[PATCH v3 0/7] PinePhone BT audio bringup
This series uses the additional DAIs added to the sun8i-codec driver to add hardware routing for BT SCO (headset) audio on the PinePhone.
The BT audio connection is represented by the "dummy" bt-sco codec. The connection to the Quectel EG-25G modem via AIF2 works as well, but I do not include it here because there is no appropriate codec driver in tree. We have been using an out-of-tree "dummy" codec driver for the modem similar to bt-sco, and I'm not sure if such a driver would be desired upstream.
Changes from v2: - Also accept #sound-dai-cells in the binding. Since dt-core.yaml already sets the type of this property, it is not possible to use oneOf, nor make a specific value deprecated.
Changes from v1: - Fixed DT binding example to follow new binding
Arnaud Ferraris (1): arm64: dts: allwinner: pinephone: Set audio card name
Samuel Holland (6): ASoC: dt-bindings: sun8i-codec: Increase #sound-dai-cells ARM: dts: sun8i-a33: Allow using multiple codec DAIs arm64: dts: allwinner: a64: Allow using multiple codec DAIs arm64: dts: allwinner: a64: Add pinmux nodes for AIF2/AIF3 arm64: dts: allwinner: a64: Allow multiple DAI links arm64: dts: allwinner: pinephone: Add support for Bluetooth audio
.../sound/allwinner,sun8i-a33-codec.yaml | 8 +++- arch/arm/boot/dts/sun8i-a33.dtsi | 4 +- .../dts/allwinner/sun50i-a64-pinephone.dtsi | 25 +++++++++++++ arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 37 ++++++++++++++----- 4 files changed, 60 insertions(+), 14 deletions(-)
Increase sound-dai-cells to 1 to allow using the DAIs in the codec corresponding to AIF2 and AIF3.
The generic ASoC OF code supports a #sound-dai-cells value of 0 or 1 with no impact to the driver, so this is a backward-compatible change.
Signed-off-by: Samuel Holland samuel@sholland.org --- .../bindings/sound/allwinner,sun8i-a33-codec.yaml | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/Documentation/devicetree/bindings/sound/allwinner,sun8i-a33-codec.yaml b/Documentation/devicetree/bindings/sound/allwinner,sun8i-a33-codec.yaml index 67405e6d8168..19f111f40225 100644 --- a/Documentation/devicetree/bindings/sound/allwinner,sun8i-a33-codec.yaml +++ b/Documentation/devicetree/bindings/sound/allwinner,sun8i-a33-codec.yaml @@ -12,7 +12,11 @@ maintainers:
properties: "#sound-dai-cells": - const: 0 + minimum: 0 + maximum: 1 + description: + A value of 0 is deprecated. When used, it only allows access to + the ADC/DAC and AIF1 (the CPU DAI), not the other two AIFs/DAIs.
compatible: oneOf: @@ -50,7 +54,7 @@ additionalProperties: false examples: - | audio-codec@1c22e00 { - #sound-dai-cells = <0>; + #sound-dai-cells = <1>; compatible = "allwinner,sun8i-a33-codec"; reg = <0x01c22e00 0x400>; interrupts = <0 29 4>;
On Thu, 29 Apr 2021 22:58:53 -0500, Samuel Holland wrote:
Increase sound-dai-cells to 1 to allow using the DAIs in the codec corresponding to AIF2 and AIF3.
The generic ASoC OF code supports a #sound-dai-cells value of 0 or 1 with no impact to the driver, so this is a backward-compatible change.
Signed-off-by: Samuel Holland samuel@sholland.org
.../bindings/sound/allwinner,sun8i-a33-codec.yaml | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
Acked-by: Rob Herring robh@kernel.org
Increase #sound-dai-cells on the digital codec to allow using the other DAIs provided by the codec for AIF2 and AIF3.
Signed-off-by: Samuel Holland samuel@sholland.org --- arch/arm/boot/dts/sun8i-a33.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm/boot/dts/sun8i-a33.dtsi b/arch/arm/boot/dts/sun8i-a33.dtsi index 7344c37107c6..2beddbb3c518 100644 --- a/arch/arm/boot/dts/sun8i-a33.dtsi +++ b/arch/arm/boot/dts/sun8i-a33.dtsi @@ -198,7 +198,7 @@ simple-audio-card,cpu { };
link_codec: simple-audio-card,codec { - sound-dai = <&codec>; + sound-dai = <&codec 0>; }; };
@@ -238,7 +238,7 @@ dai: dai@1c22c00 { };
codec: codec@1c22e00 { - #sound-dai-cells = <0>; + #sound-dai-cells = <1>; compatible = "allwinner,sun8i-a33-codec"; reg = <0x01c22e00 0x400>; interrupts = <GIC_SPI 29 IRQ_TYPE_LEVEL_HIGH>;
Increase #sound-dai-cells on the digital codec to allow using the other DAIs provided by the codec for AIF2 and AIF3.
Signed-off-by: Samuel Holland samuel@sholland.org --- arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi index 57786fc120c3..3061031780de 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi @@ -150,7 +150,7 @@ cpudai: simple-audio-card,cpu { };
link_codec: simple-audio-card,codec { - sound-dai = <&codec>; + sound-dai = <&codec 0>; }; };
@@ -878,7 +878,7 @@ dai: dai@1c22c00 { };
codec: codec@1c22e00 { - #sound-dai-cells = <0>; + #sound-dai-cells = <1>; compatible = "allwinner,sun50i-a64-codec", "allwinner,sun8i-a33-codec"; reg = <0x01c22e00 0x600>;
Now that the sun8i-codec driver supports AIF2 and AIF3, boards can use them in DAI links. Add the necessary pinmux nodes.
Signed-off-by: Samuel Holland samuel@sholland.org --- arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 12 ++++++++++++ 1 file changed, 12 insertions(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi index 3061031780de..a583355a3128 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi @@ -658,6 +658,18 @@ pio: pinctrl@1c20800 { interrupt-controller; #interrupt-cells = <3>;
+ /omit-if-no-ref/ + aif2_pins: aif2-pins { + pins = "PB4", "PB5", "PB6", "PB7"; + function = "aif2"; + }; + + /omit-if-no-ref/ + aif3_pins: aif3-pins { + pins = "PG10", "PG11", "PG12", "PG13"; + function = "aif3"; + }; + csi_pins: csi-pins { pins = "PE0", "PE2", "PE3", "PE4", "PE5", "PE6", "PE7", "PE8", "PE9", "PE10", "PE11";
simple-audio-card supports either a single DAI link at the top level, or subnodes with one or more DAI links. To use the secondary AIFs on the codec, we need to add additional DAI links to the same sound card, so we need to use the other binding.
Signed-off-by: Samuel Holland samuel@sholland.org --- arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 23 +++++++++++-------- 1 file changed, 14 insertions(+), 9 deletions(-)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi index a583355a3128..5734b4825f64 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi @@ -131,12 +131,10 @@ psci { };
sound: sound { + #address-cells = <1>; + #size-cells = <0>; compatible = "simple-audio-card"; simple-audio-card,name = "sun50i-a64-audio"; - simple-audio-card,format = "i2s"; - simple-audio-card,frame-master = <&cpudai>; - simple-audio-card,bitclock-master = <&cpudai>; - simple-audio-card,mclk-fs = <128>; simple-audio-card,aux-devs = <&codec_analog>; simple-audio-card,routing = "Left DAC", "DACL", @@ -145,12 +143,19 @@ sound: sound { "ADCR", "Right ADC"; status = "disabled";
- cpudai: simple-audio-card,cpu { - sound-dai = <&dai>; - }; + simple-audio-card,dai-link@0 { + format = "i2s"; + frame-master = <&link0_cpu>; + bitclock-master = <&link0_cpu>; + mclk-fs = <128>;
- link_codec: simple-audio-card,codec { - sound-dai = <&codec 0>; + link0_cpu: cpu { + sound-dai = <&dai>; + }; + + link0_codec: codec { + sound-dai = <&codec 0>; + }; }; };
The PinePhone has a Bluetooth chip with its PCM interface connected to AIF3. Add the DAI link so headeset audio can be routed in hardware.
Even though the link is 16 bit PCM, configuring the link a 32-bit slot is required for compatibility with AIF2, which also uses a 32-bit slot, and which shares clock dividers with AIF3. Using equal clock frequencies allows the modem and headset to be used at the same time.
Signed-off-by: Samuel Holland samuel@sholland.org --- .../dts/allwinner/sun50i-a64-pinephone.dtsi | 24 +++++++++++++++++++ 1 file changed, 24 insertions(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi index 9f69d489a81d..51cbfdc12936 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi @@ -25,6 +25,11 @@ backlight: backlight { /* Backlight configuration differs per PinePhone revision. */ };
+ bt_sco_codec: bt-sco-codec { + #sound-dai-cells = <1>; + compatible = "linux,bt-sco"; + }; + chosen { stdout-path = "serial0:115200n8"; }; @@ -91,6 +96,8 @@ vibrator { };
&codec { + pinctrl-names = "default"; + pinctrl-0 = <&aif3_pins>; status = "okay"; };
@@ -447,6 +454,23 @@ &sound { "MIC1", "Internal Microphone", "Headset Microphone", "HBIAS", "MIC2", "Headset Microphone"; + + simple-audio-card,dai-link@2 { + format = "dsp_a"; + frame-master = <&link2_codec>; + bitclock-master = <&link2_codec>; + bitclock-inversion; + + link2_cpu: cpu { + sound-dai = <&bt_sco_codec 0>; + }; + + link2_codec: codec { + sound-dai = <&codec 2>; + dai-tdm-slot-num = <1>; + dai-tdm-slot-width = <32>; + }; + }; };
&uart0 {
From: Arnaud Ferraris arnaud.ferraris@collabora.com
Add the "PinePhone" name to the sound card: this will make upstreaming an ALSA UCM config easier as we can use a unique name.
It also avoids an issue where the default card name is truncated.
Signed-off-by: Arnaud Ferraris arnaud.ferraris@collabora.com [Samuel: Split out change, updated commit message] Signed-off-by: Samuel Holland samuel@sholland.org --- arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi | 1 + 1 file changed, 1 insertion(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi index 51cbfdc12936..02712f85f6bd 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi @@ -433,6 +433,7 @@ ®_rtc_ldo {
&sound { status = "okay"; + simple-audio-card,name = "PinePhone"; simple-audio-card,aux-devs = <&codec_analog>, <&speaker_amp>; simple-audio-card,widgets = "Microphone", "Headset Microphone", "Microphone", "Internal Microphone",
Hi,
On Thu, Apr 29, 2021 at 10:58:59PM -0500, Samuel Holland wrote:
From: Arnaud Ferraris arnaud.ferraris@collabora.com
Add the "PinePhone" name to the sound card: this will make upstreaming an ALSA UCM config easier as we can use a unique name.
It also avoids an issue where the default card name is truncated.
Signed-off-by: Arnaud Ferraris arnaud.ferraris@collabora.com [Samuel: Split out change, updated commit message] Signed-off-by: Samuel Holland samuel@sholland.org
arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi | 1 + 1 file changed, 1 insertion(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi index 51cbfdc12936..02712f85f6bd 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi @@ -433,6 +433,7 @@ ®_rtc_ldo {
&sound { status = "okay";
- simple-audio-card,name = "PinePhone"; simple-audio-card,aux-devs = <&codec_analog>, <&speaker_amp>; simple-audio-card,widgets = "Microphone", "Headset Microphone", "Microphone", "Internal Microphone",
Isn't that reported to the userspace? I'm not sure we can just change it without breaking it.
Maxime
On 5/7/21 3:09 AM, Maxime Ripard wrote:
Hi,
On Thu, Apr 29, 2021 at 10:58:59PM -0500, Samuel Holland wrote:
From: Arnaud Ferraris arnaud.ferraris@collabora.com
Add the "PinePhone" name to the sound card: this will make upstreaming an ALSA UCM config easier as we can use a unique name.
It also avoids an issue where the default card name is truncated.
Signed-off-by: Arnaud Ferraris arnaud.ferraris@collabora.com [Samuel: Split out change, updated commit message] Signed-off-by: Samuel Holland samuel@sholland.org
arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi | 1 + 1 file changed, 1 insertion(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi index 51cbfdc12936..02712f85f6bd 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi @@ -433,6 +433,7 @@ ®_rtc_ldo {
&sound { status = "okay";
- simple-audio-card,name = "PinePhone"; simple-audio-card,aux-devs = <&codec_analog>, <&speaker_amp>; simple-audio-card,widgets = "Microphone", "Headset Microphone", "Microphone", "Internal Microphone",
Isn't that reported to the userspace? I'm not sure we can just change it without breaking it.
Yes, this is seen by userspace.
Ideally, we would add this property with a unique value when enabling audio for any board, but as you mention, it would be a breaking change to add those properties now.
In practice, all distributions supporting the PinePhone are already carrying this patch, because the ALSA setup needed for the PinePhone is very different from most other boards. So it would be similarly breaking for them (and their users) to drop this patch.
Regards, Samuel
On Sun, May 09, 2021 at 08:54:36PM -0500, Samuel Holland wrote:
On 5/7/21 3:09 AM, Maxime Ripard wrote:
Hi,
On Thu, Apr 29, 2021 at 10:58:59PM -0500, Samuel Holland wrote:
From: Arnaud Ferraris arnaud.ferraris@collabora.com
Add the "PinePhone" name to the sound card: this will make upstreaming an ALSA UCM config easier as we can use a unique name.
It also avoids an issue where the default card name is truncated.
Signed-off-by: Arnaud Ferraris arnaud.ferraris@collabora.com [Samuel: Split out change, updated commit message] Signed-off-by: Samuel Holland samuel@sholland.org
arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi | 1 + 1 file changed, 1 insertion(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi index 51cbfdc12936..02712f85f6bd 100644 --- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi
g> >> +++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pinephone.dtsi
@@ -433,6 +433,7 @@ ®_rtc_ldo {
&sound { status = "okay";
- simple-audio-card,name = "PinePhone"; simple-audio-card,aux-devs = <&codec_analog>, <&speaker_amp>; simple-audio-card,widgets = "Microphone", "Headset Microphone", "Microphone", "Internal Microphone",
Isn't that reported to the userspace? I'm not sure we can just change it without breaking it.
Yes, this is seen by userspace.
Ideally, we would add this property with a unique value when enabling audio for any board, but as you mention, it would be a breaking change to add those properties now.
In practice, all distributions supporting the PinePhone are already carrying this patch, because the ALSA setup needed for the PinePhone is very different from most other boards. So it would be similarly breaking for them (and their users) to drop this patch.
Ack :)
Since no one else complained, I merged this patch too
Thanks! Maxime
On Thu, Apr 29, 2021 at 10:58:52PM -0500, Samuel Holland wrote:
This series uses the additional DAIs added to the sun8i-codec driver to add hardware routing for BT SCO (headset) audio on the PinePhone.
The BT audio connection is represented by the "dummy" bt-sco codec. The connection to the Quectel EG-25G modem via AIF2 works as well, but I do not include it here because there is no appropriate codec driver in tree. We have been using an out-of-tree "dummy" codec driver for the modem similar to bt-sco, and I'm not sure if such a driver would be desired upstream.
I've applied patches 1-6
Maxime
On Thu, 29 Apr 2021 22:58:52 -0500, Samuel Holland wrote:
This series uses the additional DAIs added to the sun8i-codec driver to add hardware routing for BT SCO (headset) audio on the PinePhone.
The BT audio connection is represented by the "dummy" bt-sco codec. The connection to the Quectel EG-25G modem via AIF2 works as well, but I do not include it here because there is no appropriate codec driver in tree. We have been using an out-of-tree "dummy" codec driver for the modem similar to bt-sco, and I'm not sure if such a driver would be desired upstream.
[...]
Applied to
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git for-next
Thanks!
[1/7] ASoC: dt-bindings: sun8i-codec: Increase #sound-dai-cells commit: 880e007f15a31f446b9e1713720c6ae5a539f3f4
All being well this means that it will be integrated into the linux-next tree (usually sometime in the next 24 hours) and sent to Linus during the next merge window (or sooner if it is a bug fix), however if problems are discovered then the patch may be dropped or reverted.
You may get further e-mails resulting from automated or manual testing and review of the tree, please engage with people reporting problems and send followup patches addressing any issues that are reported if needed.
If any updates are required or you are submitting further changes they should be sent as incremental updates against current git, existing patches will not be replaced.
Please add any relevant lists and maintainers to the CCs when replying to this mail.
Thanks, Mark
participants (4)
-
Mark Brown -
Maxime Ripard -
Rob Herring -
Samuel Holland