March 2018 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH v2 0/2] ASoC: Add support for DAC PCM1789
by Mylène Josserand 08 Mar '18

08 Mar '18

Hello everyone, The current series is the second version to add the support of Texas Instrument's DAC PCM1789. This DAC is very minimalist and does not have many registers. It is important to notice that this DAC needs to always have clocks enabled (even without any data) otherwise it will be in a "desynchronized" state and can not send data correctly. This issue has been solved by performing a reset each time a sound is played. This reset can produce a "pop" noise. Depending on your DAI, you will need to provide and enable the MCLK to be able to communicate with this codec throught i2c. Changes since v1: - Create a new file to support pcm1789 instead of converting the existing pcm179x driver. All the patches are merged into one patch. - Update the code to use gpiod for the reset. - Add some fixes according to Thomas Petazzoni's reviews - Create a new patch to add device-tree bindings for this new DAC. Thank you in advance for any review. Best regards, Mylène Mylène Josserand (2): ASoC: codecs: Add support for PCM1789 ASoC: Add bindings for PCM1789 .../devicetree/bindings/sound/pcm1789.txt | 21 ++ sound/soc/codecs/Kconfig | 12 + sound/soc/codecs/Makefile | 4 + sound/soc/codecs/pcm1789-i2c.c | 76 ++++++ sound/soc/codecs/pcm1789.c | 288 +++++++++++++++++++++ sound/soc/codecs/pcm1789.h | 28 ++ 6 files changed, 429 insertions(+) create mode 100644 Documentation/devicetree/bindings/sound/pcm1789.txt create mode 100644 sound/soc/codecs/pcm1789-i2c.c create mode 100644 sound/soc/codecs/pcm1789.c create mode 100644 sound/soc/codecs/pcm1789.h -- 2.11.0

4 8

[alsa-devel] [PATCH 1/2] dell-led: Change dell-led.h to dell-common.h
by Kai-Heng Feng 08 Mar '18

08 Mar '18

This header will be used for more than just led. Change it to a more generic name. Signed-off-by: Kai-Heng Feng <kai.heng.feng(a)canonical.com> --- drivers/platform/x86/dell-laptop.c | 2 +- include/linux/{dell-led.h => dell-common.h} | 4 ++-- sound/pci/hda/dell_wmi_helper.c | 2 +- 3 files changed, 4 insertions(+), 4 deletions(-) rename include/linux/{dell-led.h => dell-common.h} (61%) diff --git a/drivers/platform/x86/dell-laptop.c b/drivers/platform/x86/dell-laptop.c index c52c6723374b..8ba820e6c3d0 100644 --- a/drivers/platform/x86/dell-laptop.c +++ b/drivers/platform/x86/dell-laptop.c @@ -29,7 +29,7 @@ #include <linux/mm.h> #include <linux/i8042.h> #include <linux/debugfs.h> -#include <linux/dell-led.h> +#include <linux/dell-common.h> #include <linux/seq_file.h> #include <acpi/video.h> #include "dell-rbtn.h" diff --git a/include/linux/dell-led.h b/include/linux/dell-common.h similarity index 61% rename from include/linux/dell-led.h rename to include/linux/dell-common.h index 92521471517f..37e4b614dd74 100644 --- a/include/linux/dell-led.h +++ b/include/linux/dell-common.h @@ -1,6 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ -#ifndef __DELL_LED_H__ -#define __DELL_LED_H__ +#ifndef __DELL_COMMON_H__ +#define __DELL_COMMON_H__ int dell_micmute_led_set(int on); diff --git a/sound/pci/hda/dell_wmi_helper.c b/sound/pci/hda/dell_wmi_helper.c index 1b48a8c19d28..56050cc3c0ee 100644 --- a/sound/pci/hda/dell_wmi_helper.c +++ b/sound/pci/hda/dell_wmi_helper.c @@ -4,7 +4,7 @@ */ #if IS_ENABLED(CONFIG_DELL_LAPTOP) -#include <linux/dell-led.h> +#include <linux/dell-common.h> enum { MICMUTE_LED_ON, -- 2.15.1

4 6

[alsa-devel] [PATCH 1/2] ALSA: vmaster: Propagate slave error
by Takashi Iwai 08 Mar '18

08 Mar '18

In slave_update() of vmaster code ignores the error from the slave get() callback and copies the values. It's not only about the missing error code but also that this may potentially lead to a leak of uninitialized variables when the slave get() don't clear them. This patch fixes slave_update() not to copy the potentially uninitialized values when an error is returned from the slave get() callback, and to propagate the error value properly. Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/core/vmaster.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/sound/core/vmaster.c b/sound/core/vmaster.c index 8632301489fa..b67de2bb06a2 100644 --- a/sound/core/vmaster.c +++ b/sound/core/vmaster.c @@ -68,10 +68,13 @@ static int slave_update(struct link_slave *slave) return -ENOMEM; uctl->id = slave->slave.id; err = slave->slave.get(&slave->slave, uctl); + if (err < 0) + goto error; for (ch = 0; ch < slave->info.count; ch++) slave->vals[ch] = uctl->value.integer.value[ch]; + error: kfree(uctl); - return 0; + return err < 0 ? err : 0; } /* get the slave ctl info and save the initial values */ -- 2.16.2

1 1

[alsa-devel] [PATCH 3/6] ASoC: max98095: Add #sound-dai-cells property documentation
by Sylwester Nawrocki 08 Mar '18

08 Mar '18

MAX98095 CODEC supports 3 input DAIs, adding the #sound-dai-cells property allows to reference selected DAI in DT. Signed-off-by: Sylwester Nawrocki <s.nawrocki(a)samsung.com> --- Documentation/devicetree/bindings/sound/max98095.txt | 3 +++ 1 file changed, 3 insertions(+) diff --git a/Documentation/devicetree/bindings/sound/max98095.txt b/Documentation/devicetree/bindings/sound/max98095.txt index 318a4c82f17f..944efa06487b 100644 --- a/Documentation/devicetree/bindings/sound/max98095.txt +++ b/Documentation/devicetree/bindings/sound/max98095.txt @@ -14,6 +14,9 @@ Optional properties: - clock-names: Should be "mclk" +- #sound-dai-cells : should be 1. + + Example: max98095: codec@11 { -- 2.14.2

2 1

[alsa-devel] [PATCH 1/6] ASoC: Add snd_soc_of_put_dai_link_codecs() helper function
by Sylwester Nawrocki 07 Mar '18

07 Mar '18

The code for dereferencing device nodes in the 'codecs' array is moved to a separate function so we can avoid repeating the code in drivers. Signed-off-by: Sylwester Nawrocki <s.nawrocki(a)samsung.com> --- include/sound/soc.h | 1 + sound/soc/soc-core.c | 32 +++++++++++++++++++++++--------- 2 files changed, 24 insertions(+), 9 deletions(-) diff --git a/include/sound/soc.h b/include/sound/soc.h index 6a11b0239f74..ad266d7e9553 100644 --- a/include/sound/soc.h +++ b/include/sound/soc.h @@ -1814,6 +1814,7 @@ int snd_soc_of_get_dai_name(struct device_node *of_node, int snd_soc_of_get_dai_link_codecs(struct device *dev, struct device_node *of_node, struct snd_soc_dai_link *dai_link); +void snd_soc_of_put_dai_link_codecs(struct snd_soc_dai_link *dai_link); int snd_soc_add_dai_link(struct snd_soc_card *card, struct snd_soc_dai_link *dai_link); diff --git a/sound/soc/soc-core.c b/sound/soc/soc-core.c index 39876737c336..e815be0d5863 100644 --- a/sound/soc/soc-core.c +++ b/sound/soc/soc-core.c @@ -4358,6 +4358,26 @@ int snd_soc_of_get_dai_name(struct device_node *of_node, } EXPORT_SYMBOL_GPL(snd_soc_of_get_dai_name); +/* + * snd_soc_of_put_dai_link_codecs - Dereference device nodes in the codecs array + * @dai_link: DAI link + * + * Dereference device nodes acquired by snd_soc_of_get_dai_link_codecs(). + */ +void snd_soc_of_put_dai_link_codecs(struct snd_soc_dai_link *dai_link) +{ + struct snd_soc_dai_link_component *component = dai_link->codecs; + int index; + + for (index = 0; index < dai_link->num_codecs; index++, component++) { + if (!component->of_node) + break; + of_node_put(component->of_node); + component->of_node = NULL; + } +} +EXPORT_SYMBOL_GPL(snd_soc_of_put_dai_link_codecs); + /* * snd_soc_of_get_dai_link_codecs - Parse a list of CODECs in the devicetree * @dev: Card device @@ -4367,7 +4387,8 @@ EXPORT_SYMBOL_GPL(snd_soc_of_get_dai_name); * Builds an array of CODEC DAI components from the DAI link property * 'sound-dai'. * The array is set in the DAI link and the number of DAIs is set accordingly. - * The device nodes in the array (of_node) must be dereferenced by the caller. + * The device nodes in the array (of_node) must be dereferenced by calling + * snd_soc_of_put_dai_link_codecs() on @dai_link. * * Returns 0 for success */ @@ -4415,14 +4436,7 @@ int snd_soc_of_get_dai_link_codecs(struct device *dev, } return 0; err: - for (index = 0, component = dai_link->codecs; - index < dai_link->num_codecs; - index++, component++) { - if (!component->of_node) - break; - of_node_put(component->of_node); - component->of_node = NULL; - } + snd_soc_of_put_dai_link_codecs(dai_link); dai_link->codecs = NULL; dai_link->num_codecs = 0; return ret; -- 2.14.2

1 0

[alsa-devel] [RFC PATCH] ASoC: dapm: add support to pinctrl dapm
by srinivas.kandagatla＠linaro.org 07 Mar '18

07 Mar '18

From: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> Purpose of having pinctrl dapm is to dynamically put the pins in low power state when they are not actively used by the audio and saving power. Without this each driver has to set the pinctrl states, either during probe or dynamically depending on the callbacks received from ASoC core. This patch adds a single macro SND_SOC_DAPM_PINCTRL which takes, widget, active and sleep state names, and the dapm would use active and sleep state pinctrl names to put the pins in active and sleep state respectively. Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- Hi All, Purpose of this patch is to discuss the idea of having pinctrl part of dapm. As an example on Qualcomm SoCs there are multiple I2S Interfaces, each of these interfaces have there own pinctrls setting, It does not make sense for the machine/platform/componet driver to put them in default/active state durning probe/init as it would waste power. If we make these pinctrl part of the dapm sequence, these would only become active when the stream is active which would save power. This patch adds a single macro SND_SOC_DAPM_PINCTRL which takes, widget, active and sleep state names, and the dapm would use active and sleep state pinctrl names to put the pins in active and sleep state respectively. Thanks, Srini include/sound/soc-dapm.h | 16 ++++++++++++++++ sound/soc/soc-dapm.c | 44 ++++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 60 insertions(+) diff --git a/include/sound/soc-dapm.h b/include/sound/soc-dapm.h index 344b96c206a3..a6ce2de4e20a 100644 --- a/include/sound/soc-dapm.h +++ b/include/sound/soc-dapm.h @@ -269,6 +269,13 @@ struct device; .reg = SND_SOC_NOPM, .shift = wdelay, .event = dapm_regulator_event, \ .event_flags = SND_SOC_DAPM_PRE_PMU | SND_SOC_DAPM_POST_PMD, \ .on_val = wflags} +#define SND_SOC_DAPM_PINCTRL(wname, active, sleep) \ +{ .id = snd_soc_dapm_pinctrl, .name = wname, \ + .priv = (&(struct snd_soc_dapm_pinctrl_priv) \ + { .active_state = active, .sleep_state = sleep,}), \ + .reg = SND_SOC_NOPM, .event = dapm_pinctrl_event, \ + .event_flags = SND_SOC_DAPM_PRE_PMU | SND_SOC_DAPM_POST_PMD } + /* dapm kcontrol types */ @@ -374,6 +381,8 @@ int dapm_regulator_event(struct snd_soc_dapm_widget *w, struct snd_kcontrol *kcontrol, int event); int dapm_clock_event(struct snd_soc_dapm_widget *w, struct snd_kcontrol *kcontrol, int event); +int dapm_pinctrl_event(struct snd_soc_dapm_widget *w, + struct snd_kcontrol *kcontrol, int event); /* dapm controls */ int snd_soc_dapm_put_volsw(struct snd_kcontrol *kcontrol, @@ -500,6 +509,7 @@ enum snd_soc_dapm_type { snd_soc_dapm_pre, /* machine specific pre widget - exec first */ snd_soc_dapm_post, /* machine specific post widget - exec last */ snd_soc_dapm_supply, /* power/clock supply */ + snd_soc_dapm_pinctrl, /* pinctrl */ snd_soc_dapm_regulator_supply, /* external regulator */ snd_soc_dapm_clock_supply, /* external clock */ snd_soc_dapm_aif_in, /* audio interface input */ @@ -581,6 +591,7 @@ struct snd_soc_dapm_widget { void *priv; /* widget specific data */ struct regulator *regulator; /* attached regulator */ + struct pinctrl *pinctrl; /* attached pinctrl */ const struct snd_soc_pcm_stream *params; /* params for dai links */ unsigned int num_params; /* number of params for dai links */ unsigned int params_select; /* currently selected param for dai link */ @@ -683,6 +694,11 @@ struct snd_soc_dapm_stats { int neighbour_checks; }; +struct snd_soc_dapm_pinctrl_priv { + const char *active_state; + const char *sleep_state; +}; + /** * snd_soc_dapm_init_bias_level() - Initialize DAPM bias level * @dapm: The DAPM context to initialize diff --git a/sound/soc/soc-dapm.c b/sound/soc/soc-dapm.c index 40085a4c8f92..babff1cd2849 100644 --- a/sound/soc/soc-dapm.c +++ b/sound/soc/soc-dapm.c @@ -35,6 +35,7 @@ #include <linux/debugfs.h> #include <linux/pm_runtime.h> #include <linux/regulator/consumer.h> +#include <linux/pinctrl/consumer.h> #include <linux/clk.h> #include <linux/slab.h> #include <sound/core.h> @@ -72,6 +73,7 @@ snd_soc_dapm_new_control_unlocked(struct snd_soc_dapm_context *dapm, static int dapm_up_seq[] = { [snd_soc_dapm_pre] = 0, [snd_soc_dapm_regulator_supply] = 1, + [snd_soc_dapm_pinctrl] = 1, [snd_soc_dapm_clock_supply] = 1, [snd_soc_dapm_supply] = 2, [snd_soc_dapm_micbias] = 3, @@ -119,6 +121,7 @@ static int dapm_down_seq[] = { [snd_soc_dapm_dai_link] = 11, [snd_soc_dapm_supply] = 12, [snd_soc_dapm_clock_supply] = 13, + [snd_soc_dapm_pinctrl] = 13, [snd_soc_dapm_regulator_supply] = 13, [snd_soc_dapm_post] = 14, }; @@ -1287,6 +1290,33 @@ int dapm_regulator_event(struct snd_soc_dapm_widget *w, } EXPORT_SYMBOL_GPL(dapm_regulator_event); +/* + * Handler for pinctrl widget. + */ +int dapm_pinctrl_event(struct snd_soc_dapm_widget *w, + struct snd_kcontrol *kcontrol, int event) +{ + struct snd_soc_dapm_pinctrl_priv *priv = w->priv; + struct pinctrl *p = w->pinctrl; + struct pinctrl_state *s; + + if (!p || !priv) + return -EIO; + + soc_dapm_async_complete(w->dapm); + + if (SND_SOC_DAPM_EVENT_ON(event)) + s = pinctrl_lookup_state(p, priv->active_state); + else + s = pinctrl_lookup_state(p, priv->sleep_state); + + if (IS_ERR(s)) + return PTR_ERR(s); + + return pinctrl_select_state(p, s); +} +EXPORT_SYMBOL_GPL(dapm_pinctrl_event); + /* * Handler for clock supply widget. */ @@ -1900,6 +1930,7 @@ static int dapm_power_widgets(struct snd_soc_card *card, int event) break; case snd_soc_dapm_supply: case snd_soc_dapm_regulator_supply: + case snd_soc_dapm_pinctrl: case snd_soc_dapm_clock_supply: case snd_soc_dapm_micbias: if (d->target_bias_level < SND_SOC_BIAS_STANDBY) @@ -2313,6 +2344,7 @@ static ssize_t dapm_widget_show_component(struct snd_soc_component *cmpnt, case snd_soc_dapm_mixer_named_ctl: case snd_soc_dapm_supply: case snd_soc_dapm_regulator_supply: + case snd_soc_dapm_pinctrl: case snd_soc_dapm_clock_supply: if (w->name) count += sprintf(buf + count, "%s: %s\n", @@ -3462,6 +3494,17 @@ snd_soc_dapm_new_control_unlocked(struct snd_soc_dapm_context *dapm, w->name, ret); } break; + case snd_soc_dapm_pinctrl: + w->pinctrl = devm_pinctrl_get(dapm->dev); + if (IS_ERR_OR_NULL(w->pinctrl)) { + ret = PTR_ERR(w->pinctrl); + if (ret == -EPROBE_DEFER) + return ERR_PTR(ret); + dev_err(dapm->dev, "ASoC: Failed to request %s: %d\n", + w->name, ret); + return NULL; + } + break; case snd_soc_dapm_clock_supply: #ifdef CONFIG_CLKDEV_LOOKUP w->clk = devm_clk_get(dapm->dev, w->name); @@ -3541,6 +3584,7 @@ snd_soc_dapm_new_control_unlocked(struct snd_soc_dapm_context *dapm, break; case snd_soc_dapm_supply: case snd_soc_dapm_regulator_supply: + case snd_soc_dapm_pinctrl: case snd_soc_dapm_clock_supply: case snd_soc_dapm_kcontrol: w->is_supply = 1; -- 2.15.1

3 4

[alsa-devel] [PATCH] ASoC: dt-bindings: wm8524: Remove unit address
by Fabio Estevam 07 Mar '18

07 Mar '18

WM8524 does not use I2C nor SPI, hence no 'reg' property is used, so remove the unit address from the bindings example to conform. This avoids users getting dtc warnings when they copy the binding example to a real dts file. Signed-off-by: Fabio Estevam <festevam(a)gmail.com> --- Documentation/devicetree/bindings/sound/wm8524.txt | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/sound/wm8524.txt b/Documentation/devicetree/bindings/sound/wm8524.txt index 20c6200..0f05535 100644 --- a/Documentation/devicetree/bindings/sound/wm8524.txt +++ b/Documentation/devicetree/bindings/sound/wm8524.txt @@ -10,7 +10,7 @@ Required properties: Example: -codec: wm8524@0 { +codec: wm8524 { compatible = "wlf,wm8524"; wlf,mute-gpios = <&gpio1 8 GPIO_ACTIVE_LOW>; }; -- 2.7.4

2 1

[alsa-devel] [PATCH v2 00/32] ASoC: rt5651: jack-detect fixes and improvements
by Hans de Goede 07 Mar '18

07 Mar '18

Hi All, Here is v2 of my rt5651 codec and Intel machine-driver series which mainly consists of various jack-detect related fixes and improvements. As discussed in the review of v1 I've made the following changes: 1) Get rid of platform_data and use device-properties instead 2) Split up a couple of the micbias / OVCD / jack-detect patches into smaller patches, as they were doing too much in a single patch 3) Improved commit messages 4) Concentrate the knowledge that SSP0 is 16 bits on BYTCR in the fixup function and extract the sample-format from the hw_params in other places Regards, Hans

4 64

[alsa-devel] [PATCH] slimbus: core: use put_device() instead of kfree()
by Arvind Yadav 07 Mar '18

07 Mar '18

Never directly free @dev after calling device_register(), even if it returned an error! Always use put_device() to give up the reference initialized. Signed-off-by: Arvind Yadav <arvind.yadav.cs(a)gmail.com> --- drivers/slimbus/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/slimbus/core.c b/drivers/slimbus/core.c index 4988a8f..7ddfc67 100644 --- a/drivers/slimbus/core.c +++ b/drivers/slimbus/core.c @@ -141,7 +141,7 @@ static struct slim_device *slim_alloc_device(struct slim_controller *ctrl, sbdev->e_addr = *eaddr; ret = slim_add_device(ctrl, sbdev, node); if (ret) { - kfree(sbdev); + put_device(&sbdev->dev); return NULL; } -- 1.9.1

2 1

Re: [alsa-devel] Dell new AIO platform
by Hui Wang 07 Mar '18

07 Mar '18

This is the alsa-info.txt for the machine of zuma-p (the machine kailang sent the patch for): https://pastebin.ubuntu.com/p/V4QgWWY5Nj/ And we have a similar bug with this one, but it is not for output, it is for input. On one of Dell desktops, there is no internal microphone, the only input devices are "Headphone Mic=0x1a" and "Headset Mic=0x19", so the imux_pins[spec->gen.cur_mux[0]] is headphone mic or headset mic, let us suppose it is headset mic, and we are unlucky that the auto-mute is disabled, and then unplug a headset from the jack, the audio is now in unplugged mode (alc_headset_mode_unplugged(codec);), then next time we plug the headset into the jack, the "what-did-you-plug" dialogue will pop up, we select headset-mic, since the imux_pins[spec->gen.cur_mux[0]] is already the headset mic (not internal microphone since there is no internal microphone on this machine), the alc_update_headset_mode() will not be called and then alc_determine_headset_type(), alc_headset_mode_ctia() and alc_headset_mode_omtp() will not be called, as a result the headset mic can't record sound. So far, I have no idea how to fix this problem. And the alsa-info.txt for my machine is as below shown: https://pastebin.ubuntu.com/p/SNMNjbRgnT/ Regards, Hui. On 2018年03月06日 17:13, Kailang wrote: > > >> -----Original Message----- >> From: Takashi Iwai [mailto:tiwai@suse.de] >> Sent: Tuesday, March 06, 2018 5:10 PM >> To: Kailang >> Cc: Hui Wang; (alsa-devel(a)alsa-project.org)" >> Subject: Re: Dell new AIO platform >> >> On Tue, 06 Mar 2018 10:03:38 +0100, >> Kailang wrote: >>> >>> >>>> -----Original Message----- >>>> From: Takashi Iwai [mailto:tiwai@suse.de] >>>> Sent: Tuesday, March 06, 2018 4:49 PM >>>> To: Kailang >>>> Cc: (alsa-devel(a)alsa-project.org) >>>> Subject: Re: Dell new AIO platform >>>> >>>> On Tue, 06 Mar 2018 08:54:39 +0100, >>>> Kailang wrote: >>>>> Hi Takashi, >>>>> >>>>> This platform was only one phone Jack. >>>>> Add dummy lineout verb to fix automute mode disable. >>>>> This just the workarround. >>>> I'm not sure why this phantom line pin is needed. >>>> What happens without that? >>> Looks like I find the root cause for this issue, since >> there is only one output device (headphone）， so the Auto-mute >> is not supported on this machine, as a result, when users >> plug headphone to the jack, audomute_hook or hp_automute_hook >> will not be called, then the alc_update_headset_mode() will >> not be called, and alc_headset_mode_default() will not be >> called. the headset jack is always in the >> alc_headset_mode_unplugged() state. >> >> OK, now understood. >> >> I can take the patch as a tentative workaround. >> But for the proper fix, could you give alsa-info.sh output? > I had no machine in my site. > > Hi Hui Wang, > > Could you help to send alsa-info.sh output to Takashi? > >> >> thanks, >> >> Takashi >> >> ------Please consider the environment before printing this e-mail.

1 1