Sometime a component or topology may configure a DAI widget with no
private data leading to a dev_dbg() dereferencne of this data.

Fix this to check for non NULL private data and let users know if widget
is missing DAI.
Signed-off-by: NLiam Girdwood <liam.r.girdwood@linux.intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

dapm_kcontrol_data is freed as part of dapm_kcontrol_free(), leaving the
paths pointer dangling in the list.

This leads to system crash when we try to unload and reload sound card.
I hit this bug during ADSP crash/reboot test case on Dragon board DB410c.

Without this patch, on SLAB Poisoning enabled build, kernel crashes with
"BUG kmalloc-128 (Tainted: G        W        ): Poison overwritten"
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>
Cc: stable@vger.kernel.org

skl-tplg-interface.h describes firmware format details for Skylake
topology files. It is part of the ABI and should reside in the uapi
directory.

While moving the file, also replace the license boilerplate with
the SPDX License Identifier.

No functional change.
Signed-off-by: NGuenter Roeck <groeck@chromium.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

Topology manifest v4 is still part of the ABI. Move its data structures
into the uapi header file.

No functional change.
Signed-off-by: NGuenter Roeck <groeck@chromium.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

Commit dc31e741 ("ASoC: topology: ABI - Add the types for BE
DAI") introduced sound topology files version 5. Initially, this
change made the topology code incompatible with v4 topology files.
Backwards compatibility with v4 configuration files was
subsequently added with commit 288b8da7 ("ASoC: topology:
Support topology file of ABI v4").

Unfortunately, backwards compatibility was never fully implemented.

First, the manifest size in (Skylake) v4 configuration files is set
to 0, which causes manifest_new_ver() to bail out with error messages
similar to the following.

snd_soc_skl 0000:00:1f.3: ASoC: invalid manifest size
snd_soc_skl 0000:00:1f.3: tplg component load failed-22
snd_soc_skl 0000:00:1f.3: Failed to init topology!
snd_soc_skl 0000:00:1f.3: ASoC: failed to probe component -22
skl_n88l25_m98357a skl_n88l25_m98357a: ASoC: failed to instantiate card -22
skl_n88l25_m98357a: probe of skl_n88l25_m98357a failed with error -22

After this problem is fixed, the following error message is seen instead.

snd_soc_skl 0000:00:1f.3: ASoC: old version of manifest
snd_soc_skl 0000:00:1f.3: Invalid descriptor token 1093938482
snd_soc_skl 0000:00:1f.3: ASoC: failed to load widget media0_in cpr 0
snd_soc_skl 0000:00:1f.3: tPlg component load failed-22

This message is seen because backwards compatibility for loading widgets
was never implemented.

The lack of audio support when running the upstream kernel on recent
Chromebooks has been reported in various forums, and can be traced back
to this problem. Attempts to fix the problem, usually by providing v5
configuration files, were only partially successful.

Let's implement backward compatibility properly to solve the problem
for good.
Signed-off-by: NGuenter Roeck <groeck@chromium.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

Currently there is no support for Tempo Semiconductor's TSCS454 CODEC.

Add support for it.
Signed-off-by: NSteven Eckhoff <steven.eckhoff.opensource@gmail.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

On APL, commit fd0f2375
("ASoC: Intel: bxt: Move codec sysclk config to codec_init function")
fixed an issue related to jack detection.

The MCLK for DA7219 does not change in this platform, but is
currently being configured everytime as part of the platform_clock
event handler for DAPM. The upshot of this is that we have
unnecessary calls to this function, and it also means that if
a stream hasn't yet been started, DA7219 driver does not have the
correct MCLK rates programmed and so the HP detection feature does
not operate as expected.

The same fix is needed on KBL.

This patch rectifies this issue by moving the sysclk call to
codec_init function so it's only called once at initialisation.
Signed-off-by: NMac Chiang <mac.chiang@intel.com>
Acked-by: NPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

The simple-card driver currently accepts a clock node in the cpu dai
sub-node and only uses it as an alternative to the
'system-clock-frequency' property to get the current frequency.

This patch adds another use of the passed clock node. If mclk-fs is
specified, the clocks in cpu and codec dai sub-nodes will be set to
the calculated rate (stream rate * mclk_fs) in hw_params.

This allows platforms to pass tuneable clocks as phandle that will
automatically be set to the right rates.
Signed-off-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

match_string() returns the index of an array for a matching string,
which can be used instead of open coded variant.
Reviewed-by: NAndy Shevchenko <andy.shevchenko@gmail.com>
Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: alsa-devel@alsa-project.org
Signed-off-by: NYisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

match_string() returns the index of an array for a matching string,
which can be used instead of open coded variant.

Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: alsa-devel@alsa-project.org
Signed-off-by: NYisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

match_string() returns the index of an array for a matching string,
which can be used instead of open coded variant.

Cc: Liam Girdwood <lgirdwood@gmail.com>
Cc: Mark Brown <broonie@kernel.org>
Cc: Jaroslav Kysela <perex@perex.cz>
Cc: Takashi Iwai <tiwai@suse.com>
Cc: alsa-devel@alsa-project.org
Signed-off-by: NYisheng Xie <xieyisheng1@huawei.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

Many X86 devices using a BYT SoC + RT5651 codec are cheap devices with
generic DMI strings, causing snd_soc_set_dmi_name() to fail to set a
long_name, making it impossible for userspace to have a correct UCM
profile which knowns which input is connected to the internal mic,
which input is connected to the hsmic (for correct jack-based switching)
and which inputs are unused.

Our quirks already specify which inputs the internal and headset mic
are connected to.

This commit sets a long_name based on the quirks so that userspace can
have UCM profiles doing the right thing based on the long_name.

Note that if we ever encounter the need for a special UCM profile for
some device we can add a quirk to set a specific long_name for the
device,
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

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

add path for hosltess lpbk from ADDA Capture to ADDA Playback
add path for hostless phone call between ADDA DAI and PCM DAI
Signed-off-by: NKaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

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

The new mt6797-afe driver uses some functions in a common file, which
works for a built-in driver but fails for a loadable module:

ERROR: "mtk_afe_pcm_free" [sound/soc/mediatek/mt6797/snd-soc-mt6797-afe.ko] undefined!
ERROR: "mtk_afe_add_sub_dai_control" [sound/soc/mediatek/mt6797/snd-soc-mt6797-afe.ko] undefined!
ERROR: "mtk_afe_pcm_new" [sound/soc/mediatek/mt6797/snd-soc-mt6797-afe.ko] undefined!
ERROR: "mtk_afe_combine_sub_dai" [sound/soc/mediatek/mt6797/snd-soc-mt6797-afe.ko] undefined!
ERROR: "mtk_afe_pcm_ops" [sound/soc/mediatek/mt6797/snd-soc-mt6797-afe.ko] undefined!

This exports the five symbols above for modules.

Fixes: b3c702f5 ("ASoC: mt6797: combine DAI to register component")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NMark Brown <broonie@kernel.org>

When CONFIG_GPIOLIB is disabled, this codec fails to build
because gpio/consumer.h is not included implicitly.

sound/soc/codecs/pcm1789.c: In function 'pcm1789_common_init':
sound/soc/codecs/pcm1789.c:247:19: error: implicit declaration of function 'devm_gpiod_get_optional'; did you mean 'devm_gpio_request_one'? [-Werror=implicit-function-declaration]
  pcm1789->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
                   ^~~~~~~~~~~~~~~~~~~~~~~

Fixes: 4ae340d1 ("ASoC: codecs: Add support for PCM1789")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NMark Brown <broonie@kernel.org>

When the call to hw_params for a component fails, the error code is held
by the variable '__ret' but the error message displays the value held by
the variable 'ret'. Fix the return code shown when hw_params fails for
a component.

Fixes: b8135864 ("ASoC: snd_soc_component_driver has snd_pcm_ops")
Signed-off-by: NJon Hunter <jonathanh@nvidia.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

This patch adds support for the Atmel I2S controller embedded into
sama5d2x SoCs.
Signed-off-by: NCyrille Pitchen <cyrille.pitchen@atmel.com>
Signed-off-by: NCodrin Ciubotariu <codrin.ciubotariu@microchip.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

This patch adds TX and RX TDM mixers for 40 TDM ports.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

All the mixer controls are pretty much same from all the afe ports.

Make these as proper macros for 2 reasons.

1> To avoid any typos in adding new mixer controls for each port.
2> Easy to edit from single place, easy to add new ports

This also prepares the routing driver to accomdate 40 tdm dais.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

This patch adds support to 40 TDM ports supported in AFE.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

Use common q6afe_dai_prepare() for MI2S dais, this will remove
some code duplication. Also make the if statement to switch to
make the code look neater.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

This patch adds support to tdm ports in AFE.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

We should set BE symmetric constraint on FE substream.

in case one BE is used by two FE1/FE2,
the first BE runtime will use FE1's substream->runtime.
hence the FE1's will be constrained by BE symmetry property.

Though, second FE2 call dpcm_apply_symmetry,
the be_substream->runtime == FE1's substream->runtime.
The FE2's substream->runtime will not be constrained
by BE's symmetry property.
Signed-off-by: NKaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

In case, one BE is used by two FE1/FE2
FE1--->BE-->
       |
FE2----]
when FE1/FE2 call dpcm_be_dai_hw_free() together
the BE users will be 2 (> 1), hence cannot be hw_free
the be state will leave at, ex. SND_SOC_DPCM_STATE_STOP

later FE1/FE2 call dpcm_be_dai_shutdown(),
will be skip due to wrong state.
leaving the BE not being hw_free and shutdown.

The BE dai will be hw_free later when calling
dpcm_be_dai_shutdown() if still in invalid state.
Signed-off-by: NKaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

channel 1: SYSMEM<->ACP
channel 2: ACP<->I2S
Instead of waiting on period interrupt of ch 2 and then starting
dma on ch1, we make ch1 dma as circular.
This removes dependency of period granularity on hw pointer.
Signed-off-by: NAkshu Agrawal <akshu.agrawal@amd.com>
Reviewed-by: NDaniel Kurtz <djkurtz@chromium.org>
Tested-by: NDaniel Kurtz <djkurtz@chromium.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

Add a quirk for the ARCHOS 80 Cesium 8" windows tablet, this device mostly
works with the default settings, except that it has only one speaker.
So add a quirk with the default settings + the mono-speaker flag.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Reviewed-by: NPierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

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

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

In MediaTek SoC chip we have multiple DAI,
such as I2S, ADDA, PCM, etc.

Organize each DAI in to one sub dai,
with its dai driver, controls, widgets, routes.

add mtk_afe_combine_sub_dai() to combine
dai driver from each DAI.

add mtk_afe_add_sub_dai_control() to register
the control, widget, routes to component.
Signed-off-by: NKaiChieh Chuang <kaichieh.chuang@mediatek.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

The newly introduced driver causes a harmless Kconfig warning when
compile-testing random configurations:

WARNING: unmet direct dependencies detected for SND_SDMA_SOC
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && DMA_OMAP [=n]
  Selected by [y]:
  - SND_OMAP_SOC [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && (ARCH_OMAP [=y] && DMA_OMAP [=n] || ARM [=y] && COMPILE_TEST [=y])

By simply allow build testing without DMA_OMAP, we can shut up that warning.

Fixes: dde637f2 ("ASoC: omap: Introduce the generic_dmaengine_pcm based sdma-pcm")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NMark Brown <broonie@kernel.org>

Before commit 3b5b899c ("ALSA: hda: Make use of core codec functions
to sync power state"), hda_set_power_state() returned the response to
the Get Power State verb, a 32-bit unsigned integer whose expected value
is 0x233 after transitioning a codec to D3, and 0x0 after transitioning
it to D0.

The response value is significant because hda_codec_runtime_suspend()
does not clear the codec's bit in the codec_powered bitmask unless the
AC_PWRST_CLK_STOP_OK bit (0x200) is set in the response value.  That in
turn prevents the HDA controller from runtime suspending because
azx_runtime_idle() checks that the codec_powered bitmask is zero.

Since commit 3b5b899c, hda_set_power_state() only returns 0x0 or
0x1, thereby breaking runtime PM for any HDA controller.  That's because
an inline function introduced by the commit returns a bool instead of a
32-bit unsigned int.  The change was likely erroneous and resulted from
copying and pasting snd_hda_check_power_state(), which is immediately
preceding the newly introduced inline function.  Fix it.

Link: https://bugs.freedesktop.org/show_bug.cgi?id=106597
Fixes: 3b5b899c ("ALSA: hda: Make use of core codec functions to sync power state")
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Abhijeet Kumar <abhijeet.kumar@intel.com>
Reported-and-tested-by: NGunnar Krüger <taijian@posteo.de>
Signed-off-by: NLukas Wunner <lukas@wunner.de>
Acked-by: NAlex Deucher <alexander.deucher@amd.com>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

The nau8824 codec can detect whether a headset or plain headphones is
inserted (as well as button presses on the headset) as such the jack_type
passed to snd_soc_card_jack_new() should include SND_JACK_MICROPHONE.
Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

some monitors care about the parity bit in the sub-frame of I2S,
but the cdn-dp always set this bit to "1", so these monitors
do not have sound output if use i2s, use spdif can fix this issue.
Signed-off-by: NChris Zhong <zyw@rock-chips.com>
Signed-off-by: NLin Huang <hl@rock-chips.com>
Signed-off-by: NMark Brown <broonie@kernel.org>

FOr platforms that use the simple-card driver, the codec cannot be selected
through 'select' magic in Kconfig. So turn this into a real config option.
Signed-off-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

This is needed when the codec is instanciated from from a device tree.
Signed-off-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

There's no need to read the register again prior to writing it, we did
that in the beginning of the function.
Signed-off-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

The pxa-ssp driver currently assumes that .set_fmt() is called before
.set_clkdiv(), .set_pll() etc.

Commit a8bd0ee5 ("ASoC: raumfeld: Use static DAI format setup") broke
support for Raumfeld hardware (and possible other PXA based ones) because
it effectively changed the order of these calls. Also, as the call to
.set_fmt() is now done at probe time, the port clock is not yet enabled.

To fix this, strip all hardware register access code from the .set_fmt()
callback and memorize the desired value, so we can use it from the
.hw_params() callback. Also make the .set_fmt() callback less destructive
by reading all registers that it writes to in the beginning and only
masking out the bits that it possibly fiddles with.
Signed-off-by: NDaniel Mack <daniel@zonque.org>
Signed-off-by: NMark Brown <broonie@kernel.org>

This patch adds support to DB820c machine driver.
Signed-off-by: NSrinivas Kandagatla <srinivas.kandagatla@linaro.org>
Reviewed-by: NBanajit Goswami <bgoswami@codeaurora.org>
Signed-off-by: NMark Brown <broonie@kernel.org>