[alsa-devel] [PATCH 1/5] ASoC: atmel: enable SSC_PCM_DMA in Kconfig

[Alexandre Belloni]

On 24/07/2019 01:25:05+0200, mirq-linux at rere.qmqm.pl wrote:
> On Tue, Jul 23, 2019 at 08:39:15PM +0200, Alexandre Belloni wrote:
> > On 23/07/2019 18:43:12+0200, mirq-linux at rere.qmqm.pl wrote:
> > > On Tue, Jul 23, 2019 at 01:36:37PM +0000, Codrin.Ciubotariu at microchip.com wrote:
> > > > On 22.07.2019 21:27, Michał Mirosław wrote:
> > > > > Allow SSC to be used on platforms described using audio-graph-card
> > > > > in Device Tree.
> > > > > 
> > > > > Signed-off-by: Michał Mirosław <mirq-linux at rere.qmqm.pl>
> > > > > ---
> > > > >   sound/soc/atmel/Kconfig | 2 +-
> > > > >   1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/sound/soc/atmel/Kconfig b/sound/soc/atmel/Kconfig
> > > > > index 06c1d5ce642c..9ef9d25bb517 100644
> > > > > --- a/sound/soc/atmel/Kconfig
> > > > > +++ b/sound/soc/atmel/Kconfig
> > > > > @@ -25,7 +25,7 @@ config SND_ATMEL_SOC_DMA
> > > > >   	default y if SND_ATMEL_SOC_SSC_DMA=y || (SND_ATMEL_SOC_SSC_DMA=m && SND_ATMEL_SOC_SSC=y)
> > > > >   
> > > > >   config SND_ATMEL_SOC_SSC_DMA
> > > > > -	tristate
> > > > > +	tristate "SoC PCM DAI support for AT91 SSC controller using DMA"
> > > > 
> > > > Could you please make something similar for SND_ATMEL_SOC_SSC_PDC? Also, 
> > > > I think that it should select ATMEL_SSC, to be able to use 
> > > > simple/graph-card with SSC.
> > > 
> > > Hmm. The Kconfig dependencies seems overly complicated, do you mind if I
> > > get rid of most of the entries in the process?
> > > 
> > 
> > Unfortunately, it is just complicated enough. This is done to support
> > all the possible configurations. Removing them will lead to linking
> > errors.
> > 
> > After having that discussion back in March, I had a very quick look and
> > didn't send a patch because I still had linking issues. It is not
> > impossible but it required more time than I had.
> 
> Can you try patch below if it covers the configurations you mention?
> This uses Kconfig's m/y resolution instead of open-coded defaults.
> 

Seems good to me, thanks.


-- 
Alexandre Belloni, Bootlin
Embedded Linux and Kernel engineering
https://bootlin.com