On Tue, Nov 26, 2019 at 04:55:39PM +0000, Adam Thomson wrote: > On 21 November 2019 21:49, Adam Thomson wrote: > > On 20 November 2019 15:24, Sebastian Reichel wrote: > > I've thought more about this and for the scenario where a machine driver > > controls the PLL through a DAPM widget associated with this codec (like some > > Intel boards do), then the PLL will be configured once here and then again > > when the relevant widget is called. I don't think that will matter but I will > > take a further look just in case this might cause some oddities. > So I don't see any issues per say with the PLL function being called twice in > the example I mentioned. However it still feels a bit clunky; You either live > with it or you have something in the machine driver to call the codec's PLL > function early doors to prevent the bias_level() code of the codec controlling > the PLL automatically. Am wondering though if there would be some use in having > an indicator that simple-card is being used so we can avoid this? I guess we If we're special casing simple-card we're doing it wrong - there's nothing stopping any other machine driver behaving in the same way.
Attachment:
signature.asc
Description: PGP signature
_______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx https://mailman.alsa-project.org/mailman/listinfo/alsa-devel
