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

hdajackretask doesn't check for script's exit code
by GitHub issues - opened 06 Sep '21

06 Sep '21

alsa-project/alsa-tools issue #8 was opened from kroq-gar78: `hdajackretask` creates a script in `/tmp` before executing it. When `/tmp` is mounted with `noexec`, running that script will fail, but the GUI gives no indication that it failed. The GUI should check for the exit code of the script, and notify the user appropriately if the script failed to execute. Also, if this isn't the appropriate place to report a bug, please let me know! Issue URL : https://github.com/alsa-project/alsa-tools/issues/8 Repository URL: https://github.com/alsa-project/alsa-tools

1 0

alsa-lib 1.2.5 and 1.2.5.1 fail to build with --enable-static
by GitHub issues - edited 06 Sep '21

06 Sep '21

alsa-project/alsa-lib issue #177 was edited from pjbroad: This build successfully: `./configure && make clean && make` This fails to build: `./configure --enable-shared=no --enable-static=yes && make clean && make` ``` make[1]: Entering directory '/home/paul/el/el-build-scripts/linux-static/libs/alsa-lib-1.2.5.1/aserver' CC aserver.o CCLD aserver /usr/bin/ld: ../src/.libs/libasound.a(control_symbols.o):(.data.rel+0x8): undefined reference to `_snd_module_control_empty' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:398: aserver] Error 1 make[1]: Leaving directory '/home/paul/el/el-build-scripts/linux-static/libs/alsa-lib-1.2.5.1/aserver' make: *** [Makefile:405: all-recursive] Error 1 ``` `gcc (Ubuntu 11.2.0-3ubuntu1) 11.2.0` Building static worked find for 1.2.4. Hopefully someone can help resolve this. Let me know if you want me to try anything. Issue URL : https://github.com/alsa-project/alsa-lib/issues/177 Repository URL: https://github.com/alsa-project/alsa-lib

1 0

alsa-lib 1.2.5 and 1.2.5.1 fail to build with --enable-static
by GitHub issues - opened 06 Sep '21

06 Sep '21

alsa-project/alsa-lib issue #177 was opened from pjbroad: This build successfully: ``` ./configure && make clean && make ``` This fails to build: ``` ./configure --enable-shared=no --enable-static=yes && make clean && make make[1]: Entering directory '/home/paul/el/el-build-scripts/linux-static/libs/alsa-lib-1.2.5.1/aserver' CC aserver.o CCLD aserver /usr/bin/ld: ../src/.libs/libasound.a(control_symbols.o):(.data.rel+0x8): undefined reference to `_snd_module_control_empty' collect2: error: ld returned 1 exit status make[1]: *** [Makefile:398: aserver] Error 1 make[1]: Leaving directory '/home/paul/el/el-build-scripts/linux-static/libs/alsa-lib-1.2.5.1/aserver' make: *** [Makefile:405: all-recursive] Error 1 ``` Building static worked find for 1.2.4. Hopefully someone can help resolve this. Let me know if you want me to try anything. Issue URL : https://github.com/alsa-project/alsa-lib/issues/177 Repository URL: https://github.com/alsa-project/alsa-lib

1 0

Asus Xonar SE front-panel audio support quirks snd_usb_audio
by smesgr 05 Sep '21

05 Sep '21

Greetings, I have bourght a Asus Xonar SE (90YA00T0-M0UA00) which is only working partially. While the rear connectors seems to work the front header is not. I found some reports scattered around the internet with the same issue but without a solution. It is a PCIe card with an pcie-to-usb interface. Thus the card is reported as a usb device: ID 0b05:189d ASUSTek Computer, Inc. Xonar SoundCard The card has several audio Endpoints EP 5 to EP 9 and an additonal HID EP 4. All the rear connectors seems to work. Can't test the SPDIF but found a SE specfic quirk in the alsa USB configuration for the SDPIF PCM thus assume this issue was already fixed. EP 5 -> Audio Out EP 6 -> SDPIF EP 8 -> Line In The rear connector don't have a plug-in-detector, the front-panel has and is working: node hw:1, #16 (0,0,0,Headphone - Output Jack,0) VALUE node hw:1, #31 (0,0,0,Mic - Input Jack,1) VALUE In stereo configuration the card has "headphones" and "speaker". The Front Panel input nor output is working. If headphones is selected the rear is disabled but the front header is not working. Side note the default Microsoft driver has the same issue so I assume the card usb configuration is faulty. I installed the ASUS driver and tracked the USB messages with wireshark in a Windows VM. I noticed the Asus driver uses those additional posts EP 7 -> Front Audio Out EP 9 -> Front Audio In Neither the default Windows, nor alsa use those EP. Is there a way to force the alsa driver to use EP7/9 to further analyse this issue? Best Regards Smesgr Sent with ProtonMail Secure Email.

2 1

pulseaudio: a52 sink doesn't stream audio from bluez source
by GitHub issues - edited 05 Sep '21

05 Sep '21

alsa-project/alsa-plugins issue #32 was edited from quequotion: Sorry to beat a dead horse, and double if this is in fact a pulseaudio issue. I recently configured bluez and pulseaudio to receive streams from devices over bluetooth (eg, to play audio from a phone over the PC's speaker system). This works when the PC's sound card is set to profiles *other than* the A52 plugin (ie, pulseaudio's provided digital stereo output and also the [DTS alsa plugin](https://github.com/darealshinji/dcaenc)). Only when set to use the AC3 output profile, audio from the bluez source is not produced. pavucontrol shows the bluez _source_ is producing sound (the level bounces appropriately), but no audio comes out of the _sink_ (no level bounce in pavucontrol either). Any ideas? Issue URL : https://github.com/alsa-project/alsa-plugins/issues/32 Repository URL: https://github.com/alsa-project/alsa-plugins

1 0

No sound after kernel 5.8.15-301
by Сурков Павел 05 Sep '21

05 Sep '21

Hello! Â I lost sound from laptop speakers after updating the kernel. Sound via HDMI, AUX or Bluetooth works. Details on the links: [1]https://bbs.archlinux.org/viewtopic.php?id=269301 [2]https://ask.fedoraproject.org/t/no-sound-after-kernel-5-8-15-301/164 41 Â Help me please. --Â Pavel Surkov Â References 1. https://bbs.archlinux.org/viewtopic.php?id=269301 2. https://ask.fedoraproject.org/t/no-sound-after-kernel-5-8-15-301/16441

2 4

[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