Merge series "ASoC: rsnd: tidyup adg and header" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark

I noticed that adg can be more clean code.
And rsnd.h header comment was not so good
because patch has been randomly added.

This patch tidyup these.

Kuninori Morimoto (5):
  ASoC: rsnd: adg: supply __printf(x, y) formatting for dbg_msg()
  ASoC: rsnd: adg: tidyup rsnd_adg_get_clkin/out() parameter
  ASoC: rsnd: adg: use more simple method for null_clk
  ASoC: rsnd: adg: check return value for rsnd_adg_get_clkin/out()
  ASoC: rsnd: tidyup __rsnd_mod_xxx macro comments

 sound/soc/sh/rcar/adg.c  | 139 +++++++++++++++++++++++++--------------
 sound/soc/sh/rcar/rsnd.h |  21 ++----
 2 files changed, 97 insertions(+), 63 deletions(-)

--
2.25.1

ASoC: img: Fix PM reference leak in img_i2s_in_probe()

pm_runtime_get_sync will increment pm usage counter even it failed.
Forgetting to putting operation will result in reference leak here.
Fix it by replacing it with pm_runtime_resume_and_get to keep usage
counter balanced.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yufen Yu <yuyufen@huawei.com>
Link: https://lore.kernel.org/r/20210524093521.612176-1-yuyufen@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: imx-es8328: use devm_snd_soc_register_card()

Using devm_snd_soc_register_card() can make the code
shorter and cleaner.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210602133359.310647-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: check for zero node count

Most callers of_get_child_count() check that "nr" is non-zero so it
causes a static checker warning when we don't do that here.  This
does not cause a problem or a crash, but having zero SSUIes does not
make sense either so let's add a check.

Addresses-Coverity: ("Unchecked return value")
Fixes: c413983eb66a ("ASoC: rsnd: adjust disabled module")
Acked-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Link: https://lore.kernel.org/r/20210603110315.81146-1-colin.king@canonical.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: Convert imx-audmux binding to json schema

Convert the imx-audmux binding to DT schema format using json-schema

Signed-off-by: Oleksij Rempel <o.rempel@pengutronix.de>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20210531064752.8809-1-o.rempel@pengutronix.de
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-audio-rpmsg: use module_rpmsg_driver to simplify the code

module_rpmsg_driver() makes the code simpler by eliminating
boilerplate code.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210602082610.3828408-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: atmel: sam9x5_wm8731: use devm_snd_soc_register_card()

Using devm_snd_soc_register_card() can make the code
shorter and cleaner.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210602141619.323286-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup __rsnd_mod_xxx macro comments

status and __rsnd_mod_xxx were updated, but some related comments were
not. And it has verbose comments. This patch cleanup/tidyup these.

1) adds missing "D" to status sample
2) remove verbose list for "H"
3) add "needs protect" to __rsnd_mod_call_xxx

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87o8cpi1zs.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: adg: check return value for rsnd_adg_get_clkin/out()

Current rsnd_adg_get_clkin/out() are void function,
thus adg->clk/clkout[i] might be NULL.

But, for_each_rsnd_clk/clkout() macros are assuming
all clks are non NULL.

Because of this mismatch, code can be complex and/or buggy.
These functions return error by this patch,
and make sure all clks are non NULL.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87pmx5i20m.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: adg: use more simple method for null_clk

commit 965386c97616c ("ASoC: rsnd: call unregister for null_hw when
removed") tried unregister null_clk, but it has some issues.

1st issue is kernel will indicate below message when unregistering,
because of its timing. unregistering should be happen after clk_disable().

clk_unregister: unregistering prepared clock: rsnd_adg_null

2nd issue is, it is using priv->null_clk, but it should be adg->null_clk.

3rd issue is it is using very complex clk registering method.
more simple clk_register/unregister_fixed_rate() should be OK.

This patch fixes these.

Fixes: 965386c97616c ("ASoC: rsnd: call unregister for null_hw when removed")
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87r1hli215.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: adg: tidyup rsnd_adg_get_clkin/out() parameter

set priv->adg before rsnd_adg_get_clkin/out() to be more simple code.
Nothing is changed, but is preparation for
next "ASoC: rsnd: adg: use more simple method for null_clk" patch

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87sg21i21j.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: adg: supply __printf(x, y) formatting for dbg_msg()

Fixes the following W=1 kernel build warning(s):

sound/soc/sh/rcar/adg.c: In function 'dbg_msg':
sound/soc/sh/rcar/adg.c:594:2: warning: function 'dbg_msg' might \
 be a candidate for 'gnu_printf' format attribute\
 [-Wsuggest-attribute=format]

Fixes: 1f9c82b5ab83 ("ASoC: rsnd: add debugfs support")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87tumhi21r.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: Skylake: use DEVICE_ATTR_RO macro

Use DEVICE_ATTR_RO() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20210524115506.35724-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: tfa989x: Add support for optional vddd-supply

Allow specifying Vddd regulator/supply to be enabled on I2C probing.

Signed-off-by: Vincent Knecht <vincent.knecht@mailoo.org>
Reviewed-by: Stephan Gerhold <stephan@gerhold.net>
Link: https://lore.kernel.org/r/20210528105101.508254-4-vincent.knecht@mailoo.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: nxp, tfa989x: Add vddd-supply property

Add optional vddd-supply property to allow regulator control.

Signed-off-by: Vincent Knecht <vincent.knecht@mailoo.org>
Link: https://lore.kernel.org/r/20210528105101.508254-3-vincent.knecht@mailoo.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: tfa989x: Add support for tfa9897

Add specific init function to poke needed registers & values for this IC

Signed-off-by: Vincent Knecht <vincent.knecht@mailoo.org>
Reviewed-by: Stephan Gerhold <stephan@gerhold.net>
Link: https://lore.kernel.org/r/20210528105101.508254-2-vincent.knecht@mailoo.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: nxp,tfa989x: Add tfa9897 support

Document TFA9897 bindings.

Signed-off-by: Vincent Knecht <vincent.knecht@mailoo.org>
Reviewed-by: Stephan Gerhold <stephan@gerhold.net>
Link: https://lore.kernel.org/r/20210528105101.508254-1-vincent.knecht@mailoo.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sigmadsp: Remove the repeated declaration

Function 'sigmadsp_reset' is declared twice, so remove the repeated
declaration.

Cc: Mark Brown <broonie@kernel.org>
Signed-off-by: Shaokun Zhang <zhangshaokun@hisilicon.com>
Link: https://lore.kernel.org/r/1622279690-3740-1-git-send-email-zhangshaokun@hisilicon.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: Constify snd_compress_ops" from Rikard Falkeborn <rikard.falkeborn@gmail.com>:

The only use of the static and global snd_compress_ops structs is to
assign their address to the compress_ops field in the
snd_soc_component_driver struct which is a pointer to const. Make them
const to allow the compiler to put them in read-only memory.

Rikard Falkeborn (5):
  ASoC: cs47125: Constify static struct snd_compress_ops
  ASoC: wm5102: Constify static struct snd_compress_ops
  ASoC: wm5110: Constify static struct snd_compress_ops
  ASoC: qcom: q6asm-dai: Constify static struct snd_compress_ops
  ASoC: SOF: Intel: Constify sof_probe_compressed_ops

 sound/soc/codecs/cs47l24.c       | 2 +-
 sound/soc/codecs/wm5102.c        | 2 +-
 sound/soc/codecs/wm5110.c        | 2 +-
 sound/soc/qcom/qdsp6/q6asm-dai.c | 2 +-
 sound/soc/sof/compress.c         | 2 +-
 sound/soc/sof/compress.h         | 2 +-
 6 files changed, 6 insertions(+), 6 deletions(-)

--
2.31.1

Merge series "ASoC: rsnd: adjust disabled module for R-Car D3" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark, Geert

Renesas Sound driver is assuming that all SSI/SRC
are connected.
But some SSI/SRC are not connected on Some Renesas SoC.
ex)
H2 E2

SRC0       <=
SRC1 SRC1
SRC2 SRC2
... ...

We accepted it by using "status = disabled" on DT before.

ex)
rcar_sound,src {
src-0 {
=> status = "disabled";
};
src1: src-1 {
...
};
...

But R-Car D3 have many disabled modules (It has SSI3/SSI4, SRC5/SRC6),
and Renesas SoC maintainer don't want above style on DT.

ex)
rcar_sound,src {
=> src0: src-0 { status = "disabled"; };
=> src1: src-1 { status = "disabled"; };
=> src2: src-2 { status = "disabled"; };
=> src3: src-3 { status = "disabled"; };
=> src4: src-4 { status = "disabled"; };
src5: src-5 {
...
};
...

This patch-set adjust to this situation, and enables to intuitive DT settings.

rcar_sound,src {
src5: src-5 {
...
};
src6: src-6 {
...
};
};

Kuninori Morimoto (4):
  ASoC: rsnd: tidyup rsnd_parse_connect_common()
  ASoC: rsnd: tidyup rsnd_dma_request_channel()
  ASoC: rsnd: tidyup rsnd_parse_connect_xxx()
  ASoC: rsnd: adjust disabled module

 sound/soc/sh/rcar/core.c | 58 ++++++++++++++++++++++++++++++++++++++--
 sound/soc/sh/rcar/dma.c  |  8 +++---
 sound/soc/sh/rcar/dvc.c  |  2 +-
 sound/soc/sh/rcar/rsnd.h | 16 ++++++-----
 sound/soc/sh/rcar/src.c  |  6 +++--
 sound/soc/sh/rcar/ssi.c  | 12 ++++++---
 sound/soc/sh/rcar/ssiu.c | 10 ++++---
 7 files changed, 91 insertions(+), 21 deletions(-)

--
2.25.1

ASoC: SOF: Intel: pci-tgl: add ADL-M support

Add PCI DID for Intel AlderLake-M.

Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210528184153.18251-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: don't print ROM status if cl_dsp_init() fails

cl_dsp_init() dumps the ROM status if it fails after max
attempts before powering off the DSP. Remove the duplicate
log to print the ROM status and error in
hda_dsp_cl_boot_firmware(). These values are invalid anyway
as the DSP is already powered off.

Co-developed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210528160551.10145-2-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: clean up hda_dsp_dump()

Clean up the hda_dsp_dump() function to avoid duplicating
the ROM status and error.

Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210528160551.10145-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: ti: omap-mcbsp: use DEVICE_ATTR_RW macro

Use DEVICE_ATTR_RW() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@gmail.com>
Link: https://lore.kernel.org/r/20210528063033.19904-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: Constify sof_probe_compressed_ops

The only usage of sof_probe_compressed_ops is to assign its address to
the compress_ops field in the snd_soc_component_driver struct, which is
a pointer to const. The assignment is done in sound/soc/sof/pcm.c. Make
it const to allow the compiler to put it in read-only memory.

Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210526231013.46530-6-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: q6asm-dai: Constify static struct snd_compress_ops

The snd_compress_ops structs are only stored in the compress_ops field
of a snd_soc_component_driver struct, so make it const to allow the
compiler to put it in read-only memory.

Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210526231013.46530-5-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm5110: Constify static struct snd_compress_ops

The snd_compress_ops structs are only stored in the compress_ops field
of a snd_soc_component_driver struct, so make it const to allow the
compiler to put it in read-only memory.

Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210526231013.46530-4-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm5102: Constify static struct snd_compress_ops

The snd_compress_ops structs are only stored in the compress_ops field
of a snd_soc_component_driver struct, so make it const to allow the
compiler to put it in read-only memory.

Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210526231013.46530-3-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs47125: Constify static struct snd_compress_ops

The snd_compress_ops structs are only stored in the compress_ops field
of a snd_soc_component_driver struct, so make it const to allow the
compiler to put it in read-only memory.

Signed-off-by: Rikard Falkeborn <rikard.falkeborn@gmail.com>
Link: https://lore.kernel.org/r/20210526231013.46530-2-rikard.falkeborn@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: adjust disabled module

In general Renesas SoC's SSI/SRC are all enabled, but some SoC is not.

H2 E2

SRC0       <=
SRC1 SRC1
SRC2 SRC2
... ...

Renesas Sound driver is assuming that *all* modules are
enabled, and thus it is using *data array* to access each modules.
Because of it, we have been using "status = disabled" at DT,
and using *full size* array but avoiding disabled module.

ex)
rcar_sound,src {
src-0 {
=> status = "disabled";
};
src1: src-1 {
...
};
...

But R-Car D3 have many disabled modules (It has SSI3/SSI4, SRC5/SRC6),
and Renesas SoC maintainer don't want above style on DT.

ex)
rcar_sound,src {
=> src0: src-0 { status = "disabled"; };
=> src1: src-1 { status = "disabled"; };
=> src2: src-2 { status = "disabled"; };
=> src3: src-3 { status = "disabled"; };
=> src4: src-4 { status = "disabled"; };
src5: src-5 {
...
};
src6: src-6 {
...
};
};

rcar_sound,ssi {
=> ssi0: ssi-0 { status = "disabled"; };
=> ssi1: ssi-1 { status = "disabled"; };
=> ssi2: ssi-2 { status = "disabled"; };
ssi3: ssi-3 {
...
};
ssi4: ssi-4 {
...
};
};

To adjust it, it needs to care about related for_each_child_of_node()
loop on rsnd driver, and it is used from...

> grep -l for_each_child_of_node sound/soc/sh/rcar/*
sound/soc/sh/rcar/core.c
sound/soc/sh/rcar/ctu.c
sound/soc/sh/rcar/dma.c
sound/soc/sh/rcar/dvc.c
sound/soc/sh/rcar/mix.c
sound/soc/sh/rcar/src.c
sound/soc/sh/rcar/ssi.c
sound/soc/sh/rcar/ssiu.c

This patch adjust to this situation.
By this patch, we can avoid disabled modules on DT

rcar_sound,src {
src5: src-5 {
...
};
src6: src-6 {
...
};
};

rcar_sound,ssi {
ssi3: ssi-3 {
...
};
ssi4: ssi-4 {
...
};
};

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/875yyzk017.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup rsnd_parse_connect_xxx()

This patch tidyup rsnd_parse_connect_xxx() style.
Nothing is changed, but is preparation for
next "ASoC: rsnd: adjust disabled module" patch

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/877djfk01l.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup rsnd_dma_request_channel()

This patch adds "char *name" to rsnd_dma_request_channel().
It is not yet used so far, but is preparation for
next "ASoC: rsnd: adjust disabled module" patch

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/878s3vk01q.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup rsnd_parse_connect_common()

This patch adds "char *name" to rsnd_parse_connect_common().
It is not yet used so far, but is preparation for
next "ASoC: rsnd: adjust disabled module" patch

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87a6obk01v.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-rpmsg: fix platform_no_drv_owner.cocci warnings

./sound/soc/fsl/imx-rpmsg.c:140:3-8: 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

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zou Wei <zou_wei@huawei.com>
Message-Id: <1622113652-56646-1-git-send-email-zou_wei@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup rsnd_ssiu_busif_err_irq_ctrl()

rsnd_ssiu_busif_err_irq_ctrl() has very similar duplicated code.
This patch merge and tidyup the code.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87sg28lwxw.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup rsnd_ssiu_busif_err_status_clear()

rsnd_ssiu_busif_err_status_clear() has very similar duplicated code.
This patch merge and tidyup the code.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87tumolwy3.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: implement BUSIF related code in ssiu.c

BUSIF is SSIU feature, but its related code is
implemented at ssi.c today.
This patch moves it to ssiu.c

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87v974lwy9.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: protect mod->status

Renesas Sound uses many modules (SSI/SSIU/SRC/CTU/MIX/DVC/DMA),
and supports complex connections/path.
Thus each modules needs to save its status to correctly control it.
This status is updated when by .trigger, and .hw_params/.hw_free.

Renesas Sound is protecting modules by using lock when .trigger,
but it was not enough to protecting each modules "status" if it was
used from many paths.

1) .hw_params/.hw_free update status
2) another doesn't update status, but overwrites by same value

This patch do
1) protects .hw_params/.hw_free by lock
2) do nothing if no status update

Without this patch, protected mod->status (= .trigger) might be
overwrote by non protected mod->status (= .hw_params / .hw_free),
and in such case, CTU/MIX/DVC/SSIU/SSI which are used from
many paths might get damage.

If above issue happens, Renesas Sound will be hung (= silence)
and never be recoverd.
I could reproduce this issue by continue playing very short sound
with loop very long term (3-4 hours) through 2 inputs (= MIXer).

For updating rsnd_status_update(), this patch removes rsnd_dai_call()
debug message. Because we already have debugfs support, and is not
good match to new code.

Reported-by: Linh Phung T. Y <linh.phung.jy@renesas.com>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87wnrklwyh.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: add debugfs support

Current rsnd supports #define DEBUG, but it is not helpful
if issue happen after 4-5 hours.
This patch adds debugfs support for it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87y2c0lwyn.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: incidate irq error message

Current rsnd is using dev_dbg() if irq error happen,
but it makes debug very difficult if some strange things happen.
This patch uses dev_info() for it, and rename the macro name.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <87zgwglwyv.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: indicate unknown error at rsnd_dai_call()

Current rsnd_dai_call() doesn't indicate error message,
thus it is very difficult to know the issue
when strange things happen.
This patch indicates error for it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <871r9snbji.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: check BUIF error everytime

Current ssi.c checks BUSIF when TDM mode, but it should be checked
everytime.
This patch do it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <8735u8nbjr.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: attach SSIU when SSI was DMA mode

SSIU is not needed if SSI was PIO mode.
This patch ignores such case.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <874keonbkg.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: ignore runtime NULL case at rsnd_runtime_channel_original_with_params()

runtime might be NULL. Let's ignore such case.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <875yz4nbkt.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: call unregister for null_hw when removed

commit d6956a7dde6fb ("ASoC: rsnd: add null CLOCKIN support")
added null_clk, but it is using local static valuable.
It will be leaked if rsnd driver was removed.
This patch moves it to priv, and call unregister when removing.

Fixes: d6956a7dde6fb ("ASoC: rsnd: add null CLOCKIN support")
Link: https://lore.kernel.org/r/87tumsoe2p.wl-kuninori.morimoto.gx@renesas.com
Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Message-Id: <877djknbl5.wl-kuninori.morimoto.gx@renesas.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42l52: use DEVICE_ATTR_WO macro

Use DEVICE_ATTR_WO() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Message-Id: <20210524114239.7960-1-yuehaibing@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs35l3x: Use neutral language in amp drivers

Revise variable names and comments in
cs35l35 and cs35l36 amp drivers.

Signed-off-by: David Rhodes <drhodes@opensource.cirrus.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Message-Id: <20210525194439.2232908-1-drhodes@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs43130: Use DEVICE_ATTR_RO macro

Use DEVICE_ATTR_RO() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Message-Id: <20210524114017.18672-1-yuehaibing@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8962: Use DEVICE_ATTR_WO macro

Use DEVICE_ATTR_WO() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Message-Id: <20210524114753.39544-1-yuehaibing@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42l56: use DEVICE_ATTR_WO macro

Use DEVICE_ATTR_WO() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Message-Id: <20210524114503.26460-1-yuehaibing@huawei.com>
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-card: Make some symbols static

The sparse tool complains as follows:

sound/soc/fsl/imx-card.c:121:27: warning:
 symbol 'ak4458_fs_mul' was not declared. Should it be static?
sound/soc/fsl/imx-card.c:138:31: warning:
 symbol 'ak4458_tdm_fs_mul' was not declared. Should it be static?
sound/soc/fsl/imx-card.c:149:27: warning:
 symbol 'ak4497_fs_mul' was not declared. Should it be static?
sound/soc/fsl/imx-card.c:166:27: warning:
 symbol 'ak5558_fs_mul' was not declared. Should it be static?
sound/soc/fsl/imx-card.c:180:31: warning:
 symbol 'ak5558_tdm_fs_mul' was not declared. Should it be static?

Those symbols are not used outside of imx-card.c, so marks
them static.

Fixes: aa736700f42f ("ASoC: imx-card: Add imx-card machine driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Message-Id: <20210524133553.2366502-1-weiyongjun1@huawei.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

ASoC: tlv320aic26: use DEVICE_ATTR_RW macro

Use DEVICE_ATTR_RW() helper instead of plain DEVICE_ATTR(),
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Message-Id: <20210524115131.46288-1-yuehaibing@huawei.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

ASoC: cs42l42: Check jack status before reporting button events

Jack must be connected before reporting button events and
if the jack is disconnected button release must be reported

Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Message-Id: <20210525090822.64577-4-tanureal@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

ASoC: cs42l42: Add support for 2400000 Bit clock

Add support for 2.4MHz clock source

Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Message-Id: <20210525090822.64577-3-tanureal@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

ASoC: cs42l42: Add support for 2304000 Bit clock

Add support for 24bits, 2 channels, 48k Sample rate bit clock

Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Message-Id: <20210525090822.64577-2-tanureal@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

ASoC: cs42l42: Fix 1536000 Bit Clock instability

The 16 Bits, 2 channels, 48K sample rate use case needs
to configure a safer pll_divout during the start of PLL
After 800us from the start of PLL the correct pll_divout
can be set

Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Reviewed-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Message-Id: <20210525090822.64577-1-tanureal@opensource.cirrus.com>
Signed-off-by: Mark Brown <broonie@sirena.org.uk>

Merge branch 'for-5.13' of https://git./linux/kernel/git/broonie/sound into asoc-5.14

Merge series "ASoC: rsnd: add D3 support" from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:

Hi Mark, Rob

These adds R-Car D3 support for rsnd driver.
[1/3] is tidyup patch for dt-bindings (not only for D3).
[2/3], [3/3] are for R-Car D3.

Kuninori Morimoto (3):
  ASoC: dt-bindings: renesas: rsnd: tidyup properties
  ASoC: rsnd: tidyup loop on rsnd_adg_clk_query()
  ASoC: rsnd: add null CLOCKIN support

 .../bindings/sound/renesas,rsnd.yaml          | 10 ++++-
 sound/soc/sh/rcar/adg.c                       | 37 ++++++++++++++++---
 2 files changed, 41 insertions(+), 6 deletions(-)

--
2.25.1

ASoC: codecs: Add driver for NXP/Goodix TFA989x (TFA1) amplifiers

NXP's TFA98xx (now part of Goodix) are fairly popular speaker amplifiers
used in many smartphones and tablets. Most of them are sold as "smart
amplifiers" with built-in "CoolFlux DSP" that is used for volume control,
plus a "sophisticated speaker-boost and protection algorithm".

Unfortunately, they are also almost entirely undocumented. The short
datasheets (e.g. [1] for TFA9897) describe the available features,
but do not provide any information about the registers or how to use
the "CoolFlux DSP".

The amplifiers are most often configured through proprietary userspace
libraries. There are also some (rather complex) kernel drivers (e.g. [2])
but even those rely on obscure firmware blobs for configuration (so-called
"containers"). They seem to contain different "profiles" with tuned speaker
settings, sample rates and volume steps (which would be better exposed
as separate ALSA mixers).

The format of the firmware files seems to have changed a lot over the time,
so it's not even possible to simply re-use the firmware originally provided
by the vendor.

Overall, it seems close to impossible to develop a proper mainline driver
for these amplifiers that could make proper use of the built-in DSP.

This commit implements a compromise: At least the TFA1 family of the
TFA98xx amplifiers (usually called TFA989x) provide a way to *bypass*
the DSP using a special register sequence. The register sequence can be
found in similar variations in the kernel drivers from lots of vendors
e.g. in [3] and was probably mainly used for factory testing.

With the DSP bypassed, the amplifier acts mostly like a dumb standard
speaker amplifier, without (hardware) volume control. However, the setup
is much simpler and it works without any obscure firmware.

This driver implements the DSP bypass combined with chip-specific
initialization sequences adapted from [2]. Only TFA9895 is supported in
this initial commit. Except for the lack of volume control I can not hear
any difference with or without the DSP, it works just fine.

This driver allows the speaker to work on mainline Linux running on the
Samsung Galaxy A3/A5 (2015) [TFA9895] and Alcatel Idol 3 [TFA9897].
TFA9897 support will be added in separate patch set later.

[1]: https://product.goodix.com/en/docview/TFA9897%20SDS_Rev.3.1?objectId=47&objectType=document&version=78
[2]: https://source.codeaurora.org/external/mas/tfa98xx
[3]: https://github.com/sonyxperiadev/kernel/blob/57b5050e340f40a88e1ddb8d16fd9adb44418923/sound/soc/codecs/tfa98xx.c#L1422-L1462

Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Link: https://lore.kernel.org/r/20210513104129.36583-2-stephan@gerhold.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: codecs: Add bindings for nxp, tfa989x

NXP/Goodix TFA989X (TFA1) amplifiers are controlled via an I2C bus.
Add simple device tree bindings that describe how to set them up
in the device tree.

Right now only nxp,tfa9895 is supported but this will be extended
to at least nxp,tfa9897 in the near future.

Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20210513104129.36583-1-stephan@gerhold.net
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: add null CLOCKIN support

Some Renesas SoC doesn't have full CLOCKIN.
This patch add null_clk, and accepts it.

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87tumsoe2p.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rsnd: tidyup loop on rsnd_adg_clk_query()

commit 06e8f5c842f2d ("ASoC: rsnd: don't call clk_get_rate() under
atomic context") used saved clk_rate, thus for_each_rsnd_clk()
is no longer needed. This patch fixes it.

Fixes: 06e8f5c842f2d ("ASoC: rsnd: don't call clk_get_rate() under atomic context")
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87v978oe2u.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: renesas: rsnd: tidyup properties

1) resets/reset-names needs minItems
2) It can use ports, not only port
3) It is not using audio-graph properties

Without this patch, we will get warnings

Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87wnrooe2z.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: core: Fix Null-point-dereference in fmt_single_name()

Check the return value of devm_kstrdup() in case of
Null-point-dereference.

Fixes: 45dd9943fce0 ("ASoC: core: remove artificial component and DAI name constraint")
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Kefeng Wang <wangkefeng.wang@huawei.com>
Link: https://lore.kernel.org/r/20210524024941.159952-1-wangkefeng.wang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_sdw: add quirk support for Brya and BT-offload

Brya is another ADL-P product.

AlderLake has support for Bluetooth audio offload capability.
Enable the BT-offload quirk for ADL-P Brya and the Intel RVP.

Signed-off-by: Vamshi Krishna Gopal <vamshi.krishna.gopal@intel.com>
Signed-off-by: Yong Zhi <yong.zhi@intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210521155632.3736393-2-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: common: Add entries for sdw codecs in ADL match table

RT5682 and Max98373 are added with SDW0,SDW2 links respectively.

Signed-off-by: Vamshi Krishna Gopal <vamshi.krishna.gopal@intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210521155632.3736393-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: Remove conditions against CONFIG_PCI

The HDA support can only be compiled when SND_SOC_SOF_PCI is enabled which
depends on CONFIG_PCI.

This makes the IS_ENABLED(CONFIG_PCI) checks redundant in the code, they
will resolve to true all the time.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@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/20210521092804.3721324-8-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ops: don't return void value

Sparse throws the following warning:

sound/soc/sof/ops.h:247:17: error: returning void-valued expression

Remove the useless returns.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210521092804.3721324-7-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: topology: fix assignment to use le32_to_cpu

Fix sparse warning by using le32_to_cpu.

Signed-off-by: Jaska Uimonen <jaska.uimonen@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210521092804.3721324-6-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: loader: Use snd_sof_dsp_block_read() instead sof_block_read()

SOF core should use the IO functions via callbacks and not directly to
ensure that it remains platform independent.

Fixes: 83ee7ab1627b7 ("ASoC: SOF: Intel: byt: Refactor fw ready / mem windows creation")
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Daniel Baluta <daniel.baluta@gmail.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/20210521092804.3721324-5-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ops: print out the polling register

Print the register offset out to provide more useful information for the
register polling debugging.

Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@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/20210521092804.3721324-4-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: pci: No need to cast second time to save the desc

At the start of the function we already have the desc, no need to cast it
again from pci_id->driver_data to save it to sof_pdata.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@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/20210521092804.3721324-3-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Check desc->ops directly in acpi/pci/of probe functions

We can check for the desc->ops directly in the probe functions, the ops
is not used directly in the functions.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@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/20210521092804.3721324-2-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: qcom: lpass-cpu: Use optional clk APIs

This driver spits out a warning for me at boot:

 sc7180-lpass-cpu 62f00000.lpass: asoc_qcom_lpass_cpu_platform_probe() error getting optional null: -2

but it looks like it is all an optional clk. Use the optional clk APIs
here so that we don't see this message and everything else is the same.

Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Banajit Goswami <bgoswami@codeaurora.org>
Fixes: 3e53ac8230c1 ("ASoC: qcom: make osr clock optional")
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210520014807.3749797-1-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-card: Add imx-card machine driver

Add machine driver for i.MX boards, which supports
AK4458/AK5558/AK4497/AK5552 DAC/ADC attached to
SAI interface currently, but these DAC/ADCs are not
only supported codecs. This machine driver is designed
to be a more common machine driver for i.MX platform,
it can support widely cpu dai interface and codec
dai interface.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Signed-off-by: Mihai Serban <mihai.serban@nxp.com>
Signed-off-by: Cosmin-Gabriel Samoila <cosmin.samoila@nxp.com>
Signed-off-by: Viorel Suman <viorel.suman@nxp.com>
Signed-off-by: Daniel Baluta <daniel.baluta@nxp.com>
Signed-off-by: Adrian Alonso <adrian.alonso@nxp.com>
Link: https://lore.kernel.org/r/1621247488-21412-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: imx-card: Add binding doc for imx sound card

Imx-card is a new added machine driver for supporting
ak4458/ak5558/ak5552/ak4497 codec on i.MX platforms. But these
DAC/ADCs are not only supported codecs. This machine driver is
designed to be a more common machine driver for i.MX platform,
it can support widely cpu dai interface and codec dai interface.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/1621247488-21412-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: topology: Fix using uninitialized pointer

The original patch changed kcontrol_type to a pointer. In some goto
cases the pointer is assigned into a struct member as uninitialized and
this will cause a runtime error with UBSan even if it isn't a real bug.
So initialize the pointer to NULL.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: d29d41e28eea ("ASoC: topology: Add support for multiple kcontrol types to a widget")
Signed-off-by: Jaska Uimonen <jaska.uimonen@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210519100713.879958-1-jaska.uimonen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs35l33: fix an error code in probe()

This error path returns zero (success) but it should return -EINVAL.

Fixes: 3333cb7187b9 ("ASoC: cs35l33: Initial commit of the cs35l33 CODEC driver.")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/YKXuyGEzhPT35R3G@mwanda
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: meson: g12a-toacodec: add support for SM1" from Neil Armstrong <narmstrong@baylibre.com>:

This patchset adds Amlogic SM1 support on the TOACODEC driver, first by switching
to regmap fields for some bit fields to avoid code duplication, and then by adding
the corresponding bits & struct for the SM1 changed bits.

Changes since v2 at [2]:
- use raw values instead of defines in REG_FIELD() for g12a_toacodec_match_data

Changes since v1 at [1]:
- switch to regmap field

[1] https://lore.kernel.org/r/20210429170147.3615883-1-narmstrong@baylibre.com
[2] https://lore.kernel.org/r/20210505072607.3815442-1-narmstrong@baylibre.com

Neil Armstrong (2):
  ASoC: meson: g12a-toacodec: use regmap fields to prepare SM1 support
  ASoC: meson: g12a-toacodec: add support for SM1 TOACODEC

 sound/soc/meson/g12a-toacodec.c | 143 +++++++++++++++++++++++++++-----
 1 file changed, 124 insertions(+), 19 deletions(-)

--
2.25.1

ASoC: soc-dai: fix up hw params only if it is needed

If fixed hw params won't be used, fixing up isn't needed also.

Signed-off-by: Gyeongtaek Lee <gt82.lee@samsung.com>
Link: https://lore.kernel.org/r/000401d748bc$fa466d50$eed347f0$@samsung.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8750: convert to the json-schema

This helps validating DTS files.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/20210512205926.780-1-zajec5@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: meson: g12a-toacodec: add support for SM1 TOACODEC

This adds support for the TOACODEC found in Amlogic SM1 SoCs.

The bits are shifted for more selection of clock sources, so this only
maps the same support for G12A to the SM1 bits.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lore.kernel.org/r/20210511074829.4110036-3-narmstrong@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: meson: g12a-toacodec: use regmap fields to prepare SM1 support

Switch usage to regmap field for bits handled by the g12a_toacodec_mux_put_enum()
function to avoid uselesss code duplication when adding SM1 variant support.

Signed-off-by: Neil Armstrong <narmstrong@baylibre.com>
Link: https://lore.kernel.org/r/20210511074829.4110036-2-narmstrong@baylibre.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: don't send DAI_CONFIG IPC for older firmware

BE hw_params op was recently added for SSP type DAIs.
But sending the DAI_CONFIG IPC during hw_params
is not supported with older firmware. So add an ABI check
to avoid sending the IPC if the firmware ABI is older than
3.18.

Fixes: e12be9fbfb91 ('ASoC: SOF: Intel: HDA: add hw params callback for SSP DAIs')
Tested-by: Yong Zhi <yong.zhi@intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20210518174121.151601-1-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: hisilicon: fix missing clk_disable_unprepare() on error in hi6210_i2s_startup()

After calling clk_prepare_enable(), clk_disable_unprepare() need
be called when calling clk_set_rate() failed.

Fixes: 0bf750f4cbe1 ("ASoC: hisilicon: Add hi6210 i2s audio driver")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210518044514.607010-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-core: use DEVICE_ATTR_RO macro

Use DEVICE_ATTR_RO helper instead of plain DEVICE_ATTR,
which makes the code a bit shorter and easier to read.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20210514081100.16196-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rk3328: fix missing clk_disable_unprepare() on error in rk3328_platform_probe()

Fix the missing clk_disable_unprepare() before return
from rk3328_platform_probe() in the error handling case.

Fixes: c32759035ad2 ("ASoC: rockchip: support ACODEC for rk3328")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20210518075847.1116983-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: fix SND_SOC_IMX_RPMSG dependency

Kconfig produces a warning with SND_SOC_FSL_RPMSG=y and SND_IMX_SOC=m:

WARNING: unmet direct dependencies detected for SND_SOC_IMX_RPMSG
  Depends on [m]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && SND_IMX_SOC [=m] && RPMSG [=y]
  Selected by [y]:
  - SND_SOC_FSL_RPMSG [=y] && SOUND [=y] && !UML && SND [=y] && SND_SOC [=y] && COMMON_CLK [=y] && RPMSG [=y] && SND_IMX_SOC [=m]!=n

Add a dependency to prevent this configuration.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com>
Link: https://lore.kernel.org/r/20210514213118.630427-1-arnd@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: topology: Add support for multiple kcontrol types to a widget

Current dapm widget has a single variable to describe its kcontrol's
type. As there can be many kcontrols in one widget it is inherently
presumed that the types are the same.

Lately there has been use cases where different types of kcontrols would
be needed for a single widget. Thus add pointer to dapm widget to hold
an array for different kcontrol types and modify the kcontrol creation
to operate in a loop based on individual kcontrol type.

Change control creation and deletion to use individual kcontrol types in
SOF driver. This is done in the same patch for not breaking bisect. SOF
driver is also currently the only one using the dapm widget
kcontrol_type.

Signed-off-by: Jaska Uimonen <jaska.uimonen@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20210507070246.404446-1-jaska.uimonen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: intel/boards: add missing MODULE_DEVICE_TABLE

This patch adds missing MODULE_DEVICE_TABLE definition which generates
correct modalias for automatic loading of this driver when it is built
as an external module.

Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zou Wei <zou_wei@huawei.com>
Link: https://lore.kernel.org/r/1620791647-16024-1-git-send-email-zou_wei@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Merge series "ASoC: SOF/Intel: fix and remove Sparse warnings" from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

We updated our SOF CI settings last week (see below) to use more options
for Sparse, and sure enough it detected one nasty single-character bug
in one of my previous patches, along with more trivial issues with
string/integer sizes and signed/unsigned confusions.

export ARCH=x86_64
export CF="-Wsparse-error -Wsparse-all -Wno-bitwise-pointer -Wno-pointer-arith -Wno-typesign -Wnoshadow"
make -k sound/soc/sof/ C=2
make -k sound/soc/intel/common/ C=2
make -k sound/soc/intel/boards/ C=2
make -k drivers/soundwire/ C=2

Pierre-Louis Bossart (9):
  ASoC: SOF: Intel: hda: fix index used in inner loop
  ASoC: Intel: sof_rt5682: shrink platform_id names below 20 characters
  ASoC: Intel: sof_cs42l42: shrink platform id below 20 characters
  ASoC: Intel: bxt_da7219_max98357a: shrink platform_id below 20
    characters
  ASoC: Intel: glk_rt5682_max98357a: shrink platform_id below 20
    characters
  ASoC: Intel: kbl_da7219_max98357a: shrink platform_id below 20
    characters
  ASoC: Intel: sof_da7219_max98373: shrink platform_id below 20
    characters
  ASoC: Intel: sof_sdw: fix signed/unsigned warning
  ASoC: Intel: soc-acpi: add ull suffix for SoundWire _ADR values

 sound/soc/intel/boards/bxt_da7219_max98357a.c | 12 +++---
 sound/soc/intel/boards/glk_rt5682_max98357a.c |  4 +-
 sound/soc/intel/boards/kbl_da7219_max98357a.c |  4 +-
 sound/soc/intel/boards/sof_cs42l42.c          |  2 +-
 sound/soc/intel/boards/sof_da7219_max98373.c  |  8 ++--
 sound/soc/intel/boards/sof_rt5682.c           | 20 +++++-----
 sound/soc/intel/boards/sof_sdw.c              |  2 +-
 .../intel/common/soc-acpi-intel-adl-match.c   | 28 +++++++-------
 .../intel/common/soc-acpi-intel-bxt-match.c   |  2 +-
 .../intel/common/soc-acpi-intel-cml-match.c   | 22 +++++------
 .../intel/common/soc-acpi-intel-cnl-match.c   |  2 +-
 .../intel/common/soc-acpi-intel-glk-match.c   |  6 +--
 .../intel/common/soc-acpi-intel-icl-match.c   | 12 +++---
 .../intel/common/soc-acpi-intel-jsl-match.c   |  6 +--
 .../intel/common/soc-acpi-intel-kbl-match.c   |  2 +-
 .../intel/common/soc-acpi-intel-tgl-match.c   | 38 +++++++++----------
 sound/soc/sof/intel/hda.c                     |  2 +-
 17 files changed, 86 insertions(+), 86 deletions(-)

--
2.25.1

ASoC: cs42l42: make HSBIAS_SENSE_EN optional

HSBIAS_SENSE_EN configures HSBIAS output current sense through
the external 2.21-k resistor. HSBIAS_SENSE is hardware feature to reduce
the potential pop noise during the headset plug out slowly. But on some
platforms ESD voltage will affect it causing test to fail, especially
with CTIA headset type. For different hardware setups, a designer might
want to tweak default behavior.

Signed-off-by: Vitaly Rodionov <vitalyr@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210511145220.125760-1-vitalyr@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: q6dsp: Undo buggy warning fix

This reverts commit 5f1b95d08de712327e452d082a50fded435ec884.

The warnings that commit 5f1b95d08de7 ("ASoC: q6dsp: q6afe: remove
unneeded dead-store initialization") was trying to fix were already
fixed in commit 12900bacb4f3 ("ASoC: qcom: q6afe: remove useless
assignments"). With both commits in the tree, port_id is uninitialized,
as pointed out by clang:

sound/soc/qcom/qdsp6/q6afe.c:1213:18: warning: variable 'port_id' is
uninitialized when used here [-Wuninitialized]
        stop->port_id = port_id;
                        ^~~~~~~
sound/soc/qcom/qdsp6/q6afe.c:1186:13: note: initialize the variable
'port_id' to silence this warning
        int port_id;
                   ^
                    = 0
1 warning generated.

Bring back the initialization so that everything works as intended.

Fixes: 5f1b95d08de7 ("ASoC: q6dsp: q6afe: remove unneeded dead-store initialization")
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Link: https://lore.kernel.org/r/20210511190306.2418917-1-nathan@kernel.org
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm2200: remove include of wmfw.h

We want all wm_adsp clients to use the wm_adsp.h header as they
shouldn't need to include internal sub-headers.

Signed-off-by: Simon Trimmer <simont@opensource.cirrus.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210511171514.270219-1-simont@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm_adsp: mark more data structures with the const qualifier

The callback structures and memory region type table can be marked as
const as they will not change during use.

Fix checkpatch warning against wm_adsp_find_region function by moving
const keyword to form the 'static const struct' pattern.

Signed-off-by: Simon Trimmer <simont@opensource.cirrus.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210511171459.270169-1-simont@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: bxt_da7219_max98357a: shrink platform_id below 20 characters

Sparse throwns the following warnings:

sound/soc/intel/boards/bxt_da7219_max98357a.c:843:19: error: too long
initializer-string for array of char(no space for nul char)

sound/soc/intel/boards/bxt_da7219_max98357a.c:844:19: error: too long
initializer-string for array of char(no space for nul char)

sound/soc/intel/boards/bxt_da7219_max98357a.c:845:19: error: too long
initializer-string for array of char(no space for nul char)

Fix by using the 'mx' acronyn for Maxim

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20210511213707.32958-5-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_cs42l42: shrink platform id below 20 characters

The platform_id is too long and is flagged by a sparse warning:

sound/soc/intel/boards/sof_cs42l42.c:483:25: error: too long
initializer-string for array of char(no space for nul char)

fix by using the 'mx' acronym for Maxim

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Link: https://lore.kernel.org/r/20210511213707.32958-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: hda: fix index used in inner loop

With more warnings than the default, Sparse throws the following
warning:

sound/soc/sof/intel/hda.c:1127:49: error: self-comparison always
evaluates to true
sound/soc/sof/intel/hda.c:1128:49: error: self-comparison always
evaluates to true
sound/soc/sof/intel/hda.c:1129:48: error: self-comparison always
evaluates to true

This looks like an obvious error, with a likely copy-pasted line
leading to the use of the wrong index in an inner loop. One of the
worst single-character bugs in a long time.

This problem was not detected in our tests since in practice SoundWire
platforms only have identical devices per link and the index mistake
did not change the results.

Fixes: 6f5d506d7ff1dq ('ASoC: SOF: Intel: SoundWire: refine ACPI match')
Reviewed-by: Paul Olaru <paul.olaru@oss.nxp.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210511213707.32958-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42l52: Minor tidy up of error paths

Fixup a needlessly initialised variable and an unchecked return
value.

Reported-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210511175718.15416-5-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs35l32: Add missing regmap use_single config

This device requires single register transactions, this will
definely cause problems with the new device ID parsing which uses
regmap_bulk_read but might also show up in the cache sync sometimes.
Add the missing flags to the regmap_config.

Fixes: 283160f1419d ("ASoC: cs35l32: Minor error paths fixups")
Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210511175718.15416-4-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>