| Age | Commit message (Collapse) | Author |
|---|
|
# Conflicts:
# net/ceph/osd_client.c
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git
|
|
# Conflicts:
# drivers/gpu/drm/xe/xe_pagefault.c
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound.git
|
|
express audio
The RME HDSPe AIO PCI express audio card has a new PCI vendor ID (1d18)
while keeping the same device ID (3fc6). The card seems working fine by
just adding the new ID.
While we're at it, use the standard macro to cleann up the existing PCI
ID entry, too.
Reported-by: AtmanActive <alsa.atmanactive.3dfqt@slmail.me>
Closes: https://lore.kernel.org/178674392532.7.10140952469564861620.1550442282@slmail.me
Tested-by: AtmanActive <alsa.atmanactive.3dfqt@slmail.me>
Link: https://patch.msgid.link/20260910144204.973359-1-tiwai@suse.de
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
|
|
Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> says:
This patch-set moves soc-dai related functions from soc-core.c to
soc-dai.c. We should have snd_soc_dai_xxx() functions on soc-dai.c.
No functional changes.
We have 2 type of function defines in include/sound/xx.h
A is for ASoC drivers
B is for ASoC framwork
I think include/sound/xx.h is only for drivers (A).
In other words, ASoC framework function (B) is not needed for drivers,
or *shouldn't* indicate to drivers.
This patch adds new sound/soc/soc-internal.h, and the functions
for ASoC framework will be moved to here.
Link: https://patch.msgid.link/8733vuu75t.wl-kuninori.morimoto.gx@renesas.com
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87ik4qssjl.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87jyp6ssjo.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
It renames function name
snd_soc_dai_tdm_mask_get() -> snd_soc_dai_stream_tdm_mask_get()
snd_soc_dai_tdm_mask_set() -> snd_soc_dai_stream_tdm_mask_set()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87ld9mssjt.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
These macros are very similar. Will be handled by same macro
snd_soc_dai_dma_data_get() -> snd_soc_dai_stream_dma_data_get()
snd_soc_dai_dma_data_set() -> snd_soc_dai_stream_dma_data_set()
snd_soc_dai_get_dma_data() -> snd_soc_dai_stream_dma_data_get()
snd_soc_dai_set_dma_data() -> snd_soc_dai_stream_dma_data_set()
This macro is not much helps code, will be removed
snd_soc_dai_init_dma_data()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87mru2ssjx.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
It renames function name
snd_soc_dai_get_widget() -> snd_soc_dai_stream_widget_get()
snd_soc_dai_set_widget() -> snd_soc_dai_stream_widget_set()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87o6eissk1.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI parameter will be capsuled soon, will be not enable to access from
soc-dai.h. Move it to soc-dai.c
It renames function name
snd_soc_dai_get_pcm_stream() snd_soc_dai_pcm_stream_get()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87pkyyssk4.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
soc_pcm_{apply/params/update}_symmetry() are DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-pcm.c. Move it to soc-dai.c
It renames function name
- soc_pcm_xxx_symmetry()
+ snd_soc_dai_symmetric_xxx()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87qzjessk7.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
soc_pcm_set_dai_params() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-pcm.c. Move it to soc-dai.c
It renames function name
- soc_pcm_set_dai_params()
+ snd_soc_dai_symmetric_set_params()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87se3usska.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
snd_soc_{un}register_dai() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-core.c. Move it to soc-dai.c
Note1
snd_soc_{un}register_dai() needs to use client_mutex which is soc-core.c
local. This patch removes static from it, and allow to use it from ASoC
internal.
It renames function name
- snd_soc_register_dai()
+ snd_soc_dai_register()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87tsoasske.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
snd_soc_is_matching_dai() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-core.c. Move it to soc-dai.c
It renames function name
- snd_soc_is_matching_dai()
+ snd_soc_dai_matches_dlc()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87v78qsskh.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
snd_soc_is_match_dai_args() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-core.c. Move it to soc-dai.c
It renames function name
- snd_soc_is_match_dai_args()
+ snd_soc_dai_matches_args()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87wlt6sskk.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
snd_soc_dai_name_get() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-core.c. Move it to soc-dai.c
It renames function name
- snd_soc_dai_name_get()
+ snd_soc_dai_name()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87y0dmssko.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
snd_soc_add_dai_controls() is DAI related function.
And its parameter will be capsuled soon, will be not enable to access from
soc-core.c. Move it to soc-dai.c
It renames function name
- snd_soc_add_dai_controls()
+ snd_soc_dai_add_controls()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/87zey2ssku.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
DAI/Component will be capsuled soon, but fmt_{single/multiple}_name()
will be used form soc-dai.c / soc-component.c.
Let's makes them to non static function.
ASoC functions / macros are defined at linux/include/sound/xxx.h (A).
ASoC framework and each vender drivers are including it.
Some functions / macros which are only for ASoC framework. It shouldn't
be used from vender drivers. But we can't foce it because sharing same
headers (A).
This patch adds new soc-internal.h (B) which is for ASoC framework
internal use. Since this patch, we can separate global functions (for
vender drivers) and local function (for ASoC framework internal use).
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/871pbeu75e.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
nau8822_suspend() disables all codec supplies and marks the register
cache dirty, but leaves regmap in live I/O mode. Resume then attempts
the cache replay without first leaving a corresponding cache-only state.
Enter cache-only mode after the regulators have been disabled
successfully, leave it after the supplies are restored, and stop the
local resume sequence if cache replay fails. Do not power the codec back
down on a resume error, because deferred component resume remains best
effort and continues after the callback.
regulator_bulk_disable() restores regulators already disabled by the
same bulk operation if a later disable fails. Only publish cache-only
state after the whole power-off operation succeeds, so the regmap state
continues to match the hardware state on either outcome.
Fixes: 1e3cb6c321be ("ASoC: nau8822: new codec driver")
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260830141828.34419-1-pengpeng@iscas.ac.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-linus
ASoC: Fixes for v7.3
An unusually large batch of fixes here, partly due to me not sending a
pull request last week and partly due to Linus' fixes for the Ericsson
systems. I'm fairly sure he's the only user there so it seemed low risk
to handle as fixes. We also have a batch of error handling fixes from
Cezary and issues with uninitialised stream_config->type from Shuming
both of which had user error reports attached. Otherwise everything is
a fairly normal combination of quirks and device specific fixes.
|
|
sta32x_set_bias_level() enables the codec supplies, restores the
coefficient and register state, and then starts the watchdog. Register
access errors in the mute and coefficient helpers and the cache replay
are discarded, allowing DAPM to publish STANDBY for a partially restored
codec.
Propagate every operation in the local restore transaction, including
the final standby-state write. Start the watchdog only after that write
succeeds. On failure, hold reset, mark the cache dirty, disable the
supplies and report a cleanup failure while preserving the primary
error. The DAPM bias callback consumes the returned error independently
of component resume handling.
Fixes: 54dc6cabe684 ("ASoC: sta32x: preserve coefficient RAM")
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260830141521.33246-1-pengpeng@iscas.ac.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Andrey Golovko <andrey.golovko@gmail.com> says:
The 1/2 of v2 is in for-7.3 as 0c7aeb0f5ece, so this is the rest of it,
rebased on that, with one more patch in front.
Mark asked whether marking the read-only Controls volatile works for
multi-byte reads. It does not, on its own. Twenty of the Controls that
patch marks volatile are wider than one byte: nineteen of four - the
Latency of FU23, FU26 and FU127, XU22 0x06, SAPU29 0x05, PPU21 and PPU26
0x06, the 0x08 Control of IT21, IT26, IT28, IT29, OT23, OT24, OT25, OT28
and OT127, MU26 0x06, MFPU21 and MFPU26 0x08 - and XU22 0x07 of two.
With val_bits = 8 the MBQ layer refuses each of them before anything
reaches the bus, so the patch on its own would turn a read that used to
be answered from the cache into -EINVAL. What the cache answered was
the placeholder from tas2783_reg_default[] rather than anything the
device had said, and the driver performs no register reads at all, so
nothing in tree can tell the difference today - but -EINVAL is not what
the patch should leave behind.
1/2 widens val_bits to 32 and repairs the one place that relied on the
old width, the calibration write, which used regmap_bulk_write() to
spread a four-byte value over four one-byte registers.
Both patches have been running on an ASUS ProArt PX13 (HN7306EAC, two
TAS2783 on one AMD ACP link) since 13 August - the 2/2 in that kernel is
the pre-v2 one, differing only in the File Download Status handling.
Playback, suspend to idle and the calibration path are unaffected.
Link: https://patch.msgid.link/20260815112000.4180-1-andrey.golovko@gmail.com
|
|
The read-only SDCA Controls are readings of device state, but the regmap
caches them and tas2783_reg_default[] gives each one a placeholder, so a
read is answered from the cache and never reaches the peripheral. The
answer is whatever the cache happens to hold: on an ASUS ProArt PX13 a
read of the PDE23 actual power state returns the 0x3 placeholder, i.e.
PS3, while the Function is powered and a cache-bypassing read of the
same Control over the bus answers 0x0. Once the cache has been given a
value it keeps it, so the Control reports a state the device left long
ago.
Mark those Controls volatile and drop their entries from the defaults.
regcache_read() and regcache_write() skip volatile registers, so a
read-only Control no longer enters the cache at all: reads reach the
peripheral, and a sync cannot attempt to write a reading back. That is
what regcache_sync() asks for in the first place - "any registers that
should not be synced should be marked as volatile".
The list is the same one the writeable_reg callback needs, so move it
into a helper both callbacks use. The File Download Status is not on
it, because the host writes that Control, but it is volatile all the
same: the device raises its own bits on its own schedule, and
sdca_fdl_process() reads the Control to see them. A cached answer would
hide the device half of the handshake.
Nothing in the driver reads any of these Controls - it performs no
register reads at all - so the change is visible only through the
regmap.
Signed-off-by: Andrey Golovko <andrey.golovko@gmail.com>
Link: https://patch.msgid.link/20260815112000.4180-3-andrey.golovko@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
tas2783_sdca_mbq_size() declares registers of one, two and four bytes,
but the regmap is created with val_bits = 8. The SoundWire MBQ bus
rejects any register whose declared size exceeds the configured value
width:
ctx->val_size = config->val_bits / BITS_PER_BYTE;
...
size = ctx->cfg.mbq_size(ctx->dev, reg);
if (!size || size > ctx->val_size)
return -EINVAL;
Both regmap_sdw_mbq_read() and regmap_sdw_mbq_write() bail out on that,
so every register the driver itself declares as two or four bytes wide
is unreachable through the regmap, in both directions, and the failure
happens before anything reaches the bus.
That is measurable. On an ASUS ProArt PX13 HN7306EAC with two TAS2783
aggregated on one link, a regmap read of XU22 Control 0x07, declared two
bytes, and of XU22 Control 0x12, declared four bytes, returns -EINVAL on
both amplifiers. With val_bits widened the same reads reach the
peripheral, which answers -ENODATA for both, powered or not - this part
appears not to implement them. So the change makes the registers
addressable; whether a given peripheral answers is up to the peripheral.
Widen val_bits to cover the largest size the callback returns. The
callback keeps deciding the size of each individual transfer, so
single-byte registers are unaffected.
The calibration path needs adjusting along with it.
regmap_bulk_write(..., buf, sizeof(u32)) relied on val_bits = 8 to
spread one calibration value over four consecutive single-byte
registers; with a four-byte value width the same call would ask for four
four-byte values and read past the end of the four-byte buffer. Write
the four registers explicitly instead, keeping the most-significant-byte
-first layout. Read back over the bus, bypassing the cache, the five
calibration values of both amplifiers are byte for byte what they were
before this patch.
What this does not do is repair regcache_sync(). On this machine a sync
aborts earlier, at a single-byte Control - the FU23 Mute of channel 0,
which tas2783_reg_default[] gives 0x1 while the driver's init sequence
writes 0x00, so the sync tries to restore it and the peripheral answers
-ENODATA - and no multi-byte register is reached at all.
Signed-off-by: Andrey Golovko <andrey.golovko@gmail.com>
Link: https://patch.msgid.link/20260815112000.4180-2-andrey.golovko@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Use devm_clk_get_enabled() to let the clock be enabled automatically
and cleaned up on unbind, dropping the manual clk_prepare_enable() and
clk_disable_unprepare() calls in probe and remove. As a result, move the
clk struct member into probe, which is the only place it's used.
Simplify the probe error paths by returning directly instead of
jumping to clk_err, and use devm_snd_soc_register_component() return
value directly. Also remove error message from ioremap as it throws one
itself. Same with devm_request_irq().
Assisted-by: opencode:deepseek-v4-flash-free
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Reviewed-by: Michal Simek <michal.simek@amd.com>
Reviewed-by: Vincenzo Frascino <vincenzo.frascino@arm.com>
Link: https://patch.msgid.link/20260824203024.15574-1-rosenp@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
sma1307_setting_loaded() checks a byte count against an element count, then
reads an eight-int header and a fixed default table from the firmware. It
also trusts the mode count while indexing a five-entry mode_set array.
Validate the byte-to-int conversion, fixed header/default extent, mode
count, and exact mode table layout before parsing the setting file.
Fixes: 576c57e6b4c1 ("ASoC: sma1307: Add driver for Iron Device SMA1307")
Signed-off-by: Pengpeng Hou <pengpeng@iscas.ac.cn>
Link: https://patch.msgid.link/20260830134428.9550-1-pengpeng@iscas.ac.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com> says:
Move all four LPASS codec macros (rx/tx/va/wsa) from REGCACHE_FLAT to
REGCACHE_MAPLE, and fix a latent v2.5 Compander1 addressing bug that
the flat cache had been absorbing.
REGCACHE_FLAT hides missing-default reads by returning the zero-
initialised cache slot instead of doing a bus read. The caller can't
tell that value apart from a real hardware value, so any un-defaulted,
readable, non-volatile register that gets read silently returns wrong
data. Since commit e062bdfdd6ad ("regmap: warn users about
uninitialized flat cache") regmap dev_warn_once()s the first such
miss, but any subsequent miss stays silent.
REGCACHE_MAPLE reports a cache miss as -ENOENT and the regmap core
falls back to a real bus read, so the caller always gets the true
hardware value. All four macros are switched in the same series
for consistency.
The switch surfaced a latent bug in wsa_macro_set_spkr_mode() on
v2.5+ silicon: two Compander1 register addresses moved on that
generation and the driver was writing to the v2.1 addresses on all
versions, taking a silent bus error. Patch 1/6 fixes that and is
tagged Cc: stable.
Series contents:
1/6: wsa-macro: use v2.5 Compander1 addresses on v2.5+ silicon
(fix, tagged for stable).
2/6: rx-macro + wsa-macro: sort concatenated reg_defaults before
regmap init so regmap does not sort a copy at runtime.
3/6: wsa-macro: switch cache to REGCACHE_MAPLE.
4/6, 5/6, 6/6: rx-macro / tx-macro / va-macro: cache switch,
same rationale as 3/6.
Tested on Glymur (SC8480XP CRD, WCD9378 SDCA compute mode + WSA8845
speakers): HPH playback, mic capture and WSA speaker playback all
work with the full series applied; the "using zero-initialized flat
cache" and "Unsorted reg_defaults" warnings both disappear from the
boot log.
Link: https://patch.msgid.link/20260908082806.648797-1-srinivas.kandagatla@oss.qualcomm.com
|
|
REGCACHE_FLAT hides missing-default reads by returning the zero-
initialised cache slot instead of doing a bus read. The caller
can't tell that value apart from a real hardware value, so any
un-defaulted, readable, non-volatile register that gets read
silently returns wrong data. Since commit e062bdfdd6ad ("regmap:
warn users about uninitialized flat cache") regmap dev_warn_once()s
the first such miss, but any subsequent miss stays silent.
Switch to REGCACHE_MAPLE. A cache miss returns -ENOENT and the
regmap core falls back to a real bus read, so the caller always
gets the true hardware value. Write-cache behaviour is unchanged.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-7-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
REGCACHE_FLAT hides missing-default reads by returning the zero-
initialised cache slot instead of doing a bus read. The caller
can't tell that value apart from a real hardware value, so any
un-defaulted, readable, non-volatile register that gets read
silently returns wrong data. Since commit e062bdfdd6ad ("regmap:
warn users about uninitialized flat cache") regmap dev_warn_once()s
the first such miss, but any subsequent miss stays silent.
Switch to REGCACHE_MAPLE. A cache miss returns -ENOENT and the
regmap core falls back to a real bus read, so the caller always
gets the true hardware value. Write-cache behaviour is unchanged.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-6-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
REGCACHE_FLAT hides missing-default reads by returning the zero-
initialised cache slot instead of doing a bus read. The caller
can't tell that value apart from a real hardware value, so any
un-defaulted, readable, non-volatile register that gets read
silently returns wrong data. Since commit e062bdfdd6ad ("regmap:
warn users about uninitialized flat cache") regmap dev_warn_once()s
the first such miss, but any subsequent miss stays silent.
Switch to REGCACHE_MAPLE. A cache miss returns -ENOENT and the
regmap core falls back to a real bus read, so the caller always
gets the true hardware value. Write-cache behaviour is unchanged.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-5-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
REGCACHE_FLAT hides missing-default reads by returning the zero-
initialised cache slot instead of doing a bus read. The caller
can't tell that value apart from a real hardware value, so any
un-defaulted, readable, non-volatile register that gets read
silently returns wrong data. Since commit e062bdfdd6ad ("regmap:
warn users about uninitialized flat cache") regmap dev_warn_once()s
the first such miss, but any subsequent miss stays silent.
Switch to REGCACHE_MAPLE. A cache miss returns -ENOENT and the
regmap core falls back to a real bus read, so the caller always
gets the true hardware value. Write-cache behaviour is unchanged.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-4-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Both drivers build the reg_defaults array by concatenating a base
table with a codec-version-specific one; the resulting array is not in
register-address order because the version-specific ranges overlap the
base range. Regmap detects this and prints
wsa_macro 6c90000.codec: Driver needs fixing: Unsorted reg_defaults,
sorting the copy
rx_macro 6a70000.codec: Driver needs fixing: Unsorted reg_defaults,
sorting the copy
then sorts its own copy at runtime. Call regcache_sort_defaults() on
the concatenated array before handing it to regmap so the warning goes
away and regmap does not have to sort a second time. Keeping the
base/version-specific split is deliberate for readability, so a static
reorder is not appropriate here.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-3-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
wsa_macro_set_spkr_mode() unconditionally writes CDC_WSA_COMPANDER1_CTL3
(0x5CC) and CDC_WSA_COMPANDER1_CTL7 (0x5DC). On v2.5+ silicon these
registers moved to CDC_2_5_WSA_COMPANDER1_CTL3 (0x5EC) and
CDC_2_5_WSA_COMPANDER1_CTL7 (0x5FC); the v2.1 addresses are dead on
that hardware, so the update_bits() calls take a bus error:
wsa_macro 6c90000.codec: ASoC error (-5): at
snd_soc_component_update_bits() on 6c90000.codec for
register: [0x000005dc]
The flat regcache had been hiding the read side of this by returning
the zero-initialised cache slot instead of doing a bus read, so the
error only becomes visible once the cache reports the miss honestly
and regmap falls back to a real bus read.
Select the correct address for Compander1 CTL3/CTL7 based on
codec_version. The Compander0 and Boost registers used in the same
function have the same address across versions, so they stay
unchanged. Also collapse the two mode branches to a single sequence
with mode-selected values, which is what the switch was doing anyway.
Fixes: 727de4fbc546 ("ASoC: codecs: lpass-wsa-macro: Correct support for newer v2.5 version")
Cc: stable@vger.kernel.org
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>
Link: https://patch.msgid.link/20260908082806.648797-2-srinivas.kandagatla@oss.qualcomm.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The "AMDI5682" ACPI HID is matched by two AMD ASoC machine drivers:
cz-da7219-max98357a (this driver, Carrizo/Stoney) and
acp3x-alc5682-max98357 (Raven/Picasso). cz-da7219-max98357a is linked
first and probes the platform device first; its DAI links reference the
Stoney ACP, which is absent on Raven/Picasso, so its card can never be
instantiated there.
This was harmless until commit 42d99857d6f0 ("ASoC: core: Move all users
to deferrable card binding"): devm_snd_soc_register_card() now returns 0
for a card left pending instead of propagating -EPROBE_DEFER, so
cz_probe() succeeds and permanently binds AMDI5682. acp3x-alc5682-max98357
never binds and the internal speakers and headphone jack get no card.
Detect Raven/Picasso (and later) by the ACP3.x audio coprocessor's
dedicated PCI function (1022:15e2); Carrizo/Stoney reach the ACP through
the GPU driver and have no such device. Return -ENODEV so the driver core
continues probing AMDI5682 with acp3x-alc5682-max98357.
Fixes: 42d99857d6f0 ("ASoC: core: Move all users to deferrable card binding")
Cc: stable@vger.kernel.org
Signed-off-by: Yorick Rommers <yorick-rommers@hotmail.com>
Tested-by: Yorick Rommers <yorick-rommers@hotmail.com>
Link: https://patch.msgid.link/20260907121228.13754-1-yorick-rommers@hotmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The put() callback of "IEC958 Playback Default" returns whether the
converted register value moved. The convert_from_spdif_status() helper
reads part of the first two channel status bytes and none of the last
two, while the get() callback returns all four. So a write that lands
only in the bits it does not read changes what userspace reads back and
reports no change. Of the 31 bits above the mode bit, 20 are such bits
in consumer mode and 29 in professional mode. The core notifies only on
a positive return.
Toggling status[2] bit 0 on an HDA HDMI codec moves the read-back from
04 00 00 00 to 04 00 01 00 with no event. Toggling the non-audio bit
in status[0] gives one.
Compare the stored status as well, the way the ac97 code does. The
write to the codec stays gated on the converted value.
Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: HyeongJun An <sammiee5311@gmail.com>
Assisted-by: Claude:claude-opus-5
Link: https://patch.msgid.link/20260908134153.1614273-1-sammiee5311@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
The hwdep mmap callback rejects read-buffer mappings that are initially
writable, but leaves VM_MAYWRITE set on mappings created with PROT_READ.
A process that can open the hwdep node O_RDWR can later use mprotect() to
make the mapping writable.
The read allocation begins with struct usb_stream. Its read_size member is
used by the fault handler to decide which pages belong to the read buffer.
The read VMA intentionally remains expandable because pcm_usb_stream uses
mremap() after reading that size. Changing read_size first can therefore
map and access pages beyond the allocation. The same member is also
consumed by usb_stream_free(), where changing it can make
free_pages_exact() release pages outside the allocation.
Clear VM_MAYWRITE for read-buffer mappings after rejecting an initially
writable VMA. This keeps the separate output-buffer mapping writable while
preventing later permission upgrades.
Fixes: 030a07e44129 ("ALSA: Add USB US122L driver")
Cc: stable@vger.kernel.org
Signed-off-by: Kazuki Hanai <hnkz.64@gmail.com>
Link: https://patch.msgid.link/20260908110053.2950767-1-hnkz.64@gmail.com
Signed-off-by: Takashi Iwai <tiwai@suse.de>
|
|
bui duc phuc <phucduc.bui@gmail.com> says:
Hi all,
This series fixes and cleans up error handling in the MT6797 ASoC driver.
Compile tested only.
Link: https://patch.msgid.link/20260821122840.41266-1-phucduc.bui@gmail.com
|
|
The errors handled here are already reported by the called functions,
either directly or deeper in the call chain. Therefore, the additional
dev_err() calls are redundant and can be removed.
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260821122840.41266-4-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Replace dev_err() with dev_err_probe() to prevent log spam when probe
returns -EPROBE_DEFER.
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260821122840.41266-3-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The error paths in mt6797_afe_enable_clock() use incorrect goto labels,
causing clocks that failed to enable to be disabled during cleanup.
Fix the goto labels to only unwind clocks that were successfully
enabled, and return the actual error code.
Fixes: c5e7fca928d3 ("ASoC: mt6797: add structure define and clock control function for 6797")
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260821122840.41266-2-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
bui duc phuc <phucduc.bui@gmail.com> says:
This series fixes several error handling issues in the mt8186 ASoC driver.
The patches mainly address failures during APLL/mux, clock, MCK, GPIO,
and regcache operations. In particular, they make sure that partial
failures are handled correctly and that resources are left in a
consistent state.
The series also removes redundant probe error messages where the errors
are already reported by the called functions.
The changes were found by manual code inspection and compile-tested only.
Link: https://patch.msgid.link/20260903051150.178057-1-phucduc.bui@gmail.com
|
|
The errors handled here are already reported by the called functions,
either directly or deeper in the call chain. Therefore, the additional
dev_err() calls are redundant and can be removed.
platform_get_irq() returns a negative value on error and does not
return zero. Therefore, only check for a negative error code and
return it directly.
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260903051150.178057-9-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Check the return value of regcache_sync() and restore cache-only mode
if synchronization fails.
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260903051150.178057-8-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
If the second mt8186_afe_gpio_select() call in adda_dl()/adda_ul() fails,
the first one's state is left applied, leaving CLK/DAT in an inconsistent
state. Roll back the first select on failure.
Also fix a copy-pasted "CLK ON" debug message that should say "CLK OFF"
in the _dl disable path.
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260903051150.178057-7-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The mt8186_mck_enable() function returns immediately when a clock
operation fails, leaving previously enabled clocks active.
Add error paths to disable previously enabled clocks before returning
the error.
Fixes: 55b423d5623c ("ASoC: mediatek: mt8186: support audio clock control in platform driver")
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260903051150.178057-6-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Currently, the mt8186_apll*_enable() functions call mux_setting(afe, true)
but do not check its return value to handle failures.
In addition, the cleanup paths of mt8186_apll*_enable() do not call
mux_setting(afe, false) when the enable operation fails, while the
mt8186_apll*_disable() functions do.
Add error handling for apll*_mux_setting() and call mux_setting(afe, false)
in the cleanup paths when mt8186_apll*_enable() fails.
Fixes: 55b423d5623c ("ASoC: mediatek: mt8186: support audio clock control in platform driver")
Signed-off-by: bui duc phuc <phucduc.bui@gmail.com>
Link: https://patch.msgid.link/20260903051150.178057-5-phucduc.bui@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
