[PATCH 0/3] arm64: dts: meson-axg: add pdm support

[PATCH 0/3] arm64: dts: meson-axg: add pdm support

[Jerome Brunet]

This patchset adds pdm support for amlogic AXG SoC.
It applies on top of the DT clean-up [0] I sent earlier

The related ASoC PDM patches can be found here [1]. While the ASoC patches
[2] are obviously needed to for the PDM to work, this series can be safely
applied without it.

[0]: https://lkml.kernel.org/r/20180829154551.26729-1-jbrunet@baylibre.com
[1]: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git/log/?h=for-4.20
[2]: https://lkml.kernel.org/r/20180829150051.18190-1-jbrunet@baylibre.com

Jerome Brunet (3):
  arm64: dts: meson-axg: add pdm
  arm64: dts: meson-axg: s400: add dmic codec
  arm64: dts: meson-axg: s400: add pdm to the sound card

 .../arm64/boot/dts/amlogic/meson-axg-s400.dts | 27 +++++++++++++++++++
 arch/arm64/boot/dts/amlogic/meson-axg.dtsi    | 12 +++++++++
 2 files changed, 39 insertions(+)

-- 
2.17.1

[PATCH 1/3] arm64: dts: meson-axg: add pdm

[Jerome Brunet]

Add the PDM device of the axg audio subsystem

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
---
 arch/arm64/boot/dts/amlogic/meson-axg.dtsi | 12 ++++++++++++
 1 file changed, 12 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
index 36be63d69e7f..883ddfce2374 100644
--- a/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
+++ b/arch/arm64/boot/dts/amlogic/meson-axg.dtsi
@@ -152,6 +152,18 @@
 			status = "disabled";
 		};
 
+		pdm: audio-controller@ff632000 {
+			compatible = "amlogic,axg-pdm";
+			reg = <0x0 0xff632000 0x0 0x34>;
+			#sound-dai-cells = <0>;
+			sound-name-prefix = "PDM";
+			clocks = <&clkc_audio AUD_CLKID_PDM>,
+				 <&clkc_audio AUD_CLKID_PDM_DCLK>,
+				 <&clkc_audio AUD_CLKID_PDM_SYSCLK>;
+			clock-names = "pclk", "dclk", "sysclk";
+			status = "disabled";
+		};
+
 		periphs: bus@ff634000 {
 			compatible = "simple-bus";
 			reg = <0x0 0xff634000 0x0 0x2000>;
-- 
2.17.1

[PATCH 2/3] arm64: dts: meson-axg: s400: add dmic codec

[Jerome Brunet]

There are 7 digital mics on the MIC daughter board attached
to the s400 board, so add the digital microphone codec to
its DTS

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
---
 arch/arm64/boot/dts/amlogic/meson-axg-s400.dts | 9 +++++++++
 1 file changed, 9 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
index ff64c429d432..f3e16cbbc61e 100644
--- a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
+++ b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
@@ -86,6 +86,15 @@
 		sound-name-prefix = "DIT";
 	};
 
+	dmics: audio-codec@3 {
+		#sound-dai-cells = <0>;
+		compatible = "dmic-codec";
+		num-channels = <7>;
+		wakeup-delay-ms = <50>;
+		status = "okay";
+		sound-name-prefix = "MIC";
+	};
+
 	emmc_pwrseq: emmc-pwrseq {
 		compatible = "mmc-pwrseq-emmc";
 		reset-gpios = <&gpio BOOT_9 GPIO_ACTIVE_LOW>;
-- 
2.17.1

Re: [PATCH 2/3] arm64: dts: meson-axg: s400: add dmic codec

[Fabio Estevam]

Hi Jerome,

On Fri, Aug 31, 2018 at 12:02 PM, Jerome Brunet <jbrunet@baylibre.com> wrote:
> There are 7 digital mics on the MIC daughter board attached
> to the s400 board, so add the digital microphone codec to
> its DTS
>
> Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
> ---
>  arch/arm64/boot/dts/amlogic/meson-axg-s400.dts | 9 +++++++++
>  1 file changed, 9 insertions(+)
>
> diff --git a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> index ff64c429d432..f3e16cbbc61e 100644
> --- a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> +++ b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> @@ -86,6 +86,15 @@
>                 sound-name-prefix = "DIT";
>         };
>
> +       dmics: audio-codec@3 {

You pass @3 without a corresponding reg = <3>, which causes dtc
warnings with W=1.

Re: [PATCH 2/3] arm64: dts: meson-axg: s400: add dmic codec

[Jerome Brunet]

On Sat, 2018-09-01 at 18:24 -0300, Fabio Estevam wrote:
> Hi Jerome,
> 
> On Fri, Aug 31, 2018 at 12:02 PM, Jerome Brunet <jbrunet@baylibre.com> wrote:
> > There are 7 digital mics on the MIC daughter board attached
> > to the s400 board, so add the digital microphone codec to
> > its DTS
> > 
> > Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
> > ---
> >  arch/arm64/boot/dts/amlogic/meson-axg-s400.dts | 9 +++++++++
> >  1 file changed, 9 insertions(+)
> > 
> > diff --git a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> > index ff64c429d432..f3e16cbbc61e 100644
> > --- a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> > +++ b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
> > @@ -86,6 +86,15 @@
> >                 sound-name-prefix = "DIT";
> >         };
> > 
> > +       dmics: audio-codec@3 {
> 
> You pass @3 without a corresponding reg = <3>, which causes dtc
> warnings with W=1.

Hi Fabio,

Thanks for pointing this out.
Martin also raised concerns about this but, unfortunately, we did not get to the
bottom of it:

https://lkml.kernel.org/r/
90780f5a6b974ee9a4f9f9f2580fc83ef6bdd805.camel@baylibre.com

In a nutshell, I numbered nodes (w/o a reg prop) to keep generic names as
suggested in the DT spec.

The idea came after working on sound cards, if you check
Documentation/devicetree/bindings/sound/simple-card.txt, you'll see this kind of
numbering is even in documentation examples.

This is used by arch/arm/boot/dts/armada-370-db.dts. This is just an example I
picked randomly, I'm not trying to single out the marvell folks here (or ASoC
simple card). I'm actually doing exactly the same kind of multi link stuff on
the amlogic cards.

This looks to be a wide spread practice. Running `make dtbs W=1` on ARM and
ARM64 arch, I've seen dozen of 'node has a unit name, but no reg property'
examples, on things like clocks, buttons, leds, pinctrl, etc ...

that being said, section 2.2.1 of the DT specs says:

" If the node has no reg property, the @unit-address must be omitted and the
node-name alone differentiates the node from other nodes at the same level in
the tree "

So apparently, this is used but not ok ?
Maybe we should simply replace the '@' with a '-' everywhere this occurs ?

If possible, I'd like to get the opinion of the DT gurus about this issue.
Cheers

Jerome

[PATCH 3/3] arm64: dts: meson-axg: s400: add pdm to the sound card

[Jerome Brunet]

Enable the PDM input device on the S400 and it to the sound card

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
---
 arch/arm64/boot/dts/amlogic/meson-axg-s400.dts | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

diff --git a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
index f3e16cbbc61e..92d3064d4966 100644
--- a/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
+++ b/arch/arm64/boot/dts/amlogic/meson-axg-s400.dts
@@ -240,6 +240,9 @@
 				"TODDR_A IN 2", "TDMIN_C OUT",
 				"TODDR_B IN 2", "TDMIN_C OUT",
 				"TODDR_C IN 2", "TDMIN_C OUT",
+				"TODDR_A IN 4", "PDM Capture",
+				"TODDR_B IN 4", "PDM Capture",
+				"TODDR_C IN 4", "PDM Capture",
 				"TODDR_A IN 6", "TDMIN_LB OUT",
 				"TODDR_B IN 6", "TDMIN_LB OUT",
 				"TODDR_C IN 6", "TDMIN_LB OUT",
@@ -312,6 +315,14 @@
 				sound-dai = <&spdif_dit>;
 			};
 		};
+
+		dai-link@8 {
+			sound-dai = <&pdm>;
+
+			codec {
+				sound-dai = <&dmics>;
+			};
+		};
 	};
 
 	wifi32k: wifi32k {
@@ -394,6 +405,13 @@
 	};
 };
 
+&pdm {
+	pinctrl-0 = <&pdm_dclk_a14_pins>, <&pdm_din0_pins>,
+		    <&pdm_din1_pins>, <&pdm_din2_pins>, <&pdm_din3_pins>;
+	pinctrl-names = "default";
+	status = "okay";
+};
+
 &pwm_ab {
 	status = "okay";
 	pinctrl-0 = <&pwm_a_x20_pins>;
-- 
2.17.1

Re: [PATCH 0/3] arm64: dts: meson-axg: add pdm support

[Kevin Hilman]

Jerome Brunet <jbrunet@baylibre.com> writes:

> This patchset adds pdm support for amlogic AXG SoC.
> It applies on top of the DT clean-up [0] I sent earlier

Applied to v4.20/dt64.

> The related ASoC PDM patches can be found here [1]. While the ASoC patches
> [2] are obviously needed to for the PDM to work, this series can be safely
> applied without it.

Thanks for the clear explanation of dependencies.

Kevin