ASoC: Intel: avs: Driver core and PCM operations

Merge series from Cezary Rojewski <cezary.rojewski@intel.com>:

Part three of main AVS driver series. This series was originally part of
the initial series which was later divided [1] into smaller,
easier-to-review chunks. Thus, many patches found here were already
present on the list.

This series consists of code typical to many drivers - PCI driver
operations, trace ability, PM operations - as well as PCM handlers for
all standard audio interfaces, that is, HDA, I2S and DMIC are found
here.

Series starts with updating firmware boot flow - libraries are no longer
ignored. This change is dependent on already merged topology code [2]
and because of that could not be part of the initial series [1].

PCM operations are split into four changes. First component operations
alone i.e. operations which are usually agnostic towards path position
(FE/BE). Then it continues with "generic" FE operations - there is no
interface split here as from Intel ADSP point of view, FE, or HOST side
as it's called in the specs, involves HD-Audio operations only.
BE (also known as LINK) side on the other hand is divided into
"non-HD-Audio" and HD-Audio part. The former represents transfer over
DMIC and I2S interfaces both.

While patches implementing standard PCI driver operations along (again
standard) HD-Audio initialization routines followed up by power
management handlers are two major ones, series covers also other
important subjects such as:

While patches implementing standard PCI driver operations along (again
standard) HD-Audio initialization routines followed up by power
management handlers are two major ones, series covers also other
important subjects such as:

- event tracing
- preparation for firmware tracing (debugability)
- coredump (debugability)
- recovery flow (attempt recovery after IPC timeout or exception)
- D0ix (D0 device substate, complements standard power management)

Series is finalized by actual addition of supported platforms: SKL and
APL-based. Platform-specific files are limited to firmware-specific
bits, that is, bits that are specific to given firmware generation.
Everything else is shared and is part of already upstream messaging
code found in ipc.c, messages.c and messages.h files.

Changes in v3:
- addressed (hopefully) trace-code compilation under .configs with
  CONFIG_FTRACE dropped

Changes in v2:
- usage of avs_releast_last_firmware() dropped in error path for library
  loading procedure as suggested by Pierre
- 'link_mask' usage replaced with 'i2s_link_mask' as requested by
  Pierre. Existing code addressed with new patch:
  "ASoC: Intel: avs: Replace link_mask usage with i2s_link_mask"
- fixed possible race during recovery flow (->recovering flag is now
  atomic and tested in single location only).
- dropped platform prefixes for basically all i2s board descriptors
- 'ssp_test' renamed to 'i2s_test' to match naming convention of other
  boards
- simplified PM implementation for current series, 'low_power' bits
  moved to future series
- replaced SND_INTEL_DSP_DRIVER_SST with _AVS as suggested by Mark.
  Required changes to intel-dspcfg will be added in future series
- number of typos across commit messages addressed

[1]: https://lore.kernel.org/all/20220311153544.136854-1-cezary.rojewski@intel.com/
[2]: https://lore.kernel.org/all/20220331135246.993089-1-cezary.rojewski@intel.com/

Cezary Rojewski (14):
  ASoC: Intel: avs: Account for libraries when booting basefw
  ASoC: Intel: avs: Generic soc component driver
  ASoC: Intel: avs: Generic PCM FE operations
  ASoC: Intel: avs: non-HDA PCM BE operations
  ASoC: Intel: avs: HDA PCM BE operations
  ASoC: Intel: avs: Coredump and recovery flow
  ASoC: Intel: avs: Prepare for firmware tracing
  ASoC: Intel: avs: D0ix power state support
  ASoC: Intel: avs: Event tracing
  ASoC: Intel: avs: Machine board registration
  ASoC: Intel: avs: PCI driver implementation
  ASoC: Intel: avs: Power management
  ASoC: Intel: avs: SKL-based platforms support
  ASoC: Intel: avs: APL-based platforms support

Piotr Maziarz (1):
  ASoC: Intel: avs: Replace link_mask usage with i2s_link_mask

 include/sound/intel-dsp-config.h      |    3 +-
 include/sound/soc-acpi.h              |    2 +
 sound/soc/intel/Kconfig               |    2 +
 sound/soc/intel/avs/Makefile          |    7 +-
 sound/soc/intel/avs/apl.c             |  250 ++++++
 sound/soc/intel/avs/avs.h             |   79 ++
 sound/soc/intel/avs/board_selection.c |  501 +++++++++++
 sound/soc/intel/avs/core.c            |  631 +++++++++++++
 sound/soc/intel/avs/dsp.c             |   27 +-
 sound/soc/intel/avs/ipc.c             |  253 +++++-
 sound/soc/intel/avs/loader.c          |   84 ++
 sound/soc/intel/avs/messages.c        |   35 +-
 sound/soc/intel/avs/messages.h        |   51 ++
 sound/soc/intel/avs/pcm.c             | 1182 +++++++++++++++++++++++++
 sound/soc/intel/avs/registers.h       |    8 +
 sound/soc/intel/avs/skl.c             |  125 +++
 sound/soc/intel/avs/topology.c        |   14 +-
 sound/soc/intel/avs/trace.c           |   33 +
 sound/soc/intel/avs/trace.h           |  154 ++++
 sound/soc/intel/avs/utils.c           |   23 +
 20 files changed, 3443 insertions(+), 21 deletions(-)
 create mode 100644 sound/soc/intel/avs/apl.c
 create mode 100644 sound/soc/intel/avs/board_selection.c
 create mode 100644 sound/soc/intel/avs/pcm.c
 create mode 100644 sound/soc/intel/avs/skl.c
 create mode 100644 sound/soc/intel/avs/trace.c
 create mode 100644 sound/soc/intel/avs/trace.h

--
2.25.1

ASoC: ux500: mop500: Check before clk_put() not needed

clk_put() already checks the clk ptr using !clk and IS_ERR()
so there is no need to check it again before calling it.

Signed-off-by: Yihao Han <hanyihao@vivo.com>
Link: https://lore.kernel.org/r/20220517033050.5191-1-hanyihao@vivo.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ipc-msg-injector: Cap the rmaining to count in IPC4 mode

If user space provides smaller buffer than the IPC4 reply then it is
possible that we corrupt user space memory since the IPC4 dfs_read function
is not using the count directly in copy_to_user() due to the nature of
an IPC4 message.

Cap the remaining counter to make sure that we are not writing too much to
the user space provided buffer.

Add a check also to make sure that the buffer is at least the size of the
IPC4 header.

Fixes: 066c67624d8c: "ASoC: SOF: ipc-msg-injector: Add support for IPC4 messages"
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220516092442.17027-1-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: APL-based platforms support

Define handlers specific to cAVS 1.5+ platforms, that is, APL and
similar platforms. These differ from SKL-alike ones in terms of AudioDSP
firmware generation and thus the '+' suffix. Introduciton of IMR,
removal of CLDMA, D0IX support and monolithic-ation of library/module
code are most impactful but are not the only changes brought with this
newer generation. Some generic and 1.5 operations are being re-used to
reduce code size.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-16-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: SKL-based platforms support

Define handlers specific to cAVS 1.5 platforms, that is SKL, KBL, AML
and all other variants based on this very version of AudioDSP
architecture. Most are specific to SKL-alike platforms with only
skl_log_buffer_offset() being exposed and used later by younger
equivalents.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-15-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Power management

To preserve power during sleep operations, handle suspend (S3),
hibernation (S4) and runtime (RTD3) transitions. As flow for all of
is shared, define common handlers to reduce code size.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-14-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: PCI driver implementation

HD-Audio bus is a PCI device. Add all functions necessary to probe such
device along with its removal sequence. Behaviour implemented for all
standard operations is similar to existing solutions: sound/pci/hda and
sound/soc/intel/skylake.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-13-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Machine board registration

AVS driver operates with granular audio card division in mind.
Super-card approach (e.g.: I2S, DMIC and HDA DAIs combined) is
deprecated in favour of individual cards - one per each device. This
provides necessary dynamism, especially for configurations with number
of codecs present and makes it easier to survive auxiliary devices
failures - one card failing to probe does not prevent others from
succeeding.

All boards spawned by AVS are unregistered on ->remove(). This includes
dummy codecs such as DMIC.

As all machine boards found in sound/soc/intel/boards are irreversibly
tied to 'super-card' approach, new boards are going to be introduced.
This temporarily increases number of boards available under /intel
directory until skylake-driver becomes deprecated and removed.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-12-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Replace link_mask usage with i2s_link_mask

'link_mask' field is intended for SoundWire, I2S should use
'i2s_link_mask' instead.

Signed-off-by: Piotr Maziarz <piotrx.maziarz@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-11-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Event tracing

Define tracing macros for easy avs debug. These cover all IPC message
types: requests, replies and notifications as well as DSP-core
operations and d0ix toggling.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-10-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: D0ix power state support

Audio DSP device supports D0 substates in form of D0ix, allowing for
preserving more power even when device is still considered active (D0).
When entered, certain domains which are not being currently used become
power gated. Entering and leaving D0ix is a complex process and differs
between firmware generations.

Conditions that disallow D0i3 and require immediate D0i0 transition
include but may not be limited to: IPC traffic, firmware tracing and
SRAM I/O. To make D0ix toggling sane, delay D0i3 transition and refresh
the timer each time an IPC is requested.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-9-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Prepare for firmware tracing

Firmware provides its own debug functionality. While coredump is one of
these, traces are the main area of interest. kfifo is enlisted to cache
log data that is being pumped to driver through SRAM. Separate DSP
operations are declared as actual feature implementation differs between
firmware generations.

As log gathering involves usage of IPCs, add all necessary: ENABLE_LOGS
and SYSTEM_TIME.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-8-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Coredump and recovery flow

In rare occasions, under stress conditions or hardware malfunction, DSP
firmware may fail. Software is notified about such situation with
EXCEPTION_CAUGHT notification. IPC timeout is also counted as critical
device failure. More often than not, driver can recover from such
situations by performing full reset: killing and restarting ADSP.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-7-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: HDA PCM BE operations

HDA streaming in DSP world means enlisting HDAudio links as BE
interfaces. Another difference when compared to its DMIC and I2S friends
is lack of NHLT blob usage - no additional hardware configuration is
needed.

Similarly to I2S component, HDA populates its DAIs dynamically, here by
the means of codec->pcm_list_head. Allows for cutting the number of soc
components required to support the interface.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-6-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: non-HDA PCM BE operations

DMIC and I2S interfaces differ in DMA operations from the HDAudio
interface. With that in mind, implement all DAI operations to handle
non-HDA BE interfaces.

To prevent code duplication in newly added code, I2S platform
registering is dynamic - makes use of specified port_mask and TDMs
array to populate as many DAIs as required.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-5-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Generic PCM FE operations

Each stream in AVS is represented by FE and BE domain. FE path stands
for HOST part of the stream while BE stands for LINK (hardware) one.
While BE portion is interface specific, FE is not. Handle all standard
DAI operations to implement FE part of the stream.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-4-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Generic soc component driver

Prepare for concrete PCM operations over HDA, DMIC and I2S interfaces by
providing generic soc component implementation. Interface-specific
components re-use this code as majority of flow is shared.

Signed-off-by: Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com>
Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-3-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: avs: Account for libraries when booting basefw

Not all modules are part of base firmware. Some are part of loadable
libraries. These need to be loaded after base firmware reports ready
status through FW_READY notification.

Their loading process is similar to the base firmware's one. Request the
binary file, verify and strip the manifest and load the actual code into
DSP memory with help of CLDMA or HD-Audio render stream, depending on
audio device generation.

List of libraries needed for loading is obtained through the topology -
vendor sections specifying the name of firmware files to request.

Signed-off-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220516101116.190192-2-cezary.rojewski@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Add support for i.MX8MM, MP, ULP

Merge series from Shengjiu Wang <shengjiu.wang@nxp.com>:

ASoC: fsl_sai: Add support for i.MX8MM, MP, ULP platforms

firmware: mtk: add adsp ipc protocol for SOF

Merge series from Tinghan Shen <tinghan.shen@mediatek.com>:

This patch provides mediatek adsp ipc support for SOF.
ADSP IPC protocol offers (send/recv) interfaces using
mediatek-mailbox APIs.

This patch was tested and confirmed to work with SOF fw on
MT8195 cherry board and MT8186 krabby board.

changes since v8:
- fix patchset 2 and 3.
  move "depends on MTK_ADSP_IPC" from SND_SOC_SOF_MTK_COMMON
  to SND_SOC_SOF_MT8195/MT8186 to prevent generating wrong
  config.

changes since v7:
- rebase to linux-next/next-22020504
- use EXPORT_SYMBOL_GPL instead of EXPORT_SYMBOL in mtk-adsp-ipc.c
- move mtk-adsp-ipc.c out from driver/firmware/mediatek
- add user of mtk-adsp-ipc.h in patchset 2 and 3.

changes since v6:
- rebase to matthias.bgg/linux.git, v5.18-next/soc
- Prefer "GPL" over "GPL v2" for MODULE_LICENSE

changes since v5:
- fix WARNING: modpost: missing MODULE_LICENSE() in drivers/mailbox
  /mtk-adsp-mailbox.o. Add MODULE_LICENSE in the last line.
- Due to WARNING: Missing or malformed SPDX-License-Identifier tag
  in line 1 in checkpatch, we don't remove SPDX-License in line 1.

changes since v4:
- add error message for wrong mbox chan

changes since v3:
- rebase on v5.16-rc8
- update reviewers

changes since v2:
- add out tag for two memory free phases

changes since v1:
- add comments for mtk_adsp_ipc_send and mtk_adsp_ipc_recv
- remove useless MODULE_LICENSE
- change label name to out_free

Allen-KH Cheng (1):
  ASoC: SOF: mediatek: Add ipc support for mt8195

TingHan Shen (1):
  firmware: mediatek: add adsp ipc protocol interface

Tinghan Shen (1):
  ASoC: SOF: mediatek: Add mt8186 ipc support

 drivers/firmware/Kconfig                      |   9 +
 drivers/firmware/Makefile                     |   1 +
 drivers/firmware/mtk-adsp-ipc.c               | 157 ++++++++++++++++++
 .../linux/firmware/mediatek/mtk-adsp-ipc.h    |  65 ++++++++
 sound/soc/sof/mediatek/Kconfig                |   2 +
 sound/soc/sof/mediatek/adsp_helper.h          |  12 +-
 sound/soc/sof/mediatek/mt8186/mt8186-loader.c |   5 +
 sound/soc/sof/mediatek/mt8186/mt8186.c        | 141 ++++++++++++++++
 sound/soc/sof/mediatek/mt8195/mt8195.c        | 138 ++++++++++++++-
 9 files changed, 519 insertions(+), 11 deletions(-)
 create mode 100644 drivers/firmware/mtk-adsp-ipc.c
 create mode 100644 include/linux/firmware/mediatek/mtk-adsp-ipc.h

--
2.18.0

ASoC: rt5645: Fix errorenous cleanup order

There is a logic error when removing rt5645 device as the function
rt5645_i2c_remove() first cancel the &rt5645->jack_detect_work and
delete the &rt5645->btn_check_timer latter. However, since the timer
handler rt5645_btn_check_callback() will re-queue the jack_detect_work,
this cleanup order is buggy.

That is, once the del_timer_sync in rt5645_i2c_remove is concurrently
run with the rt5645_btn_check_callback, the canceled jack_detect_work
will be rescheduled again, leading to possible use-after-free.

This patch fix the issue by placing the del_timer_sync function before
the cancel_delayed_work_sync.

Signed-off-by: Lin Ma <linma@zju.edu.cn>
Link: https://lore.kernel.org/r/20220516092035.28283-1-linma@zju.edu.cn
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: soc-component: Update handling to component delays

Currently the checking for if a component sits on the CPU or CODEC side
of the DAI link is done with a helper function that checks if the
component defines legacy_dai_naming. However, there are already a couple
of CPU side components that explicitly opt in to non-legacy DAI naming
and it doesn't seem like a very robust solution. Rather than looking for
the flag check if the component is attached to any of the CODEC DAIs on
the DAI link. This is more robust and helps to bring the core further in
the direction of a component being a generic block rather than being
classified as platform or CODEC drivers.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220513090532.1450944-3-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: core: Pass legacy_dai_naming flag directly

Passing the result of the helper function snd_soc_component_is_codec
to snd_soc_register_dai is less clear than just passing the DAI
naming flag directly. snd_soc_register_dai wants to know if it
should use the legacy DAI naming. The CODEC distinction is more
of a historical thing and not obviously directly related, and there
are already a couple of CPU side components that explicitly opt in
to non-legacy DAI naming.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220513090532.1450944-2-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: core: Correct spelling fliped -> flipped

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220513090532.1450944-1-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: ti: j721e-evm: Fix refcount leak in j721e_soc_probe_*

of_parse_phandle() returns a node pointer with refcount
incremented, we should use of_node_put() on it when not needed anymore.
Add missing of_node_put() to avoid refcount leak.

Fixes: 6748d0559059 ("ASoC: ti: Add custom machine driver for j721e EVM (CPB and IVI)")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220512111331.44774-1-linmq006@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: amd: vangogh: Remove duplicate include files

Remove duplicated includes.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20220514023806.34768-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: mediatek: Add mt8186 ipc support

mt8186 DSP uses two hardware mailbox IP to communicate with AP.
One mailbox is used for requests coming from AP, and the other
one is for requests from DSP.

Signed-off-by: Allen-KH Cheng <Allen-KH.Cheng@mediatek.com>
Signed-off-by: Tinghan Shen <tinghan.shen@mediatek.com>
Link: https://lore.kernel.org/r/20220512082215.3018-4-tinghan.shen@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: mediatek: Add ipc support for mt8195

This patch adds mt8195 IPC support by using mailbox.

On mt8195 resource, there are two mboxes used to handle ipc request
and reply. We create a mtk-adsp-ipc client device to request mbox
controllers.

Signed-off-by: Allen-KH Cheng <Allen-KH.Cheng@mediatek.com>
Reported-by: kernel test robot <lkp@intel.com>
Link: https://lore.kernel.org/r/20220512082215.3018-3-tinghan.shen@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

firmware: mediatek: Add adsp ipc protocol interface

Some of mediatek processors contain
the Tensilica HiFix DSP for audio processing.

The communication between Host CPU and DSP firmware is
taking place using a shared memory area for message passing.

ADSP IPC protocol offers (send/recv) interfaces using
mediatek-mailbox APIs.

We use two mbox channels to implement a request-reply protocol.

Signed-off-by: Allen-KH Cheng <allen-kh.cheng@mediatek.com>
Signed-off-by: TingHan Shen <tinghan.shen@mediatek.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Curtis Malainey <cujomalainey@chromium.org>
Reviewed-by: Tzung-Bi Shih <tzungbi@google.com>
Reviewed-by: YC Hung <yc.hung@mediatek.com>
Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com>
Link: https://lore.kernel.org/r/20220512082215.3018-2-tinghan.shen@mediatek.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Add support for i.MX8ULP

Add i.MX8ULP specific soc data, the max register is FSL_SAI_RTCAP
the IP version is also 0x0301, So version can't be used for the
condition of register FSL_SAI_MCTL setting.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/1652688372-10274-4-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Add support for i.MX8M Plus

Add i.MX8M Plus specific soc data, the max register is FSL_SAI_MDIV.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/1652688372-10274-3-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_sai: Add support for i.MX8MM

On i.MX8MM the max register is FSL_SAI_MCTL, which is
different with previous platform, so add max_register in
soc data to distinguish platforms.
And add specific soc data for i.MX8MM

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/1652688372-10274-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Add IPC4 FW loader support

Merge series from Ranjani Sridharan <ranjani.sridharan@linux.intel.com>:

The patches in this series add support for FW loading for IPC4 in the SOF
driver.

ASoC: max98396: Fix build error for implicit function declaration

sound/soc/codecs/max98396.c: In function ‘max98396_i2c_probe’:
sound/soc/codecs/max98396.c:1555:25: error: implicit declaration of function ‘devm_gpiod_get_optional’; did you mean ‘devm_regulator_get_optional’? [-Werror=implicit-function-declaration]
  max98396->reset_gpio = devm_gpiod_get_optional(&i2c->dev,
                         ^~~~~~~~~~~~~~~~~~~~~~~
                         devm_regulator_get_optional
sound/soc/codecs/max98396.c:1556:23: error: ‘GPIOD_OUT_HIGH’ undeclared (first use in this function); did you mean ‘GPIOF_INIT_HIGH’?
              "reset", GPIOD_OUT_HIGH);
                       ^~~~~~~~~~~~~~
                       GPIOF_INIT_HIGH
sound/soc/codecs/max98396.c:1556:23: note: each undeclared identifier is reported only once for each function it appears in
sound/soc/codecs/max98396.c:1565:3: error: implicit declaration of function ‘gpiod_set_value_cansleep’; did you mean ‘gpio_set_value_cansleep’? [-Werror=implicit-function-declaration]
   gpiod_set_value_cansleep(max98396->reset_gpio, 0);
   ^~~~~~~~~~~~~~~~~~~~~~~~
   gpio_set_value_cansleep
cc1: all warnings being treated as errors

Include header file <linux/gpio/consumer.h>

Fixes: b58581136770 ("ASoC: max98396: add amplifier driver")
Signed-off-by: Hui Tang <tanghui20@huawei.com>
Link: https://lore.kernel.org/r/20220512074640.75550-2-tanghui20@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: amd: add missing platform_device_unregister in acp_pci_rn_probe

acp_pci_rn_probe misses a call platform_device_unregister in error path,
this patch fixes that.

Signed-off-by: Zheng Bin <zhengbin13@huawei.com>
Link: https://lore.kernel.org/r/20220512013728.4128903-1-zhengbin13@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: samsung: Fix refcount leak in aries_audio_probe

of_parse_phandle() returns a node pointer with refcount
incremented, we should use of_node_put() on it when done.
If extcon_find_edev_by_node() fails, it doesn't call of_node_put()
Calling of_node_put() after extcon_find_edev_by_node() to fix this.

Fixes: 7a3a7671fa6c ("ASoC: samsung: Add driver for Aries boards")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20220512043828.496-1-linmq006@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: HDA: Set up sof_ipc4_fw_data for IPC4

Allocate the sof_ipc4_fw_data struct for IPC4 and set the fw header offset
for the platforms which will be used by the core when loading the firmware
image.

The core expects that the "private" field in struct snd_sof_dev (which is
unused today with IPC3) is used to save this data.

Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220511171648.1622993-6-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: IPC4: Add FW loader ops

Define and add the FW loader ops for IPC4. Also, introduce a new
structure, struct sof_ipc4_private_data that will be used to define some
IPC4-sepcific data.

Co-developed-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Co-developed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Co-developed-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220511171648.1622993-5-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Add header for IPC4 manifest

Add the header for the IPC4 manifest.

Co-developed-by: Rander Wang <rander.wang@intel.com>
Signed-off-by: Rander Wang <rander.wang@intel.com>
Co-developed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@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: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220511171648.1622993-4-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Add IPC4 private header

Add a struct sof_ipc4_fw_data to hold the firmware module data and
manifest FW header offset.
The FW reports data about the modules supported by the base FW in its
manifest and the FW header offset is platform dependent information.

This structure will be allocated when the ops are initialized for each
platform and populated when the FW is loaded.

Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220511171648.1622993-3-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: Intel: HDA: Set IPC4-specific DSP ops for CaVS platforms

Add implementation of low level, platform dependent IPC4 message handling
and set the DSP ops for IPC4 for APL, CNL and TGL platforms.

Co-developed-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220511171648.1622993-2-ranjani.sridharan@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASOC: Fix the error handling code of the probe

Merge series from Zheyu Ma <zheyuma97@gmail.com>:

These drivers mishandle the regulator resource in the probe function,
failing to disable the regulator for probing failure.

ASoC: wm9090: Remove redundant endianness flag

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device has no DAI links and as such the flag would have
no effect, remove the redundant flag.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220510153843.1029540-2-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm2000: Remove redundant endianness flag

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device has no DAI links and as such the flag would have
no effect, remove the redundant flag.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220510153843.1029540-1-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mxs-saif: Fix refcount leak in mxs_saif_probe

of_parse_phandle() returns a node pointer with refcount
incremented, we should use of_node_put() on it when done.

Fixes: 08641c7c74dd ("ASoC: mxs: add device tree support for mxs-saif")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220511133725.39039-1-linmq006@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42l56: Fix the error handling of cs42l56_i2c_probe()

The driver should goto label 'err_enable' when failing at regmap_read().

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220511015514.1777923-1-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: imx-hdmi: Fix refcount leak in imx_hdmi_probe

of_find_device_by_node() takes reference, we should use put_device()
to release it. when devm_kzalloc() fails, it doesn't have a
put_device(), it will cause refcount leak.
Add missing put_device() to fix this.

Fixes: 6a5f850aa83a ("ASoC: fsl: Add imx-hdmi machine driver")
Fixes: f670b274f7f6 ("ASoC: imx-hdmi: add put_device() after of_find_device_by_node()")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220511052740.46903-1-linmq006@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_ssi: Add support multi fifo script

With dual fifo enabled, the case that recording mono sound
in the background, playback mono sound twice in parallal,
at second time playback sound may distort, the possible
reason is using dual fifo to playback mono sound is not
recommended.

This patch is to provide a option to use multi fifo script,
which can be dynamically configured as one fifo or two fifo
mode.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1652183808-3745-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl: Fix refcount leak in imx_sgtl5000_probe

of_find_i2c_device_by_node() takes a reference,
In error paths, we should call put_device() to drop
the reference to aviod refount leak.

Fixes: 81e8e4926167 ("ASoC: fsl: add sgtl5000 clock support for imx-sgtl5000")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Fabio Estevam <festevam@gmail.com>
Link: https://lore.kernel.org/r/20220511065803.3957-1-linmq006@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs42l46: Fix the error handling of cs42l56_i2c_probe()

The driver should goto label 'err_enable' when failing at regmap_read().

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-2-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8903: Fix the error handling of wm8903_i2c_probe()

The driver should goto label 'err' when failing to request the irq.

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-7-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas6424: Fix the error handling of tas6424_i2c_probe()

After enabling the regulator, The driver should disable the regulator
when failing at probing.

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-6-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tas571x: Fix the error handling of tas571x_i2c_probe()

After enabling the regulator, The driver should disable the regulator
when failing at probing.

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-5-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt5645: Fix the error handling of rt5645_i2c_probe()

After enabling the regulator, The driver should disable the regulator
when failing at probing.

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-4-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: cs35l36: Fix the error handling of cs35l36_i2c_probe()

The driver should goto label 'err' when failing at regmap_read().

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220510153251.1741210-3-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: boards: Chromebook configuration updates

Merge series from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:

Two minor changes to enable DMIC and capture for CS35L41, and one new
configuration for AlderLake hardware.

ASoC: dt-bindings: fsl,micfil: Add compatible string for imx8mp

Add compatible string for imx8mp, which has micfil module

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1652087663-1908-2-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: fsl_micfil: Add support for i.MX8MPlus

On i.MX8Plus there are two updates for micfil module.

One is that the output format is S32_LE, only the 24 more
significative bits have information, the other bits are always
zero. Add 'formats' variable in soc data to distinguish the
format on different platform.
Another is that the fifo depth is 32 entries.

Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
Link: https://lore.kernel.org/r/1652087663-1908-1-git-send-email-shengjiu.wang@nxp.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ipc: introduce cont_update_posn in sof_ipc_stream_params struct

The host stream position is updated when no_stream_position is set as 0.
However current implementation updates host stream position only when
report data is larger than or equal to host period size which is decided
by the period size of host side. It maybe cause host stream position
update not in time. Therefore this patch introduces another field
"cont_update_posn", a boolean value aimed to update host stream position
continuously and based on period size of pipeline. It can get better
precise when need to update host stream position from firmware.

Reviewed-by: Daniel Baluta <daniel.baluta@nxp.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Signed-off-by: YC Hung <yc.hung@mediatek.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220509170425.54640-1-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: samsung: spdif: remove unnecessary check of mem_res

The resource is checked in probe function, so there is
no need do this check in remove function.

Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Link: https://lore.kernel.org/r/20220510124749.2663874-1-yangyingliang@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_ssp_amp: enable capture stream from cs35l41

Enable capture stream of the cs35l41 dai link to support feedback
stream from amplifier.

Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Brent Lu <brent.lu@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220509170922.54868-4-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_ssp_amp: fix no DMIC BE Link on Chromebooks

The SOF topology supports 2 BE Links(dmic01 and dmic16k) and each
link supports up to four DMICs. However, Chromebook does not implement
ACPI NHLT table so the mach->mach_params.dmic_num is always zero. We
add a quirk so machine driver knows it's running on a Chromebook and
need to create BE Links for DMIC.

Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Signed-off-by: Brent Lu <brent.lu@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220509170922.54868-3-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: Intel: sof_rt5682: Add support for adl_rt1019_rt5682s

This patch adds the driver data for two rt1019 speaker amplifiers on
SSP1 and rt5682s on SSP0 for ADL platform

Reviewed-by: Brent Lu <brent.lu@intel.com>
Signed-off-by: Vamshi Krishna <vamshi.krishna.gopal@intel.com>
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220509170922.54868-2-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: tlv320adcx140: Register a callback to disable the regulator_disable

The driver should register a callback that will deal with the disabling
when it fails to probe.

Signed-off-by: Zheyu Ma <zheyuma97@gmail.com>
Link: https://lore.kernel.org/r/20220510053031.1685337-1-zheyuma97@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: codecs: rt715-sdca: remove useless assignment of ops

The ops are already part of the 'struct sdw_driver', it's unclear why
this was copied into the 'slave' structure - no other driver does so.

Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220509185729.59884-1-pierre-louis.bossart@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Clean up usage of the endianness flag

Merge series from Charles Keepax <ckeepax@opensource.cirrus.com>:

Before componentisation any part registered as a CODEC would have
automatically supported both little and big endian, ie. the core
would duplicate any supported LE or BE PCM format to support the other
endian as well. As componentisation removed the distinction between
CODEC drivers and platform drivers, a flag was added to specify
if this behaviour is required for a particular component. However,
as most systems tend to use little endian the absence of the flag
is rarely noticed. Also the naming of the flag "endianness" is a
little unobvious as to if it should be applied to a particular
component.

This series adds a comment to better explain the meaning of the
flag and then tidys up the usage of the flag. A couple of uses
of the flag are removed where is has been used inappropriately
on the CPU side of the DAI link, this is clearly not valid in the
cases it has been used, and I suspect never would be valid. Then
some redundant formats are removed, since they would be covered by
existing endianness flags. And finally a bunch of devices that are
missing the flag have it added.

It is worth noting that since componenisation there are now a couple
of cases where it is not entire clear to me that the flag should
be applied to all CODECs as it was before. In those cases I haven't
updated the driver to add the flag and they are outlined here:

1) Build into the AP CODECs, these are actual silicon inside the main
processor and they typically receive audio directly from an internal
bus. It is not obvious to me that these can happily ignore endian. On
the CODEC side these include: jz4725b.c, jz4760.c, jz4770.c,
rk3328_codec.c, lpass-va-macro.c, lpass-rx-macro.c, lpass-tx-macro.c,
lpass-wsa-macro.c. There are also some examples of this scattered
around the various platform support directories in sound/soc.

2) Devices behind non-audio buses, SPI just moves bits and doesn't
really define an endian for audio data on the bus. Thus it seems the
CODEC probably can care about the endian. The only devices that fall
into this group (mostly for AoV) are: rt5514-spi.c, rt5677-spi.c,
cros_ec_codec.c (only the AoV).

3) CODECs with no DAIs, these could specify the flag and plenty of
them do; CODECs from the initial conversion to componentisation. But
the flag makes no difference here since there is nothing for it to
apply to. This includes purely analogue CODECs: aw8738.c, ssm2305.c,
tpa6130a2.c, tda7419.c, max9759.c, max9768.c, max9877.c, lm4857.c,
simple-mux.c, simple-amplifier.c. And devices that only do jack
detection: ts3a227e.c, mt6359-accdet.c.

If there are any opinions on adding the flag to any of those three
groups they would be greatfully received. But I am leaning towards
leaving 1,2 without endianness flags since it feels inappropriate,
and removing the endian flag from devices in catagory 3 that already
have it. Assuming no one objects to that I will do a follow up
series for that.

ASoC: SOF: sof-client: Update for different IPC versions

Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:

Hi,

The current IPC client infrastructure can only be used with IPC3.
This series carries updates for the core side of the client support to handle
IPC4 messages and updates the ipc message injector to be usable with IPC4.

The IPC flood test is only supported by SOF_IPC (IPC3), we are not going to
create the aux device for it at all if the firmware is using IPC4.

Regards,
Peter
---
Peter Ujfalusi (8):
  ASoC: SOF: sof-client: Add API to get the maximum IPC payload size
  ASoC: SOF: ipc-msg-injector: Query the maximum IPC payload size
  ASoC: SOF: sof-client-probes: Query the maximum IPC payload size
  ASoC: SOF: sof-client: Add API to get the ipc_type
  ASoC: SOF: sof-client: Add support IPC4 message sending
  ASoC: SOF: ipc-msg-injector: Separate the message sending
  ASoC: SOF: ipc-msg-injector: Add support for IPC4 messages
  ASoC: SOF: sof-client: IPC flood test can only work with SOF_IPC

 sound/soc/sof/sof-client-ipc-msg-injector.c | 181 ++++++++++++++++++--
 sound/soc/sof/sof-client-probes.c           |   5 +-
 sound/soc/sof/sof-client.c                  |  66 ++++++-
 sound/soc/sof/sof-client.h                  |   2 +
 4 files changed, 227 insertions(+), 27 deletions(-)

--
2.36.0

ASoC: dt-bindings: mediatek: mt8192: Drop i2s-share properties

This reverts commit e056cf4341 ("ASoC: dt-bindings: mediatek: mt8192: Add
i2s-share properties") which was merged while the property name and
definition were still being discussed. Revert the commit for now and a
follow up commit will re-add the property after it is further discussed
and reviewed.

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Link: https://lore.kernel.org/r/20220509185625.580811-1-nfraprado@collabora.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: trace: The dtrace is only available with SOF_IPC

Currently the dtrace only supported with SOF_IPC.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220506130229.23354-1-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sam9g20_wm8731: Use dev_err_probe() for snd_soc_register_card()

Log the error code when snd_soc_regster_card() fails, but fold in the
silencing of deferred probe errors.

Signed-off-by: Mark Brown <broonie@kernel.org>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20220506130349.451452-1-broonie@kernel.org

ASoC: mediatek: mt8195: Fix build warning without CONFIG_OF

sound/soc/mediatek/mt8195/mt8195-mt6359.c:1639:32: warning: ‘mt8195_mt6359_max98390_rt5682_card’ defined but not used [-Wunused-variable]
 1639 | static struct mt8195_card_data mt8195_mt6359_max98390_rt5682_card = {
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sound/soc/mediatek/mt8195/mt8195-mt6359.c:1634:32: warning: ‘mt8195_mt6359_rt1011_rt5682_card’ defined but not used [-Wunused-variable]
 1634 | static struct mt8195_card_data mt8195_mt6359_rt1011_rt5682_card = {
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
sound/soc/mediatek/mt8195/mt8195-mt6359.c:1629:32: warning: ‘mt8195_mt6359_rt1019_rt5682_card’ defined but not used [-Wunused-variable]
 1629 | static struct mt8195_card_data mt8195_mt6359_rt1019_rt5682_card = {
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Since all users of this driver do need CONFIG_OF anyway, there is no
need to save a few bytes on kernel builds while CONFIG_OF disabled, so
just remove the #ifdef to fix this warning.

Fixes: 86a6b9c9dfff ("ASoC: mediatek: mt8195: add machine support for max98390 and rt5682")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20220509120918.9000-1-yuehaibing@huawei.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wm8940: add devicetree support

This adds devicetree support to the wm8940 codec driver.
With a DT-based kernel, there is no board-specific setting
to select the driver so allow it to be manually chosen.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220509121055.31103-1-lukma@denx.de
Signed-off-by: Mark Brown <broonie@kernel.org>

doc: dts: Add device tree information regarding wm8940 codec

This commit provides documentation entry for wm8940 codec.

Signed-off-by: Lukasz Majewski <lukma@denx.de>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220509121313.31216-1-lukma@denx.de
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: sof-client: IPC flood test can only work with SOF_IPC

Currently the ipc flood test is only supported with SOF_IPC.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Rander Wang <rander.wang@intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-9-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ipc-msg-injector: Add support for IPC4 messages

The IPC message representation of an IPC4 differs from the IPC3 version
significantly.

The message for IPC4 should be written to the debugfs file in this form:
0-7 IPC4 header (2x u32)
8-  additional payload, if any

The reply is given back in the same form.

The message size limitation is the same as with the IPC3, only messages
which can fit to the mailbox can be injected (and received).

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-8-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ipc-msg-injector: Separate the message sending

Move out the code for sending the IPC message into a separate helper
function in preparation for support for handling IPC4 communication.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-7-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: sof-client: Add support IPC4 message sending

In order to be able to send an IPC4 message, the
sof_client_ipc_tx_message() needs to parse the tx message differently to
extract the size.

The IPC notification registration is done by providing the notification
type and the whole message is passed to the client when a match is found.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-6-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: sof-client: Add API to get the ipc_type

Provide a way for the client drivers to query the ipc_type used by the
firmware.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-5-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: sof-client-probes: Query the maximum IPC payload size

Instead of using the SOF_IPC_MSG_MAX_SIZE as the maximum payload size for
and IPC message, use the provided API to query it.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-4-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: ipc-msg-injector: Query the maximum IPC payload size

Instead of using the SOF_IPC_MSG_MAX_SIZE as the maximum payload size for
and IPC message, use the provided API to query it.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-3-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: SOF: sof-client: Add API to get the maximum IPC payload size

Provide a way for the client drivers to query the maximum payload size of
an IPC message.
Currently clients do not have access to this information and they can only
use the SOF_IPC_MSG_MAX_SIZE defined value.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20220506132647.18690-2-peter.ujfalusi@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>

Add missing dt-binding properties for audio components on mt8192-asurada

Merge series from Nícolas F. R. A. Prado <nfraprado@collabora.com>:

These patches add properties that were missing on the dt-bindings of the
audio components used by mt8192-asurada. Namely the i2s-share
properties for the sound platform and the #sound-dai-cells on the
rt1015p and rt5682 codecs when they're referenced by the machine sound
node.

Nícolas F. R. A. Prado (3):
  ASoC: dt-bindings: mediatek: mt8192: Add i2s-share properties
  ASoC: dt-bindings: rt1015p: Add #sound-dai-cells
  ASoC: dt-bindings: rt5682: Add #sound-dai-cells

 Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml  | 5 +++++
 Documentation/devicetree/bindings/sound/realtek,rt1015p.yaml | 3 +++
 Documentation/devicetree/bindings/sound/rt5682.txt           | 2 ++
 3 files changed, 10 insertions(+)

--
2.36.0

ASoC: pcm186x: simplify the return expression of pcm186x_power_off()

Simplify the return expression.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220505022102.54650-1-chi.minghao@zte.com.cn
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: mediatek: simplify the return expression of mtk_dai_pcm_prepare()

Simplify the return expression.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220505021808.54337-1-chi.minghao@zte.com.cn
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: uniphier: simplify the return expression of uniphier_aio_compr_set_params()

Simplify the return expression.

Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Minghao Chi <chi.minghao@zte.com.cn>
Link: https://lore.kernel.org/r/20220505021733.54275-1-chi.minghao@zte.com.cn
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: rt5682: Add #sound-dai-cells

The rt5682 codec can be pointed to through a sound-dai property to be
used as part of a machine sound driver. dtc expects #sound-dai-cells to
be defined in the codec's node in those cases, so add it in the
dt-binding and set it to 0.

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Link: https://lore.kernel.org/r/20220429203039.2207848-4-nfraprado@collabora.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: rt1015p: Add #sound-dai-cells

The rt1015p codec can be pointed to through a sound-dai property to be
used as part of a machine sound driver. dtc expects #sound-dai-cells to
be defined in the codec's node in those cases, so add it in the
dt-binding and set it to 0.

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Link: https://lore.kernel.org/r/20220429203039.2207848-3-nfraprado@collabora.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: dt-bindings: mediatek: mt8192: Add i2s-share properties

The Mediatek AFE PCM controller for MT8192 allows sharing of an I2S bus
between two busses. Add a pattern for these properties in the
dt-binding.

Signed-off-by: Nícolas F. R. A. Prado <nfraprado@collabora.com>
Link: https://lore.kernel.org/r/20220429203039.2207848-2-nfraprado@collabora.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: sdw-mockup: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-39-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wsa881x: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-38-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wcd938x: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-37-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1316-sdw: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-36-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt1308-sdw: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-35-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt715-sdca: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-34-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt715: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-33-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt711-sdca: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-32-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt711: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-31-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: rt700: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SoundWire DAI and as such
should have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-30-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wcd9335: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SLIMbus DAI and as such should
have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-29-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>

ASoC: wcd934x: Add endianness flag in snd_soc_component_driver

The endianness flag is used on the CODEC side to specify an
ambivalence to endian, typically because it is lost over the hardware
link. This device receives audio over a SLIMbus DAI and as such should
have endianness applied.

Signed-off-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220504170905.332415-28-ckeepax@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>