22 Jan
2010
22 Jan
'10
11:35 a.m.
On Fri, Jan 22, 2010 at 09:35:44AM +0100, Guennadi Liakhovetski wrote:
On Wed, 20 Jan 2010, Mark Brown wrote:
That seems wrong - if they're being managed in the bias level configuration they ought to be being turned off at some point. If they should be set all the time then set them during chip init.
Right, a couple of lines below:
- case SND_SOC_BIAS_OFF:
snd_soc_write(codec, WM8978_POWER_MANAGEMENT_1, 0);
Yeah, I saw that. This is one of those cases where numeric bitmask based updates make things much harder to follow - without referring to the datasheet and thinking about it it's not clear what's going on and if it's intentional.
I'd suggest changing all this code to use snd_soc_update_bits() and symbolic names for the bits that are getting twiddled - it really is a lot easier to follow.