When a copier exposes a single format, we can fixup the BE dailink with
that format. This is helpful when some codec have format restrictions and
e.g. don't support a 32-bit format. In that case, the copier output
formats mirror that restriction in the topology file.

An alternate solution was suggested earlier using a dedicated topology
token. When specified, the token would be used to fix-up the dailink. The
main reason why this solution was chosen is that there is a risk of a
disconnect between token definition and copier format. With a single piece
of information as suggested in this patch, there are fewer risks of a bad
configuration.
Signed-off-by: Bard Liao <yung-chuan.liao@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: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230915093507.7242-3-peter.ujfalusi@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

We will use the sof_ipc4_copier_is_single_format() function to check if a
ipc4 copier has single format available in ipc4-pcm.c in the next patch.
Signed-off-by: Bard Liao <yung-chuan.liao@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: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230915093507.7242-2-peter.ujfalusi@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

New code uses ENOTSUPP as per checkpatch recommendation:
ENOTSUPP is not a SUSV4 error code, prefer EOPNOTSUPP
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20230914124943.24399-1-peter.ujfalusi@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

Fix prototype missing warning for acp3x_es83xx_init_ops() by
including the header acp3x-es83xx.h
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202309111220.g63yHDfH-lkp@intel.com/Signed-off-by: Marian Postevca <posteuca@mutex.one>
Link: https://lore.kernel.org/r/20230913210916.2523-1-posteuca@mutex.oneSigned-off-by: Mark Brown <broonie@kernel.org>

Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>:

The PCI device registers contain a subsystem ID (SSID), that is
separate from the silicon ID. The PCI specification defines it thus:

"They provide a mechanism for board vendors to distiguish their
 boards from one another even thought the boards may have the same
 PCI controller on them."

This allows the driver for the silicon part to apply board-speficic
settings based on this SSID.

The CS35L56 driver uses this to select the correct firmware file for
the board. The actual ID is part of the PCI register set of the
host audio interface so this set of patches includes extracting the
SSID from the Intel audio controller and passing it to the machine
driver and then to ASoC components. Other PCI audio controllers
will have the same SSID registers, so can use the same mechanism to
pass the SSID.

The patch helps save power by control MICBIAS. The headset's
MICBIAS should be disabled without button requirement.
Signed-off-by: Seven Lee <wtli@nuvoton.com>
Link: https://lore.kernel.org/r/20230913064003.2925997-1-wtli@nuvoton.comSigned-off-by: Mark Brown <broonie@kernel.org>

There is no need to use temporary strings to construct the kcontrol names,
devm_kasprintf can be used to replace the snprintf + devm_kstrdup pairs.

This change will also fixes the following compiler warning/error (W=1):

sound/soc/codecs/hdac_hdmi.c: In function ‘hdac_hdmi_jack_port_init’:
sound/soc/codecs/hdac_hdmi.c:1793:63: error: ‘ Switch’ directive output may be truncated writing 7 bytes into a region of size between 1 and 32 [-Werror=format-truncation=]
 1793 |                         snprintf(kc_name, sizeof(kc_name), "%s Switch", xname);
      |                                                               ^~~~~~~
In function ‘create_fill_jack_kcontrols’,
    inlined from ‘hdac_hdmi_jack_port_init’ at sound/soc/codecs/hdac_hdmi.c:1871:8:
sound/soc/codecs/hdac_hdmi.c:1793:25: note: ‘snprintf’ output between 8 and 39 bytes into a destination of size 32
 1793 |                         snprintf(kc_name, sizeof(kc_name), "%s Switch", xname);
      |                         ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
cc1: all warnings being treated as errors

The warnings got brought to light by a recent patch upstream:
commit 6d4ab2e9 ("extrawarn: enable format and stringop overflow warnings in W=1")
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20230913091325.16877-1-peter.ujfalusi@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

If the driver properties do not define a cirrus,firmware-uid try to get the
PCI SSID as the UID.

On PCI-based systems the PCI SSID is used to uniquely identify the specific
sound hardware. This is the standard mechanism for x86 systems and is the
way to get a unique system identifier for systems that use the CS35L56 on
SoundWire.

For non-SoundWire systems there is no Windows equivalent of the ASoC driver
in I2C/SPI mode. These would be:

1. HDA systems, which are handled by the HDA subsystem.
2. Linux-specific systems.
3. Composite devices where the cs35l56 is not present in ACPI and is
   configured using software nodes.

Case 2 can use the firmware-uid property, though the PCI SSID is supported
as an alternative, as it is the standard PCI mechanism.

Case 3 is a SoundWire system where some other codec is the SoundWire bridge
device and CS35L56 is not listed in ACPI. As these are SoundWire systems
they will normally use the PCI SSID.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20230912163207.3498161-5-rf@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

If the PCI SSID has been set in the struct snd_soc_acpi_mach_params,
copy this to struct snd_soc_card so that it can be used by other
ASoC components.

This is important for components that must apply system-specific
configuration.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20230912163207.3498161-4-rf@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

Pass the PCI SSID of the audio interface through to the machine driver.
This allows the machine driver to use the SSID to uniquely identify the
specific hardware configuration and apply any platform-specific
configuration.

struct snd_sof_pdata is passed around inside the SOF code, but it then
passes configuration information to the machine driver through
struct snd_soc_acpi_mach and struct snd_soc_acpi_mach_params. So SSID
information has been added to both snd_sof_pdata and
snd_soc_acpi_mach_params.

PCI does not define 0x0000 as an invalid value so we can't use zero to
indicate that the struct member was not written. Instead a flag is
included to indicate that a value has been written to the
subsystem_vendor and subsystem_device members.

sof_pci_probe() creates the struct snd_sof_pdata. It is passed a struct
pci_dev so it can fill in the SSID value.

sof_machine_check() finds the appropriate struct snd_soc_acpi_mach. It
copies the SSID information across to the struct snd_soc_acpi_mach_params.
This done before calling any custom set_mach_params() so that it could be
used by the set_mach_params() callback to apply variant params.

The machine driver receives the struct snd_soc_acpi_mach as its
platform_data.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20230912163207.3498161-3-rf@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

Add members to struct snd_soc_card to store the PCI subsystem ID (SSID)
of the soundcard.

The PCI specification provides two registers to store a vendor-specific
SSID that can be read by drivers to uniquely identify a particular
"soundcard". This is defined in the PCI specification to distinguish
products that use the same silicon (and therefore have the same silicon
ID) so that product-specific differences can be applied.

PCI only defines 0xFFFF as an invalid value. 0x0000 is not defined as
invalid. So the usual pattern of zero-filling the struct and then
assuming a zero value unset will not work. A flag is included to
indicate when the SSID information has been filled in.

Unlike DMI information, which has a free-format entirely up to the vendor,
the PCI SSID has a strictly defined format and a registry of vendor IDs.

It is usual in Windows drivers that the SSID is used as the sole identifier
of the specific end-product and the Windows driver contains tables mapping
that to information about the hardware setup, rather than using ACPI
properties.

This SSID is important information for ASoC components that need to apply
hardware-specific configuration on PCI-based systems.

As the SSID is a generic part of the PCI specification and is treated as
identifying the "soundcard", it is reasonable to include this information
in struct snd_soc_card, instead of components inventing their own custom
ways to pass this information around.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20230912163207.3498161-2-rf@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

For the benefit of CI.

The loop is not using "node", of_node_put(node) is not needed.

Cc: Julia Lawall <julia.lawall@inria.fr>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/8734zlilmd.wl-kuninori.morimoto.gx@renesas.comSigned-off-by: Mark Brown <broonie@kernel.org>

Merge series from Konrad Dybcio <konrad.dybcio@linaro.org>:

Like most Qualcomm SoCs, SM6115 has a TX Macro.

Only some minor changes were required.

Like the Lenovo 82TL and 82V2, the Lenovo 82QF (Yoga 7 14ARB7) and 82UG
(Legion S7 16ARHA7) both need a quirk entry for the internal microphone to
function. Commit c008323f ("ASoC: amd: yc: Fix a non-functional mic on
Lenovo 82SJ") restricted the quirk that previously matched "82" to "82V2",
breaking microphone functionality on these devices. Fix this by adding
specific quirks for these models, as was done for the Lenovo 82TL.

Fixes: c008323f ("ASoC: amd: yc: Fix a non-functional mic on Lenovo 82SJ")
Closes: https://github.com/tomsom/yoga-linux/issues/51
Link: https://bugzilla.kernel.org/show_bug.cgi?id=208555#c780
Cc: stable@vger.kernel.org
Signed-off-by: August Wikerfors <git@augustwikerfors.se>
Link: https://lore.kernel.org/r/20230911213409.6106-1-git@augustwikerfors.seSigned-off-by: Mark Brown <broonie@kernel.org>

For some reason the JH7110 PWM DAC driver made it through build testing
in spite of not being updated for the move of probe() to the ops struct.
Make the required update.
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series from Biju Das <biju.das.jz@bp.renesas.com>:

This patch series aims to add match data improvements for wm8580 driver.

This patch series is only compile tested.

Merge series from Linus Walleij <linus.walleij@linaro.org>:

The Maxim devices are pretty straight-forward to convert
over to use GPIO descriptors, so let's do it.

Merge series from Biju Das <biju.das.jz@bp.renesas.com>:

This patch series aims to add match data improvements for tlv320aic32x4
driver.

This patch series is only compile tested.

Merge series from Linus Walleij <linus.walleij@linaro.org>:

The Maxim devices are pretty straight-forward to convert
over to use GPIO descriptors, so let's do it.

Merge series from Trevor Wu <trevor.wu@mediatek.com>:

This series introduces dynamic pinctrl and adds support for the SOF on
the mt8188-mt6359 machine driver.

Merge series from Biju Das <biju.das.jz@bp.renesas.com>:

This patch series aims to add match data improvements for ak4642 driver.

This patch series is only compile tested.

Merge series from Marian Postevca <posteuca@mutex.one>:

This series adds support for a line of HUAWEI laptops with
AMD CPUs that connect using the ACP3x module to a ES8336 CODEC.

The CODEC driver must be extended to support the S32 LE format
and the MCLK div by 2 option. MCLK div by 2 is needed for one specific
SKU, which uses a 48Mhz MCLK, which seems to be too high of a frequency
for the CODEC and must be divided by 2.

The acp legacy driver must also be extended by using callbacks so that
the more complicated handling of this specific CODEC can be moved
outside the more generic ACP code.

Merge series from Cristian Ciocaltea <cristian.ciocaltea@collabora.com>:

This patch series contains several fixes and improvements to drivers
based on the CS35l41 audio codec.

It has been verified on Valve's Steam Deck, except the HDA related patches.

Merge series from Hal Feng <hal.feng@starfivetech.com>:

This patchset adds PWM-DAC audio support for the StarFive JH7110 SoC.
The PWM-DAC module does not require a hardware codec, but a dummy codec is
needed for the driver. The dummy spdif codec driver, which is already
upstream, is compatible with the one which JH7110 PWM-DAC needed. So we
use it as the dummy codec driver for the JH7110 PWM-DAC module.

Merge series from Sameer Pujar <spujar@nvidia.com>:

This small series fixes redundant PLLA updates that happen for
each DAI link in the audio path. This helps to resolve DMIC clock
issue seen on Jetson TX2 platform.

The rpmsg pcm device is a device which should support
double buffering.

Found this issue with pipewire. When there is no
SNDRV_PCM_INFO_BATCH flag in driver, the pipewire will
set headroom to be zero, and because rpmsg pcm device
don't support residue report, when the latency setting
is small, the "delay" always larger than "target" in
alsa-pcm.c, that reading next period data is not
scheduled on time.

With SNDRV_PCM_INFO_BATCH flag in driver, the pipewire
will select a smaller period size for device, then
the task of reading next period data will be scheduled
on time.
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1694414287-13291-1-git-send-email-shengjiu.wang@nxp.comSigned-off-by: Mark Brown <broonie@kernel.org>

Replace the remaining dev_err() calls in probe() with dev_err_probe(),
to improve consistency.
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-12-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

According to the documentation, drivers are responsible for undoing at
removal time all runtime PM changes done during probing.

Hence, add the missing calls to pm_runtime_dont_use_autosuspend(), which
are necessary for undoing pm_runtime_use_autosuspend().

Fixes: 1873ebd3 ("ALSA: hda: cs35l41: Support Hibernation during Suspend")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-11-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

If component_add() fails, probe() returns without calling
pm_runtime_put(), which leaves the runtime PM usage counter incremented.

Fix the issue by jumping to err_pm label and drop the now unnecessary
pm_runtime_disable() call.

Fixes: 7b2f3eb4 ("ALSA: hda: cs35l41: Add support for CS35L41 in HDA systems")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-10-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

Make use of the recently introduced EXPORT_GPL_DEV_PM_OPS() macro, to
conditionally export the runtime/system PM functions.

Replace the old SET_{RUNTIME,SYSTEM_SLEEP,NOIRQ_SYSTEM_SLEEP}_PM_OPS()
helpers with their modern alternatives and get rid of the now
unnecessary '__maybe_unused' annotations on all PM functions.

Additionally, use the pm_ptr() macro to fix the following errors when
building with CONFIG_PM disabled:

ERROR: modpost: "cs35l41_pm_ops" [sound/soc/codecs/snd-soc-cs35l41-spi.ko] undefined!
ERROR: modpost: "cs35l41_pm_ops" [sound/soc/codecs/snd-soc-cs35l41-i2c.ko] undefined!
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-9-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

Use dev_err_probe() helper where possible, to simplify error handling
during probe.
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-8-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

According to the documentation, drivers are responsible for undoing at
removal time all runtime PM changes done during probing.

Hence, add the missing calls to pm_runtime_dont_use_autosuspend(), which
are necessary for undoing pm_runtime_use_autosuspend().

Note this would have been handled implicitly by
devm_pm_runtime_enable(), but there is a need to continue using
pm_runtime_enable()/pm_runtime_disable() in order to ensure the runtime
PM is disabled as soon as the remove() callback is entered.

Fixes: f517ba49 ("ASoC: cs35l41: Add support for hibernate memory retention mode")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-7-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

The interrupt handler invokes pm_runtime_get_sync() without checking the
returned error code.

Add a proper verification and switch to pm_runtime_resume_and_get(), to
avoid the need to call pm_runtime_put_noidle() for decrementing the PM
usage counter before returning from the error condition.

Fixes: f517ba49 ("ASoC: cs35l41: Add support for hibernate memory retention mode")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-6-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

Enabling the active/passive shared boosts requires setting SYNC_EN, but
*not* before receiving the PLL Lock signal.

Due to improper error handling, it was not obvious that waiting for the
completion operation times out and, consequently, the shared boost is
never activated.

Further investigations revealed the signal is triggered while
snd_pcm_start() is executed, right after receiving the
SNDRV_PCM_TRIGGER_START command, which happens long after the
SND_SOC_DAPM_PRE_PMU event handler is invoked as part of
snd_pcm_prepare().  That is where cs35l41_global_enable() is called
from.

Increasing the wait duration doesn't help, as it only causes an
unnecessary delay in the invocation of snd_pcm_start().  Moving the wait
and the subsequent regmap operations to the SNDRV_PCM_TRIGGER_START
callback is not a solution either, since they would be executed in an
IRQ-off atomic context.

Solve the issue by setting the SYNC_EN bit in PWR_CTRL3 register right
after receiving the PLL Lock interrupt.

Additionally, drop the unnecessary writes to PWR_CTRL1 register, part of
the original mdsync_up_seq, which would have toggled GLOBAL_EN with
unwanted consequences on PLL locking behavior.

Fixes: f5030564 ("ALSA: cs35l41: Add shared boost feature")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Reviewed-by: David Rhodes <david.rhodes@cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-5-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

Technically, an interrupt handler can be called before probe() finishes
its execution, hence ensure the pll_lock completion object is always
initialized before being accessed in cs35l41_irq().

Fixes: f5030564 ("ALSA: cs35l41: Add shared boost feature")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-4-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

The return code of regmap_multi_reg_write() call related to "MDSYNC up"
sequence is shadowed by the subsequent regmap_read_poll_timeout()
invocation, which will hit a timeout in case the write operation above
fails.

Make sure cs35l41_global_enable() returns the correct error code instead
of -ETIMEDOUT.

Additionally, to be able to distinguish between the timeouts of
wait_for_completion_timeout() and regmap_read_poll_timeout(), print an
error message for the former and return immediately.  This also avoids
having to wait unnecessarily for the second time.

Fixes: f8264c75 ("ALSA: cs35l41: Poll for Power Up/Down rather than waiting a fixed delay")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-3-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

The return code of regmap_multi_reg_write() call related to "MDSYNC
down" sequence is shadowed by the subsequent
wait_for_completion_timeout() invocation, which is expected to time
timeout in case the write operation failed.

Let cs35l41_global_enable() return the correct error code instead of
-ETIMEDOUT.

Fixes: f5030564 ("ALSA: cs35l41: Add shared boost feature")
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Reviewed-by: Takashi Iwai <tiwai@suse.de>
Link: https://lore.kernel.org/r/20230907171010.1447274-2-cristian.ciocaltea@collabora.comSigned-off-by: Mark Brown <broonie@kernel.org>

This driver is already using solely GPIO descriptors and
do not need to include the legacy headers <linux/gpio.h>
or <linux/of_gpio.h>. Drop them.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20230911-descriptors-asoc-max-v2-7-b9d793fb768e@linaro.orgSigned-off-by: Mark Brown <broonie@kernel.org>

This driver is already using solely GPIO descriptors and
do not need to include the legacy headers <linux/gpio.h>
or <linux/of_gpio.h>. Drop them.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20230911-descriptors-asoc-max-v2-6-b9d793fb768e@linaro.orgSigned-off-by: Mark Brown <broonie@kernel.org>