Re: [PATCH 1/2] ASoC: Pay attention to driver supplied DAI IDs

[Mark Brown <broonie@opensource.wolfsonmicro.com>]

On Tue, Oct 12, 2010 at 01:49:52PM -0500, Timur Tabi wrote:
> On Tue, Oct 12, 2010 at 9:57 AM, Mark Brown

> > -               dai->id = i;
> > +               if (dai->driver->id)
> > +                       dai->id = dai->driver->id;
> > +               else
> > +                       dai->id = i;

> I haven't looked at the code, but doesn't this assume that a DAI ID of
> 0 is invalid?

> I have a suspicion I need to retest my driver to make sure it still works.

This makes no change to the behaviour when the DAI ID is zero.
Previously it would be overwritten with i, with the new code it is still
overwritten with i.  You can specify a DAI ID of zero, but only for the
first DAI.  This is a limitation but avoids having to manually set the
DAI ID in the simple case where they're all numbered sequentially from
zero.