April 2024 - Alsa-devel - mailman.alsa-project.org

[PATCH v1] ASoC: cs35l41: Update DSP1RX5/6 Sources for DSP config
by Stefan Binding 12 Apr '24

12 Apr '24

Currently, all ASoC systems are set to use VPMON for DSP1RX5_SRC, however, this is required only for internal boost systems. External boost systems require VBSTMON instead of VPMON to be the input to DSP1RX5_SRC. Shared Boost Active acts like Internal boost (requires VPMON). Shared Boost Passive acts like External boost (requires VBSTMON) All systems require DSP1RX6_SRC to be set to VBSTMON. Signed-off-by: Stefan Binding <sbinding(a)opensource.cirrus.com> --- sound/soc/codecs/cs35l41.c | 26 ++++++++++++++++++++------ 1 file changed, 20 insertions(+), 6 deletions(-) diff --git a/sound/soc/codecs/cs35l41.c b/sound/soc/codecs/cs35l41.c index dfb4ce53491b..f8e57a2fc3e3 100644 --- a/sound/soc/codecs/cs35l41.c +++ b/sound/soc/codecs/cs35l41.c @@ -1094,6 +1094,7 @@ static int cs35l41_handle_pdata(struct device *dev, struct cs35l41_hw_cfg *hw_cf static int cs35l41_dsp_init(struct cs35l41_private *cs35l41) { struct wm_adsp *dsp; + uint32_t dsp1rx5_src; int ret; dsp = &cs35l41->dsp; @@ -1113,16 +1114,29 @@ static int cs35l41_dsp_init(struct cs35l41_private *cs35l41) return ret; } - ret = regmap_write(cs35l41->regmap, CS35L41_DSP1_RX5_SRC, - CS35L41_INPUT_SRC_VPMON); + switch (cs35l41->hw_cfg.bst_type) { + case CS35L41_INT_BOOST: + case CS35L41_SHD_BOOST_ACTV: + dsp1rx5_src = CS35L41_INPUT_SRC_VPMON; + break; + case CS35L41_EXT_BOOST: + case CS35L41_SHD_BOOST_PASS: + dsp1rx5_src = CS35L41_INPUT_SRC_VBSTMON; + break; + default: + dev_err(cs35l41->dev, "wm_halo_init failed - Invalid Boost Type: %d\n", + cs35l41->hw_cfg.bst_type); + goto err_dsp; + } + + ret = regmap_write(cs35l41->regmap, CS35L41_DSP1_RX5_SRC, dsp1rx5_src); if (ret < 0) { - dev_err(cs35l41->dev, "Write INPUT_SRC_VPMON failed: %d\n", ret); + dev_err(cs35l41->dev, "Write DSP1RX5_SRC: %d failed: %d\n", dsp1rx5_src, ret); goto err_dsp; } - ret = regmap_write(cs35l41->regmap, CS35L41_DSP1_RX6_SRC, - CS35L41_INPUT_SRC_CLASSH); + ret = regmap_write(cs35l41->regmap, CS35L41_DSP1_RX6_SRC, CS35L41_INPUT_SRC_VBSTMON); if (ret < 0) { - dev_err(cs35l41->dev, "Write INPUT_SRC_CLASSH failed: %d\n", ret); + dev_err(cs35l41->dev, "Write CS35L41_INPUT_SRC_VBSTMON failed: %d\n", ret); goto err_dsp; } ret = regmap_write(cs35l41->regmap, CS35L41_DSP1_RX7_SRC, -- 2.34.1

3 2

[PATCH][next] ASoC: codecs: tas2780: remove redundant assignments to variable ret
by Colin Ian King 12 Apr '24

12 Apr '24

Variable ret is being assigned a value that is never read in a couple of places. The variable is being re-assigned later on. The assignments are redundant and can be removed. Cleans up clang scan build warning: sound/soc/codecs/tas2780.c:84:2: warning: Value stored to 'ret' is never read [deadcode.DeadStores] Signed-off-by: Colin Ian King <colin.i.king(a)gmail.com> --- sound/soc/codecs/tas2780.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/sound/soc/codecs/tas2780.c b/sound/soc/codecs/tas2780.c index 41076be23854..972e8ea5ebde 100644 --- a/sound/soc/codecs/tas2780.c +++ b/sound/soc/codecs/tas2780.c @@ -71,7 +71,7 @@ static int tas2780_codec_resume(struct snd_soc_component *component) { struct tas2780_priv *tas2780 = snd_soc_component_get_drvdata(component); - int ret = 0; + int ret; ret = snd_soc_component_update_bits(component, TAS2780_PWR_CTRL, TAS2780_PWR_CTRL_MASK, TAS2780_PWR_CTRL_ACTIVE); @@ -81,7 +81,6 @@ static int tas2780_codec_resume(struct snd_soc_component *component) __func__, ret); goto err; } - ret = 0; regcache_cache_only(tas2780->regmap, false); ret = regcache_sync(tas2780->regmap); err: -- 2.39.2

2 1

[PATCH] MAINTAINERS: adjust file entry in TEXAS INSTRUMENTS AUDIO (ASoC/HDA) DRIVERS
by Lukas Bulwahn 12 Apr '24

12 Apr '24

Commit 8167bd1c8a45 ("ASoC: dt-bindings: ti,pcm1681: Convert to dtschema") converts ti,pcm1681.txt to ti,pcm1681.yaml, but misses to adjust the file entry in TEXAS INSTRUMENTS AUDIO (ASoC/HDA) DRIVERS. Hence, ./scripts/get_maintainer.pl --self-test=patterns complains about a broken reference. Adjust the file entry in TEXAS INSTRUMENTS AUDIO (ASoC/HDA) DRIVERS after this conversion. Signed-off-by: Lukas Bulwahn <lukas.bulwahn(a)redhat.com> --- MAINTAINERS | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/MAINTAINERS b/MAINTAINERS index e5431f06ab55..d39e0f2556f7 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -21976,7 +21976,7 @@ F: Documentation/devicetree/bindings/sound/tas2552.txt F: Documentation/devicetree/bindings/sound/tas2562.yaml F: Documentation/devicetree/bindings/sound/tas2770.yaml F: Documentation/devicetree/bindings/sound/tas27xx.yaml -F: Documentation/devicetree/bindings/sound/ti,pcm1681.txt +F: Documentation/devicetree/bindings/sound/ti,pcm1681.yaml F: Documentation/devicetree/bindings/sound/ti,pcm3168a.yaml F: Documentation/devicetree/bindings/sound/ti,tlv320*.yaml F: Documentation/devicetree/bindings/sound/tlv320adcx140.yaml -- 2.44.0

2 1

[PATCH v2] soundwire: qcom: allow multi-link on newer devices
by Krzysztof Kozlowski 11 Apr '24

11 Apr '24

Newer Qualcomm SoCs like X1E80100 might come with four speakers spread over two Soundwire controllers, thus they need a multi-link Soundwire stream runtime. Cc: Mark Brown <broonie(a)kernel.org> Cc: alsa-devel(a)alsa-project.org Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski(a)linaro.org> --- Changes in v2: 1. Only rebase (slightly different context) --- drivers/soundwire/qcom.c | 22 ++++++++++++++++++++++ 1 file changed, 22 insertions(+) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index fb70afe64fcc..ce5cf3ecceb5 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -905,6 +905,18 @@ static int qcom_swrm_init(struct qcom_swrm_ctrl *ctrl) return 0; } +static int qcom_swrm_read_prop(struct sdw_bus *bus) +{ + struct qcom_swrm_ctrl *ctrl = to_qcom_sdw(bus); + + if (ctrl->version >= SWRM_VERSION_2_0_0) { + bus->multi_link = true; + bus->hw_sync_min_links = 3; + } + + return 0; +} + static enum sdw_command_response qcom_swrm_xfer_msg(struct sdw_bus *bus, struct sdw_msg *msg) { @@ -1056,6 +1068,7 @@ static const struct sdw_master_port_ops qcom_swrm_port_ops = { }; static const struct sdw_master_ops qcom_swrm_ops = { + .read_prop = qcom_swrm_read_prop, .xfer_msg = qcom_swrm_xfer_msg, .pre_bank_switch = qcom_swrm_pre_bank_switch, }; @@ -1173,6 +1186,15 @@ static int qcom_swrm_stream_alloc_ports(struct qcom_swrm_ctrl *ctrl, mutex_lock(&ctrl->port_lock); list_for_each_entry(m_rt, &stream->master_list, stream_node) { + /* + * For streams with multiple masters: + * Allocate ports only for devices connected to this master. + * Such devices will have ports allocated by their own master + * and its qcom_swrm_stream_alloc_ports() call. + */ + if (ctrl->bus.id != m_rt->bus->id) + continue; + if (m_rt->direction == SDW_DATA_DIR_RX) { maxport = ctrl->num_dout_ports; port_mask = &ctrl->dout_port_mask; -- 2.34.1

2 4

[PATCH v2] slimbus: qcom-ngd-ctrl: Add timeout for wait operation
by Viken Dadhaniya 11 Apr '24

11 Apr '24

In current driver qcom_slim_ngd_up_worker() indefinitely waiting for ctrl->qmi_up completion object. This is resulting in workqueue lockup on Kthread. Added wait_for_completion_interruptible_timeout to allow the thread to wait for specific timeout period and bail out instead waiting infinitely. Fixes: a899d324863a ("slimbus: qcom-ngd-ctrl: add Sub System Restart support") Cc: stable(a)vger.kernel.org Reviewed-by: Konrad Dybcio <konrad.dybcio(a)linaro.org> Signed-off-by: Viken Dadhaniya <quic_vdadhani(a)quicinc.com> --- v1 -> v2: - Remove macro and add value inline. - add fix, cc and review tag. --- --- drivers/slimbus/qcom-ngd-ctrl.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index efeba8275a66..a09a26bf4988 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -1451,7 +1451,11 @@ static void qcom_slim_ngd_up_worker(struct work_struct *work) ctrl = container_of(work, struct qcom_slim_ngd_ctrl, ngd_up_work); /* Make sure qmi service is up before continuing */ - wait_for_completion_interruptible(&ctrl->qmi_up); + if (!wait_for_completion_interruptible_timeout(&ctrl->qmi_up, + msecs_to_jiffies(MSEC_PER_SEC))) { + dev_err(ctrl->dev, "QMI wait timeout\n"); + return; + } mutex_lock(&ctrl->ssr_lock); qcom_slim_ngd_enable(ctrl, true); -- QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

2 1

[PATCH] slimbus: qcom-ctrl: fix module autoloading
by Krzysztof Kozlowski 11 Apr '24

11 Apr '24

Add MODULE_DEVICE_TABLE(), so the module could be properly autoloaded based on the alias from of_device_id table. Pin controllers are considered core components, so usually they are built-in, however these Signed-off-by: Krzysztof Kozlowski <krzk(a)kernel.org> --- drivers/slimbus/qcom-ctrl.c | 1 + 1 file changed, 1 insertion(+) diff --git a/drivers/slimbus/qcom-ctrl.c b/drivers/slimbus/qcom-ctrl.c index 400b7b385a44..f7d04b50c973 100644 --- a/drivers/slimbus/qcom-ctrl.c +++ b/drivers/slimbus/qcom-ctrl.c @@ -718,6 +718,7 @@ static const struct of_device_id qcom_slim_dt_match[] = { { .compatible = "qcom,slim", }, {} }; +MODULE_DEVICE_TABLE(of, qcom_slim_dt_match); static struct platform_driver qcom_slim_driver = { .probe = qcom_slim_probe, -- 2.34.1

3 2

[PATCH] slimbus: Convert to platform remove callback returning void
by Uwe Kleine-König 11 Apr '24

11 Apr '24

The .remove() callback for a platform driver returns an int which makes many driver authors wrongly assume it's possible to do error handling by returning an error code. However the value returned is ignored (apart from emitting a warning) and this typically results in resource leaks. To improve here there is a quest to make the remove callback return void. In the first step of this quest all drivers are converted to .remove_new(), which already returns void. Eventually after all drivers are converted, .remove_new() will be renamed to .remove(). Trivially convert the slimbus drivers from always returning zero in the remove callback to the void returning variant. Signed-off-by: Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de> --- drivers/slimbus/qcom-ctrl.c | 5 ++--- drivers/slimbus/qcom-ngd-ctrl.c | 11 ++++------- 2 files changed, 6 insertions(+), 10 deletions(-) diff --git a/drivers/slimbus/qcom-ctrl.c b/drivers/slimbus/qcom-ctrl.c index 400b7b385a44..7d632fad1300 100644 --- a/drivers/slimbus/qcom-ctrl.c +++ b/drivers/slimbus/qcom-ctrl.c @@ -626,7 +626,7 @@ static int qcom_slim_probe(struct platform_device *pdev) return ret; } -static int qcom_slim_remove(struct platform_device *pdev) +static void qcom_slim_remove(struct platform_device *pdev) { struct qcom_slim_ctrl *ctrl = platform_get_drvdata(pdev); @@ -635,7 +635,6 @@ static int qcom_slim_remove(struct platform_device *pdev) clk_disable_unprepare(ctrl->rclk); clk_disable_unprepare(ctrl->hclk); destroy_workqueue(ctrl->rxwq); - return 0; } /* @@ -721,7 +720,7 @@ static const struct of_device_id qcom_slim_dt_match[] = { static struct platform_driver qcom_slim_driver = { .probe = qcom_slim_probe, - .remove = qcom_slim_remove, + .remove_new = qcom_slim_remove, .driver = { .name = "qcom_slim_ctrl", .of_match_table = qcom_slim_dt_match, diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index efeba8275a66..a5aa364f562e 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -1675,14 +1675,12 @@ static int qcom_slim_ngd_ctrl_probe(struct platform_device *pdev) return ret; } -static int qcom_slim_ngd_ctrl_remove(struct platform_device *pdev) +static void qcom_slim_ngd_ctrl_remove(struct platform_device *pdev) { platform_driver_unregister(&qcom_slim_ngd_driver); - - return 0; } -static int qcom_slim_ngd_remove(struct platform_device *pdev) +static void qcom_slim_ngd_remove(struct platform_device *pdev) { struct qcom_slim_ngd_ctrl *ctrl = platform_get_drvdata(pdev); @@ -1697,7 +1695,6 @@ static int qcom_slim_ngd_remove(struct platform_device *pdev) kfree(ctrl->ngd); ctrl->ngd = NULL; - return 0; } static int __maybe_unused qcom_slim_ngd_runtime_idle(struct device *dev) @@ -1740,7 +1737,7 @@ static const struct dev_pm_ops qcom_slim_ngd_dev_pm_ops = { static struct platform_driver qcom_slim_ngd_ctrl_driver = { .probe = qcom_slim_ngd_ctrl_probe, - .remove = qcom_slim_ngd_ctrl_remove, + .remove_new = qcom_slim_ngd_ctrl_remove, .driver = { .name = "qcom,slim-ngd-ctrl", .of_match_table = qcom_slim_ngd_dt_match, @@ -1749,7 +1746,7 @@ static struct platform_driver qcom_slim_ngd_ctrl_driver = { static struct platform_driver qcom_slim_ngd_driver = { .probe = qcom_slim_ngd_probe, - .remove = qcom_slim_ngd_remove, + .remove_new = qcom_slim_ngd_remove, .driver = { .name = QCOM_SLIM_NGD_DRV_NAME, .pm = &qcom_slim_ngd_dev_pm_ops, base-commit: 33e1d31873f87d119e5120b88cd350efa68ef276 -- 2.43.0

2 2

[PATCH v2] slimbus: qcom-ngd-ctrl: Reduce auto suspend delay
by Viken Dadhaniya 11 Apr '24

11 Apr '24

Currently we have auto suspend delay of 1s which is very high and it takes long time to driver for runtime suspend after use case is done. Hence to optimize runtime PM ops, reduce auto suspend delay to 100ms. Signed-off-by: Viken Dadhaniya <quic_vdadhani(a)quicinc.com> --- v1 -> v2: - Remove macro and add value inline. - Link to v1: https://lore.kernel.org/linux-arm-msm/d784e46d-974d-4bf3-a2d3-491e7ad19701@… --- --- drivers/slimbus/qcom-ngd-ctrl.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/drivers/slimbus/qcom-ngd-ctrl.c b/drivers/slimbus/qcom-ngd-ctrl.c index efeba8275a66..5e132273c6c4 100644 --- a/drivers/slimbus/qcom-ngd-ctrl.c +++ b/drivers/slimbus/qcom-ngd-ctrl.c @@ -81,7 +81,6 @@ #define SLIM_USR_MC_DISCONNECT_PORT 0x2E #define SLIM_USR_MC_REPEAT_CHANGE_VALUE 0x0 -#define QCOM_SLIM_NGD_AUTOSUSPEND MSEC_PER_SEC #define SLIM_RX_MSGQ_TIMEOUT_VAL 0x10000 #define SLIM_LA_MGR 0xFF @@ -1571,7 +1570,7 @@ static int qcom_slim_ngd_probe(struct platform_device *pdev) platform_set_drvdata(pdev, ctrl); pm_runtime_use_autosuspend(dev); - pm_runtime_set_autosuspend_delay(dev, QCOM_SLIM_NGD_AUTOSUSPEND); + pm_runtime_set_autosuspend_delay(dev, 100); pm_runtime_set_suspended(dev); pm_runtime_enable(dev); pm_runtime_get_noresume(dev); -- QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation

3 2

[PATCH v1 0/7] Add features, fixes and laptops for CS35L41 HDA
by Stefan Binding 11 Apr '24

11 Apr '24

This chain adds a number of improvements for CS35L41, including a new firmware loading feature, as well as some additional cleanup. Patch 1 adds support for a new firmware tuning file which is used to set special tuning parameters used by the firmware. The HP Omen laptop added in patches 2 and 3 use such a feature. Patch 4 changes the DSP1RX5/6 Source used by the firmware to depend on the boost type. A similar change in the ASoC driver will be needed later. Patches 5 and 6 are cleanup patches, with the first of those re-using cs-amp-lib for calibration which as added for CS35L56 drivers, ensuring that all laptops are using the correct calibration values. Patch 7 fixes a small error in the CS35L41 Property table for the Lenovo Legion slim 7 16ARHA7. Richard Fitzgerald (1): ALSA: hda: cs35l41: Remove redundant argument to cs35l41_request_firmware_file() Stefan Binding (6): ALSA: hda: cs35l41: Set the max PCM Gain using tuning setting ALSA: hda: cs35l41: Support HP Omen models without _DSD ALSA: hda/realtek: Add quirks for HP Omen models using CS35L41 ALSA: hda: cs35l41: Update DSP1RX5/6 Sources for DSP config ALSA: hda: cs35l41: Use shared cs-amp-lib to apply calibration ALSA: hda: cs35l41: Remove Speaker ID for Lenovo Legion slim 7 16ARHA7 include/sound/cs35l41.h | 5 + sound/pci/hda/Kconfig | 2 + sound/pci/hda/cs35l41_hda.c | 347 +++++++++++++++++++-------- sound/pci/hda/cs35l41_hda.h | 6 + sound/pci/hda/cs35l41_hda_property.c | 8 +- sound/pci/hda/patch_realtek.c | 4 + 6 files changed, 265 insertions(+), 107 deletions(-) -- 2.34.1

2 9

[PATCH] ASoC: wm_adsp: Include array_size.h
by Richard Fitzgerald 10 Apr '24

10 Apr '24

Explicitly #include array_size.h for the ARRAY_SIZE() macro. Signed-off-by: Richard Fitzgerald <rf(a)opensource.cirrus.com> --- sound/soc/codecs/wm_adsp.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/codecs/wm_adsp.c b/sound/soc/codecs/wm_adsp.c index 7d5c096e06cd..9a89682ebc1f 100644 --- a/sound/soc/codecs/wm_adsp.c +++ b/sound/soc/codecs/wm_adsp.c @@ -7,6 +7,7 @@ * Author: Mark Brown <broonie(a)opensource.wolfsonmicro.com> */ +#include <linux/array_size.h> #include <linux/ctype.h> #include <linux/module.h> #include <linux/moduleparam.h> -- 2.39.2

2 1