In article <Pine.NEB.4.21.0009250953480.10166-100000@ribston.astrakan.hig.se>,
=?iso-8859-1?Q?P=E5llen?=  <pollen@astrakan.hig.se> wrote:
> Some days ago I posted about my auvia that didn't get attached by the
> kernel, complaining about some variable samplerate. Someone told me to try
[...]
> realized that the audio seems to be stuck at 48kHz. I'm willing to do some

I can't find your first post on mail-index, but I'm guessing that the
message reported was "codec must support AC'97 2.0 Variable Rate Audio".
If that is the case, it's no surprise that when you disable that, you
get stuck at a fixed sample rate :-)

That message is there for a reason -- because the driver currently
doesn't handle such hardware. I don't have access to such a system, so I
didn't try to support it.

In order to make it work will have to arrange for software sample rate
conversion to be done, which kinda sucks.  Also, I don't think Jaromir's
test of rev H is correct -- I don't see how the chip rev could have
anything to do with it, as it is a codec function.  It should be a codec
quirk or something.