Re: [PATCH] pcm: support for audio wallclock

[Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>]

> You need to provide a function to indicate whether htstamp_wallclk is
> available or not.  Since the function returning tstamp_wallclk is a
> void function and it itself can't return the error, user can't know
> whether it's a bogus value or not.

Yes, this was planned but I wasn't sure how to go about it. These types 
of routines usually rely on hw_params, I guess I could query them and 
check the .info field. Not sure if there's another way.
Also maybe I should change the return type to provide -EINVAL is somehow 
the audio timestamp is not available for a specific device (digital 
input as Clemens mentioned)


> And that function should check whether the running PCM protocol
> version is high enough to support the new ABI.
Yep


\
>> +#ifndef DOXYGEN
>> +void INTERNAL(snd_pcm_status_get_htstamp_wallclk)(const snd_pcm_status_t *obj, snd_htimestamp_t *ptr)
>> +#else
>> +void snd_pcm_status_get_htstamp_wallclk(const snd_pcm_status_t *obj, snd_htimestamp_t *ptr)
>> +#endif
>
> I think you can avoid this hack for the new functions.

Ok, I have no idea what I am doing here, just copy-pasted to make it work...