No ack() calback in component driver

No ack() calback in component driver

[Sameer Pujar]

Hi Morimoto-san,

With your commit e2cb4a14541d ("ASoC: soc-core: merge snd_pcm_ops member 
to component driver") snd_pcm_ops is replaced with callback functions in 
component driver. In doing so, ack() callback was not retained in 
component driver.

Was this intentional because there were no users of ack() or if there 
were other reasons?

Thanks,
Sameer.

Re: No ack() calback in component driver

[Mark Brown]

[-- Attachment #1: Type: text/plain, Size: 509 bytes --]

On Mon, Sep 21, 2020 at 06:32:48PM +0530, Sameer Pujar wrote:

> With your commit e2cb4a14541d ("ASoC: soc-core: merge snd_pcm_ops member to
> component driver") snd_pcm_ops is replaced with callback functions in
> component driver. In doing so, ack() callback was not retained in component
> driver.

> Was this intentional because there were no users of ack() or if there were
> other reasons?

I can't see any reason not to allow components to have an ack()
operation so I'd guess it was just lack of use.

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: No ack() calback in component driver

[Kuninori Morimoto]

Hi Sameer, Mark

> > With your commit e2cb4a14541d ("ASoC: soc-core: merge snd_pcm_ops member to
> > component driver") snd_pcm_ops is replaced with callback functions in
> > component driver. In doing so, ack() callback was not retained in component
> > driver.
> 
> > Was this intentional because there were no users of ack() or if there were
> > other reasons?
> 
> I can't see any reason not to allow components to have an ack()
> operation so I'd guess it was just lack of use.

I think I replaced existing callback only.
Thus, I think I did nothing to it if there was no users.

Thank you for your help !!

Best regards
---
Kuninori Morimoto

Re: No ack() calback in component driver

[Sameer Pujar]

Hi Morimoto-san,

>>> With your commit e2cb4a14541d ("ASoC: soc-core: merge snd_pcm_ops member to
>>> component driver") snd_pcm_ops is replaced with callback functions in
>>> component driver. In doing so, ack() callback was not retained in component
>>> driver.
>>> Was this intentional because there were no users of ack() or if there were
>>> other reasons?
>> I can't see any reason not to allow components to have an ack()
>> operation so I'd guess it was just lack of use.
> I think I replaced existing callback only.
> Thus, I think I did nothing to it if there was no users.

I was infact referring to [0] in which you added callbacks directly in 
component driver. Later in the same series [1] 'snd_pcm_ops' was removed 
from 'snd_soc_component_driver'. It looks like at that point of time 
'snd_pcm_ops' had ack() callback. The reason for asking this is, for a 
driver (not yet in upstream) some implementation depended on this 
callback() and while moving to component it was noticed that ack() is 
not present in component. I guess without an user it is not recommended 
to add this callback. In that case I will push it along with the driver 
once it is ready.

[0] Link: 
https://lore.kernel.org/r/87tv8raf3r.wl-kuninori.morimoto.gx@renesas.com
[1] 
https://lore.kernel.org/alsa-devel/8736gb90by.wl-kuninori.morimoto.gx@renesas.com


Thanks,
Sameer.