diff mbox series

ASoC: atmel: fix atmel_ssc_set_audio link failure

Message ID 20191001142116.1172290-1-arnd@arndb.de (mailing list archive)
State New, archived
Headers show
Series ASoC: atmel: fix atmel_ssc_set_audio link failure | expand

Commit Message

Arnd Bergmann Oct. 1, 2019, 2:20 p.m. UTC
The ssc audio driver can call into both pdc and dma backends.  With the
latest rework, the logic to do this in a safe way avoiding link errors
was removed, bringing back link errors that were fixed long ago in commit
061981ff8cc8 ("ASoC: atmel: properly select dma driver state") such as

sound/soc/atmel/atmel_ssc_dai.o: In function `atmel_ssc_set_audio':
atmel_ssc_dai.c:(.text+0xac): undefined reference to `atmel_pcm_pdc_platform_register'

Fix it this time using Makefile hacks and a comment to prevent this
from accidentally getting removed again rather than Kconfig hacks.

Fixes: 18291410557f ("ASoC: atmel: enable SOC_SSC_PDC and SOC_SSC_DMA in Kconfig")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
 sound/soc/atmel/Kconfig  |  4 ++--
 sound/soc/atmel/Makefile | 10 ++++++++--
 2 files changed, 10 insertions(+), 4 deletions(-)

Comments

Alexandre Belloni Oct. 1, 2019, 2:27 p.m. UTC | #1
Hi,

On 01/10/2019 16:20:55+0200, Arnd Bergmann wrote:
> The ssc audio driver can call into both pdc and dma backends.  With the
> latest rework, the logic to do this in a safe way avoiding link errors
> was removed, bringing back link errors that were fixed long ago in commit
> 061981ff8cc8 ("ASoC: atmel: properly select dma driver state") such as
> 
> sound/soc/atmel/atmel_ssc_dai.o: In function `atmel_ssc_set_audio':
> atmel_ssc_dai.c:(.text+0xac): undefined reference to `atmel_pcm_pdc_platform_register'
> 
> Fix it this time using Makefile hacks and a comment to prevent this
> from accidentally getting removed again rather than Kconfig hacks.
> 
> Fixes: 18291410557f ("ASoC: atmel: enable SOC_SSC_PDC and SOC_SSC_DMA in Kconfig")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  sound/soc/atmel/Kconfig  |  4 ++--
>  sound/soc/atmel/Makefile | 10 ++++++++--
>  2 files changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/sound/soc/atmel/Kconfig b/sound/soc/atmel/Kconfig
> index f118c229ed82..25c31bf64936 100644
> --- a/sound/soc/atmel/Kconfig
> +++ b/sound/soc/atmel/Kconfig
> @@ -10,11 +10,11 @@ config SND_ATMEL_SOC
>  if SND_ATMEL_SOC
>  
>  config SND_ATMEL_SOC_PDC
> -	tristate
> +	bool
>  	depends on HAS_DMA
>  
>  config SND_ATMEL_SOC_DMA
> -	tristate
> +	bool
>  	select SND_SOC_GENERIC_DMAENGINE_PCM
>  
>  config SND_ATMEL_SOC_SSC
> diff --git a/sound/soc/atmel/Makefile b/sound/soc/atmel/Makefile
> index 1f6890ed3738..c7d2989791be 100644
> --- a/sound/soc/atmel/Makefile
> +++ b/sound/soc/atmel/Makefile
> @@ -6,8 +6,14 @@ snd-soc-atmel_ssc_dai-objs := atmel_ssc_dai.o
>  snd-soc-atmel-i2s-objs := atmel-i2s.o
>  snd-soc-mchp-i2s-mcc-objs := mchp-i2s-mcc.o
>  
> -obj-$(CONFIG_SND_ATMEL_SOC_PDC) += snd-soc-atmel-pcm-pdc.o
> -obj-$(CONFIG_SND_ATMEL_SOC_DMA) += snd-soc-atmel-pcm-dma.o
> +# pdc and dma need to both be built-in if any user of
> +# ssc is built-in.
> +ifdef CONFIG_SND_ATMEL_SOC_PDC
> +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-pdc.o
> +endif
> +ifdef CONFIG_SND_ATMEL_SOC_DMA
> +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-dma.o
> +endif

Doesn't that prevent them to be built as a module at all?
I'm not sure there is a use case though.
Michał Mirosław Oct. 1, 2019, 2:56 p.m. UTC | #2
On Tue, Oct 01, 2019 at 04:20:55PM +0200, Arnd Bergmann wrote:
> The ssc audio driver can call into both pdc and dma backends.  With the
> latest rework, the logic to do this in a safe way avoiding link errors
> was removed, bringing back link errors that were fixed long ago in commit
> 061981ff8cc8 ("ASoC: atmel: properly select dma driver state") such as
> 
> sound/soc/atmel/atmel_ssc_dai.o: In function `atmel_ssc_set_audio':
> atmel_ssc_dai.c:(.text+0xac): undefined reference to `atmel_pcm_pdc_platform_register'
> 
> Fix it this time using Makefile hacks and a comment to prevent this
> from accidentally getting removed again rather than Kconfig hacks.
> 
> Fixes: 18291410557f ("ASoC: atmel: enable SOC_SSC_PDC and SOC_SSC_DMA in Kconfig")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  sound/soc/atmel/Kconfig  |  4 ++--
>  sound/soc/atmel/Makefile | 10 ++++++++--
>  2 files changed, 10 insertions(+), 4 deletions(-)
> 
> diff --git a/sound/soc/atmel/Kconfig b/sound/soc/atmel/Kconfig
> index f118c229ed82..25c31bf64936 100644
> --- a/sound/soc/atmel/Kconfig
> +++ b/sound/soc/atmel/Kconfig
> @@ -10,11 +10,11 @@ config SND_ATMEL_SOC
>  if SND_ATMEL_SOC
>  
>  config SND_ATMEL_SOC_PDC
> -	tristate
> +	bool
>  	depends on HAS_DMA
>  
>  config SND_ATMEL_SOC_DMA
> -	tristate
> +	bool
>  	select SND_SOC_GENERIC_DMAENGINE_PCM
>  
>  config SND_ATMEL_SOC_SSC
> diff --git a/sound/soc/atmel/Makefile b/sound/soc/atmel/Makefile
> index 1f6890ed3738..c7d2989791be 100644
> --- a/sound/soc/atmel/Makefile
> +++ b/sound/soc/atmel/Makefile
> @@ -6,8 +6,14 @@ snd-soc-atmel_ssc_dai-objs := atmel_ssc_dai.o
>  snd-soc-atmel-i2s-objs := atmel-i2s.o
>  snd-soc-mchp-i2s-mcc-objs := mchp-i2s-mcc.o
>  
> -obj-$(CONFIG_SND_ATMEL_SOC_PDC) += snd-soc-atmel-pcm-pdc.o
> -obj-$(CONFIG_SND_ATMEL_SOC_DMA) += snd-soc-atmel-pcm-dma.o
> +# pdc and dma need to both be built-in if any user of
> +# ssc is built-in.
> +ifdef CONFIG_SND_ATMEL_SOC_PDC
> +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-pdc.o
> +endif
> +ifdef CONFIG_SND_ATMEL_SOC_DMA
> +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-dma.o
> +endif
>  obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel_ssc_dai.o
>  obj-$(CONFIG_SND_ATMEL_SOC_I2S) += snd-soc-atmel-i2s.o
>  obj-$(CONFIG_SND_MCHP_SOC_I2S_MCC) += snd-soc-mchp-i2s-mcc.o

I was just exploring similar solution, using $(if X,Y) instead, but your
fix will work just as well.

Reviewed-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>

Best Regards,
Michał Mirosław
Arnd Bergmann Oct. 1, 2019, 3:59 p.m. UTC | #3
On Tue, Oct 1, 2019 at 4:27 PM Alexandre Belloni
<alexandre.belloni@bootlin.com> wrote:

> > -obj-$(CONFIG_SND_ATMEL_SOC_PDC) += snd-soc-atmel-pcm-pdc.o
> > -obj-$(CONFIG_SND_ATMEL_SOC_DMA) += snd-soc-atmel-pcm-dma.o
> > +# pdc and dma need to both be built-in if any user of
> > +# ssc is built-in.
> > +ifdef CONFIG_SND_ATMEL_SOC_PDC
> > +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-pdc.o
> > +endif
> > +ifdef CONFIG_SND_ATMEL_SOC_DMA
> > +obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-dma.o
> > +endif
>
> Doesn't that prevent them to be built as a module at all?
> I'm not sure there is a use case though.

It should not: the idea was that snd-soc-atmel-pcm-{pdc,dma}.o
get built as modules if CONFIG_SND_ATMEL_SOC_SSC=m.

      arnd
Mark Brown Oct. 1, 2019, 5:55 p.m. UTC | #4
On Tue, Oct 01, 2019 at 04:20:55PM +0200, Arnd Bergmann wrote:
> The ssc audio driver can call into both pdc and dma backends.  With the
> latest rework, the logic to do this in a safe way avoiding link errors
> was removed, bringing back link errors that were fixed long ago in commit
> 061981ff8cc8 ("ASoC: atmel: properly select dma driver state") such as

This doesn't apply against current code, please check and resend.
Arnd Bergmann Oct. 1, 2019, 7:02 p.m. UTC | #5
On Tue, Oct 1, 2019 at 7:55 PM Mark Brown <broonie@kernel.org> wrote:
>
> On Tue, Oct 01, 2019 at 04:20:55PM +0200, Arnd Bergmann wrote:
> > The ssc audio driver can call into both pdc and dma backends.  With the
> > latest rework, the logic to do this in a safe way avoiding link errors
> > was removed, bringing back link errors that were fixed long ago in commit
> > 061981ff8cc8 ("ASoC: atmel: properly select dma driver state") such as
>
> This doesn't apply against current code, please check and resend.

I looked at "git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git
for-next"
as well as the for-linus branch in the same place, but found no
conflicting changes
in there compared to v5.4-rc1. Am I looking at the right tree?

       Arnd
Mark Brown Oct. 2, 2019, 11:33 a.m. UTC | #6
On Tue, Oct 01, 2019 at 09:02:22PM +0200, Arnd Bergmann wrote:
> On Tue, Oct 1, 2019 at 7:55 PM Mark Brown <broonie@kernel.org> wrote:

> > This doesn't apply against current code, please check and resend.

> I looked at "git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git
> for-next"
> as well as the for-linus branch in the same place, but found no
> conflicting changes
> in there compared to v5.4-rc1. Am I looking at the right tree?

Yes.  I'm just going by what git said here...
diff mbox series

Patch

diff --git a/sound/soc/atmel/Kconfig b/sound/soc/atmel/Kconfig
index f118c229ed82..25c31bf64936 100644
--- a/sound/soc/atmel/Kconfig
+++ b/sound/soc/atmel/Kconfig
@@ -10,11 +10,11 @@  config SND_ATMEL_SOC
 if SND_ATMEL_SOC
 
 config SND_ATMEL_SOC_PDC
-	tristate
+	bool
 	depends on HAS_DMA
 
 config SND_ATMEL_SOC_DMA
-	tristate
+	bool
 	select SND_SOC_GENERIC_DMAENGINE_PCM
 
 config SND_ATMEL_SOC_SSC
diff --git a/sound/soc/atmel/Makefile b/sound/soc/atmel/Makefile
index 1f6890ed3738..c7d2989791be 100644
--- a/sound/soc/atmel/Makefile
+++ b/sound/soc/atmel/Makefile
@@ -6,8 +6,14 @@  snd-soc-atmel_ssc_dai-objs := atmel_ssc_dai.o
 snd-soc-atmel-i2s-objs := atmel-i2s.o
 snd-soc-mchp-i2s-mcc-objs := mchp-i2s-mcc.o
 
-obj-$(CONFIG_SND_ATMEL_SOC_PDC) += snd-soc-atmel-pcm-pdc.o
-obj-$(CONFIG_SND_ATMEL_SOC_DMA) += snd-soc-atmel-pcm-dma.o
+# pdc and dma need to both be built-in if any user of
+# ssc is built-in.
+ifdef CONFIG_SND_ATMEL_SOC_PDC
+obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-pdc.o
+endif
+ifdef CONFIG_SND_ATMEL_SOC_DMA
+obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel-pcm-dma.o
+endif
 obj-$(CONFIG_SND_ATMEL_SOC_SSC) += snd-soc-atmel_ssc_dai.o
 obj-$(CONFIG_SND_ATMEL_SOC_I2S) += snd-soc-atmel-i2s.o
 obj-$(CONFIG_SND_MCHP_SOC_I2S_MCC) += snd-soc-mchp-i2s-mcc.o