[LAU] Fwd: [linux-audio-user] Real-time kernel

[Ken Restivo]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Tue, Jun 26, 2007 at 10:35:20AM -0400, Paul Coccoli wrote:
> On 6/26/07, Chuckk Hubbard <badmuthahubbard at gmail.com> wrote:
> >64studio:/mnt/win/Linux# cat /proc/asound/cards
> > 0 [SB             ]: HDA-Intel - HDA ATI SB
> >                      HDA ATI SB at 0xc0000000 irq 16
> >
> 
> Well there's your problem.  Maybe.  Has anyone got an hda-intel setup
> that works well with jack?  IIRC, hda-intel is problematic in that
> each manufacturer's implementation of this "standard" is different.  I
> could be totally wrong here though.


hda-intel is, I'm told, complete shit. ALSA developers basicaly told me to forget about it ever working right; the chip sucks and is poorly-documented. Each laptop manufacturer implements it their own way, and doesn't document it, just hacks around it in their proprietary drivers. Look around the ALSA list to find out more.


Also,  there is nothing wrong with the ac, battery, temperature, and other ACPI modules. In fact you may need them in order for your CPU to run safely (i.e. temperature warnings).

It's the cpufreq stuff that you should definitely remove from the kernel. Or blacklist it in udev/hotplug. Basically yank all the governors and run without any. There is no governor anywhere.

> 
> Also, your eth0 shares an interrupt with the hda-intel.  If you unplug
> eth0, does anything change?

I had firewire sharing an interrupt with eth0, for a while, and no problems. I wasn't doing heavy network activity while recording or mixing though.

- -ken
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (GNU/Linux)

iD8DBQFGgiqie8HF+6xeOIcRApGkAJkB4+0aw+WEupTqmnuTq5VgSa2flwCgkb0j
Js7IsSpXrMN+ze2MDUpdZ44=
=rhby
-----END PGP SIGNATURE-----