.pcm.0.buffersize kernel hint, not sysctl variable. After
> > setting it (using kenv), you must reload the sound module or reboot
> > if it going through /boot/device.hints. The sysctl buffersize is
> > there purely as indicator.
> >   
> 
> I forgot to mention that enabling/disabling vchans are matters as
> well, so please take note on that.

ok - back to the klds from the src tree.

[betom at ayiin] [Mon Jun  4 22:30:37 2007]
/usr/home/betom
$ sysctl -a | grep snd
hw.snd.report_soft_formats: 1
hw.snd.targetirqrate: 32
hw.snd.verbose: 1
hw.snd.maxautovchans: 4
hw.snd.unit: 0
hw.snd.pcm0.buffersize: 4096
hw.snd.pcm0.vchans: 1

[betom at ayiin] [Mon Jun  4 22:30:43 2007]
/usr/home/betom
$ kenv | grep buffer
hint.pcm.0.buffersize="4096"

I tried 4096, 8192, 16384. In all cases, stress -m 1  brings up immediate
choppiness/chirping. :(


I'm now back to the binary blob... I repeated the tests (with default
buffersize of 16K, and 8 vchans as before), and it seems that with stress -m 2
there is an ever so slight chirpiness, more noticeable with -m 3 , definitely
there @ -m 4 - but definitely better than the other one.

thanks!!
B

_________________________
{Beto|Norberto|Numard} Meijome

Life is not measured by the number of breaths we take, but by the moments that
take our breath away.

I speak for myself, not my employer. Contents may be hot. Slippery when wet.
Reading disclaimers makes you go blind. Writing them is worse. You have been
Warned.


More information about the freebsd-multimedia mailing list