Merge series from Cristian Ciocaltea <cristian.ciocaltea@collabora.com>:
This patch series handles a few issues related to the ES8316 audio
codec, discovered while doing some testing on the Rock 5B board.
In the case of IPC4, a pipeline is only paused during STOP/PAUSE/SUSPEND
triggers and the FW keeps the host DMA running when a pipeline is
paused. The start/stop tests iterate through STOP/START triggers without
involving a hw_free. This means that the pipeline state will only toggle
between PAUSED (during the STOP trigger) and RUNNING (during the START
trigger). So this test should be treated in the same way as a
PAUSE_PUSH/PAUSE_RELEASE test and the DMA should be kept running when
toggling the pipeline states between PAUSED and RUNNING.
Since there is no way to tell if a STOP trigger will be followed by hw_free
or not, this patch proposes to always skip DMA stop during the STOP trigger
and handle it later during hw_free. Introduce a new flag in struct
sof_ipc_pcm_ops, delayed_platform_trigger, that will be used to ensure that
the host DMA will not be stopped during the STOP/PAUSE/RELEASE triggers
and set it for IPC4. The platform_trigger call to stop the DMA will be
invoked during PCM hw_free instead when the pipeline is reset.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230420114137.27613-3-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
The recommended sequence for triggering the host DMA is to first program
the DMA in the FW before setting the RUN bit to start the stream in the
host. With IPC3, this sequence is honored because the FW programs the
DMA when the HW_PARAMS IPC is sent during PCM hw_params and then the host
sets the RUN bit during sof_pcm_trigger(). But with IPC4,
sof_pcm_trigger() sends the SET_PIPELINE_STATE IPC to program the DMA in
the FW after the DMA RUN bit is set.
In order to minimize the impact for IPC3, introduce a new flag as part
of struct sof_ipc_pcm_ops, ipc_first_on_start, which will be set for IPC4
only. With this flag set, the SET_PIPELINE_STATE IPC will be sent before
the DMA RUN bit is set by the host during the START/PAUSE_RELEASE
triggers.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230322094346.6019-4-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
IPC3 and IPC4 have different requirements for the order in which the FE
CPU and BE CPU DAI trigger callbacks must be invoked. With a regular PCM
start/stop, pipeline widgets are set up during hw_params and freed
during hw_free.
But when the system is suspended when a PCM is running,
pipeline widgets are freed during the SUSPEND trigger callback for the
FE CPU DAI. In order to avoid freeing the pipeline widgets before the BE
CPU DAI trigger is executed, the trigger order was modified in previous
contributions in the PCM dai_link_fixup callback to make sure that the BE
CPU DAI trigger stop/suspend is always invoked before the FE CPU DAI
trigger. But this contradicts the firmware requirement for IPC4 w.r.t.
ordering of pipeline triggers.
So, remove the freeing of pipeline widgets during FE CPU DAI suspend
trigger and handle it during system suspend when the
tear_down_all_pipelines() IPC op is invoked. This will be followed up
with a patch to fix the trigger order for IPC4.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230127120031.10709-6-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
From current design in sof_machine_check and snd_sof_new_platform_drv,
the SOF can only support ACPI type machine.
1. In sof_machine_check if there is no ACPI machine exist, the function
will return -ENODEV directly, that's we don't expected if we do not
base on ACPI machine.
2. In snd_sof_new_platform_drv the component driver need a driver name
to do ignore_machine, currently the driver name is obtained from
machine->drv_name, and the type of machine is snd_soc_acpi_mach.
So we add a new function named sof_of_machine_select that we can pass
sof_machine_check and obtain info required by snd_sof_new_platform_drv.
Signed-off-by: Chunxu Li <chunxu.li@mediatek.com>
Link: https://lore.kernel.org/r/20220805070449.6611-2-chunxu.li@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Before initiating IPC and/or bus transactions when loading the
topology during a component probe, which happens on card
registration/creation, make sure the device for the SOF driver is
pm_runtime active.
The SOF probe is not necessarily followed by the component probe, such
a timing assumption can be broken in driver bind/unbind tests. This
can be artifially shown if the module for the machine driver is
'blacklisted' and the SOF device becomes pm_runtime_suspended before
manually calling modprobe to register the card.
In an initial experiment, pm_resume_and_get() was called from
soc-component.c, since the current ASoC component model is arguably
missing dependencies between component status and device
status. However this approach proved too invasive and breaks all
existing HDMI playback solutions on Intel platforms.
While this will result in duplication of code, generating pm_runtime
transitions only if strictly required for a given component makes more
sense overall. This patch adds the pm_runtime resume transition for
SOF only.
BugLink: https://github.com/thesofproject/linux/issues/3651
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220616210825.132093-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
In order to set up a pipeline with IPC4, the total memory usage for the
pipeline needs to be calculated based on the list of connected widgets.
Add a new ipc_prepare() op to struct sof_ipc_tplg_widget_ops that will be
used to calculate the memory usage for each widget in the pipelines
associated with a PCM and prepare the widget for getting set up in the
DSP. The prepare step will be used to allocate memory for the IPC
payload, assign instance ID and update the config data for the widget
based on the runtime PCM params. Once prepared, the setup step is used
to send the IPC to create the widget in the DSP.
Add an ipc_unprepare() op to unprepare the widget i.e free the memory
allocated during prepare, free the instance ID etc. This should be
invoked after the widget is freed.
A new flag "prepared" is added to struct snd_sof_widget to track the
prepared status of widgets.
Also, IPC4 requires the platform_params and the runtime PCM params in
order to prepare a widget for set up. So modify the signature of
sof_pcm_setup_connected_widgets() and sof_widget_list_setup() to accept
these as arguments.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220426171743.171061-12-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Add a new client driver for probes support and move
all the probes-related code from the core to the
client driver.
The probes client driver registers a component driver
with one CPU DAI driver for extraction and creates a
new sound card with one DUMMY DAI link with a dummy codec
that will be used for extracting audio data from specific
points in the audio pipeline.
The probes debugfs ops are based on the initial
implementation by Cezary Rojewski and have been moved
out of the SOF core into the client driver making it
easier to maintain. This change will make it easier
for the probes functionality to be added for all platforms
without having the need to modify the existing(15+) machine
drivers.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220210150525.30756-10-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Merge series from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:
The use of the SPIB register helps reduce power consumption - though
to a smaller degree than DMI_L1. This hardware capability is however
incompatible with userspace-initiated rewinds typically used by
PulseAudio.
In the past (2015..2017) Intel suggested an API extension to let
applications disable rewinds. At the time the feedback was that such a
capability was too Intel-specific and SPIB remained unused except for
loading DSP code. We now see devices with smaller batteries being
released, and it's time to revisit Linux support for SPIB to extend
battery life.
In this update the rewinds are disabled via an opt-in kernel
parameter. In the previous reviews, there was consensus that a Kconfig
option was too complicated for distributions to set, and we are
missing a TBD API to expose such capabilities to user-space.
The debate on whether or not to use rewinds, and the impact of
disabling rewinds, will likely be closed when Intel releases the
'deep-buffer' support, currently under development [2][3]. With this
solution, rewinds will not be needed, ever. When an application deals
with content that is not latency-sensitive (e.g. music playback), it
will be able to reduce power consumption by selecting a different PCM
device with increased buffering capabilities. Low-latency streams
will be handled by the 'regular' path. In other words, the impossible
compromise between power and latency will be handled with different
PCM devices/profiles for the same endpoint, and we can push the design
of capability negotiation to a later time when all the building blocks
(firmware topology, kernel, userspace) are ready - we still have
firmware xruns, DPCM race conditions to solve, and a need to describe
these alternate PCM devices with UCM using 'modifiers'.
This patch prepares the introduction of the compress API with SOF.
After each fragment is accepted by the DSP we need to inform
the userspace applications that they can send the next fragment.
This is done via snd_compr_fragment_elapsed.
Similar with the PCM case, in order to avoid sending an IPC before
the previous IPC is handled we need to schedule a delayed work to
call snd_compr_fragment_elapsed().
See snd_sof_pcm_period_elapsed.
To sum up this patch offers the following API to SOF code:
* snd_sof_compr_init_elapsed_work
* snd_sof_compr_fragment_elapsed
Note that implementation for compressed function is in a new file
selected via CONFIG_SND_SOC_SOF_COMPRESS invisible config option.
This option is automatically selected for platforms that support
the compress interface. For now only i.MX8 platforms support this.
For symmetry we introduce snd_sof_pcm_init_elapsed_work to setup
the work struct for PCM case.
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Signed-off-by: Bud Liviu-Alexandru <budliviu@gmail.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20211004152147.1268978-5-daniel.baluta@oss.nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>
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 <ranjani.sridharan@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20210927120517.20505-12-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
