| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Merge series from Krzysztof Kozlowski <[email protected]>:
Improve a bit the Qualcomm LPASS RX macro driver and align similar parts
of code with LPASS WSA macro driver for consistency.
No external dependencies.
|
|
both tas27871 and tas2563
Set "Speaker Force Firmware Load" as the common kcontrol
for both tas27871 and tas2563 and move it into newly-created
tasdevice_snd_controls, and keep the digital gain and analog
gain in tas2781_snd_controls.
Signed-off-by: Shenghao Ding <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
acpi_get_first_physical_node() can return NULL in several cases (no such
device, ACPI table error, reference count drop to 0, etc).
Existing check just emit error message, but doesn't perform return.
Then this NULL pointer is passed to devm_acpi_dev_add_driver_gpios()
where it is dereferenced.
Adjust this error handling by adding error code return.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 02527c3f2300 ("ASoC: amd: add Machine driver for Jadeite platform")
Signed-off-by: Aleksandr Mishin <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Driver does not unregister typec structures (typec_mux_dev and
typec_switch_desc) during removal leading to leaks. Fix this by moving
typec registering parts to separate function and using devm interface to
release them. This also makes code a bit simpler:
- Smaller probe() function with less error paths and no #ifdefs,
- No need to store typec_mux_dev and typec_switch_desc in driver state
container structure.
Cc: [email protected]
Fixes: 10f514bd172a ("ASoC: codecs: Add WCD939x Codec driver")
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Reviewed-by: Neil Armstrong <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
commit c721f189e89c0 ("reset: Instantiate reset GPIO controller for
shared reset-gpios") check if there is no "resets" property
will fallback to "reset-gpios".
So don't need to handle "reset-gpios" separately in the driver,
the "reset-gpios" handler is duplicated with "resets" control handler,
remove it.
Signed-off-by: Shengjiu Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
There is no need to keep separate token list for dai and 'common' copier
token list when the 'common' list is actually the aif list, the
SOF_COPIER_DEEP_BUFFER_TOKENS are not applicable for buffers.
We could have separate lists for all types but it is probably simpler to
just use a single list for all types of copiers. Function specific tokens
will be only parsed by function specific code anyways.
Reviewed-by: Bard Liao <[email protected]>
Reviewed-by: Ranjani Sridharan <[email protected]>
Signed-off-by: Peter Ujfalusi <[email protected]>
Signed-off-by: Pierre-Louis Bossart <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
The current fsl_qmc_audio works in interleaved mode. The audio samples
are interleaved and all data are sent to (received from) one QMC
channel.
Using several QMC channels, non interleaved mode can be easily
supported. In that case, data related to ch0 are sent to (received from)
the first QMC channel, data related to ch1 use the next QMC channel and
so on up to the last channel.
In terms of constraints and settings, the two modes are slightly
different:
- Interleaved mode:
- The sample size should fit in the number of time-slots available
for the QMC channel.
- The number of audio channels should fit in the number of
time-slots (taking into account the sample size) available for the
QMC channel.
- Non-interleaved mode:
- The number of audio channels is the number of available QMC
channels.
- Each QMC channel should have the same number of time-slots.
- The sample size equals the number of time-slots of one QMC
channel.
Add support for the non-interleaved mode allowing multiple QMC channel
per DAI. The DAI switches in non-interleaved mode when more that one QMC
channel is available.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
The QMC audio uses one QMC channel per DAI and uses this QMC channel to
transmit interleaved audio channel samples.
In order to work in non-interleave mode, a QMC audio DAI needs to use
multiple QMC channels. In that case, the DAI maps each QMC channel to
exactly one audio channel.
Allow QMC audio DAIs with multiple QMC channels attached.
Signed-off-by: Herve Codina <[email protected]>
Reviewed-by: Rob Herring (Arm) <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
No function in the QMC API is available to get the number of phandles
present in a phandle list.
Fill this lack introducing qmc_chan_count_phandles().
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
qmc_chan_get_byphandle() and the resource managed version retrieve a
channel from a simple phandle.
Extend the API and introduce qmc_chan_get_byphandles_index() and the
resource managed version in order to retrieve a channel from a phandle
list using the provided index to identify the phandle in the list.
Also update qmc_chan_get_byphandle() and the resource managed version to
use qmc_chan_get_byphandles_index() and so avoid code duplication.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Constraints are set by qmc_dai_startup(). These constraints are specific
to the interleaved mode.
With the future introduction of support for non-interleaved mode, a new
set of constraints will be set. To make the code clear and keep
qmc_dai_startup() simple, extract the current interleaved mode
constraints settings to a specific function.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Submitting data to QMC channels is done in several places: transfer
completions and DAI start. The operation done is simple and consist in
one function call.
With the future introduction of support for non-interleaved mode,
submitting data will be more complex.
To avoid copy/paste of code in several places, introduce
qmc_audio_pcm_{read,write}_submit() whose goal is to handle this
data submission.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
routines
The current QMC audio driver uses only one QMC channel per DAI. The
context used by QMC channel transfer (read and write) completion
routines does not contains any QMC channel and the only one available
per DAI is used to schedule the next transfer.
This works pretty well with only one QMC channel per DAI.
The future support for non-inlerleave mode will use several QMC channel
per DAI. In that case, QMC channel transfer completion routines need to
identify the QMC channel related to the completion.
In order to fill this lack, even if identifying the current QMC channel
among several QMC channels is not needed for the current code, add one
indirection level and introduce the qmc_dai_chan data structrure.
This structure contains the QMC channel involved in the completion and
refererences to the runtime context (capture and playback) used by the
DAI.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
The driver mixes some internal values for channel DMA buffer handling
and PCM pointer handling. In the currently supported interleaved mode,
this mix does not lead to any issues but in order to prepare the
support for the non-interleaved mode, having them clearly separated will
ease the support and avoid additional computation to convert values used
in channel DMA buffer management in values usable for PCM pointer.
Use a specific set of variable for PCM pointer handling and an other set
for channel DMA buffer.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
./scripts/checkpatch.pl --strict --codespell detected several issues
when running on the fsl_qmc_audio.c file:
- CHECK: spaces preferred around that '*' (ctx:VxV)
- CHECK: Alignment should match open parenthesis
- CHECK: Comparison to NULL could be written "!prtd"
- CHECK: spaces preferred around that '/' (ctx:VxV)
- CHECK: Lines should not end with a '('
- CHECK: Please don't use multiple blank lines
Some of them are present several times.
Fix all of these issues without any functional changes.
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
devm_kasprintf() can return a NULL pointer on failure but this returned
value is not checked.
Fix this lack and check the returned value.
Fixes: 075c7125b11c ("ASoC: fsl: Add support for QMC audio")
Cc: [email protected]
Signed-off-by: Herve Codina <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
MAX_BDL_ENTRIES
The HDaudio specification Section 3.6.2 limits the number of BDL entries to 256.
Make sure we don't allow more periods than this normative value.
Signed-off-by: Peter Ujfalusi <[email protected]>
Signed-off-by: Pierre-Louis Bossart <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Those registers will be used when JD source is RT711_JD2_1P8V_1PORT.
Signed-off-by: Bard Liao <[email protected]>
Reviewed-by: Pierre-Louis Bossart <[email protected]>
Acked-by: Shuming Fan <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
When system enters suspend with an active stream, SOF core
calls hw_params_upon_resume(). On Intel platforms with HDA DMA used
to manage the link DMA, this leads to call chain of
hda_dsp_set_hw_params_upon_resume()
-> hda_dsp_dais_suspend()
-> hda_dai_suspend()
-> hda_ipc4_post_trigger()
A bug is hit in hda_dai_suspend() as hda_link_dma_cleanup() is run first,
which clears hext_stream->link_substream, and then hda_ipc4_post_trigger()
is called with a NULL snd_pcm_substream pointer.
Fixes: 2b009fa0823c ("ASoC: SOF: Intel: hda: Unify DAI drv ops for IPC3 and IPC4")
Link: https://github.com/thesofproject/linux/issues/5080
Reviewed-by: Bard Liao <[email protected]>
Signed-off-by: Kai Vehmanen <[email protected]>
Signed-off-by: Pierre-Louis Bossart <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Driver's probe() has two allocations which are needed only within the
probe() itself - for devm_regmap_init_mmio().
Usage of devm interface is a bit misleading here, because these can be
freed right after each scope finishes.
This makes the code a bit more obvious and self documenting.
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-6-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
Driver uses ARRAY_SIZE() to get number of widgets later passed to
snd_soc_dapm_new_controls(), which is an 'unsigned int'.
Reviewed-by: Dmitry Baryshkov <[email protected]>
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-5-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
Number of widgets in array passed to snd_soc_dapm_new_controls() cannot
be negative, so make it explicit by using 'unsigned int', just like
snd_soc_add_component_controls() is doing.
Reviewed-by: Dmitry Baryshkov <[email protected]>
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-4-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
The driver has static 'struct regmap_config', which is then customized
depending on device version. This works fine, because there should not
be two devices in a system simultaneously and even less likely that such
two devices would have different versions, thus different regmap config.
However code is cleaner and more obvious when static data in the driver
is also const - it serves as a template.
Mark the 'struct regmap_config' as const and duplicate it in the probe()
with kmemdup to allow customizing per detected device variant.
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-3-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
Allocate the default register values array with scoped/cleanup.h to
reduce number of error paths and make code a bit simpler.
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-2-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
Eliminate PDS cleanup by using devm_add_action_or_reset() which results
in one less error path and smaller cleanup in remove().
Signed-off-by: Krzysztof Kozlowski <[email protected]>
Link: https://patch.msgid.link/20240701-b4-qcom-audio-lpass-codec-cleanups-v3-1-6d98d4dd1ef5@linaro.org
Signed-off-by: Mark Brown <[email protected]>
|
|
Switch the driver to use GPIO descriptors.
Notice that we let the gpiolib handle line inversion for the
active low reset line (nreset !reset).
There are no upstream device trees using the tas5086 compatible
string, if there were, we would need to ascertain that they all
set the GPIO_ACTIVE_LOW flag on their GPIO lines.
Signed-off-by: Linus Walleij <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`wsa884x_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`wsa883x_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`wsa881x_regmap_config` is not modified and can be declared as const to
move its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`jz4760_codec_regmap_config` is not modified and can be declared as
const to move its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`jz4760_codec_regmap_config` is not modified and can be declared as
const to move its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`cs53l30_regmap` is not modified and can be declared as const to move
its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`cs35l36_regmap` is not modified and can be declared as const to move
its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`cs35l35_regmap` is not modified and can be declared as const to move
its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
`cs35l34_regmap` is not modified and can be declared as const to move
its data to a read-only section.
Signed-off-by: Javier Carrasco <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Merge series from Richard Fitzgerald <[email protected]>:
These two commits set the upper limit of the Speaker Volume control
to +12dB instead of +100dB.
This should have been a simple 1-line change to the #define in the
header file, but only the HDA cs35l56 driver is using this define.
The ASoC cs35l56 driver was using hardcoded numbers instead of the
header defines.
So the first commit changes the ASoC driver to use the #defined
constants. The second commit corrects the value of the constant.
|
|
In the current flow all interrupts are disabled in runtime suspend
phase. However interrupts enablement only exists in fsl_xcvr_prepare().
After resume fsl_xcvr_prepare() may not be called so it will cause all
interrupts still disabled even if resume from suspend. Interrupts
should be explictily enabled after resume.
Also, DPATH reset setting only exists in fsl_xcvr_prepare(). After
resume from suspend DPATH should be reset otherwise there'll be channel
swap issue.
Signed-off-by: Chancel Liu <[email protected]>
Acked-by: Shengjiu Wang <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Change CS35L56_MAIN_RENDER_USER_VOLUME_MAX to 48, to limit the maximum
value of the Speaker Volume control to +12dB. The minimum value is
unchanged so that the default 0dB has the same integer control value.
The original maximum of 400 (+100dB) was the largest value that can be
mathematically handled by the DSP. The actual maximum amplification is
+12dB.
Signed-off-by: Richard Fitzgerald <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
The "Speaker Volume" control was being defined using four hardcoded magic
numbers. There are #defines in the cs35l56.h header for these numbers, so
change the code to use the defined constants.
Signed-off-by: Richard Fitzgerald <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Replace the open-coded parsing of "reg" with of_property_read_reg().
The #ifdef is also easily replaced with IS_ENABLED().
Signed-off-by: Rob Herring (Arm) <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Replace the open-coded parsing of "reg" with of_property_read_reg().
The #ifdef is also easily replaced with IS_ENABLED().
Signed-off-by: Rob Herring (Arm) <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Merge series from Amadeusz Sławiński <[email protected]>:
Code used to create standalone and widget controls is mostly same, with
with exception that in standalone case dynamic object needs to be
registered and control created directly.
Following patches clean up and unify kcontrol creation code in topology
code.
|
|
Merge series from Maxim Kochetkov <[email protected]>:
nau8824 has external MCLK pin. So add enable/disable external clock
management.
|
|
Convert the RT5650/RT5645 audio CODEC bindings to DT schema.
Reviewed-by: Rob Herring (Arm) <[email protected]>
Signed-off-by: Animesh Agarwal <[email protected]>
Cc: Daniel Baluta <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Code used to create standalone and widget enum control is same, with
exception that in standalone case dynamic object needs to be registered
and control created directly.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Code used to create standalone and widget mixer control is same, with
exception that in standalone case dynamic object needs to be registered
and control created directly.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
Code used to create standalone and widget bytes control is same, with
exception that in standalone case dynamic object needs to be registered
and control created directly.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
In preparation for following cleanups move functions around.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
In following patches it will be reused to also create standalone
kcontrol, so it makes sense to name it in more generic way.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|
|
soc_tplg_dbytes_create() missed se->dobj.index initialization, so add it
there. Additionally separate dynamic object initialization into separate
logical block code.
Signed-off-by: Amadeusz Sławiński <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Mark Brown <[email protected]>
|