On Friday 22 May 2009 13:23:55 ext Mark Brown wrote: > On Fri, May 22, 2009 at 12:57:23PM +0300, Peter Ujfalusi wrote: > > On Friday 22 May 2009 12:28:07 ext Mark Brown wrote: > > > I've applied both. It'd be really good if you could look into > > > converting the TWL4030 driver into a regular platform driver, probing > > > via the normal methods, so it could get this data as regular platform > > > data. > > > > Thanks, any pointers for such an implementation? > > wm8350 and wm8400 are similar chips with audio blocked in with other > functionality. > > > Do we need to convert the codec driver(s) to use the > > platform_device_register/platform_driver_register, or is it more feasible > > to convert the platform drivers (for example > > sound/soc/omap/omap3beagle.c)? > > The idea is that the device registration gets done either in arch/arm > or in drivers/mfd - the intention is to make things look much more like > the standard device model. For twl4030 I'd expect it to be done in > drivers/mfd. I see. After a quick check: it needs some modification for the include/linux/i2c/twl4030.h: New struct for the twl4030_codec, adding that to the existing twl4030_platform_data. Than modification to drivers/mfd/twl4030-core.c: Adding the audio codec as new child. Modifying the arch/arm/mach-omap*/ board files to fill up the new audio codec struct and include it to their twl4030_platform_data. Than some modification to the twl4030.c codec in sound/soc/codec/ Or something along these lines... I'll keep this in mind and try to prepare for it in the near future. Thanks for the clarification, Péter _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
