On Sat, Jul 12, 2008 at 06:36:10PM +0100, Mark Brown wrote: > On Sat, Jul 12, 2008 at 12:00:18AM -0600, Grant Likely wrote:
> It wouldn't be the only driver not to implement PLL configuration in > this way so that's probably be OK for an initial merge. What's expected > for PLL configuration is that you implement the DAI set_pll() operation > in the codec driver, allowing machine drivers to configure the PLL when > they wish. okay > The power configuration should be fixed, though. Normally drivers > either fully implement DAPM (including set_bias_level()) or power > everything in the codec up when the driver is loaded. At the minute > what the driver is doing appears to be powering the codec up in both > _hw_params() and _probe() but never powering anything down - if that is > the case then probably all you need to do is remove the extra power up > from hw_params(), giving you the simple option. done > > Hmmm, I haven't been able to find this; either in the code or on a live > > running system. Where is the common reg dump implemented. > > /sys/bus/platform/devices/soc-audio/codec_reg Yikes. The AIC26 has registers all over the place and most of them are empty. The codec_reg attribute handling means I need to maintain a cache of the entire register file; not just the part that is actually used. Oh well; I can work around it. Thanks, g. _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev