- 05 Oct, 2021 5 commits
-
-
Pierre-Louis Bossart authored
In order to keep the widget use_count balanced, make sure the DAI widgets are allocated once in hw_params and released in hw_free. A 'setup' status flag is used to deal with cases where the .hw_params callback is invoked multiple times, and likewise with cases where hw_free is invoked without hw_params being called first (which can happen if the FE hw_params fails). In addition, this patch frees the widgets in the suspend transition, and reallocates them in the .prepare callback. The 'setup' flag helps in this case differentiate between resume (setup needed) and xruns (setup not needed). This balanced operation was not needed previously but will be required when SOF dynamic pipelines are enabled. Co-developed-by:
Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Signed-off-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Reviewed-by:
Bard Liao <bard.liao@intel.com> Reviewed-by:
Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com> Reviewed-by:
Péter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://lore.kernel.org/r/20211004171430.103674-6-pierre-louis.bossart@linux.intel.comSigned-off-by:
Mark Brown <broonie@kernel.org>
-
Bard Liao authored
Add two clks_control bits. MCLK and/or BCLK will start during hw_params and stop during hw_free if the corresponding bit is set. While the kernel does not do anything with these bitfields, this is also tagged as part of the ABI 3.19 changes. Signed-off-by:
Bard Liao <yung-chuan.liao@linux.intel.com> Signed-off-by:
Brent Lu <brent.lu@intel.com> Link: https://lore.kernel.org/r/20211004171430.103674-5-pierre-louis.bossart@linux.intel.comSigned-off-by:
-
Pierre-Louis Bossart authored
The DAI_CONFIG is used for both hw_params and hw_free. Use flags to specify what stage the configuration applies to. the DAI_CONFIG IPC may be sent also during the widget setup so each flag is cleared after the IPC to restore the state. Signed-off-by:
-
Pierre-Louis Bossart authored
Mirror changes done in SOF tree. The changes do not rely on BIT/GENMASK on purpose to keep the structure and flags common with the firmware tree. The DAI_CONFIG IPC is currently used in multiple ways. It is sent to the DSP firmware when enabling static or dynamic pipelines, in hw_params or prepare callbacks for Intel SSP, HDaudio and ALH, on trigger_stop and hw_free. This IPC has been abused a bit in the past, i.e. the values used for some of the DAI-specific fields are used to either allocate or free resources. Two typical examples are Intel HDaudio and SoundWire/ALH DAIs, where using a zero DMA channel number or stream tag signals to the firmware the DMA channels or tags allocated earlier can be freed. Rather than add a new IPC for 'hw_params' and 'hw_free', this patch suggests supporting a 2-bit value conveying the 'stage' information in an existing IPC structure. Only 3 possible values are used. The mapping between HW_PARAMS and HW_FREE flags and ALSA definitions is not strictly 1:1, e.g. in some cases the HW_PARAM flag might be set during the .prepare callback, while the HW_FREE might be sent during the ALSA .trigger for stop/suspend. The semantics of the flags is to reserve and start/stop all needed resources, typically hardware related such as DMAs or clocks, when the HW_PARAMS is set, while the HW_FREE flag allows the firmware to release the resources allocated. The data transfers are still controlled within the firmware through the propagation of the trigger command. The driver can then pass information that the DAI_CONFIG was invoked in e.g. a pipeline/DAI setup, hw_params or hw_free stage without having to use a special DAI-specific encoding. Unfortunately we can't remove old encodings due to backwards-compatibility requirements but for new cases, such as the SSP in follow-up patches, we can make the IPC less cryptic. This change is tagged as ABI 3.19 and is completely backwards compatible. Signed-off-by:
-
Pierre-Louis Bossart authored
This was added in ABI 3.17 but never added to the kernel tree. The group_id is not currently used but this patch is required before additional changes. Signed-off-by:
-
- 04 Oct, 2021 7 commits
-
-
Kunihiko Hayashi authored
To eliminate FIXME, add descriptions of what each port number is associated with. Cc: Rob Herring <robh@kernel.org> Signed-off-by:
Kunihiko Hayashi <hayashi.kunihiko@socionext.com> Link: https://lore.kernel.org/r/1633313546-23258-1-git-send-email-hayashi.kunihiko@socionext.comSigned-off-by:
-
Hans de Goede authored
Set the card.components string using the new nau8824_components() helper which returns a components string based on the DMI quirks inside the nau8824 codec driver. Signed-off-by:
Hans de Goede <hdegoede@redhat.com> Link: https://lore.kernel.org/r/20211002211459.110124-4-hdegoede@redhat.comSigned-off-by:
-
Hans de Goede authored
Some devices using the NAU8824 have only one speaker. To still have things working properly this requires the left + right channels to both be mixed to the left speaker output. This mixer setup is done by userspace based on UCM profiles. But this requires userspace to know that there is a mono-speaker. Add a helper function (for the machine driver) to get a components string providing this info. This is done inside the codec driver because the codec driver already has a DMI quirk table. Signed-off-by:
-
Hans de Goede authored
Add a quirk mechanism to allow specifying that active-high jack-detection should be used on platforms where this info is not available in devicetree. And add an entry for the Cyberbook T116 tablet to the DMI table, so that jack-detection will work properly on this tablet. Signed-off-by:
-
Hans de Goede authored
The NAU8824_JACK_LOGIC define was wrong, for active high jack-detect to work bit 1 needs to be set, rather then bit 0. The correct bit was found in the Android kernel source dump for a Cyberbook T116 tablet; and this was also tested on that same tablet. Signed-off-by:
-
Hans de Goede authored
Use the new IRQF_NO_AUTOEN flag when requesting the IRQ, rather then disabling it immediately after requesting it. This fixes a possible race where the IRQ might trigger between requesting and disabling it; and this also leads to a small code cleanup. Signed-off-by:
-
Hans de Goede authored
Use the new IRQF_NO_AUTOEN flag when requesting the IRQ, rather then disabling it immediately after requesting it. This fixes a possible race where the IRQ might trigger between requesting and disabling it; and this also leads to a small code cleanup. Signed-off-by:
-
- 02 Oct, 2021 1 commit
-
-
Seven Lee authored
The driver is for codec NAU88L21 of Nuvoton Technology Corporation. The NAU88L21 is an ultra-low power high performance audio codec that supports both analog and digital audio functions. Signed-off-by:
Seven Lee <wtli@nuvoton.com> Link: https://lore.kernel.org/r/20211001103108.3297848-1-wtli@nuvoton.comSigned-off-by:
-
- 01 Oct, 2021 21 commits
-
-
Mark Brown authored
Merge series "Add support for on demand pipeline setup/destroy" from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>: Hi, The previous, v2 of this series was sent by Daniel Baluta: https://lore.kernel.org/alsa-devel/20210917143659.401102-1-daniel.baluta@oss.nxp.com/ We have agreed that it might be better that someone from Intel is going to take it from here as we already have the infrastructure up to test and verify the dynamic pipelines support. Changes since v2 (sent by Daniel Baluta): - patch 10: Fix NULL point dereference in hda_dai_update_config() - I have kept Daniel's SoB for the series. Changes since v1: - Signed-off-by tag added by Daniel This series implements initial support for dynamic pipelines to setup/teardown pipeline as needed when a PCM is open/closed. Initially dynamic pipelines are only supported with single core setup which will be expanded with a follow-up series. Review with SOF community at https://github.com/thesofproject/linux/pull/2794 The feature has been merged on 1st of April to sof-dev, all issues found since has been fixed and squashed to this upstream series. Regards, Peter --- Ranjani Sridharan (12): ASoC: topology: change the complete op in snd_soc_tplg_ops to return int ASoC: SOF: control: Add access field in struct snd_sof_control ASoC: SOF: topology: Add new token for dynamic pipeline ASoC: SOF: sof-audio: add helpers for widgets, kcontrols and dai config set up AsoC: dapm: export a couple of functions ASoC: SOF: Add new fields to snd_sof_route ASoC: SOF: restore kcontrols for widget during set up ASoC: SOF: Don't set up widgets during topology parsing ASoC: SOF: Introduce widget use_count ASoC: SOF: Intel: hda: make sure DAI widget is set up before IPC ASoC: SOF: Add support for dynamic pipelines ASoC: SOF: topology: Add kernel parameter for topology verification include/sound/soc-dpcm.h | 1 + include/sound/soc-topology.h | 2 +- include/uapi/sound/sof/tokens.h | 1 + sound/soc/intel/skylake/skl-topology.c | 6 +- sound/soc/soc-dapm.c | 2 + sound/soc/soc-pcm.c | 4 +- sound/soc/soc-topology.c | 10 +- sound/soc/sof/intel/hda-dai.c | 174 +++--- sound/soc/sof/intel/hda.c | 177 ++++-- sound/soc/sof/intel/hda.h | 5 + sound/soc/sof/ipc.c | 22 + sound/soc/sof/pcm.c | 58 +- sound/soc/sof/pm.c | 4 +- sound/soc/sof/sof-audio.c | 709 +++++++++++++++++++------ sound/soc/sof/sof-audio.h | 32 +- sound/soc/sof/sof-priv.h | 1 + sound/soc/sof/topology.c | 362 +++++-------- 17 files changed, 1032 insertions(+), 538 deletions(-) -- 2.33.0
-
Viorel Suman authored
Implement SPDIF bypass mode. It implies internal SoC routing of SPDIF input signal to SPDIF output signal. The test bed requires two boards: B1 configured in bypass mode, and B2 to feed B1 SPDIF RX port and read B1 SPDIF TX port: B2 TX -> B1 RX, B2 RX <- B1 TX. The test procedure: a) Boot both boards b) B2: start "arecord <spdifcard> -r 48kHz | aplay <local DAC>" c) B2: start "aplay <spdifcard> -r 48kHz <2ch 48kHz audio file>" d) B1: enable bypass mode: amixer -cimxspdif cset numid=8,iface=PCM,name='Bypass Mode' on e) B2: check DAC audio, make sure the same sample rate is used at steps b) and c), in example above the rate is 48kHz. f) B1: try to run "aplay" or "arecord" on imxspdif card while in bypass mode - both must fail until bypass mode is disabled g) B1: disable bypass mode: amixer -cimxspdif cset numid=8,iface=PCM,name='Bypass Mode' off h) B1: check the usual playback and capture on imxspdif card. During this test try to set bypass mode - must not be allowed while playback or capture is running. Signed-off-by:
Viorel Suman <viorel.suman@nxp.com> Signed-off-by:
Shengjiu Wang <shengjiu.wang@nxp.com> Link: https://lore.kernel.org/r/1632649760-1651-1-git-send-email-shengjiu.wang@nxp.comSigned-off-by:
-
Ranjani Sridharan authored
Add a kernel debug flag to enable a one-shot topology verification for all pipelines including the dynamic ones. If the debug flag is set, all the topology component loading will be verified during the complete callback. Signed-off-by:
Kai Vehmanen <kai.vehmanen@linux.intel.com> Signed-off-by:
Daniel Baluta <daniel.baluta@nxp.com> Signed-off-by:
-
Ranjani Sridharan authored
Add support for dynamic pipelines by modifying the PCM hw_params ioctl implementation to determine the widgets required for a PCM stream by querying the list of connected DAPM widgets. This list is saved as part of snd_sof_pcm_stream struct and will be used to setup the widgets. The sof_widget_list_setup/free routines setup and free connected DAPM widgets when a PCM is opened/closed. These routines accept a list of connected DAPM widgets as input and determine the SOF widgets, their corresponding pipeline widgets and connections between them that need to be setup before the PCM is triggered. Please note that the dynamic pipeline feature will only be enabled for those pipelines whose dynamic_pipeline_widget flag is set in topologies. Add a new token called SOF_TKN_SCHED_DYNAMIC_PIPELINE that when set in topology will be applied to the dynamic_pipeline_widget flag of the pipeline widget. Signed-off-by:
-
Ranjani Sridharan authored
With the implementation of the dynamic pipeline feature, widgets will only be setup when a PCM is opened during the hw_params ioctl. The BE hw_params callback is responsible for sending the DAI_CONFIG for the DAI widgets in the DSP. With dynamic pipelines, the DAI widgets will need to set up first before sending the DAI_CONFIG IPC in the BE hw_params. Update the BE hw_params/hw_free callbacks for all ALH, HDA and SSP DAIs to set up/free the DAI widget before/after DAI_CONFIG IPC. Signed-off-by:
-
Ranjani Sridharan authored
Add a new field, use_count to struct snd_sof_widget to keep track of the usage count for each widget. Since widgets can belong to multiple pipelines, this field will ensure that the widget is setup only when the first pipeline that needs it is started and freed when the last pipeline that needs it is stopped. There is no need to protect the widget use_count access as the core already handles mutual exclusion at the PCM level. Add a new helper sof_widget_free() to handle freeing the SOF widgets and export the sof_widget_setup/free() functions. Signed-off-by:
-
Ranjani Sridharan authored
In preparation for supporting dynamic pipelines, move the widget setup, DAI config IPCs to the complete callback during topology loading. For current topology where all the pipelines are static, all the pipelines will be set up during complete. For topologies with dynamic and static pipelines, this will enable setting up only the static ones during topology loading. Reuse the sof_restore_pipelines() function for this purpose and rename it to sof_set_up_pipelines(). Signed-off-by:
-
Ranjani Sridharan authored
Restore kcontrols for each widget after it has been set up successfully. Signed-off-by:
Seppo Ingalsuo <seppo.ingalsuo@intel.com> Signed-off-by:
-
Ranjani Sridharan authored
Add two new fields to save the source widget and sink widget pointers in struct snd_sof_route to make it easier to look up routes by source/sink widget. Also, add a flag to indicate if the route has been set up in the DSP. These will be used when the dynamic pipeline feature is implemented and routes will have to be set up at run time. Also, add a new sof_tear_down_pipelines() callback, that will used to reset the set up status for all routes during suspend. Signed-off-by:
-
Ranjani Sridharan authored
Export a couple of DAPM functions that can be used by ASoC drivers to determine connected widgets when a PCM is started. Signed-off-by:
-
Ranjani Sridharan authored
Refactor the existing code to use helper functions to set up/free widgets, send dai config and set up kcontrols for widgets. These will be reused later on for setting up widgets in the connected DAPM widgets list for a particular PCM when the dynamic pipeline feature is implemented. Signed-off-by:
-
Ranjani Sridharan authored
Today, we set up all widgets required for all PCM streams at the time of topology parsing even if they are not used. An optimization would be to only set up the widgets required for currently active PCM streams. This would give the FW the opportunity to power gate unused memory blocks, thereby saving power. For dynamic pipelines, the widgets in the connected DAPM path for each PCM will need to be set up at runtime. This patch introduces a new token, DYNAMIC_PIPELINE, for scheduler type widgets that indicate whether a pipeline should be set up statically during topology load or at runtime when the PCM is opened. Introduce a new field called dynamic_pipeline_widget in struct snd_sof_widget to save the value of the parsed token. The token is set only for the pipeline (scheduler type) widget and must be propagated to all widgets in the same pipeline during topology load. Introduce another field called pipe_widget in struct snd_sof_widget that saves the pointer to the scheduler widget with the same pipeline ID as that of the widget. This field is populated when the pipeline completion callback is invoked during topology loading. Signed-off-by:
-
Ranjani Sridharan authored
Add a new field to save the access setting for all controls in struct snd_sof_control. This will be used to ensure that only widgets belonging to static pipelines have volatile controls. Signed-off-by:
-
Ranjani Sridharan authored
In the SOF driver, the operations performed in the complete callback can fail and therefore topology loading should return an error in such cases. So, change the signature of the complete op in struct snd_soc_tplg_ops to return an int to return the error. Also, amend the complete callback functions in the SOF driver and the SKL driver to conform with the new signature. Signed-off-by:
-
Rikard Falkeborn authored
The struct iapq8096_ops is only assigned to the ops field in the snd_soc_dai_link struct which is a pointer to const struct snd_soc_ops. Make it const to allow the compiler to put it in read-only memory. Signed-off-by:
Rikard Falkeborn <rikard.falkeborn@gmail.com> Link: https://lore.kernel.org/r/20211001115030.10402-1-rikard.falkeborn@gmail.comSigned-off-by:
-
Derek Fang authored
When the system resumes from S3, if the system plays a beep, there is continuous "Zizi.." noise from HP that could be heard. It is caused by the SAR mode switch during the combo jack re-detection which be executed parallelly in a workqueue after the system resumes. This patch changes the behavior of SAR mode switch to avoid this issue. Signed-off-by:
Derek Fang <derek.fang@realtek.com> Link: https://lore.kernel.org/r/20211001074113.2223-2-derek.fang@realtek.comSigned-off-by:
-
Derek Fang authored
When codec's ASRC was enabled, the DA and AD filters use the reference clk which tracks the I2S clks. And if the I2S clks' timing of open and close are not expected, this patch allows the filters to switch to use sysclk if ASRC's ref clks disappeared and could fix the below possible issues: 1. Avoid DA filter to keep surplus samples. 2. Avoid that AD filter works failed during dapm's power on. For example, if I2S clks were closed before dacdat during playback off due to jack unplug, it causes ref clks disappeared and DA filter remained some samples which will produce pop noise on the next HP playback. ASRC auto-disable could clear the samples during the playback off. Signed-off-by:
-
Trevor Wu authored
platforms->of_node and codes->of_node are assigned in probe function, and of_node_put is called at the end of probe function, because of_node seems to be not used after probe functon In this patch, of_node_put is moved to platform remove function in case of_node is used at any occasions after probe function in the future. Signed-off-by:
Trevor Wu <trevor.wu@mediatek.com> Link: https://lore.kernel.org/r/20211001031601.3953-1-trevor.wu@mediatek.comSigned-off-by:
-
Malik_Hsu authored
Add a board config adl_mx98360a_rt5682 to support alc5682 headset codec and max98360a speaker amplifier. Follow Intel BT offload design by connecting alc5682 to SSP0 and max98360a to SSP1. Signed-off-by:
Malik_Hsu <malik_hsu@wistron.corp-partner.google.com> Signed-off-by:
-
Rikard Falkeborn authored
The struct tegra_machine_snd_ops is only assigned to the ops field in the snd_soc_dai_link struct which is a pointer to const struct snd_soc_ops. Make it const to allow the compiler to put it in read-only memory. Signed-off-by:
-
Simon Trimmer authored
This simplification allows the use of the standard kernel pattern of static inline dummy functions for debugfs code. Most systems will only have a small number of snd_soc_components so the memory impact is minimal. Signed-off-by:
Simon Trimmer <simont@opensource.cirrus.com> Suggested-by:
-
- 30 Sep, 2021 6 commits
-
-
Jack Yu authored
The DAI clock is only used in I2S mode, to make it clear and to fix clock resource release issue, we move CCF clock related code to rt5682_i2c_probe to fix clock register/unregister issue. Signed-off-by:
Jack Yu <jack.yu@realtek.com> Link: https://lore.kernel.org/r/20210929054344.12112-1-jack.yu@realtek.comSigned-off-by:
-
Shengjiu Wang authored
On i.MX8ULP the audio interface and codec are controlled by Cortex-M domain, Cortex-M core provides audio service over rpmsg. The rpmsg audio function is almost same as i.MX7ULP platform, so share same configuration. Signed-off-by:
-
Guo Zhengkui authored
Remove a repeated "#include <linux/firmware.h>" in line 32. Signed-off-by:
Guo Zhengkui <guozhengkui@vivo.com> Acked-by:
-
Peter Ujfalusi authored
We have growing number of options under SND_SOC_SOF_TOPLEVEL as SOF adaptation is growing (Intel, NXP, AMD and Mediatek) and new features are added. It will make the menuconfig user experience much cleaner if we move the SOF options under a separate page. Signed-off-by:
Daniel Baluta <daniel.baluta@gmail.com> Reviewed-by:
-
Derek Fang authored
Fix the device-tree schema errors that be reported by using 'make dt_binding_check'. Signed-off-by:
-
Derek Fang authored
Revise the macro RT5682S_PLLB_SRC_MASK to 0x1 because the mux is only two-source. Signed-off-by:
-
