ASoC: dt-bindings: renesas, rsnd: Fix duplicate 'allOf' entries

Commit e52f3f291152 ("ASoC: audio-graph-card: Refactor schema") added an
'allOf' entry, but one is already present in the schema. Multiple keys
is not valid and results in an error:

ruamel.yaml.constructor.DuplicateKeyError: while constructing a mapping
  in "<unicode string>", line 4, column 1
found duplicate key "allOf" with value "[]" (original value: "[]")
  in "<unicode string>", line 262, column 1

Fixes: e52f3f291152 ("ASoC: audio-graph-card: Refactor schema")
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Sameer Pujar <spujar@nvidia.com>
Cc: alsa-devel@alsa-project.org
Signed-off-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201119161848.3379929-1-robh@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: control: override volume info callback

ASoC dapm controls currently don't support more than 2 channels. This is
a problem for SOF-based devices where individual volume control cannot
be provided on the 4 DMIC input path.

If we want to provide controls for more than 2 channels, this patch
suggests a simple solution based on an override of the info callback.
For example, in the case with 4 channel DMIC PGAs, a sof_info callback
would be used. Mono and stereo cases will keep using the existing dapm
info callback.

A longer-term solution would be to remove the limits to 2 channels in
ASoC/DAPM/topology. This is a topic Intel is currently looking into,
e.g. by removing the use of 'reg' and 'rreg' fields and use arrays
instead. Such changes will be rather intrusive and touch multiple codec
and platform drivers. Removing restrictions is the right thing to do,
but this will need to be done in steps with lots of validation.

Signed-off-by: Jaska Uimonen <jaska.uimonen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20201111173105.1927466-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: codecs: add support for LPASS Codec macros" from Srinivas Kandagatla <srinivas.kandagatla@linaro.org>:

This patchset adds support for two Codec Macro blocks( WSA and VA) available in
Qualcomm LPASS (Low Power Audio SubSystem).

There are WSA, VA, TX and RX Macros on LPASS IP, each of the Macro block
has specific connectivity like WSA Macros are intended to connect
to WSA Smart speaker codecs via SoundWire. VA Macro is intended for DMICs,
and TX/RX for Analog codecs via SoundWire like other WCD Codecs to provide
headphone/ear/lineout etc ..

Most of the work is derived from downstream Qualcomm kernels.
Credits to various Qualcomm authors from Patrick Lai's team who have
contributed to this code.

This patchset has been tested on support to Qualcomm Robotics RB5 Development
Kit based on QRB5165 Robotics SoC. This board has 2 WSA881X smart speakers
with onboard DMIC connected to internal LPASS codec via WSA  and VA macros
respectively.

Thanks,
srini

-Changes since v2:
 - various unnecessary variable intializations removed, suggested by Pierre
 - fixed a static checker error
 - collected reviews for dt-bindings.
 - fixed licence headers as suggested by Pierre.

Srinivas Kandagatla (6):
  ASoC: qcom: dt-bindings: add bindings for lpass wsa macro codec
  ASoC: codecs: lpass-wsa-macro: Add support to WSA Macro
  ASoC: codecs: lpass-wsa-macro: add dapm widgets and route
  ASoC: qcom: dt-bindings: add bindings for lpass va macro codec
  ASoC: codecs: lpass-va-macro: Add support to VA Macro
  ASoC: codecs: lpass-va-macro: add dapm widgets and routes

 .../bindings/sound/qcom,lpass-va-macro.yaml   |   67 +
 .../bindings/sound/qcom,lpass-wsa-macro.yaml  |   69 +
 sound/soc/codecs/Kconfig                      |    8 +
 sound/soc/codecs/Makefile                     |    4 +
 sound/soc/codecs/lpass-va-macro.c             | 1503 ++++++++++
 sound/soc/codecs/lpass-wsa-macro.c            | 2464 +++++++++++++++++
 sound/soc/codecs/lpass-wsa-macro.h            |   17 +
 7 files changed, 4132 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/sound/qcom,lpass-va-macro.yaml
 create mode 100644 Documentation/devicetree/bindings/sound/qcom,lpass-wsa-macro.yaml
 create mode 100644 sound/soc/codecs/lpass-va-macro.c
 create mode 100644 sound/soc/codecs/lpass-wsa-macro.c
 create mode 100644 sound/soc/codecs/lpass-wsa-macro.h

--
2.21.0

ASoC: pcm: call snd_soc_dapm_stream_stop() in soc_pcm_hw_clean

Currently, the SND_SOC_DAPM_STREAM_START event is sent during
pcm_prepare() but the SND_SOC_DAPM_STREAM_STOP event is
sent only in dpcm_fe_dai_shutdown() after soc_pcm_close().
This results in an imbalance between when the DAPM widgets
receive the PRE/POST_PMU/PMD events. So call
snd_soc_dapm_stream_stop() in soc_pcm_hw_clean() before the
snd_soc_pcm_component_hw_free() to keep the stream_stop DAPM
event balanced with the stream_start event in soc_pm_prepare().

Also, in order to prevent duplicate DAPM stream events,
remove the call for DAPM STREAM_START event in dpcm_fe_dai_prepare()
and the call for DAPM STREAM_STOP event in dpcm_fe_dai_shutdown().

Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20201117215001.163107-1-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: add hw specific PCM constraints

Part of PCM constraints are set based on DSP topology, but rest
should be set based on hardware capabilities. Add PCM constraints
for Intel platforms:

- Add constraint for the period count to be integer. This avoids
  wrap-arounds of the DMA circular buffer in middle of a period.

- Align period size to dword/32bit as per HDA spec.

Both constraints are aligned with current implementation in snd-hda-intel
driver.

Signed-off-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/20201118140545.2138895-2-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: relax PCM period and buffer size constraints

Current SOF implementation limits period and buffer sizes to multiples
of period_min. Period_min is defined in topology, but is in practise set
to align with the SOF DSP timer tick (typically 1ms).

While this approach helps user-space to avoid period sizes, which are
not aligned to the DSP timer tick, it causes problems to applications
which want to align data processing size to that of ALSA period size.
One example is JACK audio server, which limits period sizes to power of
two values.

Other ALSA drivers where audio data transfer is driven by a timer tick,
like USB, do not constraint period and buffer sizes to exact multiple of
the timer tick.

To align SOF to follow the same behaviour, drop the additional alignment
constraints. As a side-effect, this patch can cause irregularity to
period wakeup timing. This happens when application chooses settings
which were previously forbidden. For example, if application configures
period size to 2^14 bytes and audio config of S32_LE/2ch/48000Hz, one
period represents 42.667ms of audio. Without this patch, this
configuration is not allowed by SOF. With the patch applied,
configuration is allowed but the wakeups are paced by the DSP timer
tick, which is typically 1ms. Application will see period wakeups with a
42/43/42/43ms repeating pattern.

Both approaches are valid within ALSA context, but relaxing the
constraints is better aligned with existing applications and other ALSA
drivers like USB audio.

Signed-off-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/20201118140545.2138895-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: select WANT_DEV_COREDUMP

Select WANT_DEV_COREDUMP for catpt driver.

Signed-off-by: Piotr Maziarz <piotrx.maziarz@linux.intel.com>
Signed-off-by: Gustaw Lewandowski <gustaw.lewandowski@linux.intel.com>
--
Changes in v2:
 - change should be added to catpt only

Acked-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201117145223.21222-1-gustaw.lewandowski@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Correct the clock source for mclk0

On VF610, mclk0 = bus_clk;
On i.MX6SX/6UL/6ULL/7D, mclk0 = mclk1;
On i.MX7ULP, mclk0 = bus_clk;
On i.MX8QM/8QXP, mclk0 = bus_clk;
On i.MX8MQ/8MN/8MM/8MP, mclk0 = bus_clk;

So add variable mclk0_is_mclk1 in fsl_sai_soc_data to
distinguish these platforms.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Acked-by: Nicolin Chen <nicoleotsuka@gmail.com>
Link: https://lore.kernel.org/r/1605768038-4582-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sunxi: do not select COMMON_CLK to fix builds

COMMON_CLK is a user-selectable option with its own dependencies.  The
most important dependency is !HAVE_LEGACY_CLK.  User-selectable drivers
should not select COMMON_CLK because they will create a dependency cycle
and build failures.  For example on MIPS a configuration with COMMON_CLK
(selected by SND_SUN8I_CODEC) and HAVE_LEGACY_CLK (selected by
SOC_RT305X) is possible:

  WARNING: unmet direct dependencies detected for COMMON_CLK
    Depends on [n]: !HAVE_LEGACY_CLK [=y]
    Selected by [y]:
    - SND_SUN8I_CODEC [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] &&
      (ARCH_SUNXI || COMPILE_TEST [=y]) && OF [=y] && (MACH_SUN8I || ARM64 && ARCH_SUNXI || COMPILE_TEST [=y])

    /usr/bin/mips-linux-gnu-ld: drivers/clk/clk.o: in function `clk_set_rate':
    (.text+0xaeb4): multiple definition of `clk_set_rate'; arch/mips/ralink/clk.o:(.text+0x88): first defined here

Signed-off-by: Krzysztof Kozlowski <krzk@kernel.org>
Reviewed-by: Samuel Holland <samuel@sholland.org>
Acked-by: Maxime Ripard <mripard@kernel.org>
Link: https://lore.kernel.org/r/20201118201420.4878-1-krzk@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: lpass-va-macro: add dapm widgets and routes

Add dapm widgets and routes for this codec.

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/20201105113458.12360-7-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: lpass-va-macro: Add support to VA Macro

Qualcomm LPASS (Low Power Audio SubSystem) has internal codec
VA macro block which is used for connecting with DMICs.

This patch adds support to the codec part of the VA Macro block

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/20201105113458.12360-6-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: dt-bindings: add bindings for lpass va macro codec

This binding is for LPASS has internal codec VA macro which is
for connecting with DMICs.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201105113458.12360-5-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: lpass-wsa-macro: add dapm widgets and route

This patch adds dapm widgets and routes on this codec

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/20201105113458.12360-4-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: lpass-wsa-macro: Add support to WSA Macro

Qualcomm LPASS (Low Power Audio SubSystem) has internal codec
WSA macro block which is used for connecting with WSA Smart
speakers over soundwire.

This patch adds support to the codec part of the WSA Macro block.

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/20201105113458.12360-3-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: dt-bindings: add bindings for lpass wsa macro codec

This binding is for LPASS has internal codec WSA macro which is
for connecting with WSA Smart speakers.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201105113458.12360-2-srinivas.kandagatla@linaro.org
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: soc-component: add snd_soc_component_xxx()" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark

These are v2 of snd_soc_component_compr_xxx() function patches.
I think component related function should be implemented at
soc-component.c, otherwise it is confusable to read.
These are for it.

v1 -> v2
- fixup function return timing on
snd_soc_component_compr_get_params()
snd_soc_component_compr_get_metadata()

- use mutex at compr side on
snd_soc_component_compr_get_caps()
snd_soc_component_compr_get_codec_caps()
snd_soc_component_compr_copy()

Link: https://lore.kernel.org/r/878sb78ac4.wl-kuninori.morimoto.gx@renesas.com
Kuninori Morimoto (12):
  ASoC: soc-component: add snd_soc_component_compr_open()
  ASoC: soc-component: add snd_soc_component_compr_free()
  ASoC: soc-component: add snd_soc_component_compr_trigger()
  ASoC: soc-component: add snd_soc_component_compr_set_params()
  ASoC: soc-component: add snd_soc_component_compr_get_params()
  ASoC: soc-component: add snd_soc_component_compr_get_caps()
  ASoC: soc-component: add snd_soc_component_compr_get_codec_caps()
  ASoC: soc-component: add snd_soc_component_compr_ack()
  ASoC: soc-component: add snd_soc_component_compr_pointer()
  ASoC: soc-component: add snd_soc_component_compr_copy()
  ASoC: soc-component: add snd_soc_component_compr_set_metadata()
  ASoC: soc-component: add snd_soc_component_compr_get_metadata()

 include/sound/soc-component.h |  22 +++
 sound/soc/soc-component.c     | 255 ++++++++++++++++++++++++++++++++
 sound/soc/soc-compress.c      | 263 ++++------------------------------
 3 files changed, 303 insertions(+), 237 deletions(-)

--
2.25.1

Merge series "ASoC: graph card schema rework" from Rob Herring <robh@>:

This series updates Sameer's patch to repartition the graph card binding
schema and incorporate the OF graph schema. The schema was also mixing
card node and DAI node properties, so I've split the DAI part (the
'port' node) into a separate schema.

There's another problem that 'frame-master' and 'bitclock-master' have
inconsistent types of boolean and phandle. Having the properties just
point to the local or remote endpoint within an endpoint node is kind of
pointless. We should have gone with just boolean, but looks like we
already have several users. MMP OLPC is the one platform using boolean,
but it happens to work because the properties are effectively ignored
and CPU DAI as the master is the default.

Rob

Rob Herring (3):
  ASoC: dt-bindings: Use OF graph schema
  ASoC: dt-bindings: marvell,mmp-sspa: Use audio-graph-port schema
  ASoC: dt-bindings: Refine 'frame-master' and 'bitclock-master' type

Sameer Pujar (1):
  ASoC: audio-graph-card: Refactor schema

 .../bindings/sound/audio-graph-card.yaml      | 106 +-----------------
 .../bindings/sound/audio-graph-port.yaml      |  72 ++++++++++++
 .../bindings/sound/audio-graph.yaml           |  45 ++++++++
 .../bindings/sound/marvell,mmp-sspa.yaml      |  25 +----
 .../bindings/sound/renesas,rsnd.yaml          |  12 +-
 .../bindings/sound/simple-card.yaml           |   6 +-
 6 files changed, 132 insertions(+), 134 deletions(-)
 create mode 100644 Documentation/devicetree/bindings/sound/audio-graph-port.yaml
 create mode 100644 Documentation/devicetree/bindings/sound/audio-graph.yaml

base-commit: e2e99930ec006c6fe1d62af339a765ade71a0d9a
--
2.25.1

ASoC: mt8192: revert "add audio afe document"

This reverts commit 1afc60e00de3abbb6c559da409c5c1bb8c1d98ec
(dt-bindings: mediatek: mt8192: add audio afe document).

 https://mailman.alsa-project.org/pipermail/alsa-devel/2020-November/176873.html
 These dependency are continuing to upstream, I will revert this patch first.
 After the dependencies are merged, I will send 'mt8192-afe-pcm.yaml' again.

[Massaged commit message to put hint about issue being fixed into there
rather than the trailer -- broonie]

Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/1605081920-11848-1-git-send-email-jiaxin.yu@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: samsung: fix platform_no_drv_owner.cocci warnings

./sound/soc/samsung/midas_wm1811.c:534:3-8: No need to set .owner here. The core will do it.

 Remove .owner field if calls are used which set it automatically

Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci

Signed-off-by: Zou Wei <zou_wei@huawei.com>
Reviewed-by: Krzysztof Kozlowski <krzk@kernel.org>
Link: https://lore.kernel.org/r/1605529740-68757-1-git-send-email-zou_wei@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mmp-sspa: clear transmit phase bit for non-stereo formats

The transmit phase register value is never cleared during hw params.
So once hw params sets this bit to handle a two channel format, it
remains configured for dual-phase, which is not desirable for mono
playback.

Signed-off-by: Kyle Russell <bkylerussell@gmail.com>
Link: https://lore.kernel.org/r/20201106145905.365903-1-bkylerussell@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wcd9335: Remove unnecessary conversion to bool

The '>=' expression itself is bool, no need to convert it to bool.
Fix the following coccicheck warning:

./sound/soc/codecs/wcd9335.c:3982:25-30: WARNING: conversion to bool not needed here

Reported-by: Tosk Robot <tencent_os_robot@tencent.com>
Signed-off-by: Kaixu Xia <kaixuxia@tencent.com>
Link: https://lore.kernel.org/r/1604652816-1330-1-git-send-email-kaixuxia@tencent.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: Refine 'frame-master' and 'bitclock-master' type

'frame-master' and 'bitclock-master' are only a single phandle, so use
the more specific 'phandle' type definition instead.

Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Sameer Pujar <spujar@nvidia.com>
Link: https://lore.kernel.org/r/20201117013349.2458416-5-robh@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: marvell, mmp-sspa: Use audio-graph-port schema

Now that we have a graph schema, reference it from the marvell,mmp-sspa
schema.

Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Lubomir Rintel <lkundrak@v3.sk>
Link: https://lore.kernel.org/r/20201117013349.2458416-4-robh@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: audio-graph-card: Refactor schema

There can be customized sound cards which are based on generic audio
graph. In such cases most of the stuff is reused from generic audio
graph. To facilitate this, refactor audio graph schema into multiple
files and the base schema can be reused for specific sound cards.

The graph card nodes and port nodes are separate entities, so they
should be separate schemas.

Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
[robh: Split out port schema. Add graph.yaml in subsequent commit]
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Sameer Pujar <spujar@nvidia.com>
Link: https://lore.kernel.org/r/20201117013349.2458416-2-robh@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_get_metadata()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_get_metadata().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87zh3l6gl8.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_set_metadata()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_set_metadata().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/871rgx7v5t.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_copy()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_copy().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87361d7v5z.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_pointer()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_pointer().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/874klt7v65.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_ack()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_ack().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/875z697v6c.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_get_codec_caps()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_get_codec_caps().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/877dqp7v6i.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_get_caps()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_get_caps().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/878sb57v6q.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_get_params()

component related function should be implemented at
soc-component.c.
This patch adds snd_soc_component_compr_get_params().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87a6vl7v6x.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_set_params()

component related function should be implemented at
soc-component.c.
This patch moves soc-compress soc_compr_components_set_params()
to soc-component as snd_soc_component_compr_set_params().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87blg17v74.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_trigger()

component related function should be implemented at
soc-component.c.
This patch moves soc-compress soc_compr_components_trigger()
to soc-component as snd_soc_component_compr_trigger().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87d00h7v7k.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_free()

component related function should be implemented at
soc-component.c.
This patch moves soc-compress soc_compr_components_free()
to soc-component as snd_soc_component_compr_free().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87eekx7v7r.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: add snd_soc_component_compr_open()

component related function should be implemented at
soc-component.c.
This patch moves soc-compress soc_compr_components_open()
to soc-component as snd_soc_component_compr_open().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/87ft5d7v7x.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Fix 7/8 spaces indentation in Kconfig

Some entries used 7 or 8 spaces instead if a single TAB.

Signed-off-by: Geert Uytterhoeven <geert@linux-m68k.org>
Link: https://lore.kernel.org/r/20201110174904.3413846-1-geert@linux-m68k.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: intel: SND_SOC_INTEL_KEEMBAY should depend on ARCH_KEEMBAY

The Intel Keem Bay audio module is only present on Intel Keem Bay SoCs.
Hence add a dependency on ARCH_KEEMBAY, to prevent asking the user about
this driver when configuring a kernel without Intel Keem Bay platform
support.

Fixes: c544912bcc2dc806 ("ASoC: Intel: Add makefiles and kconfig changes for KeemBay")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20201110145001.3280479-1-geert+renesas@glider.be
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: SND_SOC_FSL_AUD2HTX should depend on ARCH_MXC

The Freescale/NXP AUDIO TO HDMI TX module is only present on NXP i.MX 8
Series SoCs.  Hence add a dependency on ARCH_MXC, to prevent asking the
user about this driver when configuring a kernel without i.MX 8 platform
support.

Fixes: 8a24c834c053ef1b ("ASoC: fsl_aud2htx: Add aud2htx module driver")
Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Link: https://lore.kernel.org/r/20201110145120.3280658-1-geert+renesas@glider.be
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoc: adi: Kconfig: Remove depends on for ADI reference designs

Audio ADI reference designs are also used on some ZynqMP boards, and can
also be used on Intel FPGA boards and also on some more complex FPGA
combinations (FPGA cards connected through PCIe).

This change removes the dependency on Microblaze and Zynq architectures
to allow the usage of this driver for the systems described above.

Signed-off-by: Bogdan Togorean <bogdan.togorean@analog.com>
Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Link: https://lore.kernel.org/r/20201110152213.37811-1-alexandru.ardelean@analog.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: adau1977: convert text binding to yaml format

This change converts the old device-tree binding for ADAU1977 from text
format to the new yaml format.

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20201110084754.46756-1-alexandru.ardelean@analog.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: Intel: catpt: Offload fixes and code optimization" from Cezary Rojewski <cezary.rojewski@intel.com>:

First two of the series address bugs connected mainly to offload streams:
- scenarios with very low buffer sizes: RESET_STREAM IPC timeouts
- fix lp clock selection when switching between PAUSE <-> RESUME states:
  glitches on first offload when no additional stream is opened
  simultaneously

Follow ups are: code reduction and optimization oriented patches.
This has been foretold in:

[PATCH v10 00/14] ASoC: Intel: Catpt - Lynx and Wildcat point
https://www.spinics.net/lists/alsa-devel/msg116440.html
  Note: LPT power up/down sequences might get aligned with WPT once
  enough testing is done as capabilities are shared for both DSPs.

First, optimize applying of user settings - prevent redundand calls from
happening - and then as mentioned above, streamline power on/off sequence
for LPT and WPT.

Cezary Rojewski (5):
  ASoC: Intel: catpt: Skip position update for unprepared streams
  ASoC: Intel: catpt: Correct clock selection for dai trigger
  ASoC: Intel: catpt: Optimize applying user settings
  ASoC: Intel: catpt: Streamline power routines across LPT and WPT
  ASoC: Intel: catpt: Cleanup after power routines streamlining

 sound/soc/intel/catpt/core.h   |  10 ++-
 sound/soc/intel/catpt/device.c |  18 +++---
 sound/soc/intel/catpt/dsp.c    |  56 ++--------------
 sound/soc/intel/catpt/pcm.c    | 113 ++++++++++++++++-----------------
 4 files changed, 74 insertions(+), 123 deletions(-)

--
2.17.1

base-commit: 3650b228f83adda7e5ee532e2b90429c03f7b9ec

Merge series "ASoC: pcm512x: Patch series to set fmt from `set_fmt()`" from Kirill Marinushkin <kmarinushkin@birdec.com>:

Set format from `set_fmt()` func instead of `hw_params()`, plus supportive
commits

Kirill Marinushkin (4):
  ASoC: pcm512x: Fix not setting word length if DAIFMT_CBS_CFS
  ASoC: pcm512x: Rearrange operations in `hw_params()`
  ASoC: pcm512x: Move format check into `set_fmt()`
  ASoC: pcm512x: Add support for more data formats

 sound/soc/codecs/pcm512x.c | 134 ++++++++++++++++++++++++++++-----------------
 1 file changed, 84 insertions(+), 50 deletions(-)

--
2.13.6

ASoC: qcom: sc7180: Register shutdown handler for lpass platform

Register shutdown handler to stop sc7180 lpass platform driver
and to disable audio clocks.

Signed-off-by: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Signed-off-by: Srinivasa Rao Mandadapu <srivasam@codeaurora.org>
Link: https://lore.kernel.org/r/1605292702-25046-1-git-send-email-srivasam@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: lpass-sc7180: Add 32 bit format support for capture

Add 32 bit format support for capture in lpass-sc7180
snd_soc_dai_driver capabilities. Need to add contstraints
in machine driver so that only specific format allowed.

Signed-off-by: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Signed-off-by: Srinivasa Rao Manidadapu <srivasam@codeaurora.org>
Link: https://lore.kernel.org/r/1605512955-7017-1-git-send-email-srivasam@codeaurora.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: initial support for Alderlake-S

Add Kconfig entries, PCI ID and chip info for Alderlake-S product.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20201116142642.2106067-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: meson: fix COMPILE_TEST error

When compiled with CONFIG_HAVE_CLK, the kernel need to get provider for the
clock API. This is usually selected by the platform and the sound drivers
should not really care about this. However COMPILE_TEST is special and the
platform required may not have been selected, leading to this type of
error:

> aiu-encoder-spdif.c:(.text+0x3a0): undefined reference to `clk_set_parent'

Since we need a sane provider of the API with COMPILE_TEST, depends on
COMMON_CLK.

Fixes: 6dc4fa179fb8 ("ASoC: meson: add axg fifo base driver")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Link: https://lore.kernel.org/r/20201116172423.546855-1-jbrunet@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: sc7180: initialize the "no_headphone" variable

The "no_headphone" variable is never set to "false" so it could be
uninitialized.

Fixes: e936619b7ce7 ("ASoC: qcom: sc7180: Modify machine driver for sound card")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/20201116125950.GA44063@mwanda
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: Cleanup after power routines streamlining

With LPT switching to WPT-based power on/off routines, functions that
have been previously used by it are rendered redundant so remove them.

Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201116133332.8530-6-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: Streamline power routines across LPT and WPT

There is no need for separate power on/off routines for LPT and WPT as
as the protocol is shared for both platforms. Make WPT routines generic
and reuse them in LPT case too.

Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201116133332.8530-5-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: Optimize applying user settings

Initial user settings such as volume control need to be applied only
once after stream is allocated. As prepare() operation can be invoked
multiple times during the stream's lifetime, relocate
catpt_dai_apply_usettings() and call it directly within
catpt_dai_hw_params() rather than on every catpt_dai_prepare().

catpt_dai_apply_usettings() remains unchanged.

Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201116133332.8530-4-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: Correct clock selection for dai trigger

During stream start DSP firmware requires LPCS disabled as that moment in
time is resource heavy. Currently high-clock is selected on start of
second stream onwards while low-clock is re-selected before stream
actually leaves RESUME state i.e. PAUSE_STREAM call. Fix this by always
updating clock before RESUME_STREAM and directly after PAUSE_STREAM.

Fixes: a126750fc865 ("ASoC: Intel: catpt: PCM operations")
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201116133332.8530-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: catpt: Skip position update for unprepared streams

Playing with very low period sizes may lead to timeouts when awaiting
RESET_STREAM reply for offload streams. This is caused by NOTIFY_POSITION
appearing in the middle of trigger(stop).

Stream is unprepared during trigger(stop) where PAUSE_STREAM IPC gets
invoked. However, all data that is already mixed in DSP firmware's mixer
stream will still be played regardless of the pause. For offload streams,
this means possibility for another NOTIFY_POSITION to process. Keep these
notifications in check by only handling them when stream is in prepared
state.

Fixes: a126750fc865 ("ASoC: Intel: catpt: PCM operations")
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20201116133332.8530-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pcm512x: Add support for more data formats

Currently, pcm512x driver supports only I2S data format.
This commit adds RJ, LJ, DSP_A and DSP_B as well.

I don't expect regression WRT existing sound cards, because:

* default value in corresponding register of pcm512x codec is 0 ==  I2S
* existing in-tree sound cards with pcm512x codec are configured for I2S
* i don't see how existing off-tree sound cards with pcm512x codec could be
  configured differently - it would not work
* tested explicitly, that there is no regression with Raspberry Pi +
  sound card `sound/soc/bcm/hifiberry_dacplus.c`

Signed-off-by: Kirill Marinushkin <kmarinushkin@birdec.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Matthias Reichl <hias@horus.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: alsa-devel@alsa-project.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201115122306.18164-5-kmarinushkin@birdec.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pcm512x: Move format check into `set_fmt()`

I would like to describe the reasoning by quoting Peter Ujfalusi
<peter.ujfalusi@ti.com> from his review of this patch series v1 [1]:

> When you bind a link you will use set_fmt for the two sides to see if
> they can agree, that both can support what has been asked.
>
> The pcm512x driver just saves the fmt and say back to that card:
> whatever, I'm fine with it. But runtime during hw_params it can fail due
> to unsupported bus format, which it actually acked to be ok.
>
> This is the difference.
>
> Sure, some device have constraint based on the fmt towards the hw_params
> and it is perfectly OK to do such a checks and rejections or build
> rules/constraints based on fmt, but failing hw_params just because
> set_fmt did not checked that the bus format is not even supported is not
> a nice thing to do.

[1] https://patchwork.kernel.org/project/alsa-devel/patch/
    20201109212133.25869-1-kmarinushkin@birdec.com/

Signed-off-by: Kirill Marinushkin <kmarinushkin@birdec.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Matthias Reichl <hias@horus.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: alsa-devel@alsa-project.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201115122306.18164-4-kmarinushkin@birdec.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pcm512x: Rearrange operations in `hw_params()`

This commit is a preparation for the next patch in the series.
It's goal is to make format check easy-to-move-out. Theoretically, more
butifications are possile in `hw_params()` func, but my intention in this
commit is to keep behaviour unchanged.

Signed-off-by: Kirill Marinushkin <kmarinushkin@birdec.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Matthias Reichl <hias@horus.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: alsa-devel@alsa-project.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201115122306.18164-3-kmarinushkin@birdec.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: pcm512x: Fix not setting word length if DAIFMT_CBS_CFS

In `pcm512x_hw_params()`, the following switch-case:

~~~~
switch (pcm512x->fmt & SND_SOC_DAIFMT_MASTER_MASK) {
case SND_SOC_DAIFMT_CBS_CFS:
~~~~

returns 0, which was preventing word length from being written into codecs
register.

Fixed by writing it into register before checking
`SND_SOC_DAIFMT_MASTER_MASK`.
Tested with Raspberry Pi + sound card `hifiberry_dacplus` in CBS_CFS format

Signed-off-by: Kirill Marinushkin <kmarinushkin@birdec.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Matthias Reichl <hias@horus.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Cc: Peter Ujfalusi <peter.ujfalusi@ti.com>
Cc: alsa-devel@alsa-project.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Link: https://lore.kernel.org/r/20201115122306.18164-2-kmarinushkin@birdec.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: use inclusive language for bclk/fsync/topology" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

The SOF (Sound Open Firmware) tree contains a lot of references in
topology files to 'codec_slave'/'codec_master' terms, which in turn
come from alsa-lib and ALSA/ASoC topology support at the kernel
level. These terms are no longer compatible with the guidelines
adopted by the kernel community [1], standard organizations, and need
to change in backwards-compatible ways.

The main/secondary terms typically suggested in guidelines don't mean
anything for clocks, this patchset suggests instead the use of
'provider' and 'consumer' terms, with the 'codec' prefix kept to make
it clear that the codec is the reference. The CM/CS suffixes are also
replaced by CP/CC.

It can be argued that the change of suffix is invasive, but finding a
replacement that keeps the M and S shortcuts has proven difficult in
quite a few contexts.

The previous definitions are kept for backwards-compatibility so this
change should not have any functional impact. It is suggested that new
contributions only use the new terms but there is no requirement to
transition immediately to the new definitions for existing code. Intel
will however update all its past contributions related to bit
clock/frame sync configurations immediately.

This patchset contains the kernel changes only, the alsa-lib changes
were shared separately.

Feedback welcome
~Pierre

[1] https://lkml.org/lkml/2020/7/4/229

Pierre-Louis Bossart (4):
  ASoC: topology: use inclusive language for bclk and fsync
  ASoC: SOF: use inclusive language for bclk and fsync
  ASoC: Intel: atom: use inclusive language for SSP bclk/fsync
  ASoC: Intel: keembay: use inclusive language for bclk and fsync

 include/sound/soc-dai.h                  | 32 +++++++++++++++---------
 include/sound/sof/dai.h                  | 16 ++++++++----
 include/uapi/sound/asoc.h                | 22 ++++++++++------
 sound/soc/intel/atom/sst-atom-controls.c | 12 ++++-----
 sound/soc/intel/atom/sst-atom-controls.h |  4 +--
 sound/soc/intel/keembay/kmb_platform.c   | 22 ++++++++--------
 sound/soc/intel/keembay/kmb_platform.h   |  8 +++---
 sound/soc/soc-topology.c                 | 24 +++++++++---------
 sound/soc/sof/topology.c                 | 18 ++++++-------
 9 files changed, 89 insertions(+), 69 deletions(-)

--
2.25.1

ASoC: rt5682: Add a new property to the device tree.

The patch adds a new property to the drice tree for the DMIC clcok driving.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20201113055400.11242-2-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5682: Add a new property for the DMIC clock driving

The patch adds a new property to set the DMIC clock driving.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://lore.kernel.org/r/20201113055400.11242-1-oder_chiou@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: keembay: use inclusive language for bclk and fsync

Use 'clock provider' and 'clock consumer' terms.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201112163100.5081-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: atom: use inclusive language for SSP bclk/fsync

Use 'provider' and 'consumer' terms.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201112163100.5081-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: use inclusive language for bclk and fsync

Mirror alsa-lib definitions w/ codec_provider (CP) and
codec_consumer (CC).

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201112163100.5081-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: topology: use inclusive language for bclk and fsync

Mirror suggested changes in alsa-lib.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20201112163100.5081-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

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

Three cosmetic patches and a fix for a randconfig issue.

v2: rebase to remove git am issue (worked with --3way before)

Pierre-Louis Bossart (1):
  ASoC: SOF: Intel: fix Kconfig dependency for SND_INTEL_DSP_CONFIG

Randy Dunlap (3):
  ASoC: SOF: imx: fix Kconfig punctuation
  ASoC: SOF: Kconfig: fix Kconfig punctuation and wording
  ASoC: SOF: Intel: fix Kconfig punctuation and wording

 sound/soc/sof/Kconfig       | 42 ++++++++++++------------
 sound/soc/sof/imx/Kconfig   | 10 +++---
 sound/soc/sof/intel/Kconfig | 64 ++++++++++++++++++-------------------
 3 files changed, 58 insertions(+), 58 deletions(-)

--
2.25.1

Merge series "Support SC7180 sound card variation without headset" from xuyuqing <xuyuqing@huaqin.corp-partner.google.com>:

Use compatible string to let machine driver handle the case
where ALC5682 headset codec is replaced with ADAU7002 dmic codec

xuyuqing (2):
  ASoC: google: dt-bindings: add new compatible for sc7180-coachz
  ASoC: qcom: sc7180: Add machine driver for sound card

 .../bindings/sound/google,sc7180-trogdor.yaml |  4 +-
 sound/soc/qcom/Kconfig                        |  1 +
 sound/soc/qcom/sc7180.c                       | 86 ++++++++++++++++---
 3 files changed, 77 insertions(+), 14 deletions(-)

--
2.25.1

Merge series "ASoC: soundwire: SDCA device quirk and cleanups" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Add one quirk to handle new TigerLake-SDCA device and cleanup code.

Pierre-Louis Bossart (4):
  ASoC: Intel: sof_sdw: add quirk for new TigerLake-SDCA device
  ASoC: codecs: rt*.c: remove useless pointer cast
  ASoC: rt711: remove warnings
  ASoC: codecs: max98373-sdw: align regmap use with other codecs

 sound/soc/codecs/max98373-sdw.c  |  2 +-
 sound/soc/codecs/rt1308-sdw.c    |  2 +-
 sound/soc/codecs/rt5682-sdw.c    |  2 +-
 sound/soc/codecs/rt700.c         |  2 +-
 sound/soc/codecs/rt711-sdw.c     | 15 ++++++++-------
 sound/soc/codecs/rt711.c         |  2 +-
 sound/soc/codecs/rt715.c         |  2 +-
 sound/soc/intel/boards/sof_sdw.c | 10 ++++++++++
 8 files changed, 24 insertions(+), 13 deletions(-)

--
2.25.1

Merge series "ASoC: Fix error handling in wm899x" from Zhang Qilong <zhangqilong3@huawei.com>:

The first patch fixed a wrong free in wm8997_probe. The
remaining three patches fixed PM disable depth imbalance
on error handling.

Zhang Qilong (4):
  ASoC: arizona: Fix a wrong free in wm8997_probe
  ASoC: arizona: Fix PM disable depth imbalance on error
  ASoC: wm8994: Fix PM disable depth imbalance on error
  ASoC: wm8998: Fix PM disable depth imbalance on error

 sound/soc/codecs/wm8994.c | 6 +++++-
 sound/soc/codecs/wm8997.c | 8 ++++++--
 sound/soc/codecs/wm8998.c | 4 +++-
 3 files changed, 14 insertions(+), 4 deletions(-)

--
2.25.4

ASoC: imx-ssi: Remove unused driver

The imx-ssi driver was only used by i.MX non-DT platforms.

Since 5.10-rc1, i.MX has been converted to a DT-only platform and all
board files are gone.

Remove the imx-ssi audio driver as there are no more users at all.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-6-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: eukrea: Remove the SND_SOC_IMX_SSI selection

SND_SOC_IMX_SSI was only used by i.MX non-DT platforms.

SND_SOC_EUKREA_TLV320 already selects the SND_SOC_FSL_SSI symbol,
which is enough.

Remove the unneeded SND_SOC_IMX_SSI selection.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-5-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-mc13783: Remove unused driver

The imx-mc13783 was used on imx27-pdk and imx31-pdk non-DT platforms.

Since 5.10-rc1, i.MX has been converted to a DT-only platform and all
board files are gone.

Remove the imx-mc13783 audio machine driver as there is no user at all.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-4-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: phycore-ac97: Remove unused driver

Since commit e1324ece2af4 ("ARM: imx: Remove i.MX35 board files"), the
MACH_PCM043 and MACH_PCA100 non-DT platform are no longer supported,
so get rid of their machine audio driver too.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-3-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mx27vis-aic32x4: Remove unused driver

Since commit 879c0e5e0ac7 ("ARM: imx: Remove i.MX27 board files"), the
MACH_IMX27_VISSTRIM_M10 non-DT platform is no longer supported,
so get rid of its machine audio driver too.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-2-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm1133-ev1: Remove unused driver

Since commit c93197b0041d ("ARM: imx: Remove i.MX31 board files"), the
MACH_MX31ADS_WM1133_EV1 non-DT platform is no longer supported,
so get rid of its machine audio driver too.

Signed-off-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20201110203937.25684-1-festevam@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: fix Kconfig dependency for SND_INTEL_DSP_CONFIG

SND_INTEL_DSP_CONFIG is selected by the HDaudio, Skylake and SOF
drivers. When the HDaudio link is not selected as a option, this
Kconfig option is not touched and will default to whatever other
drivers selected. In the case e.g. where HDaudio is compiled as
built-in, the linker will complain:

ld: sound/soc/sof/sof-pci-dev.o: in function `sof_pci_probe':
sof-pci-dev.c:(.text+0x5c): undefined reference to
`snd_intel_dsp_driver_probe'

Adding the select for all HDaudio platforms, regardless of whether
they rely on the HDaudio link or not, solves the problem.

Reported-by: Randy Dunlap <rdunlap@infradead.org>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Fixes: 82d9d54a6c0ee ('ALSA: hda: add Intel DSP configuration / probe code')
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/20201112164425.25603-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: fix Kconfig punctuation and wording

Add periods at the end of sentences in help text.
Insert a "not" in one place.
Fix help text indentation.
Remove a duplicate "Intel(R)" in one place.

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20201112164425.25603-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Kconfig: fix Kconfig punctuation and wording

Add periods at the end of sentences in help text.
Drop "a" in one place where it is not needed.
Fix a verb.
Add some capitalization.

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20201112164425.25603-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: imx: fix Kconfig punctuation

Add periods at the end of sentences in help text.

Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20201112164425.25603-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: loader: do not warn about unknown firmware headers

The firmware extended data IPC and manifest structures are designed to
be extendable without breaking the driver-firmware ABI. Driver should
not raise a warning in case a new header type is detected at
firmware boot. There are already checks for IPC ABI compatibility in
snd_sof_ipc_valid() and if the versions are deemed compatible, extra
fields in IPC messages should not trigger warnings.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Karol Trzciński <karolx.trzcinski@linux.intel.com>
Link: https://lore.kernel.org/r/20201111173321.1933452-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: Boards: tgl_max98373: add dpcm_capture flag for speaker_smart_amp

Smart_amp_speaker device has the playback stream and capture stream
associated to it. Hence add the dpcm_capture = 1 flag while dailink
creation.
This patches fixes:
ERR kernel [timestamp] SSP1-Codec: ASoC: no backend capture stream

Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Dharageswari R <dharageswari.r@intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20201109210958.84198-1-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: max98373-sdw: align regmap use with other codecs

It's not clear why this driver has an additional call to
regmap_mark_dirty(), remove to align with others.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20201111214318.150529-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt711: remove warnings

cppcheck complains, use separate loop variable for sink and source ports

sound/soc/codecs/rt711-sdw.c:382:4: style: Variable 'i' is reassigned
a value before the old one has been used. [redundantAssignment]
 i = 0;
   ^
sound/soc/codecs/rt711-sdw.c:371:4: note: i is assigned
  i++;
   ^
sound/soc/codecs/rt711-sdw.c:382:4: note: i is overwritten
 i = 0;
   ^

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Link: https://lore.kernel.org/r/20201111214318.150529-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: rt*.c: remove useless pointer cast

Pointer cast is not necessary, remove.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Link: https://lore.kernel.org/r/20201111214318.150529-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_sdw: add quirk for new TigerLake-SDCA device

Add quirks for jack detection, rt715 DAI and number of speakers.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@linux.intel.com>
Link: https://lore.kernel.org/r/20201111214318.150529-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: sc7180: Modify machine driver for sound card

Bypass set jack because there is no jack on coachz.
Create route for dmic.

Signed-off-by: xuyuqing <xuyuqing@huaqin.corp-partner.google.com>
Link: https://lore.kernel.org/r/20201112014328.695232-3-xuyuqing@huaqin.corp-partner.google.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: google: dt-bindings: add new compatible for sc7180-coachz

Add devicetree bindings for coachz in documentation file

Signed-off-by: xuyuqing <xuyuqing@huaqin.corp-partner.google.com>
Link: https://lore.kernel.org/r/20201112014328.695232-2-xuyuqing@huaqin.corp-partner.google.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: arizona: Fix a wrong free in wm8997_probe

In the normal path, we should not free the arizona,
we should return immediately. It will be free when
call remove operation.

Fixes: 31833ead95c2c ("ASoC: arizona: Move request of speaker IRQs into bus probe")
Reported-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Acked-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20201111130923.220186-2-zhangqilong3@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "Fix PM disable depth imbalance on error" from Zhang Qilong <zhangqilong3@huawei.com>:

The pm_runtime_enable will increase power disable depth. Thus
a pairing decrement is needed on the error handling path to
keep it balanced according to context. This series of patches
fixed it.

Zhang Qilong (3):
  ASoC: wm8994: Fix PM disable depth imbalance on error
  ASoC: wm8997: Fix PM disable depth imbalance on error
  ASoC: wm8998: Fix PM disable depth imbalance on error

 sound/soc/codecs/wm8994.c | 6 +++++-
 sound/soc/codecs/wm8997.c | 9 +++++++--
 sound/soc/codecs/wm8998.c | 4 +++-
 3 files changed, 15 insertions(+), 4 deletions(-)

--
2.25.4

ASoC: mediatek: mt8192: skip first time data at the beginning of DMIC recording

We can choose to drop away any length of data from the beginning according
to project needs. Some projects don't want to throw away any data, because
they want to use recorded data to do echo cancellation, so they have to
make sure that they are aligned with the reference data as much as
possible. Or there are other algorithms in the upper layer to eliminate
this noise. Or some projects want to eliminate this noise form the kernel
layer. However, the minimum recommended value is 50ms to skip pop noise.

Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com>
Link: https://lore.kernel.org/r/1605062722-8711-1-git-send-email-jiaxin.yu@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Remove mt6359_platform_driver_remove

remove mt6359_platform_driver_remove due to it is
useless.

Signed-off-by: Shane.Chien <shane.chien@mediatek.com>
Link: https://lore.kernel.org/r/1605068096-12587-1-git-send-email-shane.chien@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tegra: Don't warn on probe deferral

Deferred probe is an expected return value for snd_soc_register_card().
Given that the driver deals with it properly, there's no need to output
a warning that may potentially confuse users.

Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Link: https://lore.kernel.org/r/20201111103245.152189-1-jonathanh@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8998: Fix PM disable depth imbalance on error

The pm_runtime_enable will increase power disable depth. Thus
a pairing decrement is needed on the error handling path to
keep it balanced according to context.

Fixes: 31833ead95c2c ("ASoC: arizona: Move request of speaker IRQs into bus probe")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20201111041326.1257558-4-zhangqilong3@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8994: Fix PM disable depth imbalance on error

The pm_runtime_enable will increase power disable depth. Thus
a pairing decrement is needed on the error handling path to
keep it balanced according to context.

Fixes: 57e265c8d71fb ("ASoC: wm8994: Move runtime PM init to platform device init")
Signed-off-by: Zhang Qilong <zhangqilong3@huawei.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20201111041326.1257558-2-zhangqilong3@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "Audio Graph Updates" from Sameer Pujar <spujar@nvidia.com>:

This series is a prepraration for using generic graph driver for Tegra210
audio. Tegra audio graph series will be sent in a separate series because
it has some dependency over other series for documentation work. This
series can focus on the generic ASoC driver updates. Below are the summary
of changes done.

 * Support multiple instances of a component. For example there can be
   multiple I2S devices which can use the same component driver.

 * Support open platforms with empty Codec endpoint. Customers can plug
   their own HW and can populate codec endpoint.

 * In a component model there can be many components which can be
   connected together. In such cases Identify no-pcm DPCM DAI links which
   can be used in BE<->BE connections.

 * Expose structures or helpers to be re-used by other similar graph
   drivers.

The series is based on following references where DPCM usgae for Tegra
Audio and simple-card driver proposal were discussed.

 * https://lkml.org/lkml/2020/4/30/519 (DPCM for Tegra)
 * https://lkml.org/lkml/2020/6/27/4 (simple-card driver)

Changelog
=========

v4 -> v5
--------
 * No changes in the core/audio-graph driver patches which are
   part of the current series.
 * Dropped Tegra audio patches and doc patches. This will be
   sent separately once the doc depdendencies are resolved.

v3 -> v4
--------
 * Added new patches to convert graph.txt and audio-graph-card.txt
   to corresponding json-schema files. Later these references
   are used in Tegra audio graph schema.

 * AHUB component binding docs are updated to reflect the usage
   of ports/port/endpoint

 * More common stuff is moved into graph_parse_of() and this is
   used by both generic and Tegra audio graph.

 * DT binding for Tegra audio graph is updated to included "ports { }"

 * As per the suggestion 'void *data' member is dropped from
   'asoc_simple_priv' and instead container method is used to
   maintain required custom data internal to Tegra audio graph.

v2 -> v3
--------
 * Dropped new compatible addition in generic graph driver
   after reviewing it with Morimoto-san. Instead added Tegra
   audio graph driver and new compatibles are added in the same.
 * Added new patches to expose new members for customization
   in audio graph driver.
 * Added new patch for Tegra audio graph driver and related
   documentation.
 * Minor change in below commit where mutex version of helper is used
   "ASoC: audio-graph: Identify 'no_pcm' DAI links for DPCM"
 * DT binding is updated to use the newly exposed compatibles
 * No changes in other patches

v1 -> v2
--------
 * Re-organized ports/endpoints description for ADMAIF and XBAR.
   Updated DT patches accordingly.
 * After above change, multiple Codec endpoint support is not
   required and hence dropped for now. This will be considered
   separately if at all required in future.
 * Re-ordered patches in the series.

Sameer Pujar (7):
  ASoC: soc-core: Fix component name_prefix parsing
  ASoC: soc-pcm: Get all BEs along DAPM path
  ASoC: audio-graph: Use of_node and DAI for DPCM DAI link names
  ASoC: audio-graph: Identify 'no_pcm' DAI links for DPCM
  ASoC: audio-graph: Support empty Codec endpoint
  ASoC: audio-graph: Expose new members for asoc_simple_priv
  ASoC: audio-graph: Expose helpers from audio graph

 include/sound/graph_card.h           |  16 ++++
 include/sound/simple_card_utils.h    |   3 +
 include/sound/soc.h                  |   1 +
 sound/soc/generic/audio-graph-card.c | 175 ++++++++++++++++++++++++-----------
 sound/soc/soc-core.c                 |   3 +-
 sound/soc/soc-pcm.c                  |   3 +-
 6 files changed, 143 insertions(+), 58 deletions(-)
 create mode 100644 include/sound/graph_card.h

--
2.7.4

ASoC: qcom: common: Fix refcounting in qcom_snd_parse_of()

There are two issues in this function.

1) We can't drop the refrences on "cpu", "codec" and "platform" before
   we take the reference.  This doesn't cause a problem on the first
   iteration because those pointers start as NULL so the of_node_put()
   is a no-op.  But on the subsequent iterations, it will lead to a use
   after free.

2) If the devm_kzalloc() allocation failed then the code returned
   directly instead of cleaning up.

Fixes: c1e6414cdc37 ("ASoC: qcom: common: Fix refcount imbalance on error")
Fixes: 1e36ea360ab9 ("ASoC: qcom: common: use modern dai_link style")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/20201105125154.GA176426@mwanda
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_xcvr: fix break condition

The break condition copied by mistake as same
as loop condition in the previous version, but must
be the opposite. So fix it.

Signed-off-by: Viorel Suman <viorel.suman@nxp.com>
Link: https://lore.kernel.org/r/20201102161810.902464-1-viorel.suman@oss.nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: audio-graph: Expose helpers from audio graph

This commit exposes following functions which can be used by a sound
card driver based on generic audio graph driver. Idea is vendors can
have a thin driver and re-use common stuff from audio graph driver.

 - graph_card_probe()
 - graph_parse_of()

In doing so a new header file is added for above. The graph_probe()
function is simplified by moving more common stuff to graph_parse_of().

Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-8-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: audio-graph: Expose new members for asoc_simple_priv

Add new members in struct 'asoc_simple_priv'. Idea is to leverage
simple or graph card driver as much as possible and vendor can
maintain a thin driver to control the behavior by populating these
newly exposed members.

Following are the members added in 'asoc_simple_priv':

  - 'ops' struct: In some cases SoC vendor drivers may want to
    implement 'snd_soc_ops' callbacks differently. In such cases
    custom callbacks would be used.

  - 'force_dpcm' flag: Right now simple or graph card drivers
    detect DAI links as DPCM links if:

      * The dpcm_selectable is set AND
      * Codec is connected to multiple CPU endpoints or aconvert
        property is used for rate/channels.

    So there is no way to directly specify usage of DPCM alone. So a
    flag is exposed to mark all links as DPCM. Vendor driver can
    set this if required.

  - 'dpcm_selectable': Currently simple or audio graph drivers
    provide a way to enable this for specific compatibles. However
    vendor driver may want to define some additional info. Thus
    expose this variable where vendor drivers can set this if
    required.

Audio graph driver is updated to consider above flags or callbacks.
Subsequent patches in the series illustrate usage for above.

Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-7-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: audio-graph: Support empty Codec endpoint

For open platforms, which can support pluggable audio cards, Codec
endpoint is not fixed always. It actually depends on the compatible
HW module that is going to be connected. From SoC side the given I/O
interface is always available. Hence such links have fixed CPU endpoint
but no Codec endpoint. This patch helps to support such links where
user can populate Codec endpoint only and its fields in Platform DT
depending on the plugged HW.

Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-6-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: audio-graph: Identify 'no_pcm' DAI links for DPCM

PCM devices are created for FE dai links with 'no-pcm' flag as '0'.
Such DAI links have CPU component which implement either pcm_construct()
or pcm_new() at component or dai level respectively. Based on this,
current patch exposes a helper function to identify such components
and populate 'no_pcm' flag for DPCM DAI link.

This helps to have BE<->BE component links where PCM devices need
not be created for CPU component involved in such links.

Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/1604329814-24779-5-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>