Merge series from Srinivas Kandagatla <srinivas.kandagatla@oss.qualcomm.com>:
Recent static calibration support patches have excersised new code path
exposing some missing checks and also accessing some variables which are
of LE type that are accessed directly without converting it to CPU type.
Thanks to Dan Carpenter and kernel test robot for reporting this.
We have 2 similar functions, both converts date from snd_kcontrol to
snd_soc_component.
(A) snd_soc_kcontrol_component()
(B) snd_soc_dapm_kcontrol_component()
(A) is just wrapper for snd_kcontrol_chip().
(B) is for more complex conversion.
Having similar functions is confusable. So (A) will be replaced to
original snd_kcontrol_chip(). (B) will be stay, but the function
name should be xx_to_xx().
And it is defined at soc-component.h. It should be implemented at
soc-dapm.c.
This patch renames it to snd_soc_dapm_kcontrol_to_component(), and
move to soc-dapm.c.
This patch keeps compatible by using define, but old name will be replaced
on each drivers and removed from ASoC in the future.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://patch.msgid.link/871poyy53x.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
This series improves the IPC4 probes implementation by improving the
report quality and adds support for querying the available probes
on the firmware side.
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
Up to this point we had a spoken rule what sof-client can access and what
they must not (they cannot use sdev directly), recently there were attempts
to break this rule as clients could access sdev via the sof client dev.
The series will re-arrange the client device struct to 'hide' the sdev (and
the list) from client drivers to prevent abuse.
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
SDW codecs use the global HDaudio WAKEEN/STS to detect wakes since LNL. But
the wakeen is handled in the SDW driver. We should filter the bus->codec_mask
to only include HDA and IDISP codecs to avoid clearing the wakeen status
before it is handled.
Add SOF_TKN_COMP_SCHED_DOMAIN and connect it to struct snd_sof_widget
comp_domain member, with new get_token_comp_domain() function.
The logic is such that if the topology attribute is not present in the
widget node the corresponding IPC4 extension value is taken from the
module's manifest like before. But if the attribute is found and
recognized its value overrides what is there in the manifest.
Signed-off-by: Jyri Sarha <jyri.sarha@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Message-ID: <20250829151101.27327-1-peter.ujfalusi@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Add another debugfs file, "probe_points_available", that shows all the
available probe points in the SOF FW at the time of query. The probe
points are there only when an active SOF stream exists in the
system. However, the stream identifiers are persistent in the sense
that the same probe point identifiers always appear with the same
playback or capture command in the same system configuration.
The output, when reading "probe_points_available", may look like this:
0x1000005,0x0,0x100 host-copier.0.playback output buf idx 0 (connected)
0x7,0x0,0x100 gain.1.1 input buf idx 0 (connected)
0x1000007,0x0,0x0 gain.1.1 output buf idx 0
0x3,0x0,0x0 mixin.1.1 input buf idx 0
0x1000003,0x0,0x0 mixin.1.1 output buf idx 0
0x4,0x0,0x0 mixout.2.1 input buf idx 0
0x1000004,0x0,0x0 mixout.2.1 output buf idx 0
0x10007,0x0,0x0 gain.2.1 input buf idx 0
0x1010007,0x0,0x0 gain.2.1 output buf idx 0
0x11,0x0,0x0 smart_amp.2.1 input buf idx 0
0x1000011,0x0,0x0 smart_amp.2.1 output buf idx 0
0x10005,0x0,0x0 dai-copier.SSP.NoCodec-0.playback input buf idx 0
The triplet at the beginning of a line can be copy-pasted as such to
"probe_points" debugfs file for adding a probe point. The rest of the
line tries to give human readable explanation of what this probe point
is.
Signed-off-by: Jyri Sarha <jyri.sarha@linux.intel.com>
Reviewed-by: Liam Girdwood <liam.r.girdwood@intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Message-ID: <20250829093022.32094-6-peter.ujfalusi@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The current output of three integers is not very human readable. Use
ipc4 functions to describe in more detail what the struct
sof_probe_point_desc buffer_id is actually referring to in an ipc4 SOF
system.
Before this commit the "probe_points" debugfs file could read as:
Id: 0x01000004 Purpose: 0 Node id: 0x100
Id: 0x00000006 Purpose: 0 Node id: 0x100
And after in the same situation in an ipc4 system it reads:
0x7,0x0,0x100 gain.1.1 input buf idx 0 (connected)
0x1000005,0x0,0x100 host-copier.0.playback output buf idx 0 (connected)
The triplet in the beginning of the line can be used to reinserted the
probe point again by writing it into "probe_points" debugfs file, if
its first removed by writing the fist number in "probe_points_remove".
The last number is ignored when creating a probe point.
Signed-off-by: Jyri Sarha <jyri.sarha@linux.intel.com>
Reviewed-by: Liam Girdwood <liam.r.girdwood@intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Message-ID: <20250829093022.32094-5-peter.ujfalusi@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
This patch enables use of 8-bit unsigned, A-law, and mu-law
sample formats with IPC4 SOF.
The ipc4-topology.h is updated with IPC4 sample types. The
purpose is to convert ALSA types to IPC4 types.
The functions of_ipc4_update_hw_params(), sof_ipc4_get_valid_bits(),
and new function sof_ipc4_get_sample_type() are updated to handle
the sample type conversions.
The function sof_ipc4_prepare_copier_module() is updated to set
the DMA SCS bit for all non 32 bits sample types.
The change to function sof_ipc4_get_valid_bits() returns 8 bits
for these ALSA formats.
The change to function sof_ipc4_prepare_copier_module() is
needed to handle properly all non 32-bit formats with SCS
bit set.
To support playback with new 8 bits types, the
sof_ipc4_init_input_audio_fmt() function is updated to get
the sample type and use it in search for copier input pin
format.
To support capture, the sof_ipc4_init_output_audio_fmt()
is updated similarly. Since the function uses separate
out_ref_type argument, instead of single parameters struct,
the out_ref_type needs to be added to every user of the
function. Therefore functions sof_ipc4_prepare_copier_module(),
sof_ipc4_prepare_gain_module(), sof_ipc4_prepare_mixer_module(),
sof_ipc4_prepare_src_module(), and sof_ipc4_prepare_process_module()
are updated to set the out_ref_type.
Signed-off-by: Seppo Ingalsuo <seppo.ingalsuo@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Liam Girdwood <liam.r.girdwood@intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Message-ID: <20250829105305.31818-3-peter.ujfalusi@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
SDW codecs use the global HDaudio WAKEEN/STS to detect wakes since LNL.
But the wake event is handled in the SDW driver. We should only clear
WAKESTS for the HDA and IDISP codecs that was detected when HDA DSP
probe. The SoundWire codec will be included in the codec_mask if we read
the STATESTS register when a SoundWire codec wake event happens.
The commit avoid updating bus->codec_mask in resume to not clear WAKESTS
of SoundWire codecs.
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Message-ID: <20250829082908.2693-3-peter.ujfalusi@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>:
Set of patches all connected to the machine-board registration process
and the board_selection.c file. Apart from LOC reduction, allows for
multiple cards of the same 'type' e.g.: avs_rt5640, to be present
simultanetously without any workarounds or tricks. The last patch
improves debugability - helps us (the devs) deal with basic I2S-boards
issues out there without forcing the user to constantly switch between
debug-friendly and non-debug options.
--
Initially I wanted to have the first two patches (new implementation and
the follow up cleanup) to be just one patch. However, the git diff, even
with --patience does not look good, at least in my opinion. Decided to
split the changes into two separate patches to make it easy to review
and comment.
As the first two patches carry heaviest load, the summary:
- provide a unified avs_register_board() wrapper for
platform_device_register_data(PLATFORM_DEVID_AUTO). The _AUTO is what
makes every device name unique and prevents any naming conflicts to
occur when matching components <-> cards in runtime
- provide avs_register_board_pdata(), a wrapper one-level above
avs_register_board() that takes all custom parameters in its argument
list so that the initialization of 'struct avs_mach_pdata' does not
need to be repeated in every function like its done today
- make use of the two above to write simpler registration functions:
1. init any interface-type-specific values e.g.: SSP port for I2S
board
2. call avs_register_board_pdata()
3. register the ASoC component with avs_register_xxx_component()
In regard to the last three patches - two are simplifications, no
functional changes just LOC reduction and readability.
The last one is to improve debug-ability, especially when working in the
open field - often we, the devs, find it useful when debugging
production systems to run i2s_test (the debug board) without impacting
the 'real' board e.g.: avs_rt5640. By default i2s_test is empty so a
typical user and their scenarios are not impacted.
Merge series from Qianfeng Rong <rongqianfeng@vivo.com>:
The 'ret' variable usually is used to store returns from some functions,
which return either zero on success or negative error codes on failure.
Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but it's ugly as pants. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.
Change "ret" from u32/unsigned int to int type. No effect on runtime.
Merge series from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:
Current Renesas MSIOF get unknown error when first used.
This patch-set will fixup this issue.
The fwnode.h is not supposed to be used by the drivers as it
has the definitions for the core parts for different device
property provider implementations. Drop it.
Note, that fwnode API for drivers is provided in property.h
which is included here.
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Message-ID: <20250826132734.278226-1-andriy.shevchenko@linux.intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Change the 'ret' variable from unsigned int to int to store negative error
codes or zero returned by regmap_field_write().
Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but it's ugly as pants. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.
No effect on runtime.
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Message-ID: <20250828075406.386208-4-rongqianfeng@vivo.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Change the 'ret' variable from unsigned int to int to store negative error
codes or zero returned by regmap_update_bits().
Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but it's ugly as pants. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.
No effect on runtime.
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Reviewed-by: Nuno Sá <nuno.sa@analog.com>
Message-ID: <20250828075406.386208-3-rongqianfeng@vivo.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Change the 'ret' variable from u32 to int to store negative error codes
directly or returned by other functions.
Storing the negative error codes in unsigned type, doesn't cause an issue
at runtime but it's ugly as pants. Additionally, assigning negative error
codes to unsigned type may trigger a GCC warning when the -Wsign-conversion
flag is enabled.
No effect on runtime.
Signed-off-by: Qianfeng Rong <rongqianfeng@vivo.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Message-ID: <20250828075406.386208-2-rongqianfeng@vivo.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
The i2s_test card serves debug purpose and is not probed by default.
Currently i2s_test and non-i2s_test sound cards exclude each other. To
increase the test coverage, allow both board types to be probed
simultaneously and share the available SSP port pool.
As 'i2s_test' module parameter is empty by default and requires manual
input from user to activate, there is no impact for standard production
scenarios.
Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Message-ID: <20250827142229.869139-6-cezary.rojewski@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Switch to the new board registration functions and remove the
then-unused code.
The SSP-number-check is still important and cannot be just dropped. Move
it from avs_register_i2s_board(), which is being removed with this
patch, to avs_register_i2s_boards() to maintain sanity when registering
a board.
Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Message-ID: <20250827142229.869139-3-cezary.rojewski@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
To support multiple instances of the same card utilize
PLATFORM_DEVID_AUTO when registering new platform devices. This
adaptation is also an opportunity to streamline the devices
registration - avs_register_board().
All the new functions are equivalents of the existing ones apart from
adjusting platform_device_register_data() call with PLATFORM_DEVID_AUTO
and code size reduction given the presence of unified register function.
Follow up changes will complete the transition and cleanup the duplicate
members.
Reviewed-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Message-ID: <20250827142229.869139-2-cezary.rojewski@intel.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Using for_each_available_child_of_node_scoped() allows us to get rid of
of_node_put() calls from early returns or breaks in the loop. It also
fixes issues with missing of_node_put() calls.
Switch to for_each_available_child_of_node_scoped() in parse_dai_link_info().
Also drop the braces around if blocks now that the inner block is just
one statement.
Signed-off-by: Chen-Yu Tsai <wenst@chromium.org>
Reviewed-by: Fei Shao <fshao@chromium.org>
Message-ID: <20250825151111.3696404-1-wenst@chromium.org>
Signed-off-by: Mark Brown <broonie@kernel.org>
If a defer probe happens for the cpu device, the return value for
cpu_pdev "ret = -EINVAL" breaks defer probe. As the cpus->dai_name
can be replaced by cpus->of_node, to simplify the code, remove related
code for cpu_pdev.
Fixes: 6a5f850aa8 ("ASoC: fsl: Add imx-hdmi machine driver")
Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Message-ID: <20250820031641.2158261-1-shengjiu.wang@nxp.com>
Signed-off-by: Mark Brown <broonie@kernel.org>
Pull perf fix from Borislav Petkov:
- Fix a case where the events throttling logic operates on inactive
events
* tag 'perf_urgent_for_v6.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf: Avoid undefined behavior from stopping/starting inactive events
Pull x86 fixes from Borislav Petkov:
- Fix the GDS mitigation detection on some machines after the recent
attack vectors conversion
- Filter out the invalid machine reset reason value -1 when running as
a guest as in such cases the reason why the machine was rebooted does
not make a whole lot of sense
- Init the resource control machinery on Hygon hw in order to avoid a
division by zero and to actually enable the feature on hw which
supports it
* tag 'x86_urgent_for_v6.17_rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
x86/bugs: Fix GDS mitigation selecting when mitigation is off
x86/CPU/AMD: Ignore invalid reset reason value
x86/cpu/hygon: Add missing resctrl_cpu_detect() in bsp_init helper
Mark Brown
Srinivas Kandagatla
Kuninori Morimoto
Jyri Sarha
Peter Ujfalusi
Seppo Ingalsuo
Bagas Sanjaya
Bard Liao
Andy Shevchenko
Qianfeng Rong
Cezary Rojewski
Chen-Yu Tsai
Matthias Schiffer
Xichao Zhao
Shengjiu Wang
Linus Torvalds