Chris Morgan [Fri, 20 Oct 2023 17:15:39 +0000 (12:15 -0500)]
ASoC: es8328: Use rounded rate for es8328_set_sysclk()
I have a board (RK3588 based) that sets the sysclk to 12287999. The
es8328 driver fails to match this to the 12288000 rate and fails to
load. Allow the rate comparison to work if the frequency is within
100hz by dividing it by 100 and rounding it, then multiplying it back
by 100.
Note the 100hz value was chosen arbitrarily by me, but it has only
been tested with a 1hz difference.
Charles Keepax [Thu, 19 Oct 2023 17:34:05 +0000 (12:34 -0500)]
ASoC: intel: sof_sdw: Move the builtin microphones to dataport 1
cs42l43 supports 4 hardwired microphones, but only supports up to 2
microphone headsets. Only dataport 1 can support 4 channel capture,
but that is currently used for the headset microphone. Switch things
around such that DP1 is used for the builtin mics and DP2 is used for
the headset microphones.
Charles Keepax [Thu, 19 Oct 2023 17:34:04 +0000 (12:34 -0500)]
ASoC: intel: sof_sdw_cs42l43: Create separate jacks for hp and mic
It makes sense to report the microphone separately from the headphones,
that way ALSA UCM can differentiate between switching the playback and
the capture. For example, still using the built-in microphone path when
a 3-pole headset is inserted.
Charles Keepax [Thu, 19 Oct 2023 17:34:02 +0000 (12:34 -0500)]
ASoC: intel: sof_sdw: Stop processing CODECs when enough are found
When adding CODECs to a DAI link, the code should stop processing more
CODECs when the expected number of CODECs are discovered. This fixes a
small corner case issue introduced when support for different devices
on the same SoundWire link was added. In the case of aggregated
devices everything is fine, as all devices intended for the DAI link
will be marked with the same group and any not intended for that DAI
are skipped by the group check. However for non-aggregated devices the
group check is bypassed and the current code does not stop after it
has found the first device. Meaning if additional non-aggregated devices
are present on the same SoundWire link they will be erroneously added
into the DAI link.
Fix this issue, and provide a small optimisation by ceasing to process
devices once we have reached the required number of devices for the
current DAI link.
Fixes: 317dcdecaf7a ("ASoC: intel: sof_sdw: Allow different devices on the same link") Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com> Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com> Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20231019173411.166759-2-pierre-louis.bossart@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org>
Gustavo A. R. Silva [Mon, 9 Oct 2023 21:24:23 +0000 (15:24 -0600)]
ASoC: sigmadsp: Add __counted_by for struct sigmadsp_data and use struct_size()
Prepare for the coming implementation by GCC and Clang of the __counted_by
attribute. Flexible array members annotated with __counted_by can have
their accesses bounds-checked at run-time via CONFIG_UBSAN_BOUNDS (for
array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
functions).
While there, use struct_size() and size_sub() helpers, instead of the
open-coded version, to calculate the size for the allocation of the
whole flexible structure, including of course, the flexible-array
member.
This code was found with the help of Coccinelle, and audited and
fixed manually.
Signed-off-by: "Gustavo A. R. Silva" <gustavoars@kernel.org> Reviewed-by: Kees Cook <keescook@chromium.org> Link: https://lore.kernel.org/r/ZSRvh1j2MVVhuOUv@work Signed-off-by: Mark Brown <broonie@kernel.org>
Peter Ujfalusi [Thu, 12 Oct 2023 19:18:48 +0000 (15:18 -0400)]
ASoC: SOF: Intel: hda-dsp: Make sure that no irq handler is pending before suspend
In the existing IPC support, the reply to each IPC message is handled in
an IRQ thread. The assumption is that the IRQ thread is scheduled without
significant delays.
On an experimental (iow, buggy) kernel, the IRQ thread dealing with the
reply to the last IPC message before powering-down the DSP can be delayed
by several seconds. The IRQ thread will proceed with register accesses
after the DSP is powered-down which results in a kernel crash.
While the bug which causes the delay is not in the audio stack, we must
handle such cases with defensive programming to avoid such crashes.
Call synchronize_irq() before proceeding to power down the DSP to make
sure that no irq thread is pending execution.
Closes: https://github.com/thesofproject/linux/issues/4608 Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Link: https://lore.kernel.org/r/20231012191850.147140-3-pierre-louis.bossart@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org>
Brent Lu [Thu, 12 Oct 2023 19:08:15 +0000 (15:08 -0400)]
ASoC: Intel: sof_hdmi: add common header for HDMI
Add a common header for Intel HDMI dai link (idisp) initialization.
Declare the sof_hdmi_private structure in machine driver private data
and use it to initialize dai link.
Balamurugan C [Thu, 12 Oct 2023 19:08:08 +0000 (15:08 -0400)]
ASoC: Intel: MTL: Add entry for HDMI-In capture support to non-I2S codec boards.
Adding HDMI-In capture support for the MTL products which doesn't have
onboard I2S codec. But need to support HDMI-In capture via I2S and
audio playback through HDMI/DP monitor.
Mark Brown [Thu, 12 Oct 2023 16:51:49 +0000 (17:51 +0100)]
Harden SSP boards and add TDM support
Merge series from Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>:
In quite a few places in code there are checks for number of SSPs
present on system, to reduce maintenance burden introduce helper
functions allowing to get SSP and TDM from machine board configuration.
Current mechanism replaces "%d" present in some routes and widget names
with SSP number. However there are also configurations which make use of
of TDM number, in which case expected behavior would be to have string
in form of SSP:TDM - see implementation of avs_i2s_platform_register()
in sound/soc/intel/avs/pcm.c. Implement custom function, which parses
string and make use of it when parsing topology. While at it make sure
that we generate dynamic names only if there is no multiple SSPs or TDMs
defined.
Nathan Chancellor [Wed, 11 Oct 2023 20:21:51 +0000 (13:21 -0700)]
ASoC: tegra: Fix -Wuninitialized in tegra210_amx_platform_probe()
Clang warns (or errors with CONFIG_WERROR=y):
sound/soc/tegra/tegra210_amx.c:553:10: error: variable 'soc_data' is uninitialized when used here [-Werror,-Wuninitialized]
553 | soc_data->regmap_conf);
| ^~~~~~~~
A refactoring removed the initialization of this variable but its use
was not updated. Use the soc_data value in the amx variable to resolve
the warning and remove the soc_data variable, as it is now entirely
unused.
Amadeusz Sławiński [Thu, 12 Oct 2023 08:35:01 +0000 (10:35 +0200)]
ASoC: Intel: avs: Improve topology parsing of dynamic strings
Current mechanism replaces "%d" present in some routes and widget names
with SSP number. However there are also configurations which make use of
TDM number, in which case expected behavior would be to have string in
form of SSP:TDM - see implementation of avs_i2s_platform_register() in
sound/soc/intel/avs/pcm.c.
Implement custom function, which parses string and make use of it when
parsing topology. While at it make sure that we generate dynamic names
only if there is no multiple SSPs or TDMs defined.
Amadeusz Sławiński [Thu, 12 Oct 2023 08:35:00 +0000 (10:35 +0200)]
ASoC: Intel: avs: Introduce helper functions for SSP and TDM handling
In quite a few places in code there are checks for number of SSPs
present on system, to reduce maintenance burden introduce helper
functions allowing to get SSP and TDM from machine board configuration.
Additionally in boards we use SSP and TDM to generate quite a few
strings, it could be done like:
if (tdms)
dl->name = devm_kasprintf(dev, GFP_KERNEL, "SSP%d:%d-Codec",
ssp_port, tdm_slot);
else
dl->name = devm_kasprintf(dev, GFP_KERNEL, "SSP%d-Codec",
ssp_port);
but quite quickly code ends up with spaghetti of similar if elses.
Instead introduce macro which can be used to generate correct string,
allowing to minimize code to something like:
dl->name = devm_kasprintf(dev, GFP_KERNEL, AVS_STRING_FMT("SSP", "-Codec",
ssp_port, tdm_slot));
Amadeusz Sławiński [Thu, 12 Oct 2023 08:34:59 +0000 (10:34 +0200)]
ASoC: Intel: avs: Only create SSP%d snd_soc_dai_driver when requested
When using TDM configuration some other device may be using SSP%d, so
don't create snd_soc_dai_driver configuration for it unless requested
by TDM configuration.
While at it adjust tdf8532 board to explicitly describe TDM
configuration.
ASoC: apple: mca: Annotate struct mca_data with __counted_by
Prepare for the coming implementation by GCC and Clang of the __counted_by
attribute. Flexible array members annotated with __counted_by can have
their accesses bounds-checked at run-time checking via CONFIG_UBSAN_BOUNDS
(for array indexing) and CONFIG_FORTIFY_SOURCE (for strcpy/memcpy-family
functions).
As found with Coccinelle[1], add __counted_by for struct mca_data.
Cc: Martin Povišer <povik+lin@cutebit.org> 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: asahi@lists.linux.dev Cc: alsa-devel@alsa-project.org Signed-off-by: Kees Cook <keescook@chromium.org> Reviewed-by: "Gustavo A. R. Silva" <gustavoars@kernel.org> Link: https://lore.kernel.org/r/20230922175050.work.819-kees@kernel.org Signed-off-by: Mark Brown <broonie@kernel.org>
Uwe Kleine-König [Mon, 9 Oct 2023 15:59:45 +0000 (17:59 +0200)]
ASoC: SOF: Convert to platform remove callback returning void
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new(), which already returns void. Eventually after all drivers
are converted, .remove_new() will be renamed to .remove().
The SOF platform drivers all use either sof_of_remove() or
sof_acpi_remove() which both return zero unconditionally. Change these
functions to return void and the drivers to use .remove_new(). There is
no semantical change.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20231009155945.285537-1-u.kleine-koenig@pengutronix.de Signed-off-by: Mark Brown <broonie@kernel.org>
Charles Keepax [Tue, 10 Oct 2023 10:24:25 +0000 (11:24 +0100)]
ASoC: cs42l43: Update values for bias sense
Due to an error in the datasheet the bias sense values currently don't
match the hardware. Whilst this is a change to the binding no devices
have yet shipped so updating the binding will not cause any issues.
Charles Keepax [Tue, 10 Oct 2023 10:24:24 +0000 (11:24 +0100)]
ASoC: dt-bindings: cirrus,cs42l43: Update values for bias sense
Due to an error in the datasheet the bias sense values currently don't
match the hardware. Whilst this is a change to the binding no devices
have yet shipped so updating the binding will not cause any issues.
Kuninori Morimoto [Mon, 9 Oct 2023 23:39:43 +0000 (23:39 +0000)]
ASoC: fsl: mpc5200_dma.c: Fix warning of Function parameter or member not described
This patch fixes the warnings of "Function parameter or member 'xxx'
not described".
>> sound/soc/fsl/mpc5200_dma.c:116: warning: Function parameter or member 'component' not described in 'psc_dma_trigger'
sound/soc/fsl/mpc5200_dma.c:116: warning: Function parameter or member 'substream' not described in 'psc_dma_trigger'
sound/soc/fsl/mpc5200_dma.c:116: warning: Function parameter or member 'cmd' not described in 'psc_dma_trigger'
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202310061914.jJuekdHs-lkp@intel.com/ Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Fixes: 6d1048bc1152 ("ASoC: fsl: mpc5200_dma: remove snd_pcm_ops") Link: https://lore.kernel.org/r/87il7fcqm8.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
Mark Brown [Mon, 9 Oct 2023 19:49:16 +0000 (20:49 +0100)]
ASoC: DT matching and header cleanups
Merge series from Rob Herring <robh@kernel.org>:
This is a series is part of ongoing clean-ups related to device
matching and DT related implicit includes. Essentially of_device.h has
a bunch of implicit includes and generally isn't needed any nore except
for of_match_device(). As we also generally want to get rid of
of_match_device() as well, I've done that so we're not updating the
includes twice.
Richard Fitzgerald [Mon, 9 Oct 2023 15:34:12 +0000 (16:34 +0100)]
ASoC: cs35l56: ASP1 DOUT must default to Hi-Z when not transmitting
The ASP1 DOUT line must be defaulted to be high-impedance when
it is not actually transmitting data for an active channel.
In non-SoundWire modes ASP1 will usually be shared by multiple
amps so each amp must only drive the line during the slot for
an enabled TX channel.
In SoundWire mode a custom firmware can use ASP1 as a secondary
chip-to-chip audio link or as GPIO. It should be defaulted to
high-impedance since by default the purpose of this pin is not
known.
Backport note:
On kernel versions before 6.6 the cs35l56->base.regmap argument
to regmap_set_bits() must be changed to cs35l56->regmap.
Mark Brown [Mon, 9 Oct 2023 17:40:09 +0000 (18:40 +0100)]
ASoC: codecs: fix widget name comparisons
Merge series from Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>:
Some codec drivers compare widget names with strcmp, ignoring the component
name prefix. If prefix is used, the comparisons start failing.
Add a helper to fix the issue.
Linus Walleij [Fri, 6 Oct 2023 13:46:31 +0000 (15:46 +0200)]
ASoC: mt8192-afe-gpio: Drop unused include
This driver includes the legacy GPIO header <linux/gpio.h> but
is not using any symbols from it. AFE has a custom GPIO
implementation that is not using the kernel GPIO framework.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20231006-descriptors-asoc-mediatek-v1-8-07fe79f337f5@linaro.org Signed-off-by: Mark Brown <broonie@kernel.org>
Linus Walleij [Fri, 6 Oct 2023 13:46:30 +0000 (15:46 +0200)]
ASoC: mt8186-mt6366-rt1019-rt5682s: Drop unused include
This driver includes the legacy GPIO header <linux/gpio.h> but
is not using any symbols from it. AFE has a custom GPIO
implementation that is not using the kernel GPIO framework,
so it need not include it either.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20231006-descriptors-asoc-mediatek-v1-7-07fe79f337f5@linaro.org Signed-off-by: Mark Brown <broonie@kernel.org>
Linus Walleij [Fri, 6 Oct 2023 13:46:29 +0000 (15:46 +0200)]
ASoC: mt8173-rt5650: Drop unused includes
This driver includes the legacy GPIO header <linux/gpio.h> and
<linux/of_gpio.h> but does not use any symbols from either of
them so drop the includes.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20231006-descriptors-asoc-mediatek-v1-6-07fe79f337f5@linaro.org Signed-off-by: Mark Brown <broonie@kernel.org>
Linus Walleij [Fri, 6 Oct 2023 13:46:28 +0000 (15:46 +0200)]
ASoC: mt8173-rt5650-rt5676: Drop unused includes
This driver includes the legacy GPIO header <linux/gpio.h> and
<linux/of_gpio.h> but does not use any symbols from either of
them so drop the includes.
Linus Walleij [Fri, 6 Oct 2023 13:46:27 +0000 (15:46 +0200)]
ASoC: mt8173-rt5650-rt5514: Drop unused includes
This driver includes the legacy GPIO header <linux/gpio.h> and
<linux/of_gpio.h> but does not use any symbols from either of
them so drop the includes.
Linus Walleij [Fri, 6 Oct 2023 13:46:26 +0000 (15:46 +0200)]
ASoC: mt8173-max98090: Drop unused include
This driver includes the legacy GPIO header <linux/gpio.h> but
is not using any symbols from it. Drop the include.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Tested-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20231006-descriptors-asoc-mediatek-v1-3-07fe79f337f5@linaro.org Signed-off-by: Mark Brown <broonie@kernel.org>
Linus Walleij [Fri, 6 Oct 2023 13:46:25 +0000 (15:46 +0200)]
ASoC: mediatek: mt2701-cs42448: Convert to GPIO descriptors
The driver is pretty straight-forward to convert to use GPIO
descriptors, however a separate patch is needed to accept
the DT GPIO resources ending with "-gpio1" and "-gpio2"
instead of the standard "-gpio" or "-gpios" name convention.
Linus Walleij [Fri, 6 Oct 2023 13:46:24 +0000 (15:46 +0200)]
gpiolib: of: Add quirk for mt2701-cs42448 ASoC sound
These gpio names are due to old DT bindings not following the
"-gpio"/"-gpios" conventions. Handle it using a quirk so the
driver can just look up the GPIOs.
Krzysztof Kozlowski [Tue, 3 Oct 2023 15:54:22 +0000 (17:54 +0200)]
ASoC: codecs: wsa-macro: fix uninitialized stack variables with name prefix
Driver compares widget name in wsa_macro_spk_boost_event() widget event
callback, however it does not handle component's name prefix. This
leads to using uninitialized stack variables as registers and register
values. Handle gracefully such case.
Krzysztof Kozlowski [Thu, 5 Oct 2023 07:52:50 +0000 (09:52 +0200)]
ASoC: qcom: reduce number of binding headers includes
Move the includes of binding headers from Qualcomm SoC sound drivers
headers to unit files actually using these bindings. This reduces the
amount of work for C preprocessor and makes usage of bindings easier to
follow. No impact expected on the final binaries.
projects / users / jedix / linux-maple.git / log