September 2021 - Alsa-devel - mailman.alsa-project.org

[PATCH v3 0/2] Add reset-gpios handling for max98927
by Alejandro 04 Sep '21

04 Sep '21

The max98927 codec on some devices (i.e. Xiaomi Mi A2 Lite phone) require hardware-resetting the codec by driving a reset-gpio. This series add support for it through an optional reset-gpios property. Alejandro Tafalla (2): ASoC: max98927: Handle reset gpio when probing i2c dt-bindings: sound: max98927: Add reset-gpios optional property .../devicetree/bindings/sound/max9892x.txt | 3 +++ sound/soc/codecs/max98927.c | 14 ++++++++++++++ 2 files changed, 17 insertions(+) -- 2.33.0

5 10

[PATCH v5 0/2] Cirrus Logic CS35L41 Amplifier
by David Rhodes 04 Sep '21

04 Sep '21

ASoC driver and devicetree documentation for a new Cirrus Logic amplifier CS35L41 v5 changes: Reverse xmas-tree style for declarations At probe, regulator enable before handle pdata At probe, set pdata before component registration Remove several DT properties and implement as mixer controls Remove some unnecessary NULL inits in otp_unpack Revise license header style David Rhodes (2): ASoC: cs35l41: CS35L41 Boosted Smart Amplifier ASoC: cs35l41: Add bindings for CS35L41 .../devicetree/bindings/sound/cs35l41.yaml | 162 ++ include/sound/cs35l41.h | 35 + sound/soc/codecs/Kconfig | 12 + sound/soc/codecs/Makefile | 4 + sound/soc/codecs/cs35l41-i2c.c | 115 ++ sound/soc/codecs/cs35l41-spi.c | 144 ++ sound/soc/codecs/cs35l41-tables.c | 611 +++++++ sound/soc/codecs/cs35l41.c | 1587 +++++++++++++++++ sound/soc/codecs/cs35l41.h | 774 ++++++++ 9 files changed, 3444 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/cs35l41.yaml create mode 100644 include/sound/cs35l41.h create mode 100644 sound/soc/codecs/cs35l41-i2c.c create mode 100644 sound/soc/codecs/cs35l41-spi.c create mode 100644 sound/soc/codecs/cs35l41-tables.c create mode 100644 sound/soc/codecs/cs35l41.c create mode 100644 sound/soc/codecs/cs35l41.h -- 2.25.1

5 10

[PATCH v3] ASoC: rockchip: i2s: Fix concurrency between tx/rx
by Sugar Zhang 03 Sep '21

03 Sep '21

This patch adds lock to fix comcurrency between tx/rx to fix 'rockchip-i2s ff070000.i2s; fail to clear' Considering the situation; tx stream rx stream | | | disable enable | | reset After this patch: lock | tx stream | enable | unlock -------- --------- lock | rx stream | disable | reset | unlock Signed-off-by: Sugar Zhang <sugar.zhang(a)rock-chips.com> --- Changes in v3: - Rebase and fix conflicts. Changes in v2: - Fix compile error. sound/soc/rockchip/rockchip_i2s.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/rockchip/rockchip_i2s.c b/sound/soc/rockchip/rockchip_i2s.c index 53e0b4a..7e89f5b 100644 --- a/sound/soc/rockchip/rockchip_i2s.c +++ b/sound/soc/rockchip/rockchip_i2s.c @@ -15,6 +15,7 @@ #include <linux/clk.h> #include <linux/pm_runtime.h> #include <linux/regmap.h> +#include <linux/spinlock.h> #include <sound/pcm_params.h> #include <sound/dmaengine_pcm.h> @@ -53,6 +54,7 @@ struct rk_i2s_dev { bool is_master_mode; const struct rk_i2s_pins *pins; unsigned int bclk_ratio; + spinlock_t lock; /* tx/rx lock */ }; static int i2s_runtime_suspend(struct device *dev) @@ -96,6 +98,7 @@ static void rockchip_snd_txctrl(struct rk_i2s_dev *i2s, int on) unsigned int val = 0; int retry = 10; + spin_lock(&i2s->lock); if (on) { regmap_update_bits(i2s->regmap, I2S_DMACR, I2S_DMACR_TDE_ENABLE, I2S_DMACR_TDE_ENABLE); @@ -136,6 +139,7 @@ static void rockchip_snd_txctrl(struct rk_i2s_dev *i2s, int on) } } } + spin_unlock(&i2s->lock); } static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) @@ -143,6 +147,7 @@ static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) unsigned int val = 0; int retry = 10; + spin_lock(&i2s->lock); if (on) { regmap_update_bits(i2s->regmap, I2S_DMACR, I2S_DMACR_RDE_ENABLE, I2S_DMACR_RDE_ENABLE); @@ -183,6 +188,7 @@ static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) } } } + spin_unlock(&i2s->lock); } static int rockchip_i2s_set_fmt(struct snd_soc_dai *cpu_dai, @@ -684,6 +690,7 @@ static int rockchip_i2s_probe(struct platform_device *pdev) if (!i2s) return -ENOMEM; + spin_lock_init(&i2s->lock); i2s->dev = &pdev->dev; i2s->grf = syscon_regmap_lookup_by_phandle(node, "rockchip,grf"); -- 2.7.4

2 1

[PATCH v2] ASoC: rockchip: i2s: Fix concurrency between tx/rx
by Sugar Zhang 03 Sep '21

03 Sep '21

This patch adds lock to fix comcurrency between tx/rx to fix 'rockchip-i2s ff070000.i2s; fail to clear' Considering the situation; tx stream rx stream | | | disable enable | | reset After this patch: lock | tx stream | enable | unlock -------- --------- lock | rx stream | disable | reset | unlock Signed-off-by: Sugar Zhang <sugar.zhang(a)rock-chips.com> --- Changes in v2: - Fix compile error. sound/soc/rockchip/rockchip_i2s.c | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/sound/soc/rockchip/rockchip_i2s.c b/sound/soc/rockchip/rockchip_i2s.c index c7dc350..88e340f 100644 --- a/sound/soc/rockchip/rockchip_i2s.c +++ b/sound/soc/rockchip/rockchip_i2s.c @@ -15,6 +15,7 @@ #include <linux/clk.h> #include <linux/pm_runtime.h> #include <linux/regmap.h> +#include <linux/spinlock.h> #include <sound/pcm_params.h> #include <sound/dmaengine_pcm.h> @@ -49,6 +50,7 @@ struct rk_i2s_dev { bool rx_start; bool is_master_mode; const struct rk_i2s_pins *pins; + spinlock_t lock; /* tx/rx lock */ }; static int i2s_runtime_suspend(struct device *dev) @@ -92,6 +94,7 @@ static void rockchip_snd_txctrl(struct rk_i2s_dev *i2s, int on) unsigned int val = 0; int retry = 10; + spin_lock(&i2s->lock); if (on) { regmap_update_bits(i2s->regmap, I2S_DMACR, I2S_DMACR_TDE_ENABLE, I2S_DMACR_TDE_ENABLE); @@ -132,6 +135,7 @@ static void rockchip_snd_txctrl(struct rk_i2s_dev *i2s, int on) } } } + spin_unlock(&i2s->lock); } static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) @@ -139,6 +143,7 @@ static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) unsigned int val = 0; int retry = 10; + spin_lock(&i2s->lock); if (on) { regmap_update_bits(i2s->regmap, I2S_DMACR, I2S_DMACR_RDE_ENABLE, I2S_DMACR_RDE_ENABLE); @@ -179,6 +184,7 @@ static void rockchip_snd_rxctrl(struct rk_i2s_dev *i2s, int on) } } } + spin_unlock(&i2s->lock); } static int rockchip_i2s_set_fmt(struct snd_soc_dai *cpu_dai, @@ -589,6 +595,7 @@ static int rockchip_i2s_probe(struct platform_device *pdev) if (!i2s) return -ENOMEM; + spin_lock_init(&i2s->lock); i2s->dev = &pdev->dev; i2s->grf = syscon_regmap_lookup_by_phandle(node, "rockchip,grf"); -- 2.7.4

2 2

[PATCH v2 02/52] ASoC: dt-bindings: Convert Bluetooth SCO Link binding to a schema
by Maxime Ripard 03 Sep '21

03 Sep '21

Bluetooth SCO Link are supported by Linux with a matching device tree binding. Now that we have the DT validation in place, let's convert the device tree bindings for that driver over to a YAML schema. Cc: alsa-devel(a)alsa-project.org Cc: devicetree(a)vger.kernel.org Cc: Liam Girdwood <lgirdwood(a)gmail.com> Cc: Mark Brown <broonie(a)kernel.org> Cc: Samuel Holland <samuel(a)sholland.org> Signed-off-by: Maxime Ripard <maxime(a)cerno.tech> --- Changes from v1: - Accept either a 0 or 1 #sound-dai-cells --- .../devicetree/bindings/sound/bt-sco.txt | 13 ------- .../bindings/sound/linux,bt-sco.yaml | 38 +++++++++++++++++++ 2 files changed, 38 insertions(+), 13 deletions(-) delete mode 100644 Documentation/devicetree/bindings/sound/bt-sco.txt create mode 100644 Documentation/devicetree/bindings/sound/linux,bt-sco.yaml diff --git a/Documentation/devicetree/bindings/sound/bt-sco.txt b/Documentation/devicetree/bindings/sound/bt-sco.txt deleted file mode 100644 index 641edf75e184..000000000000 --- a/Documentation/devicetree/bindings/sound/bt-sco.txt +++ /dev/null @@ -1,13 +0,0 @@ -Bluetooth-SCO audio CODEC - -This device support generic Bluetooth SCO link. - -Required properties: - - - compatible : "delta,dfbmcs320" or "linux,bt-sco" - -Example: - -codec: bt_sco { - compatible = "delta,dfbmcs320"; -}; diff --git a/Documentation/devicetree/bindings/sound/linux,bt-sco.yaml b/Documentation/devicetree/bindings/sound/linux,bt-sco.yaml new file mode 100644 index 000000000000..e3a1f485f664 --- /dev/null +++ b/Documentation/devicetree/bindings/sound/linux,bt-sco.yaml @@ -0,0 +1,38 @@ +# SPDX-License-Identifier: GPL-2.0 +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/sound/linux,bt-sco.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Bluetooth SCO Audio Codec Device Tree Bindings + +maintainers: + - Mark Brown <broonie(a)kernel.org> + +properties: + '#sound-dai-cells': + enum: + - 0 + + # For Wideband PCM + - 1 + + compatible: + enum: + - delta,dfbmcs320 + - linux,bt-sco + +required: + - '#sound-dai-cells' + - compatible + +additionalProperties: false + +examples: + - | + codec { + #sound-dai-cells = <0>; + compatible = "linux,bt-sco"; + }; + +... -- 2.31.1

2 1

DSD Recording on RME 2-PRO FS/BE
by David Janello 03 Sep '21

03 Sep '21

Greetings, I am testing native DSD recording using arecord with the RME 2-PRO FS/BE audio interface. using command line arecord -c 2 -Dhw:CARD=Pro53791340 -r 176400 -f DSD_U32_BE audio.dsf gives output of: Recording WAVE 'audio.dsf' : Direct Stream Digital, 4-byte (x32), big endian, oldest bits in MSB, Rate 176400 Hz, Stereo arecord: set_params:1368: Sample format non available Available formats: - S32_LE Using S32_LE parameter arecord -c 2 -Dhw:CARD=Pro53791340 -r 176400 -f S32_LE audio.dsf gives output of Recording WAVE 'audio.dsf' : Signed 32 bit Little Endian, Rate 176400 Hz, Stereo Which appears to be a PCM recording and not DSD. What is the correct command line to arecord to capture DSD audio streams and write them to a dsd file? Thanks David

1 0

[PATCH] ASoC: Intel: boards: Fix CONFIG_SND_SOC_SDW_MOCKUP select
by Nathan Chancellor 03 Sep '21

03 Sep '21

When CONFIG_SND_SOC_INTEL_SOUNDWIRE_SOF_MACH is enabled without CONFIG_EXPERT, there is a Kconfig warning about unmet dependencies: WARNING: unmet direct dependencies detected for SND_SOC_SDW_MOCKUP Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && EXPERT [=n] && SOUNDWIRE [=y] Selected by [y]: - SND_SOC_INTEL_SOUNDWIRE_SOF_MACH [=y] && ... Selecting a symbol does not account for dependencies so if symbol A selects symbol B which depends on symbol C, symbol B or its select of symbol A should depend on symbol C as well. Make the CONFIG_SND_SOC_SDW_MOCKUP select in CONFIG_SND_SOC_INTEL_SOUNDWIRE_SOF_MACH depend on CONFIG_EXPERT as the help text for CONFIG_SND_SOC_SDW_MOCKUP indicates it is intended to be a development option. Fixes: 0ccac3bcf356 ("ASoC: Intel: boards: sof_sdw: add SoundWire mockup codecs for tests") Signed-off-by: Nathan Chancellor <nathan(a)kernel.org> --- An alternative here is if CONFIG_SND_SOC_INTEL_SOUNDWIRE_SOF_MACH wants to unconditionally select CONFIG_SND_SOC_SDW_MOCKUP, the "depends on EXPERT" can be moved to the prompt (tristate "..." if EXPERT). I am happy to send a new patch if that is what is desired. sound/soc/intel/boards/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/intel/boards/Kconfig b/sound/soc/intel/boards/Kconfig index 046955bf717c..8284c46d7435 100644 --- a/sound/soc/intel/boards/Kconfig +++ b/sound/soc/intel/boards/Kconfig @@ -602,7 +602,7 @@ config SND_SOC_INTEL_SOUNDWIRE_SOF_MACH select SND_SOC_DMIC select SND_SOC_INTEL_HDA_DSP_COMMON select SND_SOC_INTEL_SOF_MAXIM_COMMON - select SND_SOC_SDW_MOCKUP + select SND_SOC_SDW_MOCKUP if EXPERT help Add support for Intel SoundWire-based platforms connected to MAX98373, RT700, RT711, RT1308 and RT715 base-commit: 170c0d7460fc4aa522995ae4096b5a442f50a1fc -- 2.32.0.264.g75ae10bc75

3 9

[PATCH] ASoC: mt8195: correct the dts parsing logic about DPTX and HDMITX
by Trevor Wu 03 Sep '21

03 Sep '21

According to the description in dt-bindings, phandle assignment of HDMI TX and DP TX are not required properties, but driver regards them as required properties. In real use case, it's expected that DP TX and HDMI TX are optional features, so correct the behavior in driver. Fixes: 40d605df0a7b ("ASoC: mediatek: mt8195: add machine driver with mt6359, rt1019 and rt5682") Signed-off-by: Trevor Wu <trevor.wu(a)mediatek.com> --- .../mt8195/mt8195-mt6359-rt1019-rt5682.c | 22 +++++++++---------- 1 file changed, 10 insertions(+), 12 deletions(-) diff --git a/sound/soc/mediatek/mt8195/mt8195-mt6359-rt1019-rt5682.c b/sound/soc/mediatek/mt8195/mt8195-mt6359-rt1019-rt5682.c index 5dc217f59bd6..c97ace7387b4 100644 --- a/sound/soc/mediatek/mt8195/mt8195-mt6359-rt1019-rt5682.c +++ b/sound/soc/mediatek/mt8195/mt8195-mt6359-rt1019-rt5682.c @@ -1018,13 +1018,12 @@ static int mt8195_mt6359_rt1019_rt5682_dev_probe(struct platform_device *pdev) of_parse_phandle(pdev->dev.of_node, "mediatek,dptx-codec", 0); if (!dai_link->codecs->of_node) { - dev_err(&pdev->dev, "Property 'dptx-codec' missing or invalid\n"); - return -EINVAL; + dev_dbg(&pdev->dev, "No property 'dptx-codec'\n"); + } else { + dai_link->codecs->name = NULL; + dai_link->codecs->dai_name = "i2s-hifi"; + dai_link->init = mt8195_dptx_codec_init; } - - dai_link->codecs->name = NULL; - dai_link->codecs->dai_name = "i2s-hifi"; - dai_link->init = mt8195_dptx_codec_init; } if (strcmp(dai_link->name, "ETDM3_OUT_BE") == 0) { @@ -1032,13 +1031,12 @@ static int mt8195_mt6359_rt1019_rt5682_dev_probe(struct platform_device *pdev) of_parse_phandle(pdev->dev.of_node, "mediatek,hdmi-codec", 0); if (!dai_link->codecs->of_node) { - dev_err(&pdev->dev, "Property 'hdmi-codec' missing or invalid\n"); - return -EINVAL; + dev_dbg(&pdev->dev, "No property 'hdmi-codec'\n"); + } else { + dai_link->codecs->name = NULL; + dai_link->codecs->dai_name = "i2s-hifi"; + dai_link->init = mt8195_hdmi_codec_init; } - - dai_link->codecs->name = NULL; - dai_link->codecs->dai_name = "i2s-hifi"; - dai_link->init = mt8195_hdmi_codec_init; } } -- 2.18.0

2 1

[PATCH 0/2] Add code to manage DSP related clocks
by Daniel Baluta 03 Sep '21

03 Sep '21

From: Daniel Baluta <daniel.baluta(a)nxp.com> DSP node on the Linux kernel side must also take care of enabling DAI/DMA related clocks. By design we choose to manage DAI/DMA clocks from the kernel side because of the architecture of some i.MX8 boards. Clocks are handled by a special M4 core which runs a special firmware called SCFW (System Controler firmware). This communicates with A cores running Linux via a special Messaging Unit and implements a custom API which is already implemented by the Linux kernel i.MX clocks implementation. Note that these clocks are optional. We can use the DSP without them. First patch was already sent to review on SOF Github: https://github.com/thesofproject/linux/pull/3131 Daniel Baluta (2): ASoC: SOF: imx: Add code to manage DSP related clocks dt-bindings: dsp: fsl: Add DSP optional clocks documentation .../devicetree/bindings/dsp/fsl,dsp.yaml | 33 ++++++++ sound/soc/sof/imx/imx-common.c | 77 +++++++++++++++++++ sound/soc/sof/imx/imx-common.h | 16 ++++ sound/soc/sof/imx/imx8.c | 32 ++++++++ sound/soc/sof/imx/imx8m.c | 33 ++++++++ 5 files changed, 191 insertions(+) -- 2.27.0

4 6

Re: [PATCH 1/2] ASoC: rt5682s: Add driver for ALC5682I-VS codec
by Mark Brown 03 Sep '21

03 Sep '21

On Fri, Aug 27, 2021 at 05:59:30AM +0000, Derek [方德義] wrote: > > > +#ifdef CONFIG_COMMON_CLK > > > + /* Check if MCLK provided */ > > > + rt5682s->mclk = devm_clk_get(component->dev, "mclk"); > > > > Perhaps factor the clock init out into a _probe_clks() function and then have a > > stub function rather than the two ifdefs? > We want to factor the codes like below to reduce ifdefs. > Could you give any comments? That looks fine.

1 0