- 27 Feb, 2024 1 commit
-
-
Richard Fitzgerald authored
Cast u8 values to u32 when using them to build a 32-bit unsigned value that is then stored in a u64. This avoids the possibility of a bad sign extension where the u8 is implicitly extended to an int, thus changing it from an unsigned to a signed value. Whether this is a real problem is debatable, but it does no harm to ensure that the u8 are cast to a suitable type for shifting. Signed-off-by:
Richard Fitzgerald <rf@opensource.cirrus.com> Fixes: e1830f66 ("ASoC: cs35l56: Add helper functions for amp calibration") Link: https://msgid.link/r/20240227100042.99-1-rf@opensource.cirrus.comSigned-off-by:
Mark Brown <broonie@kernel.org>
-
- 26 Feb, 2024 9 commits
-
-
Mark Brown authored
Merge series from Jerome Brunet <jbrunet@baylibre.com>: This are various fixes and clean up gathered while working on Amlogic audio support. These help better handle higher and unusual clock configuration for TDM, SPDIF or PDM.
-
Mark Brown authored
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>: The patchset may not cover all codecs found in the codecs/ directory - noticed a possible improvement and grepped for similar pattern across C files found in the directory. Those addressed here seem pretty straightforward. Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed.
-
Jerome Brunet authored
The rate of the stream does not matter for the fifos of the axg family. Fifos will just push or pull data to/from the DDR according to consumption or production of the downstream element, which is the DPCM backend. Drop the rate list and allow continuous rates. The lower and upper rate are set according what is known to work with the different backends This allows the PDM input backend to also use continuous rates. Signed-off-by:
Jerome Brunet <jbrunet@baylibre.com> Link: https://msgid.link/r/20240223175116.2005407-6-jbrunet@baylibre.comSigned-off-by:
-
Jerome Brunet authored
Use maximum width between 2 edges to setup spdifin thresholds and detect the input sample rate. This comes from Amlogic SDK and seems to be marginally more reliable than minimum width. This is done to align with a future eARC support. No issue was reported with minimum width so far, this is considered to be an update so no Fixes tag is set. Signed-off-by:
-
Jerome Brunet authored
ASoC stopped using CBS_CFS and CBM_CFM a few years ago but the traces in the amlogic tdm interface driver did not follow. Update this to match the new format names Signed-off-by:
-
Jerome Brunet authored
According to Amlogic datasheets for the SoCs supported by this driver, the maximum bit clock rate is 100MHz. The tdm interface allows the rates listed by the DAI driver, regardless of the number slots or their width. However, these will impact the bit clock rate. Hitting the 100MHz limit is very unlikely for most use cases but it is possible. For example with 32 slots / 32 bits wide, the maximum rate is no longer 384kHz but ~96kHz. Add the constraint accordingly if the component is not already active. If it is active, the rate is already constrained by the first stream rate. Fixes: d60e4f1e ("ASoC: meson: add tdm interface driver") Signed-off-by:
-
Jerome Brunet authored
By default, when mclk-fs is not provided, the tdm-interface driver requests an MCLK that is 4x the bit clock, SCLK. However there is no justification for this: * If the codec needs MCLK for its operation, mclk-fs is expected to be set according to the codec requirements. * If the codec does not need MCLK the minimum is 2 * SCLK, because this is minimum the divider between SCLK and MCLK can do. Multiplying by 4 may cause problems because the PLL limit may be reached sooner than it should, so use 2x instead. Fixes: d60e4f1e ("ASoC: meson: add tdm interface driver") Signed-off-by:
-
Krzysztof Kozlowski authored
Starting with Qualcomm SM8350 SoC, so Low Power Audio SubSystem (LPASS) block version v9.2, the register responsible for TX SMIC MUXn muxes is different. In earlier LPASS versions this mux had bit fields for analogue (ADCn) and digital (SWR_DMICn) MICs. Choice of ADCn was selecting the analogue path in CDC_TX_TOP_CSR_SWR_DMICn_CTL register. With LPASS v9.2 and newer, the bit fields are integrated into just SWR_MICn and there is no distinction for analogue or digital MIC in the register. Fix support for LPASS v9.2+: 1. Add new set of widgets and audio routes for LPASS v9.2. 2. Do not choose analogue or digital in CDC_TX_TOP_CSR_SWR_DMICn_CTL based on value of the mux. 3. Replace all the input widgets (TX SWR_ADCn, TX SWR_DMICn) with TX SWR_INPUTn ones. The change is not backwards compatible with older DTBs and existing mixer settings, therefore it does not change handling of older platforms with working micrphones (SC8280xp) but only the ones with issues (SM8450, SM8550) which need the fix. Signed-off-by:
Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://msgid.link/r/20240226115925.53953-3-krzysztof.kozlowski@linaro.orgSigned-off-by:
-
Krzysztof Kozlowski authored
TX macro codec differs slightly between different Qualcomm Low Power Audio SubSystem (LPASS) block versions. In LPASS version 9.2 the register responsible for TX SMIC MUXn muxes is different, thus to properly support it, the driver needs to register different widgets per different LPASS version. Prepare for supporting this register difference by refactoring existing code: 1. Move few widgets (TX SMIC MUXn, TX SWR_ADCn, TX SWR_DMICn) out of common 'tx_macro_dapm_widgets[]' array to a new per-variant specific array 'tx_macro_dapm_widgets_v9[]'. 2. Move also related audio routes into new array. 3. Store pointers to these variant-specific arrays in new variant-data structure 'tx_macro_data'. 4. Add variant-specific widgets and routes in component probe, instead of driver probe. The change should have no real impact, except re-shuffling code and registering some widgets and audio routes in component probe, instead of driver probe. Signed-off-by:
-
- 24 Feb, 2024 1 commit
-
-
Mark Brown authored
Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>: Factory calibration of the speakers stores the calibration information into an EFI variable. This set of patches adds support for applying speaker calibration data from that EFI variable. The HDA patch (#5) depends on the ASoC patches #2 and #3
-
- 23 Feb, 2024 7 commits
-
-
Varshini Rajendran authored
Add sam9x7 compatible in the DT documentation. Signed-off-by:
Varshini Rajendran <varshini.rajendran@microchip.com> Acked-by:
Conor Dooley <conor.dooley@microchip.com> Link: https://lore.kernel.org/r/20240223172657.672439-1-varshini.rajendran@microchip.comSigned-off-by:
-
Varshini Rajendran authored
Add sam9x7 compatible to DT bindings documentation. Signed-off-by:
-
Richard Fitzgerald authored
If there are factory calibration settings in EFI, extract the settings and write them to the firmware calibration controls. Signed-off-by:
-
Richard Fitzgerald authored
If there are factory calibration settings in EFI, extract the settings and write them to the firmware calibration controls. This must be done after any firmware or coefficients have been downloaded to the amp. Signed-off-by:
-
Richard Fitzgerald authored
Adds some helper functions and data for applying amp calibration. 1. cs35l56_read_silicon_uid() to get the silicon ID that is used to search for the correct calibration data entry. 2. Add the registers for the silicon ID to the readable registers. 3. cs35l56_get_calibration() wrapper around cs_amp_get_efi_calibration_data() 4. cs35l56_calibration_controls() table of the firmware controls for calibration data. 5. Added members to struct cs35l56_base to store the calibration data. Signed-off-by:
-
Richard Fitzgerald authored
Create a new library for code that is used by multiple Cirrus Logic amps. This initially implements extracting amp calibration data from EFI and writing it to firmware controls. During factory calibration of built-in speakers the firmware calibration constants are stored in an EFI file. The file contains an array of calibration constants for each of the speakers. cs_amp_get_calibration_data() searches for an entry matching the requested UID stamp, otherwise by array index. If the data is found in EFI the constants for that speaker are copied back to the caller. If EFI is not enabled, the cs_amp_get_calibration_data() implementation will compile to simply return -ENOENT and the linker can drop the code. The code to write calibration controls uses cs_dsp. Building of cs_dsp is not forced. Instead, the code will compile away the calls to cs_dsp if cs_dsp is not reachable. This strategy of conditional code allows cs-amp-lib to be shared by multiple drivers without forcing inclusion of other modules that might be unnecessary. The calls to efi.get_variable() and cs_dsp are in small wrapper functions. This is so that a KUNIT_STATIC_STUB_REDIRECT can be added in a future patch to redirect these calls to replacement functions for KUnit testing. Signed-off-by:
-
Richard Fitzgerald authored
Separate the functionality of wm_adsp_event() into two exported functions wm_adsp_start() and wm_adsp_stop(). This allows the codec driver to start and stop the DSP outside of a DAPM widget. Signed-off-by:
-
- 22 Feb, 2024 3 commits
-
-
Kuninori Morimoto authored
snd_soc_is_matching_dai() checks DAI name, which is paired function with snd_soc_dai_name_get(). It checks dlc->dai_name and dai->name (A) or dai->driver_name (B) or dai->component->name (C) static int snd_soc_is_matching_dai(...) { ... if (strcmp(dlc->dai_name, dai->name) == 0) ~~~~~~~~~~~~~ ^^^^^^^^^(A) if (... strcmp(dai->driver->name, dlc->dai_name) == 0) (B)^^^^^^^^^^^^^^^^ ~~~~~~~~~~~~~ if (... strcmp(dlc->dai_name, dai->component->name) == 0) ~~~~~~~~~~~~~ ^^^^^^^^^^^^^^^^^^(C) ... } But (B) part order is different with (A) and (C) (= ^^^^ and ~~~~). This is not a big deal, but confusable to read. Fixup it. Signed-off-by:Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Link: https://msgid.link/r/87wmqxjbcg.wl-kuninori.morimoto.gx@renesas.comSigned-off-by:
-
Vijendar Mukunda authored
Modify ACP63_REG_END macro to access all ACP registers. Signed-off-by:
Vijendar Mukunda <Vijendar.Mukunda@amd.com> Link: https://msgid.link/r/20240222102656.631144-2-Vijendar.Mukunda@amd.comSigned-off-by:
-
Vijendar Mukunda authored
To align with AMD SoundWire manager driver license, update license as GPL-2.0-only for Pink Sardine ACP PCI driver and corresponding child drivers. Signed-off-by:
-
- 21 Feb, 2024 9 commits
-
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
Amadeusz Sławiński <amadeuszx.slawinski@linux.intel.com> Signed-off-by:
Cezary Rojewski <cezary.rojewski@intel.com> Link: https://msgid.link/r/20240221152516.852353-7-cezary.rojewski@intel.comSigned-off-by:
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. rt5640_set_dai_sysclk() is an example of that - clk_set_rate() is not guarded by IS_ERR(). By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
-
Cezary Rojewski authored
Most of clk_xxx() functions do check if provided clk-pointer is non-NULL. These do not check if the pointer is an error-pointer. Providing such to a clk_xxx() results in a panic. By utilizing _optional() variant of devm_clk_get() the driver code is both simplified and more robust. There is no need to remember about IS_ERR(clk) checks each time mclk is accessed. Reviewed-by:
-
Krzysztof Kozlowski authored
On some boards with multiple WSA8840/WSA8845 speakers, the reset (shutdown) GPIO is shared between two speakers. Use the reset controller framework and its "reset-gpio" driver to handle this case. This allows bring-up and proper handling of all WSA884x speakers on X1E80100-CRD board. Cc: Bartosz Golaszewski <brgl@bgdev.pl> Cc: Sean Anderson <sean.anderson@seco.com> Reviewed-by:
Philipp Zabel <p.zabel@pengutronix.de> Signed-off-by:
-
Krzysztof Kozlowski authored
On newer Qualcomm platforms, like X1E80100-CRD, the WSA884x speakers share SD_N GPIOs between two speakers, thus a coordinated assertion is needed. Linux supports handling shared GPIO lines through "reset-gpios" property, thus allow specifying either powerdown or reset GPIOs (these are the same). Cc: Bartosz Golaszewski <brgl@bgdev.pl> Cc: Sean Anderson <sean.anderson@seco.com> Acked-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Mark Brown authored
Merge series from Cezary Rojewski <cezary.rojewski@intel.com>: The avs-driver continues to be utilized on more recent Intel machines. As TGL-based (cAVS 2.5) e.g.: RPL, inherit most of the functionality from previous platforms: SKL <- APL <- CNL <- ICL <- TGL rather than putting everything into a single file, the platform-specific bits are split into cnl/icl/tgl.c files instead. Makes the division clear and code easier to maintain. Layout of the patchset: First are two changes combined together address the sound-clipping problem, present when only one stream is running - specifically one CAPTURE stream. Follow up is naming-scheme adjustment for some of the existing functions what improves code incohesiveness. As existing IPC/IRQ code operates solely on cAVS 1.5 architecture, it needs no abstraction. The situation changes when newer platforms come into the picture. Thus the next two patches abstract the existing IPC/IRQ handlers so that majority of the common code can be re-used. The ICCMAX change stands out a bit - the AudioDSP firmware loading procedure differs on ICL-based platforms (and onwards) and having a separate commit makes the situation clear to the developers who are going to support the solution from LTS perspective. For that reason I decided not to merge it into the commit introducing the icl.c file.
-
- 20 Feb, 2024 10 commits
-
-
Cezary Rojewski authored
Update board selection with tables specifying supported I2S configurations. DMIC/HDAudio board selection require no update as dmic/hdaudio machine boards are generic and not tied to any specific codec. Reviewed-by:
-
Cezary Rojewski authored
For ICL+ platforms to avoid DMI/OPIO L1 entry during the base firmware load procedure, HW recommends to set LTRP_GB to 95us and start an additional CAPTURE stream in the background. Once the load completes, original LTRP_GB value is restored and the additional stream is released. Reviewed-by:
-
Cezary Rojewski authored
Define handlers specific to cAVS 2.5 platforms, that is TGL, ADL, RPL and all other variants based on this very version of AudioDSP architecture. Most operations are inherited from their predecessors with the major difference being AudioDSP cores management - firmware handlers that on its own so there is no need to interfere. Reviewed-by:
-
Cezary Rojewski authored
Define handlers specific to cAVS 2.0 platforms, that is ICL, JSL and all other variants based on this very version of AudioDSP architecture. Most operations are inherited from their predecessors with the major difference being firmware-logging functionality - IPC request as well as debug memory windows layout have changed. Reviewed-by:
-
Cezary Rojewski authored
Define handlers specific to cAVS 1.8 platforms, that is CNL, CFL, CML and all other variants based on this very version of AudioDSP architecture. Most operations are inherited from their predecessors. Reviewed-by:
-
Cezary Rojewski authored
Servicing IPCs on CNL platforms and onward differs from the existing one. To make room for these, relocate SKL-based platforms specific code into the skl.c file leaving only the genering irq_handler in the common code. Reviewed-by:
-
Cezary Rojewski authored
Servicing IPCs on CNL platforms and onward differs from the existing one. To make room for these, enrich platform descriptor with fields representing crucial IPC registers and utilize them throughout the code. While cleaning up device descriptors, reduce the number of code lines by assigning 'min_fw_version' within a single line. Reviewed-by:
-
Cezary Rojewski authored
Prefix members that are platform-specific with 'avs_' to improve code cohesiveness and reduce the chance for naming-conflics with other drivers. Reviewed-by:
-
Cezary Rojewski authored
To avoid sound clipping when there just one, single CAPTURE stream ongoing, disable L1SEN before it is started. Any PLAYBACK stream or additional CAPTURE allows L1SEN to be re-enabled. Reviewed-by:
-
Cezary Rojewski authored
Code loading is not the only procedure that manipulates L1SEN. Update existing mechanism so the stream starting procedure can interfere with L1SEN without causing any trouble to its other users. Reviewed-by:
-
