On Thu, Aug 20, 2009 at 04:46:15PM +0900, Kuninori Morimoto wrote: > Dear Mark, pHilipp, Magnus It's a bit easier to follow things if you reply to individual mails rather than stitching things together into a singel mail like this. > > Should this driver also be able to drive AK4641? I'm asking because I > I checked datasheet. > My codec is AK4643, and it is compatible with AK4642. > And AK4642 is not compatible AK4641 (^^; > I will re-name it AK464x -> AK4642. > Can you agree with me ? Yes, that sounds lkike a good approach. You can add entries to the I2C device ID table for both ak4642 and ak4643 so it's clear that both devices are supported. > > +#define PREALLOC_BUFFER (32 * 1024) > > +#define PREALLOC_BUFFER_MAX (32 * 1024) > > Is it worth letting machines override these in the platform data or is > > there no point? > do you mean it should be possible to change size ? > If so, is it good as future tasks ? > This time, 32k is good for me. It's OK for you but is it going to be the right choice for every system? > > If possible please try to avoid using a global variable like this. > > Future processors may come with multiple FSI blocks. > > Right now I'm quite happy that the CEU driver does not use global > hmm... I can agree with you. > But this time, is it good as future tasks ? > I will fix this problem if I can get free time > > arch/sh/boards/mach-se/7724/Makefile | 3 +- > > arch/sh/boards/mach-se/7724/fsi-ak464x.c | 90 ++++++++++++++++++++++++++++++ > > the machine driver itself should be under sound/soc/sh. > if fsi-ak464x.c is in arch/sh/boards/mach-se/7724/ > it works, and it say You will be able to load the driver, yes, this is a coding style sort of issue. Placing machine drivers outside of ASoC will create maintinance problems since the ASoC APIs change relatively often. > but when I tried to move it to sound/soc/sh, > then it be doesn't work. it say > ------------ > ALSA device list: > No soundcards found. > ------------ This needn't be anything to worry about - cards can be started at any time, this is just a snapshot of the cards that were ready at this point during system startup. > there is no "AK4642 Audio Codec 0.0.1" print, > so, snd_soc_codec_device :: probe isn't called. > can you give me advice ? If you could repost with the driver under sound/soc/sh it may be possible to identify where the problem lies. _______________________________________________ Alsa-devel mailing list Alsa-devel@xxxxxxxxxxxxxxxx http://mailman.alsa-project.org/mailman/listinfo/alsa-devel