Re: ESI Juli@ crash with external clock switch - patch

[Pavel Hofman <pavel.hofman@ivitera.com>]

Dne 16.1.2015 v 21:39 Takashi Iwai napsal(a):
> At Fri, 16 Jan 2015 21:36:10 +0100,
> Pavel Hofman wrote:
>>
>> Dne 16.1.2015 v 18:13 Takashi Iwai napsal(a):
>>> OK, I'm going to send a fix series including the relevant correction.
>>> Give it a try later.
>>
>> Thanks a lot for the patches, the external rate switching on Juli now
>> works perfectly (not tested on Quartet yet). One can tell you are
>> seasoned in kernel development, I would get lost in the synchronization
>> and workqueue facilities.
>>
>>>> I am not sure users would want/need to disable a feature which detects
>>>> incoming samplerate. IMO if the work thread is running only in the
>>>> external clock mode, nothing more is needed.
>>>
>>> Hm, but you can still see the other attributes of SPDIF input frames,
>>> right?  Or all these useless when the clock is set to internal?
>>> If so, it'd be easy to add the dynamic turn on/off per the clock
>>> mode.
>>
>> You are right, that would disable update of other controls informing
>> about incoming SPDIF details. These are useful in internal clock mode
>> too - if the soundcard is master for the spdif chain. A new control
>> would make sense then.
>
> Alright.
>
>> I am leaving for a week, then I will test quartet and the PM features.
>
> There should be no change regarding quartet, also about PM.
> The patches doesn't add the PM support to quartet, but rather
> robustify only the PM of Juli@ (and ak4113/4114 codec side support).
>
> In anyway, I'm going to merge them once when you confirm them
> working.


I confirm the patches are working both for Juli and Quartet. Juli tested 
also for pm-suspend, working fine after resume.

Thanks a lot,

Pavel.