- 28 Nov, 2022 9 commits
-
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6Core service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Signed-off-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by:
Rob Herring <robh@kernel.org> Link: https://lore.kernel.org/r/20221115120235.167812-10-krzysztof.kozlowski@linaro.orgSigned-off-by:
Mark Brown <broonie@kernel.org>
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6PRM service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Signed-off-by:
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6ASM service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Move most of the examples from its children to this new file. Signed-off-by:
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6ADM service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Move most of the examples from its children to this new file. Signed-off-by:
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6APM service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Move most of the examples from its children to this new file. Signed-off-by:
-
Krzysztof Kozlowski authored
The APR/GPR bindings with services got complicated so move out the Q6AFE service to its own binding. Previously the compatible was documented in qcom,apr.yaml. Move most of the examples from its children to this new file. Signed-off-by:
-
Krzysztof Kozlowski authored
Correct the APR/GPR example: 1. Use consistent 4-space indentation, 2. Add required properties to services nodes, so the binding check passes once schema for these services is improved, 3. Add few other properties as APR/GPR is part of a GLINK edge: qcom,glink-channels and qcom,intents. 4. Drop unnecessary services, to make the example compact. Signed-off-by:
-
Krzysztof Kozlowski authored
The APR/GPR nodes are organized like: apr-or-gpr-device-node <- qcom,apr.yaml apr-gpr-service@[0-9] <- qcom,apr.yaml service-specific-components <- /schemas/sound/qcom,q6*.yaml The schema for services (apr-gpr-service@[0-9]) already grows considerably and is still quite not specific. It allows several incorrect combinations, like adding a clock-controller to a APM device. Restricting it would complicate the schema even more. Bringing new support for sound on Qualcomm SM8450 and SC8280XP SoC would grow it as well. Simplify the qcom,apr.yaml by splitting the services to a shared file which will be: 1. Referenced by qcom,apr.yaml with additionalProperties:true, 2. Referenced by specific bindings for services with additionalProperties:false (not yet in this commit). While moving the code, add also required 'reg' and 'qcom,protection-domain' to further constrain the bindings. Signed-off-by: -
Krzysztof Kozlowski authored
On SM8450 and SC8280XP with Qualcomm GPR, the GLINK channel name (qcom,glink-channels) for ADSP is "adsp_apps". Signed-off-by:
-
- 25 Nov, 2022 16 commits
-
-
Mark Brown authored
Merge series from Nicolas Frattaroli <frattaroli.nicolas@gmail.com>: This patchset refactors the Rockchip I2S/TDM driver in order to support the RK3588 SoC, and then adds the necessary compatible string to load the driver for it. Patch 1 rectifies a problem with the bindings where we were too strict about requiring the rockchip,grf property. Most features of this audio device don't need access to the GRF to function. Patch 2 modifies the driver to adjust its behaviour to what the changed bindings now allow, namely using most things without the GRF. Patch 3 and 4 are boring compatible string stuff that enables RK3588 support. No special data is needed to initialise the driver for this instance of the I2S/TDM IP.
-
Mark Brown authored
Merge series from Maarten Zanders <maarten.zanders@mind.be>: A collection of fixes and improvements for the adau1372 driver.
-
Maarten Zanders authored
The ADAU1372 contains 24bit ADCs and DACs. Allow the driver to use its native mode which uses the same settings as the current 32 bit mode. Signed-off-by:
Maarten Zanders <maarten.zanders@mind.be> Link: https://lore.kernel.org/r/20221028152626.109603-3-maarten.zanders@mind.beSigned-off-by:
-
Maarten Zanders authored
The DAPM control for PGAx uses the PGA mute bit for power management. This bit is active high but is set to non-inverted (ie when powering, it will mute). The ALSA control "PGA x Capture Switch" uses the active high PGA_ENx bit, but is set to inverted. So when enabling this switch, the PGA gets disabled. To correct the behaviour, invert both these bits. Signed-off-by:
-
Maarten Zanders authored
"mclk" is retrieved from the configuration and assigned to adau1372->clk. However adau1372->mclk (==NULL) is used for clk_prepare_enable() and clk_disable_unprepare() which don't have any effect. Remove .clk from struct adau1372 and use .mclk throughout. This change ensures that the input clock is switched on/off when the bias level is changed. Signed-off-by:
-
Jiasheng Jiang authored
As the mtk_btcvsd_snd_write and mtk_btcvsd_snd_read may return error, it should be better to catch the exception. Fixes: 4bd8597d ("ASoC: mediatek: add btcvsd driver") Signed-off-by:
Jiasheng Jiang <jiasheng@iscas.ac.cn> Link: https://lore.kernel.org/r/20221116030750.40500-1-jiasheng@iscas.ac.cnSigned-off-by:
-
Dmitry Torokhov authored
When resetting the block, the reset line is being driven low and then high, which means that the line in DTS should be annotated as "active low". Fixes: 1877c9fd ("ASoC: dt-bindings: add dt bindings for wcd9335 audio codec") Signed-off-by:
Dmitry Torokhov <dmitry.torokhov@gmail.com> Acked-by:
-
Nicolas Frattaroli authored
This adds support for the RK3588 SoC to the I2S/TDM driver. Signed-off-by:
Nicolas Frattaroli <frattaroli.nicolas@gmail.com> Link: https://lore.kernel.org/r/20221025124132.399729-5-frattaroli.nicolas@gmail.comSigned-off-by:
-
Nicolas Frattaroli authored
This adds the compatible string for the RK3588 SoC. Signed-off-by:
-
Nicolas Frattaroli authored
Only IO Multiplex and two TRCM modes need access to the GRF, so making it a hard requirement is not a wise idea, as it complicates support for newer SoCs which do not do these things. Signed-off-by:
-
Nicolas Frattaroli authored
Turns out most things don't require the GRF, so making the property optional is the obvious move. Signed-off-by:
-
Mark Brown authored
Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>: Use EXPORT_SYMBOL_NS_GPL() instead of EXPORT_SYMBOL_GPL() and patch the three drivers that use cs_dsp to add the MODULE_IMPORT_NS(). To make the namespace more specific the KConfig symbol for cs_dsp is changed from CS_DSP to FW_CS_DSP.
-
Mark Brown authored
Merge series from Bard Liao <yung-chuan.liao@linux.intel.com>: The code in drivers/soundwire/intel_init.c is hardware-dependent and the code does not apply to new generations starting with MeteorLake. Refactor and clean-up the code to make this intel_init.c hardware-agnostic and move all hardware-dependencies in the SOF driver using chip descriptors.
-
Mark Brown authored
Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>: Writing a firmware control should be returning 1 if the control value changed, so these two patches add that. Though this is an ALSA requirement it is also useful for non-ALSA clients of cs_dsp to know if the control value changed, so the main handling is implemented in cs_dsp. TLV controls are specifically an ALSA thing so they are handled specially in wm_adsp. Simon Trimmer (2): firmware: cs_dsp: cs_dsp_coeff_write_ctrl() should report changed ASoC: wm_adsp: Return whether changed when writing controls drivers/firmware/cirrus/cs_dsp.c | 17 ++++++++++++----- sound/soc/codecs/wm_adsp.c | 27 ++++++++++++++++++--------- 2 files changed, 30 insertions(+), 14 deletions(-) -- 2.30.2
-
Richard Fitzgerald authored
Move all the exports into a namespace. This also adds the MODULE_IMPORT_NS to the 3 drivers that use the exported functions. Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Link: https://lore.kernel.org/r/20221124134556.3343784-3-rf@opensource.cirrus.comSigned-off-by:
-
Richard Fitzgerald authored
Qualify the KConfig symbol for cs_dsp by adding a FW_ prefix so that it is more explicit what is being referred to. This is preparation for using the symbol to namespace the exports. Signed-off-by:
-
- 24 Nov, 2022 10 commits
-
-
Uwe Kleine-König authored
The probe function doesn't make use of the i2c_device_id * parameter so it can be trivially converted. Signed-off-by:
Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Link: https://lore.kernel.org/r/20221118224540.619276-604-uwe@kleine-koenig.orgReviewed-by:
Matt Flax <flatmax@flatmax.com> Signed-off-by:
-
Uwe Kleine-König authored
.probe_new() doesn't get the i2c_device_id * parameter, so determine that explicitly in the probe function. Signed-off-by:
-
Uwe Kleine-König authored
The probe function doesn't make use of the i2c_device_id * parameter so it can be trivially converted. Signed-off-by:
-
Uwe Kleine-König authored
The probe function doesn't make use of the i2c_device_id * parameter so it can be trivially converted. Signed-off-by:
-
Mark Brown authored
Merge branch 'i2c/client_device_id_helper-immutable' of https://git.kernel.org/pub/scm/linux/kernel/git/wsa/linux into HEAD so we can apply I2C cleanups.
-
Jiaxin Yu authored
In mt8186 platform, I2S2 should be the main I2S port that provide the clock, on the contrary I2S3 should be the second I2S port that use this clock. Fixes: 9986bdae ("ASoC: mediatek: mt8186: Configure shared clocks") Signed-off-by:
Jiaxin Yu <jiaxin.yu@mediatek.com> Reviewed-by:
AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20221124023050.4470-1-jiaxin.yu@mediatek.comSigned-off-by:
-
David Lin authored
The audio amplifier NAU8318 is almost functionally identical to NAU8315. Adds compatible string "nuvoton,nau8318" for driver reuse. Signed-off-by:
David Lin <CTLIN0@nuvoton.com> Link: https://lore.kernel.org/r/20221124055658.53828-2-CTLIN0@nuvoton.comSigned-off-by:
-
David Lin authored
Add new acpi id and compatible id for nau8315. Signed-off-by:
-
Srinivas Kandagatla authored
SND_SOC_QCOM_COMMON depends on SOUNDWIRE for some symbols but this is not explicitly specified using Kconfig depends. On the other hand SND_SOC_QCOM_COMMON is also directly selected by the sound card Kconfigs, this could result in various combinations and some symbols ending up in modules and soundcard that uses those symbols as in-build driver. Fix these issues by explicitly specifying the dependencies of SND_SOC_QCOM_COMMON and also use imply a to select SND_SOC_QCOM_COMMON so that the symbol is selected based on its dependencies. Also remove dummy stubs in common.c around CONFIG_SOUNDWIRE Fixes: 3bd975f3 ("ASoC: qcom: sm8250: move some code to common") Reported-by:
kernel test robot <lkp@intel.com> Signed-off-by:
Srinivas Kandagatla <srinivas.kandagatla@linaro.org> Link: https://lore.kernel.org/r/20221124140351.407506-1-srinivas.kandagatla@linaro.orgSigned-off-by:
-
Yuan Can authored
As the devm_kcalloc may return NULL, the return value needs to be checked to avoid NULL poineter dereference. Fixes: 24caf8d9 ("ASoC: qcom: lpass-sc7180: Add platform driver for lpass audio") Signed-off-by:
Yuan Can <yuancan@huawei.com> Link: https://lore.kernel.org/r/20221124140510.63468-1-yuancan@huawei.comSigned-off-by:
-
- 23 Nov, 2022 5 commits
-
-
Simon Trimmer authored
Functions that update cs_dsp controls need to handle return codes that indicate whether the control value changed. A return code of 1 indicates a change, 0 indicates no-change and a negative value is an error condition. Acked controls implicitly change value when written so a successful write shall always report that the value changed. Signed-off-by:
Simon Trimmer <simont@opensource.cirrus.com> Signed-off-by:
-
Simon Trimmer authored
ALSA callers need to know whether there was a change to the value so that they can report a control write change correctly. Signed-off-by:
-
Pierre-Louis Bossart authored
There's no reason to delay the multi-link parsing, this can be done earlier before checking the SoundWire capabilities. Signed-off-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Reviewed-by:
Péter Ujfalusi <peter.ujfalusi@linux.intel.com> Reviewed-by:
Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Signed-off-by:
Bard Liao <yung-chuan.liao@linux.intel.com> Link: https://lore.kernel.org/r/20221111042653.45520-9-yung-chuan.liao@linux.intel.comSigned-off-by:
-
Pierre-Louis Bossart authored
The number of links is checked with a chip-dependent helper in the caller, remove the check in drivers/soundwire/intel_init.c This change makes intel_init.c hardware-agnostic - which is quite fitting for a layer that only creates auxiliary devices. Signed-off-by:
Vinod Koul <vkoul@kernel.org> Link: https://lore.kernel.org/r/20221111042653.45520-8-yung-chuan.liao@linux.intel.comSigned-off-by:
-
Pierre-Louis Bossart authored
The functionality is implemented with per-chip callbacks, there are no users of this symbol, remove the code. Signed-off-by:
-
