Re: [alsa-devel] [PATCH RESEND 1/2] ASoC: Intel: kbl: Remove option of choosing CH count based on pdata
On Wed, Nov 15, 2017 at 03:37:42PM +0530, Naveen Manohar wrote:
Remove option of choosing channel count based on pdata for kbl_rt5663_max98927 machine driver. As User-space expects 4ch DMIC data and uses channel-map in ucm to figure-out the desired channel to use, Hence removing driver change to allow choosing channels.
To repeat my previous question which I don't seem to see a response to:
| Which userspace and are you sure there are no boards that only have two | of the mics?
On Wed, Nov 29, 2017 at 12:33:14PM +0000, Mark Brown wrote:
On Wed, Nov 15, 2017 at 03:37:42PM +0530, Naveen Manohar wrote:
Remove option of choosing channel count based on pdata for kbl_rt5663_max98927 machine driver. As User-space expects 4ch DMIC data and uses channel-map in ucm to figure-out the desired channel to use, Hence removing driver change to allow choosing channels.
To repeat my previous question which I don't seem to see a response to:
| Which userspace and are you sure there are no boards that only have two | of the mics?
I am not sure why Naveen didn't respond, I will ask around.
FWIW the userspace in question is CRAS. CRAS wants to manage MICs and would *know* how many MICs are on the board and would work with PCM data accordingly...
On Thu, Nov 30, 2017 at 09:32:08AM +0530, Vinod Koul wrote:
On Wed, Nov 29, 2017 at 12:33:14PM +0000, Mark Brown wrote:
On Wed, Nov 15, 2017 at 03:37:42PM +0530, Naveen Manohar wrote:
Remove option of choosing channel count based on pdata for kbl_rt5663_max98927 machine driver. As User-space expects 4ch DMIC data and uses channel-map in ucm to figure-out the desired channel to use, Hence removing driver change to allow choosing channels.
To repeat my previous question which I don't seem to see a response to:
| Which userspace and are you sure there are no boards that only have two | of the mics?
I am not sure why Naveen didn't respond, I will ask around.
My Bad, my apologies to Mark. I did try replying but message wasn't sent from my outbox which I'm noticing today after vinod's mail to Mark. Sorry about missing this, it's due to mutt configuration in linux that got corrupted recently in my work-mac, making my replies for patches un-sent.
Regarding the Query: As Vinod mentioned, I'm working on Chrome OS user-space stack that uses customized audio server "CRAS". These patches are verified on a board that has only 2 dmics.
Giving the control to CRAS, enables to select the desired dmic using channel-map configuration in user-space. Ex:Rear/Front mic selection for World & Front facing camera scenario
FWIW the userspace in question is CRAS. CRAS wants to manage MICs and would
*know* how many MICs are on the board and would work with PCM data accordingly...
-- ~Vinod
--
participants (3)
-
Mark Brown -
Naveen M -
Vinod Koul