linux-toradex.git/sound/soc/codecs, branch master Linux kernel for Apalis and Colibri modules Merge tag 'asoc-fix-v7.4-merge-window' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus 2026-10-31T15:31:50+00:00 Takashi Iwai tiwai@suse.de 2026-10-31T15:31:50+00:00 74f6055d0c5997b13e6a5253fdc2263f3af90408 ASoC: Fixes for v7.4 A moderately large collection of fixes and quirks that have come in during the merge window, everything here is driver specific and none of it unusually severe. 
ASoC: Fixes for v7.4

A moderately large collection of fixes and quirks that have come in
during the merge window, everything here is driver specific and none of
it unusually severe.
ASoC: hdac_hda: Keep runtime PM state in sync across codec remove/probe 2026-10-28T09:00:27+00:00 Peter Ujfalusi peter.ujfalusi@linux.intel.com 2026-10-27T14:46:55+00:00 99203e8d8968def973e82bb712ea959c6a46c16e After the codec has been removed, its runtime PM status is RPM_SUSPENDED while hdac_hda_codec_probe() powers the codec up via snd_hda_codec_device_new(). The pm_runtime_put()/snd_hda_power_down() calls on probe and registration then see an already suspended device, the codec suspend callback is not called and the codec, its HDA link and the codec_powered bit stay set, preventing the controller from runtime suspending. The remove has a similar problem with the codec link accounting: hdac_hda_codec_remove() drops the link reference taken in probe without looking at the runtime PM state of the codec. If the codec was already runtime suspended, the link reference has been dropped by the codec suspend callback and the remove drops it a second time. Resume the codec in hdac_hda_codec_remove() before disabling runtime PM so that the link reference and the codec_powered bit are always in the same state as after the probe, and mark the device active on probe once the codec has been powered up. Finally mark the codec device suspended at the end of the remove. Its runtime PM is disabled at this point and if only the ASoC card is unbound (e.g. by unbinding the machine driver), a following system suspend would otherwise call the runtime suspend callback on the already cleaned up codec via pm_runtime_force_suspend(). This is a followup patch to address Sashiko reports. Link: https://lore.kernel.org/linux-sound/20261027083840.28813-4-peter.ujfalusi@linux.intel.com/ Fixes: c3ec8ac82105 ("ASoC: hdac_hda: fix memleak on module unload") Cc: stable@vger.kernel.org Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://patch.msgid.link/20261027144655.31988-1-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> 
After the codec has been removed, its runtime PM status is
RPM_SUSPENDED while hdac_hda_codec_probe() powers the codec up via
snd_hda_codec_device_new(). The pm_runtime_put()/snd_hda_power_down()
calls on probe and registration then see an already suspended device,
the codec suspend callback is not called and the codec, its HDA link and
the codec_powered bit stay set, preventing the controller from runtime
suspending.

The remove has a similar problem with the codec link accounting:
hdac_hda_codec_remove() drops the link reference taken in probe without
looking at the runtime PM state of the codec. If the codec was already
runtime suspended, the link reference has been dropped by the codec
suspend callback and the remove drops it a second time.

Resume the codec in hdac_hda_codec_remove() before disabling runtime PM
so that the link reference and the codec_powered bit are always in the
same state as after the probe, and mark the device active on probe once
the codec has been powered up.

Finally mark the codec device suspended at the end of the remove. Its
runtime PM is disabled at this point and if only the ASoC card is
unbound (e.g. by unbinding the machine driver), a following system
suspend would otherwise call the runtime suspend callback on the already
cleaned up codec via pm_runtime_force_suspend().

This is a followup patch to address Sashiko reports.

Link: https://lore.kernel.org/linux-sound/20261027083840.28813-4-peter.ujfalusi@linux.intel.com/
Fixes: c3ec8ac82105 ("ASoC: hdac_hda: fix memleak on module unload")
Cc: stable@vger.kernel.org
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://patch.msgid.link/20261027144655.31988-1-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: rt721: Remove invalid register defaults 2026-10-27T12:42:39+00:00 Oder Chiou oder_chiou@realtek.com 2026-10-27T06:02:47+00:00 e32773a86100a0a6e5e4b30c50b773de34b5c6c3 Egor Poderiagin reported runtime resume failures caused by rejected writes during regcache_sync(). Remove the PDE23 REQ_POWER_STATE default and the defaults for the volatile MBQ registers 0x2000046 and 0x6100006. Also remove 0x2000046 from the MBQ readable and volatile register lists. This allows cache sync to complete without suppressing sync errors. Thanks to Egor Poderiagin for the detailed report and testing of the default removals. Closes: https://lore.kernel.org/linux-sound/8a12a2d9-1b56-4e00-8ccc-cf31325f8366@crazyrussian.pro/ Fixes: 7a0886826308 ("ASoC: codecs: rt721-sdca-sdw: Propagate regcache_sync() errors") Signed-off-by: Oder Chiou <oder_chiou@realtek.com> Link: https://patch.msgid.link/20261027-rt721fix-v1-1-42cd4b1711d2@realtek.com Signed-off-by: Mark Brown <broonie@kernel.org> 
Egor Poderiagin reported runtime resume failures caused by rejected
writes during regcache_sync().

Remove the PDE23 REQ_POWER_STATE default and the defaults for the
volatile MBQ registers 0x2000046 and 0x6100006. Also remove 0x2000046
from the MBQ readable and volatile register lists.

This allows cache sync to complete without suppressing sync errors.

Thanks to Egor Poderiagin for the detailed report and testing of the
default removals.

Closes:
https://lore.kernel.org/linux-sound/8a12a2d9-1b56-4e00-8ccc-cf31325f8366@crazyrussian.pro/

Fixes: 7a0886826308 ("ASoC: codecs: rt721-sdca-sdw: Propagate regcache_sync() errors")
Signed-off-by: Oder Chiou <oder_chiou@realtek.com>
Link: https://patch.msgid.link/20261027-rt721fix-v1-1-42cd4b1711d2@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: hdac_hda: Balance the runtime PM disable in codec remove 2026-10-27T11:53:12+00:00 Peter Ujfalusi peter.ujfalusi@linux.intel.com 2026-10-27T08:38:40+00:00 f26c0d5f91d918bb66cbbc8916f41836adb38346 hdac_hda_codec_remove() disables runtime PM of the codec device before releasing the HDA link and calling the codec driver's remove callback, so that no runtime resume can race with the teardown and access the link or the already freed codec driver data. snd_hda_codec_cleanup_for_unbind() then disables runtime PM again. On the next probe snd_hda_codec_register() enables runtime PM only once, leaving it disabled after the ASoC card or the codec driver has been unbound and bound again: the codec stays powered and runtime_status reads 'unsupported'. Re-enable runtime PM at the end of hdac_hda_codec_remove() to balance its own disable. The codec remains disabled until it is registered again. Fixes: c3ec8ac82105 ("ASoC: hdac_hda: fix memleak on module unload") Cc: stable@vger.kernel.org Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://patch.msgid.link/20261027083840.28813-4-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> 
hdac_hda_codec_remove() disables runtime PM of the codec device before
releasing the HDA link and calling the codec driver's remove callback,
so that no runtime resume can race with the teardown and access the
link or the already freed codec driver data.
snd_hda_codec_cleanup_for_unbind() then disables runtime PM again.

On the next probe snd_hda_codec_register() enables runtime PM only once,
leaving it disabled after the ASoC card or the codec driver has been
unbound and bound again: the codec stays powered and runtime_status
reads 'unsupported'.

Re-enable runtime PM at the end of hdac_hda_codec_remove() to balance
its own disable. The codec remains disabled until it is registered again.

Fixes: c3ec8ac82105 ("ASoC: hdac_hda: fix memleak on module unload")
Cc: stable@vger.kernel.org
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://patch.msgid.link/20261027083840.28813-4-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: hdac_hda: Restore the codec device type on unbind 2026-10-27T11:53:12+00:00 Peter Ujfalusi peter.ujfalusi@linux.intel.com 2026-10-27T08:38:39+00:00 08a44bde821d4fe22c0b6e63da04b3c59ac243a8 hdac_hda_codec_probe() changes the type of the codec device to HDA_DEV_ASOC to prevent snd_hda_codec_unregister() from unregistering the device when the card is freed. The type is never restored, and hda_bus_match() requires the device and driver types to be equal. After unbinding the codec driver via sysfs the HDA_DEV_LEGACY codec drivers no longer match the device and bind fails with -ENODEV. Restore HDA_DEV_LEGACY with a device managed action registered before the component, so it runs after the component and the card are gone. Fixes: b60ee2e281b6 ("ASoC: hdac_hda: overwrite hdev type to HDA_DEV_ASOC") Cc: stable@vger.kernel.org Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://patch.msgid.link/20261027083840.28813-3-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> 
hdac_hda_codec_probe() changes the type of the codec device to
HDA_DEV_ASOC to prevent snd_hda_codec_unregister() from unregistering
the device when the card is freed.

The type is never restored, and hda_bus_match() requires the device and
driver types to be equal. After unbinding the codec driver via sysfs the
HDA_DEV_LEGACY codec drivers no longer match the device and bind fails
with -ENODEV.

Restore HDA_DEV_LEGACY with a device managed action registered before
the component, so it runs after the component and the card are gone.

Fixes: b60ee2e281b6 ("ASoC: hdac_hda: overwrite hdev type to HDA_DEV_ASOC")
Cc: stable@vger.kernel.org
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://patch.msgid.link/20261027083840.28813-3-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: hdac_hda: Allocate the private data at codec driver bind 2026-10-27T11:53:12+00:00 Peter Ujfalusi peter.ujfalusi@linux.intel.com 2026-10-27T08:38:38+00:00 7d357eae973a1e1933f81f55aa1b3b4e3064570f The SOF HDA code allocates struct hdac_hda_priv and stores it as drvdata of the codec device before the codec driver is bound. The driver core clears drvdata on unbind, so after unbinding the codec driver via sysfs a subsequent bind dereferences a NULL pointer in hdac_hda_dev_probe(). All information in the private data can be derived from the codec device itself, so allocate and set it up in hdac_hda_dev_probe() as a device managed resource, which is released on unbind. Remove the allocation from SOF, keeping the check for iDisp codecs without display audio component. Fixes: 3fd63658caed ("ASoC: Intel: Drop hdac_ext usage for codec device creation") Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://patch.msgid.link/20261027083840.28813-2-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org> 
The SOF HDA code allocates struct hdac_hda_priv and stores it as drvdata
of the codec device before the codec driver is bound. The driver core
clears drvdata on unbind, so after unbinding the codec driver via sysfs
a subsequent bind dereferences a NULL pointer in hdac_hda_dev_probe().

All information in the private data can be derived from the codec
device itself, so allocate and set it up in hdac_hda_dev_probe() as a
device managed resource, which is released on unbind.

Remove the allocation from SOF, keeping the check for iDisp codecs
without display audio component.

Fixes: 3fd63658caed ("ASoC: Intel: Drop hdac_ext usage for codec device creation")
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://patch.msgid.link/20261027083840.28813-2-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: codecs: es*: controls that write each other's bits 2026-10-26T20:19:23+00:00 Mark Brown broonie@kernel.org 2026-10-26T20:19:23+00:00 60aa5abdc9997f0f4d3ab766d8a4af79ce8e400e Zhang Yi <zhangyi@everest-semi.com> says: Both the Kcontrols for the ES8326 and ES8375 have an issue where they write to other bits. This patch will fix that issue. Link: https://patch.msgid.link/20261026054837.13618-1-zhangyi@everest-semi.com 
Zhang Yi <zhangyi@everest-semi.com> says:

Both the Kcontrols for the ES8326 and ES8375 have an issue where they write to other bits.
This patch will fix that issue.

Link: https://patch.msgid.link/20261026054837.13618-1-zhangyi@everest-semi.com
ASoC: codecs: es*: Modify the maximum value of DMIC_GAIN 2026-10-26T20:19:23+00:00 Zhang Yi zhangyi@everest-semi.com 2026-10-26T05:48:37+00:00 29b867d8351de378704775d3733bfb589c48ab1f The Kcontrol related to DMIC_GAIN affects bits[3:2]. However, the old code sets ES8375_DMIC_GAIN_MAX to 4, which causes the Kcontrol to affect bits [4:2],this is incorrect. Therefore, change ES8375_DMIC_GAIN_MAX to 3. Fixes: de2b3119f9f7 ("ASoC: codecs: add support for ES8375") Reviewed-by: HyeongJun An <sammiee5311@gmail.com> Signed-off-by: Zhang Yi <zhangyi@everest-semi.com> Link: https://patch.msgid.link/20261026054837.13618-3-zhangyi@everest-semi.com Signed-off-by: Mark Brown <broonie@kernel.org> 
The Kcontrol related to DMIC_GAIN affects bits[3:2].
However, the old code sets ES8375_DMIC_GAIN_MAX to 4, which causes the Kcontrol to
affect bits [4:2],this is incorrect. Therefore, change ES8375_DMIC_GAIN_MAX to 3.

Fixes: de2b3119f9f7 ("ASoC: codecs: add support for ES8375")
Reviewed-by: HyeongJun An <sammiee5311@gmail.com>
Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
Link: https://patch.msgid.link/20261026054837.13618-3-zhangyi@everest-semi.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: codecs: es*: Modify the registers related to ALC RECOVERY 2026-10-26T20:19:23+00:00 Zhang Yi zhangyi@everest-semi.com 2026-10-26T05:48:36+00:00 dd64271fe1b935252cf454a72a558cbdb7be57c5 The register related to "ALC Capture Recovery Level" is ES8326_ALC_RECOVERY, not ES8326_ALC_LEVEL. Fixes: 5c439937775d ("ASoC: codecs: add support for ES8326") Reviewed-by: HyeongJun An <sammiee5311@gmail.com> Signed-off-by: Zhang Yi <zhangyi@everest-semi.com> Link: https://patch.msgid.link/20261026054837.13618-2-zhangyi@everest-semi.com Signed-off-by: Mark Brown <broonie@kernel.org> 
The register related to "ALC Capture Recovery Level" is ES8326_ALC_RECOVERY,
not ES8326_ALC_LEVEL.

Fixes: 5c439937775d ("ASoC: codecs: add support for ES8326")
Reviewed-by: HyeongJun An <sammiee5311@gmail.com>
Signed-off-by: Zhang Yi <zhangyi@everest-semi.com>
Link: https://patch.msgid.link/20261026054837.13618-2-zhangyi@everest-semi.com
Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: codecs: wsa881x: Fix shared powerdown GPIO handling 2026-10-26T20:17:45+00:00 Marek Szyprowski m.szyprowski@samsung.com 2026-10-22T09:09:22+00:00 c80f86ed857e21dcdd389559d3e62bf53ff39f9a On Qualcomm RB5 both WSA8815 speaker amplifiers share a single powerdown GPIO (TLMM 130), which is handled by the shared GPIO proxy. Since commit b30973e8c392 ("gpio: shared: make the voting mechanism adaptable") the proxy treats the value set by the first user on request as the default and lets any single user vote for the opposite value. The driver requests the GPIO de-asserted (GPIOD_OUT_LOW), so powering the amplifier up became the default and asserting powerdown became the winning vote. Once both amplifiers are runtime suspended, resuming them deadlocks: the first one releases its powerdown vote and waits for SoundWire enumeration, but the line stays asserted, because the second amplifier is still suspended and votes for powerdown. The wait times out and playback fails: wsa881x-codec sdw:1:0:0217:2110:00:3: Initialization not complete wsa881x-codec sdw:1:0:0217:2110:00:3: ASoC error (-110): at snd_soc_pcm_component_pm_runtime_get() on sdw:1:0:0217:2110:00:3 Request the GPIO asserted, so that powerdown is the default state and the line is de-asserted as long as any amplifier needs to be powered, then de-assert it to power up the amplifier as before. Also use gpiod_set_value_cansleep() instead of gpiod_direction_output() in the runtime PM callbacks. The line is already configured as an output and the proxy resets its default value when the direction is set while it has only a single user, which can happen if the other amplifiers have not been probed yet. Fixes: bfa336cee332 ("ASoC: wsa881x: Move custom workaround to gpiolib-of") Fixes: b30973e8c392 ("gpio: shared: make the voting mechanism adaptable") Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com> Link: https://patch.msgid.link/20261022090923.769464-1-m.szyprowski@samsung.com Signed-off-by: Mark Brown <broonie@kernel.org> 
On Qualcomm RB5 both WSA8815 speaker amplifiers share a single powerdown
GPIO (TLMM 130), which is handled by the shared GPIO proxy. Since commit
b30973e8c392 ("gpio: shared: make the voting mechanism adaptable") the
proxy treats the value set by the first user on request as the default
and lets any single user vote for the opposite value.

The driver requests the GPIO de-asserted (GPIOD_OUT_LOW), so powering the
amplifier up became the default and asserting powerdown became the
winning vote. Once both amplifiers are runtime suspended, resuming them
deadlocks: the first one releases its powerdown vote and waits for
SoundWire enumeration, but the line stays asserted, because the second
amplifier is still suspended and votes for powerdown. The wait times out
and playback fails:

  wsa881x-codec sdw:1:0:0217:2110:00:3: Initialization not complete
  wsa881x-codec sdw:1:0:0217:2110:00:3: ASoC error (-110): at snd_soc_pcm_component_pm_runtime_get() on sdw:1:0:0217:2110:00:3

Request the GPIO asserted, so that powerdown is the default state and
the line is de-asserted as long as any amplifier needs to be powered,
then de-assert it to power up the amplifier as before.

Also use gpiod_set_value_cansleep() instead of gpiod_direction_output()
in the runtime PM callbacks. The line is already configured as an output
and the proxy resets its default value when the direction is set while
it has only a single user, which can happen if the other amplifiers have
not been probed yet.

Fixes: bfa336cee332 ("ASoC: wsa881x: Move custom workaround to gpiolib-of")
Fixes: b30973e8c392 ("gpio: shared: make the voting mechanism adaptable")
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Link: https://patch.msgid.link/20261022090923.769464-1-m.szyprowski@samsung.com
Signed-off-by: Mark Brown <broonie@kernel.org>