soc: qcom: Add APR bus driver

This patch adds support to APR bus (Asynchronous Packet Router) driver.
APR driver is made as a bus driver so that the apr devices can added removed
more dynamically depending on the state of the services on the dsp.
APR is used for communication between application processor and QDSP to
use services on QDSP like Audio and others.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-and-tested-by: Rohit kumar <rohitkr@codeaurora.org>
Acked-by: Andy Gross <andy.gross@linaro.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

soc: qcom dt-bindings: Add APR bus bindings

This patch add dt bindings for Qualcomm APR (Asynchronous Packet Router)
bus driver. This bus is used for communicating with DSP which provides
audio and various other services to cpu.

Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: Rob Herring <robh@kernel.org>
Reviewed-by: Bjorn Andersson <bjorn.andersson@linaro.org>
Reviewed-by: Banajit Goswami <bgoswami@codeaurora.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: removed separate byte count variables for playback and capture

Removed separate byte count variables for playback and capture.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: added byte count register offset variables to rtd

Added byte count register offset variables to audio_substream_data
structure. Modified dma pointer callback.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: dma config parameters changes

Added dma configuration parameters to rtd structure.
Moved dma configuration parameters initialization to
hw_params callback.
Removed hard coding in prepare and trigger callbacks.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Reviewed-by: Daniel Kurtz <djkurtz@chromium.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cirrus: i2s: IRQ-based stream watchdog

I2S controller on EP93xx seems to have undocumented HW issue. According to
"EP93xx User’s Guide", controller can handle underflow and either transmit
last sample or zeroes in such case until FIFO is filled again. In reality
undeflow conditions seem to confuse internal state machine from time to
time and the whole stream gets shifted by one byte (as captured by logic
analyser on the I2S outputs). One could only hear noise instead of original
stream and this continues until the FIFO is disabled and enabled again.

Work this around by watching underflow interrupt and resetting I2S TX
channel + fill FIFO with zero samples until DMA catches up again. This is
a nasty workaround, but it works. Hence, Kconfig option to disable it in
case of problems.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ARM: ep93xx: i2s: Add IRQ to platform device resources

According to "EP93xx User’s Guide" it's called I2SINTR and has number 60.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cirrus: i2s: Stop enabling I2S2 and I2S3 FIFOs

The driver never supported more than 2 channels because of
ep93xx_i2s_dma_data[] supporting only 1 DMA channel in each
direction.
Stop enabling two unused I2S controller FIFOs, this will simplify
future interrupt support.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5645: Add platform-data for Lenovo Ideapad Mixx 320

The Lenovo Ideapad Mixx 320 has a digital mic connected to DMIC2
add a DMI based quirk pointing to the intel_braswell_platform_data
for devices with a mic on DMIC2.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5645: Add platform-data for Lenovo Ideapad Mixx 310

The Lenovo Ideapad Mixx 310 has a differential internal analog mic,
add platform-data for this.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap: Delete the obsolete omap-pcm

All DAI drivers are now using the new sdma-pcm platform driver. The
omap-pcm can be removed from the tree, but we need to keep the SND_OMAP_SOC
Kconfig option until the relevant defconfigs are updated to avoid
regression due to missing audio.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: davinci-mcasp: Convert to use the sdma-pcm instead of omap-pcm

Use the new platform driver in case of sDMA.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: bytcr_rt5640: Use device-property for differential mics

Set the "realtek,in1-differential" or "realtek,in3-differential"
device-property when the BYT_RT5640_DIFF_MIC quirk is set instead of
directly poking the codec registers.

This also fixes the BYT_RT5640_DIFF_MIC quirk not working when
combined with BYT_RT5640_IN3_MAP.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: bytcr_rt5640: Configure PLL1 before using it

When platform_clock_control() first selects PLL1 as sysclk the PLL_CTRL
registers have not been setup yet and we effectively have an invalid clock
configuration until byt_rt5640_aif1_hw_params() gets called.

Add a new byt_rt5640_prepare_and_enable_pll1() helper and use that from
both platform_clock_control() and byt_rt5640_aif1_hw_params() to fix this.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Add button press support

Enable button press detection for headsets by using the ovcd IRQ to get
notified of button presses.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Add jack-detect support

Add jack-detect support, loosely based on earlier work on this by:

Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Francisco mendez <francisco.mendez@intel.com>

Note getting the OVCD to work reliable was sort of finicky, so there are
quite a few comments on this to hopefully avoid people breaking it in the
future.

This (and the follow-up button press support) has been tested on the
following devices:

Acer Iconia Tab 8 W1-810
Asus T100CHI
Asus T100TA
Asus T200TA
Axxo WT1011
Chuwi Vi8
Dell Venue 8 Pro 5830
HP Pavilion X2 10-n000nd
HP Stream 7
I.T. Works TW891
Lamina I8270
MSI S100
Peaq C1010
Pipo W4
PoV MobiiTAB-P800W (v2.0)
Toshiba Click Mini L9W-B

BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=196377
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Allow specifying dmic data pins through device-properties

Allow specifying dmic data pins through device-properties / dt. This will
allow us to stop exporting rt5640_dmic_enable() once all callers of it have
been converted to setting device-properties for this instead.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Move checking of device-properties to component probe callback

On some platforms the platform code may need to add device-properties,
rather then relying only on properties set by the firmware.

This commit moves the parsing of the device-properties from the i2c-driver
probe() function, which may be called at any time, to the component-driver
probe() function, which gets called after the platform code calls
snd_soc_register_card().

This allows the platform code to attach extra device-properties before
the device-properties are parsed.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Remove unused rt5640_platform_data

There are no in tree users of platform-data for the rt5640 codec driver,
so lets remove support for it.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Add devicetree-bindings for dmic, jack-detect

Add devicetree-bindings for the dmic, jack-detect source and overcurrent-
detect threshold settings.

The dmic bindings mirror the existing bindings for the rt5645.
The jd-src and ovcd bindings mirror the existing bindings for the rt5651.

Cc devicetree@vger.kernel.org
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5640: Remove is_sys_clk_from_pll, it has ordering issues

is_sys_clk_from_pll() is used as a snd_soc_dapm_route.connected callback,
checking RT5640_GBL_CLK to determine if the sys-clk is PLL1 and thus the
PWR_PLL bit in reg PWR_ANLG2 must be set.

RT5640_GBL_CLK is changed by rt5640_set_dai_sysclk(), which gets called by
the pre_pmu / post_pmd functions of the "Platform Clock" dapm-supply.

This creates an ordering issue, during a dapm transition first all
connected() callbacks are called to build a list of supplies to enable
and then the complete list is walked to enable the supplies. Since the
connected() check happens before enabling any supplies,
is_sys_clk_from_pll() ends up deciding if the PWR_PLL bit should be set
based on the state the "Platform Clock" supply had *before* the transition.
This sometimes results in PWR_PLL being off, even though *after* the
transition PLL1 is configured as sys-clk.

This commit removes is_sys_clk_from_pll() instead simply setting / clearing
PWR_PLL in rt5640_set_dai_sysclk() based on the selected sys-clk, which
fixes this and as a bonus results in a nice cleanup.

Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: zx-i2s: fix spelling mistake: "timeing" -> "timing"

Trivial fix to spelling mistake in dev_err message text

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: uniphier: add digital output volume for UniPhier sound system

This patch adds controllers for digital volume of PCM output. Volume
effects simply linear, not dB scale as follows:
  Gained PCM = Original * 0x4000 / Volume

The value range of volume is from 0x0001 to 0xffff. 0x0000 works as
mute. Initial value is 0x4000 (+0dB).

Signed-off-by: Katsuhiro Suzuki <suzuki.katsuhiro@socionext.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: remove legacy_dai_naming from snd_soc_register_dais()

We can get legacy dai name flag from component driver.
Thus, there is no need to have its parameter on snd_soc_register_dais().
Let's remove unneeded parameter

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: remove snd_soc_component_add_unlocked()

There is no user to call snd_soc_component_add_unlocked() anymore.
Let's merge snd_soc_component_add_unlocked() and
snd_soc_component_add().

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc.h: merge CONFIG_DEBUG_FS

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: convert platform explanation to component

commit ef050bece1b55 ("ASoC: Remove platform code now everything is
componentised") removed platform code, but it didn't care
about platform documentation.
This patch convert platform explanation to component

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: add component_list_show()

commit ef050bece1b55 ("ASoC: Remove platform code now everything is
componentised") removed platform code, then platform_list_show() was
removed, too. But we want to keep it as component_list_show.
This patch add it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: remove Codec related code

Now no one is using Codec related code.
Let's remove all

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: remove unneeded .pcm_new/free

commit ef050bece1b55 ("ASoC: Remove platform code now everything is
componentised") removed platform code, but it didn't remove
.pcm_new/free which existed only for platform.
This patch remove these

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: remove .get_regmap

To setup regmap, ALSA SoC has snd_soc_component_init_regmap() and
.get_regmap. But these are duplicated feature.
Now, no one is using .get_regmap, let's remove it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: remove codec reg_cache

Codec reg_cache is legacy feature, almost all driver are now using
common regmap, and very few driver had been used this legacy feature.
Because of this background, it is now implemented on each
driver internally now.
So now, no one is using codec reg_cache. Let's remove it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: remove codec hw_write/control_data

No one is using codec hw_write/control_data any more.
Let's remove these.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap-mcbsp: Convert to use the sdma-pcm instead of omap-pcm

Use the new platform driver.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap-mcpdm: Convert to use the sdma-pcm instead of omap-pcm

Use the new platform driver.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap-dmic: Convert to use the sdma-pcm instead of omap-pcm

Use the new platform driver.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap-hdmi-audio: Convert to use the sdma-pcm instead of omap-pcm

Use the new platform driver.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap: Introduce the generic_dmaengine_pcm based sdma-pcm

With the generic dmaengine_pcm support the omap-cpm can be replaced with a
much smaller wrapper.

CPU DAI drivers can use the:
int sdma_pcm_platform_register(struct device *dev,
       char *txdmachan, char *rxdmachan);

To register the platform driver, txdmachan/rxdmachan is only needed to be
provided if the DMA channel names are not standard tx/rx, like in case of
McPDM, or the DAI is only capable of one audio direction (DMIC, HDMI).

This patch only introduces the source file and changes to the
Kconfig/Makefile, but does not change any of the DAI drivers to use it.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: uniphier: evea: use DAPM to change source of line-in

This patch replaces mixer switch to DAPM one for changing audio source
of line-in.

Signed-off-by: Katsuhiro Suzuki <suzuki.katsuhiro@socionext.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: nau8824: fix spelling mistake: "semaphone" -> "semaphore"

Trivial fix to spelling mistake in dev_warn messages

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fix return value check in mt6351_codec_driver_probe()

In case of error, the function dev_get_regmap() returns NULL pointer not
ERR_PTR(). The IS_ERR() test in the return value check should be
replaced with NULL test.

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

media: i2c: tda1997: replace codec to component

Now we can replace Codec to Component. Let's do it.

Note:
xxx_codec_xxx() -> xxx_component_xxx()
.idle_bias_off = 0 -> .idle_bias_on = 1
.ignore_pmdown_time = 0 -> .use_pmdown_time = 1
- -> .endianness = 1
- -> .non_legacy_dai_naming = 1

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Tested-by: Tim Harvey <tharvey@gateworks.com>
Acked-by: Tim Harvey <tharvey@gateworks.com>
Acked-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5663: Optimize the power consumption

The patch optimizes the power consumption.

Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42xx8: Make the node name generic

According to Devicetree Specification v0.2 document:

"The name of a node should be somewhat generic, reflecting the function
of the device and not its precise programming model."

Do as suggested in the binding example.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8960: Make the node name generic

According to Devicetree Specification v0.2 document:

"The name of a node should be somewhat generic, reflecting the function
of the device and not its precise programming model."

Do as suggested in the binding example.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8962: Make the node name generic

According to Devicetree Specification v0.2 document:

"The name of a node should be somewhat generic, reflecting the function
of the device and not its precise programming model."

Do as suggested in the binding example.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sgtl5000: Make the node name generic

According to Devicetree Specification v0.2 document:

"The name of a node should be somewhat generic, reflecting the function
of the device and not its precise programming model."

Do as suggested in the binding example.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas6424: Allow disabling auto diagnostics for faster power-on

The TAS6424 incorporates both DC-load and AC-load diagnostics which are
used to determine the status of the load. The DC diagnostics runs when any
channel is directed to leave the Hi-Z state and enter the MUTE or PLAY
state.
The DC diagnostics are turned on by default but, if a fast startup without
diagnostics is required, the diagnostics can be disabled using a dedicated
ALSA control.

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: nau8810: change input PGA mixer stage

Organize the paths of the mixer, "Input Boost Stage", including
the routes of the mixer. The control is not used correctly before.
Besides, the driver changes the name of the mixer controls.

Signed-off-by: John Hsu <KCHSU0@nuvoton.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_spdif: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_esai: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_ssi: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge remote-tracking branch 'asoc/for-4.17' into asoc-4.18 to resolve a
conflict between a fix and new development in mtk

ASoC: dai playback and capture active may be greater than 1

At the moment playback_active and capture_active are using only 1 bit so
the maximum active count is 1.

However, snd_soc_runtime_activate() may be called several time on the
same dai. This happens when a dai is part of several dai_links. It is
often the case for "snd-soc-dummy-dai".

This is a problem if snd_soc_runtime_activate() is called an even number
of times on a dai. In this case the active count overflow back to 0. As
consequence, ASoC functions, such as soc_dpcm_runtime_update(), won't run
correctly.

Storing these usage counts on plain 'unsigned int' solves the problem.

Fixes: f0fba2ad1b6b ("ASoC: multi-component - ASoC Multi-Component Support")
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cirrus: i2s: Fix {TX|RX}LinCtrlData setup

According to "EP93xx User’s Guide", I2STXLinCtrlData and I2SRXLinCtrlData
registers actually have different format. The only currently used bit
(Left_Right_Justify) has different position. Fix this and simplify the
whole setup taking into account the fact that both registers have zero
default value.

The practical effect of the above is repaired SND_SOC_DAIFMT_RIGHT_J
support (currently unused).

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org

ASoC: cirrus: i2s: Fix LRCLK configuration

The bit responsible for LRCLK polarity is i2s_tlrs (0), not i2s_trel (2)
(refer to "EP93xx User's Guide").

Previously card drivers which specified SND_SOC_DAIFMT_NB_IF actually got
SND_SOC_DAIFMT_NB_NF, an adaptation is necessary to retain the old
behavior.

Signed-off-by: Alexander Sverdlin <alexander.sverdlin@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org

ASoC: amd: fix spelling mistake: "failer" -> "failure"

Trivial fix to spelling mistake in dev_err error message

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: add the .probe() callback in mt2701_afe_pcm_dai_component

For the sake of uniformity, this patch adds a callback mt2701_afe_pcm_probe()
in mt2701_afe_pcm_dai_component to retrieve the regmap - the canonical way to
obtain the pointer..

Doing so, we could switch to use devm_snd_soc_register_component() to register
the component driver.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: cht-bsw-rt5672: allow for topology-defined codec-dai setup

Hard-coded setups conflict with topology defined ones. Move this code to
codec_fixup so that SOF can override codec dai settings, e.g. to only use
2 channels.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: avoid using snd_soc_platform

avoid using snd_soc_platform, which is removed after 4.18

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas6424: Add support for the mute pin

mute can be connected to GPIO. In that case we have to drive it to the
correct value

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas6424: Add support for the standby pin

The standby pin can be connected to a GPIO. In that case we have to drive
it to the correct values for the TAS6424 to operate properly.

Signed-off-by: Jean-Jacques Hiblot <jjhiblot@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: core: Allow codec_conf DT lookups to match parent of_node

For devices implemented as a MFD it is common to only have a single node
in devicetree representing the whole device. As such when looking up
codec_conf mappings we should match against both the devices of_node and
the devices parent's of_node, as is already done for DAIs and platform
components.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: preallocate pages use platform device

preallocate pages should use platform device,
since we set dma mask for platform device.

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org

ASoC: mediatek: preallocate pages use platform device

preallocate pages should use platform device,
since we set dma mask for platform device.

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap: Remove OMAP_MUX dependency from Nokia N810 audio support

Commit e9f5f1e45608 ("ARM: OMAP2+: Remove legacy mux code") removed
CONFIG_OMAP_MUX making impossible to build Nokia N810 audio support. Remove
this dependency so we can do at least build tests.

Fixes: e9f5f1e45608 ("ARM: OMAP2+: Remove legacy mux code")
Signed-off-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap: n810: HS mic is not working, add a widget for it with comment

The bias for the analog HS microphone is coming from Retu/Vilma chip and
we do not have control over it, yet.

For clarity, add a new DAPM_MIC widget for the HS mic and document the
current state.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge branch 'asoc-4.17' into asoc-4.18 n810 dependencies

ASoC: omap: n810: Correct the cpu_dai, platform and codec name

The non DT boot is no longer supported and when booting with DT the device
names are different.

Fix them up for now, but the n810.c should be updated to support probing
via DT with proper bindings.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: omap: n810: Correct the card level dapm_route

Fix the capture DAPM route due to core changes regarding to mic bias.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mt6797: switch to SPDX license tag

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Reviewed-by: Matthias Brugger <matthias.bgg@gmail.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mt6797: add structure define and clock control function for 6797

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: add mt6797-mt6351 driver and config option

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mt6351 switch to SPDX license tag

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: uniphier: remove boilerplate from lisence comment

This patch removes boilerplate of GPLv2, use only SPDX identifier as
same as other recently ASoC DAI drivers.

Signed-off-by: Katsuhiro Suzuki <suzuki.katsuhiro@socionext.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_ssi: Use u32 variable type when using regmap_read()

Convert the sisr and sisr2 variable types to u32 to avoid the following
sparse warnings:

sound/soc/fsl/fsl_ssi.c:391:42: warning: incorrect type in argument 3 (different base types)
sound/soc/fsl/fsl_ssi.c:391:42:    expected unsigned int *val
sound/soc/fsl/fsl_ssi.c:391:42:    got restricted __be32 *<noident>
sound/soc/fsl/fsl_ssi.c:393:17: warning: restricted __be32 degrades to integer
sound/soc/fsl/fsl_ssi.c:393:15: warning: incorrect type in assignment (different base types)
sound/soc/fsl/fsl_ssi.c:393:15:    expected restricted __be32 [usertype] sisr2
sound/soc/fsl/fsl_ssi.c:393:15:    got unsigned int
sound/soc/fsl/fsl_ssi.c:396:50: warning: incorrect type in argument 3 (different base types)
sound/soc/fsl/fsl_ssi.c:396:50:    expected unsigned int [unsigned] val
sound/soc/fsl/fsl_ssi.c:396:50:    got restricted __be32 [usertype] sisr2
sound/soc/fsl/fsl_ssi.c:398:42: warning: incorrect type in argument 2 (different base types)
sound/soc/fsl/fsl_ssi.c:398:42:    expected unsigned int [unsigned] [usertype] sisr
sound/soc/fsl/fsl_ssi.c:398:42:    got restricted __be32 [addressable] [usertype] sisr

In other places where regmap_read() is used a u32 variable is passed
to store the register read value, so do the same here as well.

regmap API already takes care of endianness, so the usage of u32 is safe.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: rename audio_substream_data variable

In order to make audio_substream_data structure variable
consistent throughout the code, changed the name from
audio_config to rtd wherever applicable.

Signed-off-by: Vijendar Mukunda <Vijendar.Mukunda@amd.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: switch to SPDX license tag

Add SPDX identifiers to all remaining files in sound/soc/mediatek/

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: add MT7622 AFE compatible in documentation

Add support for MT7622 AFE which shares the same binding with MT2701.

Cc: devicetree@vger.kernel.org
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: add MT7622 AFE support

This patch adds support for the MT7622 AFE which reuses MT2701 driver.

We also introduce the 'struct mt2701_soc_variants' to differentiate
between the SoC generations as there might be other (existing or future)
chips that use the same binding and driver, then being a little more
abstract could help in the long run.

Cc: Jia Zeng <jia.zeng@mediatek.com>
Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Reviewed-by: Garlic Tseng <garlic.tseng@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: Add MTK_STREAM_NUM to mtk-base-afe.h

Add MTK_STREAM_NUM to common header and modify related stuff so that
the other SoCs can reuse it.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Reviewed-by: Garlic Tseng <garlic.tseng@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: simplify the control logic of MT2701 I2S

This patch adjusts the mt2701_afe_i2s_ops to simplify the control
logic of the I2S path.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Reviewed-by: Garlic Tseng <garlic.tseng@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: add mt6351 codec driver

This patch adds the MediaTek MT6351 codec driver.
MT6351 communicate with SoC through MediaTek PMIC wrapper.
MT6351 use MediaTek proprietary audio interface.

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: compress: Add helper functions for component open/free

There are 2 loops calling open and 4 loops calling free for all the
components on a DAI link. Factor out these loops into helper functions
to make the code a little clearer.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Remove platform code now everything is componentised

As all drivers have been moved over to the new generic component
code remove the now unused platform specific code.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge branch 'asoc-4.17' into asoc-4.18 for compress dependencies

ASoC: compress: Only call free for components which have been opened

The core should only call free on a component if said component has
already had open called on it. This is not presently the case and most
compressed drivers in the kernel assume it will be. This causes null
pointer dereferences in the drivers as they attempt clean up for stuff
that was never put in place.

This is fixed by aborting calling open callbacks once a failure is
encountered and then during clean up only iterating through the
component list to that point.

This is a fairly quick fix to the issue, to allow backporting. There
is more refactoring to follow to tidy the code up a little.

Fixes: 9e7e3738ab0e ("ASoC: snd_soc_component_driver has snd_compr_ops")
Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Acked-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Update email address for Vinod

Update the email address for compressed audio maintainer
Also update .mailmap.

Signed-off-by: Vinod Koul <vkoul@kernel.org>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: use snd_soc_dai_get_drvdata() to get the private data

Reduce the boilerplate code to retrieve the private data.
No functional change intended.

Signed-off-by: Ryder Lee <ryder.lee@mediatek.com>
Reviewed-by: Garlic Tseng <garlic.tseng@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: add documents for mt6797

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mt6797: add mt6797 platform driver

Signed-off-by: KaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm_adsp: Account for name prefixes when toggling preloader

Use the correct functions to allow a name prefix assigned through
codec_conf to be taken into consideration whilst enabling and disabling
the preloader widget.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sgtl5000: Switch to SPDX identifier

Adopt the SPDX license identifier headers to ease license compliance
management.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sgtl5000: Fix the spelling of 'exceed'

Fix the spelling of 'exceed' in two comments.

Signed-off-by: Fabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sh: Update menu title and platform dependency

Change the menu title to refer to "Renesas SoCs" instead of "SuperH", as
both SuperH and ARM SoCs are supported.

Since commit 9b5ba0df4ea4f940 ("ARM: shmobile: Introduce ARCH_RENESAS")
is ARCH_RENESAS a more appropriate platform dependency for Renesas ARM
SoCs than the legacy ARCH_SHMOBILE, hence use the former.
Renesas SuperH SH-Mobile SoCs are still covered by the SUPERH
dependency.

This will allow to drop ARCH_SHMOBILE on ARM and ARM64 in the near
future.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sh: Drop SUPERH platform dependency

The SIU sound peripheral is used only on SuperH SH-Mobile platforms.
As both SUPERH and ARCH_SHMOBILE are set for these platforms, the SUPERH
dependency can be dropped.

Signed-off-by: Geert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: kbl_rt5663_rt5514_max98927: Replace GFP_ATOMIC with GFP_KERNEL

In kabylake_audio_probe which is not atomic context, we use GFP_ATOMIC
flag with memory allocation, fix that by using GFP_KERNEL.

Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: kbl_rt5663_max98927: Replace GFP_ATOMIC with GFP_KERNEL

In kabylake_audio_probe which is not atomic context, we use GFP_ATOMIC
flag with memory allocation, fix that by using GFP_KERNEL.

Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: kbl_da7219_max98357a: Replace GFP_ATOMIC with GFP_KERNEL

In kabylake_audio_probe which is not atomic context, we use GFP_ATOMIC
flag with memory allocation, fix that by using GFP_KERNEL.

Signed-off-by: Vinod Koul <vinod.koul@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>