On Tue, May 20, 2014 at 02:18:30PM +0200, Marek Szyprowski wrote:
> On 2014-05-20 13:57, Mark Brown wrote:

> >> +                  else
> >> +                          dev_err(dev,
> >> +                                  "unsupported reference clock rate 
> >> (%d)\n",
> >> +                                  rate);

> >This looks like a switch statement.  Should the driver not try to set
> >the clock to a supported rate if it's not already at one rather than
> >error out - it seems like a more constructive thing to do?

> Hmm, the problem here is that you cannot determine the right rate. The rate
> is
> encoded on REF_SEL[1:0] pins (usually soldered to some resistors) and cannot
> be
> read via i2c commands. To set add support for rate setting, I would need to
> add
> one more property with correct ref clock rate. Is this okay?

Sounds reasonable to me, yes.

Attachment: signature.asc
Description: Digital signature

Reply via email to