Merge series "ASoC: SOF: Intel: add SoundWire support" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

This patchset provides the support for SoundWire support on Intel
CometLake, IcelLake and TigerLake RVP platforms and form-factor
devices to be released 'soon'.

The bulk of the code is about detecting a valid SoundWire
configuration from ACPI, and implementing the interfaces suggested in
'[PATCH 0/8] soundwire: remove platform devices, add SOF interfaces'
for interrupts, PCI wakes and clock-stop configurations.

Since that SoundWire series will not be in 5.7, the build support for
SOF w/ SoundWire is not provided for now, and fall-back functions will
be used. This code is tested on a daily basis in the SOF tree and is
not expected to change in significant ways.

Changes since v2:
Corrected error in ACPI table (thanks Amadeusz)
Added patch 11 to add reset cycle required on some SoundWire platforms

Bard Liao (1):
  ASoC: SOF: Intel: hda: merge IPC, stream and SoundWire interrupt
    handlers

Pierre-Louis Bossart (8):
  ASoC: soc-acpi: expand description of _ADR-based devices
  ASoC: SOF: Intel: add SoundWire configuration interface
  ASoC: SOF: IPC: dai-intel: move ALH declarations in header file
  ASoC: SOF: Intel: hda: add SoundWire stream config/free callbacks
  ASoC: SOF: Intel: hda: initial SoundWire machine driver autodetect
  ASoC: SOF: Intel: hda: disable SoundWire interrupts on suspend
  ASoC: SOF: Intel: hda: add parameter to control SoundWire clock stop
    quirks
  ASoC: SOF: Intel: hda-ctrl: add reset cycle before parsing
    capabilities

Rander Wang (2):
  ASoC: SOF: Intel: hda: add WAKEEN interrupt support for SoundWire
  Asoc: SOF: Intel: hda: check SoundWire wakeen interrupt in irq thread

 include/sound/soc-acpi.h                      |  39 +-
 include/sound/sof/dai-intel.h                 |  18 +-
 .../intel/common/soc-acpi-intel-cml-match.c   |  87 +++-
 .../intel/common/soc-acpi-intel-icl-match.c   |  97 ++++-
 .../intel/common/soc-acpi-intel-tgl-match.c   |  49 ++-
 sound/soc/sof/intel/hda-ctrl.c                |  25 +-
 sound/soc/sof/intel/hda-dsp.c                 |   2 +
 sound/soc/sof/intel/hda-loader.c              |  31 ++
 sound/soc/sof/intel/hda.c                     | 400 ++++++++++++++++++
 sound/soc/sof/intel/hda.h                     |  66 +++
 10 files changed, 750 insertions(+), 64 deletions(-)

--
2.20.1

ASoC: rt5682: move DAI clock registry to I2S mode

The SoundWire mode doesn't need the DAI clocks.
Therefore, the DAI clock registry moves to I2S mode case.

Signed-off-by: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200327073849.18291-1-shumingf@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pxa: magician: convert to use i2c_new_client_device()

Move away from the deprecated API and return the shiny new ERRPTR where
useful.

Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Link: https://lore.kernel.org/r/20200326211010.13471-2-wsa+renesas@sang-engineering.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda-ctrl: add reset cycle before parsing capabilities

Without this cycle, HDaudio capability parsing fails on some devices.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-12-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Asoc: SOF: Intel: hda: check SoundWire wakeen interrupt in irq thread

If pci device is in D0, wakeen interrupt will be
aggregated at cAVS level as interrupt. This commit
check the wakeen status and process it in irq thread

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-11-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add WAKEEN interrupt support for SoundWire

When a SoundWire link is in clock stop state, a Slave device may wake
up the Master for some events such as jack detection. The WAKEEN
interrupt will be triggered and processed by the audio pci device.

If audio device is in D3, the interrupt will be routed to PME, or
aggregated at cAVS level as interrupt when audio device is in D0. This
patch only supports D3 case, where the audio pci device will be
resumed by a PME event and the WAKEEN interrupt will be processed
after audio pci device is powered up and ROM is initialized
successfully.

The WAKEEN handling is only enabled after the first boot due to
dependencies on a shim_lock mutex being initialized.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-10-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add parameter to control SoundWire clock stop quirks

Add module parameter so that the different modes can be quickly tested.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-9-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: merge IPC, stream and SoundWire interrupt handlers

We have a single irq handler for SOF interrupts. We can further merge
SoundWire ones to completely remove MSI interrupts handling issues
leading to timeouts.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-8-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: disable SoundWire interrupts on suspend

Doing this avoid conflicts and errors reported on the bus.

The interrupts are only re-enabled on resume after the firmware is
downloaded, so the behavior is not fully symmetric

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-7-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: initial SoundWire machine driver autodetect

For now we have a limited number of machine driver configurations, and
we can detect them based on the link configuration returned after
checking hardware and firmware (BIOS) configurations.

The link configuration is checked with a link_mask as well as a list
of _ADR descriptors for each link.

There is a chance that in extreme cases where the BIOS contains too
much information we would need to detect which Slave devices actually
report as 'attached'. This would be more accurate than static
table-based solutions, but it also introduces timing dependencies
since we don't know when those devices might become attached, so will
only be only be looked at if we see limitations with static methods
and the usual quirks based e.g. on DMI information.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: add SoundWire stream config/free callbacks

These callbacks are invoked when a matching hw_params/hw_free() DAI
operation takes place, and will result in IPC operations with the SOF
firmware.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: IPC: dai-intel: move ALH declarations in header file

ALH was inserted in the wrong place during integration, add after DMIC
to mirror the file used by SOF firmware.

No functional change, just text move in the same file to better track
changes, if any.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: add SoundWire configuration interface

Now that the SoundWire core supports the multi-step initialization,
call the relevant APIs.

The actual hardware enablement can be done in two places, ideally we'd
want to startup the SoundWire IP as soon as possible (while still
taking power rail dependencies into account)

However when suspend/resume is implemented, the DSP device will be
resumed first, and only when the DSP firmware is downloaded/booted
would the SoundWire child devices be resumed, so there are only
marginal benefits in starting the IP earlier for the first probe.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-acpi: expand description of _ADR-based devices

For SoundWire, we need to know if endpoints needs to be 'aggregated'
(MIPI parlance, meaning logically grouped), e.g. when two speaker
amplifiers need to be handled as a single logical output.

We don't necessarily have the information at the firmware (BIOS)
level, so add a notion of endpoints and specify if a device/endpoint
is part of a group, with a position.

This may be expanded in future solutions, for now only provide a group
and position information.

Since we modify the header file, change all existing upstream tables
as well to avoid breaking compilation/bisect.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325215027.28716-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: rt1308-sdw: configure amplifier with set_tdm_slot()" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

When two (or more) amplifiers are on the same link, the integrator may:
a) assign dedicated slots for each of the amplifiers.
b) provide the same configuration to all amplifiers, and rely on
additional controls/processing in the amplifier to generate different
outputs.

case a) was the initial direction for SoundWire and is required for
amplifiers with limited capabilities, but to deal with orientation or
'posture' changes it's easier to implement case b) when the amplifier
can deal with multiple channels.

This patchset suggest the use of the set_tdm_slot() API to define
which of the channels will be consumed by what amplifiers. This maps
well with SoundWire's 'ChannelEnable' registers. The notion of
slot_width is however irrelevant here and ignored, and SoundWire ports
are typically single direction, so only one of the two masks shall be
used.

Pierre-Louis Bossart (2):
  ASoC: rt1308-sdw: add set_tdm_slot() support
  ASoC: rt1308-sdw: use slot and rx_mask to configure stream

 sound/soc/codecs/rt1308-sdw.c | 38 +++++++++++++++++++++++++++++++----
 sound/soc/codecs/rt1308-sdw.h |  2 ++
 2 files changed, 36 insertions(+), 4 deletions(-)

--
2.20.1

Merge series "ASoC: Intel: machine driver updates" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

One correction for GeminiLake and 2 additional machine drivers for
Chromebooks.

Curtis Malainey (1):
  ASoC: Intel: Make glk+rt5682 echo ref dynamic

Sathyanarayana Nujella (2):
  ASoC: Intel: sof_rt5682: Add support for tgl-max98373-rt5682
  ASoC: Intel: common: Add mach table for tgl-max98373-rt5682

Yong Zhi (1):
  ASoC: intel: sof_da7219_max98373: Add speaker switch

 sound/soc/intel/boards/Kconfig                |  1 +
 sound/soc/intel/boards/Makefile               |  2 +-
 sound/soc/intel/boards/glk_rt5682_max98357a.c |  1 +
 sound/soc/intel/boards/sof_da7219_max98373.c  | 67 +++++++++-------
 sound/soc/intel/boards/sof_maxim_common.c     | 80 +++++++++++++++++++
 sound/soc/intel/boards/sof_maxim_common.h     | 24 ++++++
 sound/soc/intel/boards/sof_rt5682.c           | 21 +++++
 .../intel/common/soc-acpi-intel-tgl-match.c   | 13 +++
 8 files changed, 178 insertions(+), 31 deletions(-)
 create mode 100644 sound/soc/intel/boards/sof_maxim_common.c
 create mode 100644 sound/soc/intel/boards/sof_maxim_common.h

--
2.20.1

Merge series "ASoC: SOF: cleanups and improvements" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Couple of small patches to improve error handling, inits, logs.

Hope we can have this for 5.7?

Guennadi Liakhovetski (2):
  ASoC: SOF: (cosmetic) use for_each_pcm_streams() in sof_dai_load()
  ASoC: SOF: fix uninitialised "work" with VirtIO

Kai Vehmanen (2):
  ASoC: SOF: Intel: hda: do not leave clock gating off upon error
  ASoC: SOF: Intel: hda: call codec wake at chip init

Ranjani Sridharan (1):
  ASoC: SOF: Intel: hda: Improve DSP state logging

 sound/soc/sof/intel/hda-ctrl.c | 15 +++++++++---
 sound/soc/sof/intel/hda-dsp.c  | 43 ++++++++++++++++++++++++++++++++--
 sound/soc/sof/pcm.c            |  4 +---
 sound/soc/sof/sof-audio.h      |  3 +++
 sound/soc/sof/topology.c       | 17 ++++++++++----
 5 files changed, 69 insertions(+), 13 deletions(-)

--
2.20.1

ASoC: pxa: Enable AC'97 bus support for PXA machines

The AC'97 based PXA machines currently don't build reliably as they don't
ensure that an AC'97 bus is built, causing at least eseries_pxa_defconfig
to fail to build. Add selects to fix this.

Reported-by: KernelCI <bot@kernelci.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20200326180116.21375-1-broonie@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pxa: Select regmap from AC'97 machines

regmap needs to be selected by users which for machine drivers that select
AC'97 CODEC drivers means that we need to also select regmap to ensure that
the CODEC driver will build if nothing else enables regmap as is likely for
such systems.

Signed-off-by: Mark Brown <broonie@kernel.org>
Link: https://lore.kernel.org/r/20200326151053.40806-1-broonie@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1308-sdw: use slot and rx_mask to configure stream

If the DAI was configured with a set_tdm_slots() call, use the information.

A platform or machine driver may configure each amplifier to extract
different bitSlots from the frame, or extract the same data and use
processing to generate the relevant output. The latter case is easier
to handle in case of orientation changes.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325212905.28145-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1308-sdw: add set_tdm_slot() support

Add ability to select which of the channels is used, or both, in case
two RT1308 amplifiers are located on the same link.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325212905.28145-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: call codec wake at chip init

Further align HDA init sequence to the legacy non-DSP HDA driver by
calling snd_hdac_set_codec_wakeup() during the chip init sequence.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-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/20200325211233.27394-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: do not leave clock gating off upon error

The misc clock gating (MISCBDCGE) is disabled for controller reset and
reenabled once reset is complete.

Fix the case when error happens during reset, and clock gating is
left disabled. The clock gating should be reenabled also in this case.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-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/20200325211233.27394-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: fix uninitialised "work" with VirtIO

In the VirtIO case the sof_pcm_open() function isn't called on the
host during guest streaming, which then leaves "work" structures
uninitialised. However it is then used to handle position update
messages from the DSP. Move their initialisation to immediately after
allocation of the containing structure.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: (cosmetic) use for_each_pcm_streams() in sof_dai_load()

Use for_each_pcm_streams() to enumerate streams in sof_dai_load()
instead of doing that manually.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: Improve DSP state logging

Improve the DSP power state logs with the state names
instead of values.

Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325211233.27394-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: common: Add mach table for tgl-max98373-rt5682

Update tgl mach table with: Maxim98373 Amp and ALC5682 hp codec.
Both of the codecs are on I2S bus.

Signed-off-by: Jairaj Arava <jairaj.arava@intel.com>
Signed-off-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_rt5682: Add support for tgl-max98373-rt5682

This patch does the below:
1. Adds the driver data and updates quirk info for TGL
        with Max98373 speaker amp and ALC5682 headset codec.
2. Added max98373 speaker related code to common file for re-use.

Signed-off-by: Jairaj Arava <jairaj.arava@intel.com>
Signed-off-by: Sathyanarayana Nujella <sathyanarayana.nujella@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: intel: sof_da7219_max98373: Add speaker switch

Add "Spk Switch" and associated widget, route to max98360a
speaker amp for power saving, also remove the speaker_amp_init()
callback with complete separated tables for max98373 and max98360a.

Signed-off-by: Bhat, Uday M <uday.m.bhat@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: Make glk+rt5682 echo ref dynamic

Without the dynamic flag to allow runtime routing, the card cannot
probe on chromebooks because SOF is constantly waiting for the link.
Adding flag back to allow upstream kernels to work on rt5682 based
chromebooks since SOF can now ignore the hard coded front end.

Signed-off-by: Curtis Malainey <cujomalainey@chromium.org>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200325213245.28247-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mchp-i2s-mcc: make signed 1 bit bitfields unsigned

The signed 1 bit bitfields should be unsigned, so make them unsigned.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20200325132913.110115-1-colin.king@canonical.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Convert jz4740-i2s doc to YAML

Convert the textual binding documentation for the AIC (AC97/I2S
Controller) of Ingenic SoCs to a YAML schema, and add the new compatible
strings in the process.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200306222931.39664-1-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: jz4740-i2s: Add support for the JZ4770

Before the JZ4770, the playback and capture sampling rates had to match.
The JZ4770 supports independent sampling rates for both.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20200306222931.39664-6-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: jz4740-i2s: Add support for the JZ4760

The change of offset for the {rx,tx}_threshold fields in the conf
register predates the JZ4780, and was first introduced in the JZ4760.

Signed-off-by: Paul Cercueil <paul@crapouillou.net>
Link: https://lore.kernel.org/r/20200306222931.39664-5-paul@crapouillou.net
Signed-off-by: Mark Brown <broonie@kernel.org>

dt-bindings: sound: rockchip-i2s: add #sound-dai-cells property

'#sound-dai-cells' is required to properly interpret
the list of DAI specified in the 'sound-dai' property,
so add them to 'rockchip-i2s.yaml'

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200324094149.6904-2-jbx6244@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

dt-bindings: sound: convert rockchip i2s bindings to yaml

Current dts files with 'i2s' nodes are manually verified.
In order to automate this process rockchip-i2s.txt
has to be converted to yaml.

Signed-off-by: Johan Jonker <jbx6244@gmail.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20200324094149.6904-1-jbx6244@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8974: remove unused variables

sound/soc/codecs/wm8974.c:200:38: warning:
 wm8974_aux_boost_controls defined but not used [-Wunused-const-variable=]
sound/soc/codecs/wm8974.c:204:38: warning:
 wm8974_mic_boost_controls defined but not used [-Wunused-const-variable=]

commit 8a123ee2a46d ("ASoC: WM8974 DAPM cleanups")
left behind this, remove them.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20200324070615.16248-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "Support built-in Mic on Tegra boards that use WM8903" from Dmitry Osipenko <digetx@gmail.com>:

Hello,

This small series adds audio route for built-in microphone on NVIDIA Tegra
boards that use WM8903 CODEC. In particular this is needed in order to unmute
internal microphone on Acer A500 tablet device. I'm planning to send out the
device tree for the A500 for 5.8, so will be nice to get the microphone
sorted out. Please review and apply, thanks in advance.

Dmitry Osipenko (2):
  dt-bindings: sound: tegra-wm8903: Document built-in microphone audio
    source
  ASoC: tegra: tegra_wm8903: Support DAPM events for built-in microphone

 .../sound/nvidia,tegra-audio-wm8903.txt        |  1 +
 sound/soc/tegra/tegra_wm8903.c                 | 18 ++++++++++++++++++
 2 files changed, 19 insertions(+)

--
2.25.1

ASoC: rt5682: Add the descriptions for the DMIC clock rate and delay settings

The patch adds the descriptions for the DMIC clock rate and delay settings.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-3-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Add a property for DMIC delay

The patch adds a property for DMIC delay (ms) to avoid pop noise and
changes the default delay setting.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-2-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Add a property for DMIC clock rate

The patch adds a property for DMIC clock rate (hz) and changes the
default to the common optimize DMIC clock rate.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200323082547.7898-1-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: MT6660: make spdxcheck.py happy

The SPDX-License-Identifier shall not be suffixed with anything further.
This makes ./scripts/spdxcheck.py complain:

  sound/soc/codecs/mt6660.c: 1:36 Invalid token: //

Clean up SPDX-License-Identifier line to make spdxcheck.py happy.

Signed-off-by: Lukas Bulwahn <lukas.bulwahn@gmail.com>
Link: https://lore.kernel.org/r/20200321114022.8545-1-lukas.bulwahn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra-wm8903: Document built-in microphone audio source

The internal microphone source is needed in order to be able to describe
the hardware audio routing for devices that have the built-in microphone
in addition to the external Mic Jack.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20200320205504.30466-2-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra: tegra_wm8903: Support DAPM events for built-in microphone

The enable-GPIO needs to be toggled on a DAPM event in order to turn
microphone ON/OFF, otherwise microphone won't work.

Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Acked-by: Stephen Warren <swarren@nvidia.com>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20200320205504.30466-3-digetx@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: Merge CPU/Codec for soc_dai_pcm_new()

Now CPU/Codec DAIs are alias for dais.
Thus, we can directly use for_each_rtd_dais() macro
for soc_dai_pcm_new().
This patch merge CPU/Codec for it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87r1xsolen.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-pcm: Merge for_each_rtd_cpu/codec_dais()

Now we can use for_each_rtd_dais().
Let's use it instead of for_each_rtd_cpu/codec_dais().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87sgi8olet.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-dapm: Merge for_each_rtd_cpu/codec_dais()

Now we can use for_each_rtd_dais().
Let's use it instead of for_each_rtd_cpu/codec_dais().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87tv2ooley.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: Merge for_each_rtd_cpu/codec_dais()

Now we can use for_each_rtd_dais().
Let's use it instead of for_each_rtd_cpu/codec_dais().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87v9n4olf4.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: Merge CPU/Codec DAIs

ALSA SoC is currently categorizing CPU/Codec DAIs,
and it works well.

But modern devices require more complex connections,
for example Codec to Codec, etc, and future devices will
enable to more complex connections.
Because of these background, CPU/Codec DAIs categorizing is
no longer good much to modern device.

Currently, rtd has both CPU/Codec DAIs pointer.

rtd->cpu_dais   = [][][][][][][][][]
rtd->codec_dais = [][][][][][][][][]

This patch merges these into DAIs pointer.

rtd->dais = [][][][][][][][][][][][][][][][][][]
            ^cpu_dais         ^codec_dais
            |--- num_cpus ---|--- num_codecs --|

Then, we can merge for_each_rtd_cpu/codec_dais() from this patch.

- for_each_rtd_cpu_dais() {
- ...
- }
- for_each_rtd_codec_dais() {
- ...
- }
+ for_each_rtd_dais() {
+ ...
+ }

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/87wo7kolfa.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge branch 'topic/ro_wordlength' of git://git./linux/kernel/git/vkoul/soundwire into asoc-5.7

soundwire: stream: Add read_only_wordlength flag to port properties

According to SoundWire Specification Version 1.2.
"A Data Port number X (in the range 0-14) which supports only one
value of WordLength may implement the WordLength field in the
DPX_BlockCtrl1 Register as Read-Only, returning the fixed value of
WordLength in response to reads."

As WSA881x interfaces in PDM mode making the only field "WordLength"
in DPX_BlockCtrl1" fixed and read-only. Behaviour of writing to this
register on WSA881x soundwire slave with Qualcomm Soundwire Controller
is throwing up an error. Not sure how other controllers deal with
writing to readonly registers, but this patch provides a way to avoid
writes to DPN_BlockCtrl1 register by providing a read_only_wordlength
flag in struct sdw_dpn_prop

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20200311113545.23773-2-srinivas.kandagatla@linaro.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

Merge series "ASoC: stm32: manage rebind issue" from Olivier Moysan <olivier.moysan@st.com>:

This patchset corrects a rebind issue on STM32 SPDIFRX and I2S drivers.

The same correction has already been applied for SAI driver:
0d6defc7e0e4 ("ASoC: stm32: sai: manage rebind issue")

The commit e894efef9ac7 ("ASoC: core: add support to card rebind")
allows to rebind the sound card after a rebind of one of its component.
With this commit, the sound card is actually rebound,
but may be no more functional.

The following problems have been seen on STM32 drivers.

1) DMA channel is not requested:

With the sound card rebind the simplified call sequence is:
    probe
        snd_soc_register_component
                snd_soc_try_rebind_card
                        snd_soc_instantiate_card
        devm_snd_dmaengine_pcm_register

The problem occurs because the pcm must be registered,
before snd_soc_instantiate_card() is called.

Modify the driver, to change the call sequence as follows:
    probe
        devm_snd_dmaengine_pcm_register
        snd_soc_register_component
                snd_soc_try_rebind_card

2) DMA channel is not released:

dma_release_channel() is not called when
devm_dmaengine_pcm_release() is executed.
This occurs because SND_DMAENGINE_PCM_DRV_NAME component,
has already been released through devm_component_release().

devm_dmaengine_pcm_release() should be called before
devm_component_release() to avoid this problem.

Call snd_dmaengine_pcm_unregister() and snd_soc_unregister_component()
explicitly from the driver, to have the right sequence.

Olivier Moysan (3):
  ASoC: stm32: spdifrx: fix regmap status check
  ASoC: stm32: spdifrx: manage rebind issue
  ASoC: stm32: i2s: manage rebind issue

 sound/soc/stm/stm32_i2s.c     | 40 ++++++++++++++++------
 sound/soc/stm/stm32_spdifrx.c | 64 +++++++++++++++++++----------------
 2 files changed, 63 insertions(+), 41 deletions(-)

--
2.17.1

Merge series "ASoC: sdm845: fix soundwire stream handling" from Srinivas Kandagatla <srinivas.kandagatla@linaro.org>:

Recent addition of SoundWire stream state-machine checks in linux-next
have shown an existing issue with handling soundwire streams in codec drivers.

In general soundwire stream prepare/enable/disable can be called from either
codec/machine/controller driver. However calling it in codec driver means
that if multiple instances(Left/Right speakers) of the same codec is
connected to the same stream then it will endup calling stream
prepare/enable/disable more than once. This will mess up the stream
state-machine checks in the soundwire core.

Moving this stream handling to machine driver would fix this issue
and also allow board/platform specfic power sequencing.

Changes since v1:
- removed false error check while setting sruntime.

Srinivas Kandagatla (2):
  ASoC: qcom: sdm845: handle soundwire stream
  ASoC: codecs: wsa881x: remove soundwire stream handling

 sound/soc/codecs/wsa881x.c | 44 +------------------------
 sound/soc/qcom/Kconfig     |  2 +-
 sound/soc/qcom/sdm845.c    | 67 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 69 insertions(+), 44 deletions(-)

--
2.21.0

ASoC: rt5682: remove noisy debug messages

Some debug messages are too noisy.
This patch removes it.

Signed-off-by: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20200317073321.12660-1-shumingf@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: sdm845: handle soundwire stream

In existing setup WSA881x codec handles soundwire stream,
however DB845c and other machines based on SDM845c have 2
instances for WSA881x codec. This will force soundwire stream
to be prepared/enabled twice or multiple times.
Handling SoundWire Stream in machine driver would fix this issue.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200317151233.8763-2-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: wsa881x: remove soundwire stream handling

There could be multiple instances of this codec on any platform,
so handling stream directly in this codec driver can lead to
multiple calls to prepare/enable/disable on the same SoundWire stream.
Move this stream handling to machine driver to fix this issue.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20200317151233.8763-3-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: stm32: i2s: manage rebind issue

The commit e894efef9ac7 ("ASoC: core: add support to card rebind")
allows to rebind the sound card after a rebind of one of its component.
With this commit, the sound card is actually rebound,
but may be no more functional.

Corrections:
- Call snd_dmaengine_pcm_register() before snd_soc_register_component().
- Call snd_dmaengine_pcm_unregister() and snd_soc_unregister_component()
explicitly from I2S driver.

Signed-off-by: Olivier Moysan <olivier.moysan@st.com>
Link: https://lore.kernel.org/r/20200318144125.9163-4-olivier.moysan@st.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: stm32: spdifrx: manage rebind issue

The commit e894efef9ac7 ("ASoC: core: add support to card rebind")
allows to rebind the sound card after a rebind of one of its component.
With this commit, the sound card is actually rebound,
but may be no more functional.

Corrections:
- Call snd_dmaengine_pcm_register() before snd_soc_register_component().
- Call snd_dmaengine_pcm_unregister() and snd_soc_unregister_component()
explicitly from SPDFIRX driver.

Signed-off-by: Olivier Moysan <olivier.moysan@st.com>
Link: https://lore.kernel.org/r/20200318144125.9163-3-olivier.moysan@st.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ALSA: compress: Add wma, alac and ape support" from Vinod Koul <vkoul@>:

This series adds more WMA profiles and WMA decoder parameters to UAPI and
then support for these in qcom driver. It also adds FLAC and APE IDs and
decoder parameters to UAPI and then support in qcom driver

This was tested on Dragon board RB3.

Last, bump up the compressed version so that userspace can check for the
support.

Since the series touches compress uapi and asoc, it would make sense to go
thru asoc tree with acks.

Changes in v3:
 - add r-b from Srini
 - use macros for FLAC channel layout tags

Changes in v2:
 - use bitflags for wma profiles

Vinod Koul (9):
  ALSA: compress: add wma codec profiles
  ALSA: compress: Add wma decoder params
  ASoC: qcom: q6asm: pass codec profile to q6asm_open_write
  ASoC: qcom: q6asm: add support to wma config
  ASoC: qcom: q6asm-dai: add support to wma decoder
  ALSA: compress: add alac & ape decoder params
  ASoC: qcom: q6asm: add support for alac and ape configs
  ASoC: qcom: q6asm-dai: add support for ALAC and APE decoders
  ALSA: compress: bump the version

 include/uapi/sound/compress_offload.h |   2 +-
 include/uapi/sound/compress_params.h  |  37 +++-
 sound/soc/qcom/qdsp6/q6asm-dai.c      | 139 ++++++++++++++-
 sound/soc/qcom/qdsp6/q6asm.c          | 243 +++++++++++++++++++++++++-
 sound/soc/qcom/qdsp6/q6asm.h          |  51 +++++-
 5 files changed, 465 insertions(+), 7 deletions(-)

--
2.24.1

ASoC: soc-dai: return proper error for get_sdw_stream()

snd_soc_dai_get_sdw_stream() returns null if dai does not support
this callback, this is no very useful for the caller to
differentiate if this is an error or unsupported call for the dai.

return -ENOTSUPP in cases where this callback is not supported.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20200316151110.2580-1-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ALSA: compress: bump the version

We have added support for bunch of new decoders and parameters for
decoders. To help users find the support bump the version up to 0,2,0.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-10-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm-dai: add support for ALAC and APE decoders

Qualcomm DSPs also supports the ALAC and APE decoders, so add support
for these and convert the snd_codec_params to qdsp format.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-9-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm: add support for alac and ape configs

Qualcomm DSPs expect ALAC and APE configs to be send for decoders,
so add the API to program the respective config to the DSP.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-8-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ALSA: compress: add alac & ape decoder params

Add ALAC (Apple Lossless Audio Codec) and APE (Monkey's Lossless Audio
Codec) defines and parameters required to configure these.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-7-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm-dai: add support to wma decoder

Qualcomm DSPs also supports the wma decoder, so add support for wma
decoder and convert the snd_codec_params to qdsp format.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-6-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm: add support to wma config

Qualcomm DSPs expect wma v9 and wma v10 configs to be set for wma
decoders, so add the API to program the respective wma config to the DSP

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-5-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm: pass codec profile to q6asm_open_write

Codec profile is required to be passed for WMA codecs so that we know
the codec profile present and tell DSP accordingly, so update this API
to pass the codec profile as argument

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-4-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ALSA: compress: Add wma decoder params

Some WMA decoders like WMAv10 etc need some additional encoder option
parameters, so add these as WMA decoder params.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-3-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ALSA: compress: add wma codec profiles

Some codec profiles were missing for WMA, like WMA9/10 lossless and
wma10 pro, so add these profiles

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20200316055221.1944464-2-vkoul@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge branch 'for-5.6' of https://git./linux/kernel/git/broonie/sound into asoc-5.7

ASoC: max98357a: Add ACPI HID MAX98360A

Maxim MAX98360A audio amplifier is functionally identical to MAX98357A,
add ACPI ID "MAX98360A" for driver reuse.

Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Link: https://lore.kernel.org/r/1584114926-29287-1-git-send-email-yong.zhi@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: SOF: multi-cpu dais, IPC and Intel improvements for 5.7" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Corrections for compatibility with the multi-cpu dai check for
min-channels, IPC simplifications, and removal of unnecessary
boot-related code for Intel platforms.

Amery Song (2):
  ASoC: SOF: Intel: hda: remove unnecessary ROM IPC filter function
  ASoC: SOF: Intel: remove unnecessary waitq before loading firmware

Bard Liao (2):
  ASoC: SOF: Intel: hda-dai: add stream capability
  ASoC: SOF: topology: connect dai widget to all cpu-dais

Karol Trzcinski (2):
  ASoC: SOF: Make sof_ipc_ext_data enum more rigid
  ASoC: SOF: Remove SOF_IPC_EXT_DMA_BUFFER

 include/sound/sof/info.h         | 22 +-------
 sound/soc/sof/intel/cnl.c        |  5 --
 sound/soc/sof/intel/hda-dai.c    | 96 ++++++++++++++++++++++++++++++++
 sound/soc/sof/intel/hda-ipc.c    | 20 +------
 sound/soc/sof/intel/hda-loader.c |  3 -
 sound/soc/sof/intel/hda.c        |  3 -
 sound/soc/sof/intel/hda.h        |  1 -
 sound/soc/sof/loader.c           |  6 --
 sound/soc/sof/sof-priv.h         |  5 --
 sound/soc/sof/topology.c         |  8 ++-
 10 files changed, 108 insertions(+), 61 deletions(-)

base-commit: 101247a3b86e1cc0e382b7e887a56176290fc957
--
2.20.1

Merge series "ASoC: Intel: machine driver updates for 5.7" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Fix HDaudio/HDMI Kconfig dependencies, cleanup code, and add support
for 2 new configs (DA7219+Maxim 98360A and RT5692+RT1015)

Note that the last patch is the first case where we added the
Reviewed-by tags automagically from the list of GitHub
approvers. Feedback/comments welcome on the format, and thanks to
Ranjani for the nice script!

Changes since v1:
Change base to broonie/for-5.7
Rebase and fix for_each_codec_dais() macro in Patch 9.
Add new patch to remove GFP_ATOMIC

Guennadi Liakhovetski (3):
  ASoC: codecs: hdac_hdmi: (cosmetic) remove redundant variable
    initialisations
  ASoC: Intel: skylake: (cosmetic) remove redundant variable
    initialisations
  ASoC: Intel: (cosmetic) simplify structure member access

Kai Vehmanen (4):
  ASoC: Intel: boards: drop reverse deps for SND_HDA_CODEC_HDMI
  ASoC: Intel: sof_pcm512x: drop reverse deps for SND_HDA_CODEC_HDMI
  ASoC: Intel: sof_pcm512x: make HDMI optional for all platforms
  ASoC: SOF: Intel: hda: remove SND_SOC_SOF_HDA_COMMON_HDMI_CODEC

Pierre-Louis Bossart (1):
  ASoC: Intel: don't use GFP_ATOMIC for machine driver contexts

Yong Zhi (2):
  ASoC: Intel: sof_da7219_max98373: Add support for max98360a speaker
    amp
  ASoC: Intel: sof_rt5682: Add rt1015 speaker amp support

 sound/soc/codecs/hdac_hdmi.c                  |   6 +-
 sound/soc/intel/boards/Kconfig                |  20 ++--
 sound/soc/intel/boards/bdw-rt5650.c           |   2 +-
 sound/soc/intel/boards/bdw-rt5677.c           |   2 +-
 sound/soc/intel/boards/broadwell.c            |   2 +-
 sound/soc/intel/boards/bxt_da7219_max98357a.c |   2 +-
 sound/soc/intel/boards/bxt_rt298.c            |   2 +-
 sound/soc/intel/boards/bytcht_da7213.c        |   2 +-
 sound/soc/intel/boards/cht_bsw_max98090_ti.c  |   2 +-
 sound/soc/intel/boards/cht_bsw_nau8824.c      |   2 +-
 sound/soc/intel/boards/cht_bsw_rt5645.c       |   2 +-
 sound/soc/intel/boards/cml_rt1011_rt5682.c    |   4 +-
 sound/soc/intel/boards/glk_rt5682_max98357a.c |   2 +-
 sound/soc/intel/boards/haswell.c              |   2 +-
 sound/soc/intel/boards/kbl_rt5663_max98927.c  |   2 +-
 .../intel/boards/kbl_rt5663_rt5514_max98927.c |   2 +-
 sound/soc/intel/boards/skl_hda_dsp_generic.c  |   2 +-
 .../soc/intel/boards/skl_nau88l25_max98357a.c |   2 +-
 sound/soc/intel/boards/skl_nau88l25_ssm4567.c |   2 +-
 sound/soc/intel/boards/sof_da7219_max98373.c  |  78 +++++++++++--
 sound/soc/intel/boards/sof_pcm512x.c          |  38 ++++--
 sound/soc/intel/boards/sof_rt5682.c           | 110 +++++++++++++++++-
 .../intel/common/soc-acpi-intel-jsl-match.c   |  34 +++++-
 sound/soc/intel/skylake/skl.c                 |   4 +-
 sound/soc/sof/intel/Kconfig                   |  11 --
 sound/soc/sof/intel/hda.c                     |   3 +-
 26 files changed, 269 insertions(+), 71 deletions(-)

base-commit: 101247a3b86e1cc0e382b7e887a56176290fc957
--
2.20.1

Merge series "ASoC: brcm: add dsl and pon chip audio driver" from Kevin Li <kevin-ke.li@broadcom.com>:

Changes in v2:
 * Make the comment a C++ one for license header
 * Remove all empty functions
 * Change all variable to use kernel coding style
 * Comment chip TX RX block independently generate I2S bus signals

Kevin Li (2):
  ASoC: brcm:  Add DSL/PON SoC audio driver
  ASoC: brcm:  DSL/PON SoC device tree bindings of audio driver

 .../bindings/sound/brcm,bcm63xx-audio.txt     |  29 ++
 sound/soc/bcm/Kconfig                         |   9 +
 sound/soc/bcm/Makefile                        |   4 +
 sound/soc/bcm/bcm63xx-i2s-whistler.c          | 317 ++++++++++++
 sound/soc/bcm/bcm63xx-i2s.h                   |  90 ++++
 sound/soc/bcm/bcm63xx-pcm-whistler.c          | 485 ++++++++++++++++++
 6 files changed, 934 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/sound/brcm,bcm63xx-audio.txt
 create mode 100644 sound/soc/bcm/bcm63xx-i2s-whistler.c
 create mode 100644 sound/soc/bcm/bcm63xx-i2s.h
 create mode 100644 sound/soc/bcm/bcm63xx-pcm-whistler.c

--
2.25.1

ASoC: rt5682: Fine tune the HP performance in soundwire mode

The setting is sync with I2C/I2S mode.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200313023850.28875-1-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: brcm: DSL/PON SoC device tree bindings of audio driver

Signed-off-by: Kevin Li <kevin-ke.li@broadcom.com>
Link: https://lore.kernel.org/r/20200312223242.2843-1-kevin-ke.li@broadcom.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: brcm: Add DSL/PON SoC audio driver

This patch adds Broadcom DSL/PON SoC audio driver
with Whistler I2S block. The SoC supported by this
patch are BCM63158B0,BCM63178 and BCM47622/6755

Signed-off-by: Kevin Li <kevin-ke.li@broadcom.com>
Link: https://lore.kernel.org/r/20200312223242.2843-2-kevin-ke.li@broadcom.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: don't use GFP_ATOMIC for machine driver contexts

We've removed GFP_ATOMIC in all machine drivers and somehow this keeps
coming back due to copy-paste. Move to GFP_KERNEL.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-11-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_rt5682: Add rt1015 speaker amp support

This patch adds jsl_rt5682_rt1015 which supports the
RT5682 headset codec and RT1015 speaker amplifier combination
on JasperLake platform.

Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-10-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_da7219_max98373: Add support for max98360a speaker amp

Add Maxim MAX98360A plug-and-play Class-D amplifier support on SSP1,
new card ID is sofda7219max98360a, name sof-da7219max98360a.

Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-9-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: (cosmetic) simplify structure member access

Fix a clumsy structure member dereference in all machine drivers.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-8-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: skylake: (cosmetic) remove redundant variable initialisations

Variables, used as loop iterators, don't need to be initialised.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-7-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: hdac_hdmi: (cosmetic) remove redundant variable initialisations

Remove several redundant variable initialisations.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: remove SND_SOC_SOF_HDA_COMMON_HDMI_CODEC

To help user-space with HDMI codec driver transition, both
a kernel module parameter and a kernel option were initially
provided to configure default behaviour of SOF on Intel hardware
with commit 139c7febad1a ("ASoC: SOF: Intel: add support for
snd-hda-codec-hdmi").

As hdac-hdmi is already now lagging in features compared to
snd-hda-codec-hdmi, move ahead with the transition and remove
the build option to select between the two, and instead default
to snd-hda-codec-hdmi if it is enabled in kernel build.

The old behaviour of using hdac-hdmi driver can still be forced
via the kernel module parameter.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_pcm512x: make HDMI optional for all platforms

Make HDMI optional for APL and later platforms. If no HDMI codec
is found on the HDA bus, the graphics side driver is missing or
correct codec driver is not part of kernel build, codec_mask
reflects this and HDMI is disabled. The DSP topology will still
have the links for HDMI, so connect these to dummy codec to avoid
failures in topology loading.

This change also fixes a kernel oops that was triggered if
sof_pcm512x was used with SOF configured to use hdac-hdmi (can be
done via "use_common_hdmi=0" or by selecting
CONFIG_SND_SOC_SOF_HDA_COMMON_HDMI_CODEC=n). This is not a supported
configuration.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_pcm512x: drop reverse deps for SND_HDA_CODEC_HDMI

Having a reverse dependency to a config that has its own additional
dependencies, is generally not recommended. And this applies to
select statements for SND_HDA_CODEC_HDMI, e.g. the case where SND_HDA
and SND_SOC_SOF_HDA are built as modules, but the machine driver is
built-in, leading to compile errors (reported as
i386-randconfig-e003-20200206).

Give up on trying to define different dependencies based on SOF/SST
selection, and simply add a "depends on" for SND_HDA_CODEC_HDMI. This
fixes the issue with randconfigs. Only downside is that SND_HDA_CODEC_HDMI
may be built unnecessarily in some cases, but this seems like the lesser
evil.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: boards: drop reverse deps for SND_HDA_CODEC_HDMI

Having a reverse dependency to a config that has its own additional
dependencies, is generally not recommended. And this applies to
select statements for SND_HDA_CODEC_HDMI, e.g. the case where SND_HDA
and SND_SOC_SOF_HDA are built as modules, but the machine driver is
built-in, leading to compile errors (reported as
i386-randconfig-e003-20200206).

Give up on trying to define different dependencies based on SOF/SST
selection, and simply add a "depends on" for SND_HDA_CODEC_HDMI. This
fixes the issue with randconfigs. Only downside is that SND_HDA_CODEC_HDMI
may be built unnecessarily in some cases, but this seems like the lesser
evil.

Fixes: aa2b4a5 ('ASoC: Intel: boards: fix incorrect HDMI Kconfig dependency')
Reported-by: kbuild test robot <lkp@intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312194859.4051-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: topology: connect dai widget to all cpu-dais

Extend code from single cpu-dai to multi-dai

Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-7-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: remove unnecessary waitq before loading firmware

The HDA_DSP_IPC_PURGE_FW IPC from ROM is already handled in
cl_dsp_init(), and it will never be received in the IRQ thread,
so the wait condition on this IPC will never be satisfied. The
wait before loading firmware is redundant and can be removed safely.

Signed-off-by: Amery Song <chao.song@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Pierre Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-6-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: remove unnecessary ROM IPC filter function

The HDA_DSP_IPC_PURGE_FW IPC from ROM is already handled in
cl_dsp_init(), and as IPC IRQ is disabled at this stage, this
IPC will be never received in the IRQ thread. The function
hda_dsp_ipc_is_sof for filtering the ROM IPC can be removed
safely.

Signed-off-by: Amery Song <chao.song@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Keyon <yang.jie@intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Remove SOF_IPC_EXT_DMA_BUFFER

This enum code, and what's more important, related structures is
unused in whole source code, so it shouldn't be kept.

Signed-off-by: Karol Trzcinski <karolx.trzcinski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Make sof_ipc_ext_data enum more rigid

It's a part of ABI interface, so enum value shouldn't change
for example after removing some old enum code.

Signed-off-by: Karol Trzcinski <karolx.trzcinski@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda-dai: add stream capability

snd_soc_dai_stream_valid() will check if the stream is valid by testing
stream->channels_min. So we do need the information in dai driver.

The stream name is not added since we want to sure
playback_widget/capture_widget will be created by topology.

Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20200312200622.24477-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Revise the DAC1 volume setting

The max volume of the DAC1 Playback Volume is 0dB.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20200313023850.28875-2-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm_adsp: Use scnprintf() for the limited buffer output

snprintf() is a hard-to-use function, it's especially difficult to use
it for concatenating substrings in a buffer with a limited size.
Since snprintf() returns the would-be-output size, not the actual
size, the subsequent use of snprintf() may point to the incorrect
position.

Use scnprintf() instead for fixing such potential errors.

Signed-off-by: Takashi Iwai <tiwai@suse.de>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20200313130334.9028-1-tiwai@suse.de
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: topology: Perform component check upfront

Function soc_tplg_dbytes_create(), calls soc_tplg_init_kcontrol() to
perform additional driver specific initialization. While
soc_tplg_init_kcontrol() ensures that component is valid before invoking
ops->control_load, there is no such check at the end of
soc_tplg_dbytes_create() where list_add() is used.

Also in quite a few places, there is reference of tplg->comp->dapm or
tplg->comp->card, without any checks for tplg->comp.

In consequence of the above this may lead to referencing NULL pointer.

This allows for removal of now unnecessary checks.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20200312122239.14489-1-amadeuszx.slawinski@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: SOF: preparatory patches" from Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>:

2 preparatory patches for the SOF virtualisation work. No functional
changes, only some code refactoring and exporting of helper functions.

v2: rebase, no changes otherwise.

Thanks
Guennadi

Guennadi Liakhovetski (2):
  ASoC: (cosmetic) simplify dpcm_prune_paths()
  ASoC: export DPCM runtime update functions

 include/sound/soc-dpcm.h |  4 +--
 sound/soc/soc-dapm.c     |  8 +++---
 sound/soc/soc-pcm.c      | 68 +++++++++++++++++++++++++-----------------------
 3 files changed, 42 insertions(+), 38 deletions(-)

--
1.9.3

ASoC: dt-bindings: google, cros-ec-codec: Fix dtc warnings in example

Extra dtc warnings (roughly what W=1 enables) are now enabled by default
when building the binding examples. These were fixed treewide in
5.6-rc5, but the newly added google,cros-ec-codec schema adds some new
warnings:

Documentation/devicetree/bindings/sound/google,cros-ec-codec.example.dts:17.28-21.11:
Warning (unit_address_vs_reg): /example-0/reserved_mem: node has a reg or ranges property, but no unit name
Documentation/devicetree/bindings/sound/google,cros-ec-codec.example.dts:22.19-32.11:
Warning (unit_address_vs_reg): /example-0/cros-ec@0: node has a unit name, but no reg property
Documentation/devicetree/bindings/sound/google,cros-ec-codec.example.dts:26.37-31.15:
Warning (unit_address_vs_reg): /example-0/cros-ec@0/ec-codec: node has a reg or ranges property, but no unit name

Fixing the above, then results in:

Documentation/devicetree/bindings/sound/google,cros-ec-codec.example.dts:26.13-23:
Warning (reg_format): /example-0/cros-ec@0:reg: property has invalid length (4 bytes) (#address-cells == 1, #size-cells == 1)
Documentation/devicetree/bindings/sound/google,cros-ec-codec.example.dts:27.37-32.15:
Warning (unit_address_vs_reg): /example-0/cros-ec@0/ec-codec: node has a reg or ranges property, but no unit name

Fixes: eadd54c75f1e ("dt-bindings: Convert the binding file google, cros-ec-codec.txt to yaml format.")
Signed-off-by: Rob Herring <robh@kernel.org>
Reviewed-by: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Cc: alsa-devel@alsa-project.org
Cc: Benson Leung <bleung@chromium.org>
Cc: Mark Brown <broonie@kernel.org>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Guenter Roeck <groeck@chromium.org>
Cc: Enric Balletbo i Serra <enric.balletbo@collabora.com>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Link: https://lore.kernel.org/r/20200311205841.2710-1-robh@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: export DPCM runtime update functions

This makes DPCM runtime update functions available for external
calling. As an example, virtualised ASoC component drivers may need
to call these when managing shared DAPM routes that are used by more
than one driver (i.e. when host driver and guest drivers have a DAPM
path from guest PCM to host DAI where some parts are owned by host
driver and others by guest driver).

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20200312095214.15126-3-guennadi.liakhovetski@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: (cosmetic) simplify dpcm_prune_paths()

Currently dpcm_prune_paths() has up to 4 nested condition and loop
levels, which forces the code to use flags for flow control.
Extracting widget status verification code from dpcm_prune_paths()
into a separate function simplifies the code.

Signed-off-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20200312095214.15126-2-guennadi.liakhovetski@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>