Add missing dmic_num mention and clarify that 'links' mean 'SoundWire
links', not to be used for other links.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220308192610.392950-2-pierre-louis.bossart@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

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

In preparation for adding support for the new IPC version that has been
introduced in the SOF firmware, this patch set includes some clean ups
and necessary modifications to commonly used functions that will be
re-used across different IPC-specific code.

Merge series from Codrin Ciubotariu <codrin.ciubotariu@microchip.com>:

This patch series adds support for Pulse Density Microphone Controller
(PDMC), present on Microchip's SAMA7G5.
The PDMC interfaces up to 4 digital microphones having Pulse Density
Modulated (PDM) outputs. It generates a single clock line and samples 1 or
2 data lines. The signal path includes an audio grade programmable
decimation filter and outputs 24-bit audio words.
The source of each channel can be independently defined as PDMC_DS0 or
PDMC_DS1, sampled at the rising or falling edge of PDMC_CLK.

The patch series starts with a fix on the ASoC DMA engine support. Then
continues with the bindings and the driver of PDMC. It is followed by the
DT nodes for SAMA7G5 and SAMA7G5-EK. In the end, the drivers for PDMC
and PDM microphones are enabled in sama7_defconfig.

This function only calls of_node_put() in the regular path.
And it will cause refcount leak in error paths.
For example, when codec_np is NULL, saif_np[0] and saif_np[1]
are not NULL, it will cause leaks.

of_node_put() will check if the node pointer is NULL, so we can
call it directly to release the refcount of regular pointers.

Fixes: e968194b ("ASoC: mxs: add device tree support for mxs-sgtl5000")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220308020146.26496-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

The Pulse Density Microphone Controller (PDMC) interfaces up to 4 digital
microphones having Pulse Density Modulated (PDM) outputs. It generates a
single clock line and samples 1 or 2 data lines. The signal path includes
an audio grade programmable decimation filter and outputs 24-bit audio
words on the APB bus.
Signed-off-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20220307122202.2251639-4-codrin.ciubotariu@microchip.comSigned-off-by: Mark Brown <broonie@kernel.org>

Add DT bindings for the new Microchip PDMC embedded in sama7g5 SoCs.
Signed-off-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
Link: https://lore.kernel.org/r/20220307122202.2251639-3-codrin.ciubotariu@microchip.comSigned-off-by: Mark Brown <broonie@kernel.org>

Even if struct snd_dmaengine_pcm_config is used, prepare_slave_config()
callback might not be set. Check if this callback is set before using it.

Fixes: fa654e08 ("ASoC: dmaengine-pcm: Provide default config")
Signed-off-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20220307122202.2251639-2-codrin.ciubotariu@microchip.comSigned-off-by: Mark Brown <broonie@kernel.org>

Fix the missing clk_disable_unprepare() before return
from rk817_platform_probe() in the error handling case.

Fixes: 0d6a04da ("ASoC: Add Rockchip rk817 audio CODEC support")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Tested-by: Chris Morgan <macromorgan@hotmail.com>
Link: https://lore.kernel.org/r/20220307090146.4104-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

The device_node pointer is returned by of_parse_phandle()  with refcount
incremented. We should use of_node_put() on it when done.

This function only calls of_node_put() in the regular path.
And it will cause refcount leak in error paths.
Fix this by calling of_node_put() in error handling too.

Fixes: 4e28491a ("ASoC: mediatek: mt8192-mt6359: fix device_node leak")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20220308015224.23585-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

The device_node pointer is returned by of_parse_phandle()  with refcount
incremented. We should use of_node_put() on it when done.

Fixes: afb93d71 ("ASoC: SOF: imx: Add i.MX8M HW support")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220308023325.31702-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

Fix the missing clk_disable_unprepare() before return
from rockchip_i2s_probe() in the error handling case.

Fixes: 01605ad1 ("ASoC: rockchip-i2s: enable "hclk" for rockchip I2S controller")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Link: https://lore.kernel.org/r/20220307083553.26009-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

In acp3x_audio_probe() platform_get_irq() may return error, but i2s_irq now
is unsigned int so the error handling is never triggered.

Fixes: 87d71a12 ("ASoC: amd: pcm-dma: Use platform_get_irq() to get the interrupt")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20220305123613.6324-1-yuehaibing@huawei.comSigned-off-by: Mark Brown <broonie@kernel.org>

In acp5x_audio_probe() platform_get_irq() may return error, but i2s_irq now
is unsigned int so the error handling is never triggered.

Fixes: 87d71a12 ("ASoC: amd: pcm-dma: Use platform_get_irq() to get the interrupt")
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Link: https://lore.kernel.org/r/20220305123705.3708-1-yuehaibing@huawei.comSigned-off-by: Mark Brown <broonie@kernel.org>

Use "bt-sco-pcm-wb" codec dai driver for wb bt audio.
Signed-off-by: Jiaxin Yu <jiaxin.yu@mediatek.com>
Reviewed-by: Tzung-Bi Shih <tzungbi@kernel.org>
Link: https://lore.kernel.org/r/20220307033056.11463-1-jiaxin.yu@mediatek.comSigned-off-by: Mark Brown <broonie@kernel.org>

Now we can use asoc_substream_to_rtd() macro,
let's use it.
Signed-off-by: Lianjie Zhang <zhanglianjie@uniontech.com>
Reviewed-by: Cezary Rojewski <cezary.rojewski@intel.com>
Link: https://lore.kernel.org/r/20220307151939.32870-1-zhanglianjie@uniontech.comSigned-off-by: Mark Brown <broonie@kernel.org>

The device_node pointer is returned by of_parse_phandle()  with refcount
incremented. We should use of_node_put() on it when done.

This function only calls of_node_put() in the regular path.
And it will cause refcount leak in error paths.
Fix this by calling of_node_put() in error handling too.

Fixes: a45f8853 ("ASoC: Add driver for PROTO Audio CODEC (with a WM8731)")
Signed-off-by: Miaoqian Lin <linmq006@gmail.com>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20220308013949.20323-1-linmq006@gmail.comSigned-off-by: Mark Brown <broonie@kernel.org>

This will avoid having to add the extended data for DAI components during
sof_widget_setup() as an extra step.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-11-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

These will be used later on by IPC-specific code.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-10-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

Modify the get_token op for string type tokens to pass the string as the
argument instead of a pointer to struct snd_soc_tplg_vendor_string_elem.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-9-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

It is always 0 and never used while parsing.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-8-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

Looks like a copy-paste error. The CPU DAI is never used.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-7-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

With the change in the commit:
"ASoC: SOF: Intel: hda: assign link DMA channel at run-time",
there is no need to find the CPU DAI in sof_link_hda_load().

Fixes: bdf4ad3f ('ASoC: SOF: Intel: hda: assign link DMA channel at
run-time')
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-6-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

cppcheck warning:

sound/soc/sof/sof-audio.c:884:0: style: The function
'snd_sof_find_spcm_pcm_id' is never used. [unusedFunction]
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@intel.com>
Reviewed-by: Péter Ujfalusi <peter.ujfalusi@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-5-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

Set the swidget's core for scheduler type widgets to match that of the
pipeline core. This simplifies the flow for core get/put during widget
setup/free.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-4-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

The commit "ASoC: SOF: core: Unregister machine driver before IPC and
debugfs" moved the call to unregister the machine driver to be done
before freeing the IPC. With this change, we can safely move the call to
snd_sof_remove() inside the same if() block just above.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-3-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

It is not used.
Signed-off-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20220307181111.49392-2-ranjani.sridharan@linux.intel.comSigned-off-by: Mark Brown <broonie@kernel.org>

Merge series from Sascha Hauer <s.hauer@pengutronix.de>:

This series has some updates for the fsl_sai driver: Some general
cleanup patches, a bugfix in the ip revision checking and finally
the mclk setting is made more accurate and support for 1:1 bclk:mclk
setting is added.

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

A couple of updates for Intel and AMD hardware, along with minor
cleanups

Ajit Kumar Pandey (4):
  ASoC: SOF: amd: Flush cache after ATU_BASE_ADDR_GRP register update
  ASoC: SOF: amd: Use semaphore register to synchronize ipc's irq
  ASoC: SOF: amd: Move group register configuration to acp-loader
  ASoC: SOF: amd: Increase ACP_HW_SEM_RETRY_COUNT value

Curtis Malainey (1):
  ASoC: SOF: fix 32 signed bit overflow

Gongjun Song (1):
  ASoC: SOF: Intel: pci-tgl: add RPL-S support

Peter Ujfalusi (2):
  ASoC: SOF: amd: acp-pcm: Take buffer information directly from runtime
  ASoC: SOF: amd: Do not set ipc_pcm_params ops as it is optional

Pierre-Louis Bossart (2):
  ASoC: SOF: debug: clarify operator precedence
  ASoC: SOF: Intel: hda: clarify operator precedence

 include/sound/sof/header.h         |  2 +-
 include/uapi/sound/sof/abi.h       |  2 +-
 sound/soc/sof/amd/acp-dsp-offset.h |  1 +
 sound/soc/sof/amd/acp-ipc.c        | 22 ++++++++++++++--------
 sound/soc/sof/amd/acp-loader.c     |  9 +++++++++
 sound/soc/sof/amd/acp-pcm.c        |  7 ++++---
 sound/soc/sof/amd/acp-stream.c     |  3 +++
 sound/soc/sof/amd/acp.c            | 29 ++++++++++++++---------------
 sound/soc/sof/amd/acp.h            |  3 +--
 sound/soc/sof/amd/renoir.c         |  1 -
 sound/soc/sof/debug.c              |  2 +-
 sound/soc/sof/intel/hda.c          |  2 +-
 sound/soc/sof/intel/pci-tgl.c      |  2 ++
 13 files changed, 52 insertions(+), 33 deletions(-)

--
2.30.2

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

Updates to clean-up the GPIOLIB dependency and a quirk for HP
SoundWire devices.

Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>:

This adds support for I2S/TDM links where the slot width varies
depending on the sample width, in a way that cannot be guessed by
component hw_params().

A typical example is:

- 16-bit samples use 16-bit slots
- 24-bit samples use 32-bit slots

There is no way for a component hw_params() to deduce from the information
it is passed that 24-bit samples will be in 32-bit slots.

Some audio hardware cannot support a fixed slot width or a slot width
equal to the sample width in all cases. This is usually due either to
limitations of the audio serial port or system clocking restrictions.

Merge series from Stephan Gerhold <stephan@gerhold.net>:

This series adds a simple driver and DT schema for the Awinic AW8738
audio amplifier. It's fairly simple - the main difference to
simple-amplifier is that there is a "one-wire pulse control" that
allows configuring the amplifier to one of a few pre-defined modes.
This can be used to configure the speaker-guard function (primarily
the power limit for the amplifier).

The "adata->i2s_irq" variable is unsigned so the error handling
will not work.

Fixes: 87d71a12 ("ASoC: amd: pcm-dma: Use platform_get_irq() to get the interrupt")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/20220304131256.GA28739@kiliSigned-off-by: Mark Brown <broonie@kernel.org>

DETECT_MODE and PLL_START must be zero while HP_PDN and ADC_PDN are
both 1. If this condition is broken it can discharge FILT+ and it
can then take up to 1 second for FILT+ to recharge.

There is no workaround required for this, simply avoiding settings
and sequences that would break the requirement. The driver already
meets the requirement.

But it is not obvious from reading the code that this requirement
exists, or what is ensuring it is met. So it would not currently be
obvious to someone changing the code that there is certain special
behaviour that must be maintained.

To avoid accidental breakage in the future:

- Add comments into the register definitions to warn about this so
  that anyone changing the code around DETECT_MODE and PLL_START is
  aware of this requirement.

- Add a comment where PLL_START is written to 1 to highlight the
  requirement and why it is satisfied.

- Add a comment in cs42l42_setup_hs_type_detect() when DETECT_MODE is
  initialized.
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220304144015.398656-1-rf@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

The parts supported by this driver can have product-specific
firmware and tunings files. Typically these have been used on
embedded systems where the manufacturer is responsible for
installing the correct product-specific firmware files into
/lib/firmware. However, the linux-firmware repository places all
available firmwares into /lib/firmware and it is up to the driver to
select the correct product-specific firmware from that directory.

For example a product containing four smart amplifiers may provide
firmware specific for that product and each of the amplifiers may
have coefficient files containing tunings for their placement in the
mechanical design.

This change extends firmware (wmfw) and coefficient (bin) filenames
to be of the general form:

<cirrus/>part-dspN-fwtype<-system_name<-asoc_component_prefix>>.type

Where the cirrus subdirectory, system_name and asoc_component_prefix
are optional.

New files will be placed in the cirrus subdirectory to avoid
polluting the main /lib/firmware/ location. The generic name must be
searched in /lib/firmware before /lib/firmware/cirrus so that a
generic file in the new location does not override existing
product-specific files in the legacy location.

The search order for firmware files is:
  - cirrus/part-dspN-fwtype-system_name-asoc_component_prefix.wmfw
  - cirrus/part-dspN-fwtype-system_name.wmfw
  - part-dspN-fwtype.wmfw
  - cirrus/part-dspN-fwtype.wmfw

- Qualifications are added to the filename so that rightwards is more
  specific.
- The system_name is provided by the codec driver.
- The asoc_component_prefix is used to identify tunings for individual
  parts because it would already exist to disambiguate the controls
  and it makes it obvious which firmware file applies to which device.

The optional coefficient file must have the same filename
construction as the discovered wmfw except:
  - where the wmfw has only system_name then the bin file can
    optionally include the asoc_component_prefix. This is to allow a
    common wmfw for all amps but separate tunings per amp.
Signed-off-by: Simon Trimmer <simont@opensource.cirrus.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220303155016.122125-1-simont@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

As the potential failure of the clk_enable(),
it should be better to check it and return error
if fails.

Fixes: ab6f6d85 ("ASoC: fsi: add master clock control functions")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Link: https://lore.kernel.org/r/20220302062844.46869-1-jiasheng@iscas.ac.cnSigned-off-by: Mark Brown <broonie@kernel.org>

SC7280 config selected WCD938X instead of WCD938X_SDW Soundwire codecs.
WCD938X_SDW actually selects WCD938X, so directly selecting WCD938X results
in unmet dependencies and below warning

WARNING: unmet direct dependencies detected for SND_SOC_WCD938X
  Depends on [n]: SOUND [=y] && !UML && SND [=y] && SND_SOC [=m]
 && SND_SOC_WCD938X_SDW [=n] && (SOUNDWIRE [=n] || !SOUNDWIRE [=n])
  Selected by [m]:
  - SND_SOC_SC7280 [=m] && SOUND [=y] && !UML && SND [=y]
 && SND_SOC [=m] && SND_SOC_QCOM [=m] && (I2C [=y] && SOUNDWIRE [=n] ||
 COMPILE_TEST [=y])

Fix this issue by selecting WCD SoundWire codecs instead of component driver.

Fixes: 57350bd4 ("ASoC: qcom: SC7280: Add machine driver")
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20220304160934.32010-1-srinivas.kandagatla@linaro.orgSigned-off-by: Mark Brown <broonie@kernel.org>

cs35l41_private is not used on cs35l41_handle_pdata
Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220304150721.3802-5-tanureal@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

This device only has 4 TX channels.

Fixes: fe1024d5 ("ASoC: cs35l41: Combine adjacent register writes")
Signed-off-by: Lucas Tanure <tanureal@opensource.cirrus.com>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220304150721.3802-3-tanureal@opensource.cirrus.comSigned-off-by: Mark Brown <broonie@kernel.org>

Changes:
1. Revise rt5682s_sar_power_mode and rt5682s_headset_detect to be more
   rational.
2. Manually set to the jack-unplugging state via rt5682s_headset_detect
   during going to suspend. Close unnecessary powers and prepare for
   re-detecting the CBJ during resuming.
3. Simplize rt5682s_resume.
Signed-off-by: Derek Fang <derek.fang@realtek.com>
Link: https://lore.kernel.org/r/20220307102154.26065-1-derek.fang@realtek.comSigned-off-by: Mark Brown <broonie@kernel.org>

As the potential failure of the wm8350_register_irq(),
it should be better to check it and return error if fails.
Also, use 'free_' in order to avoid the same code.

Fixes: a6ba2b2d ("ASoC: Implement WM8350 headphone jack detection")
Signed-off-by: Jiasheng Jiang <jiasheng@iscas.ac.cn>
Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20220304023821.391936-1-jiasheng@iscas.ac.cnSigned-off-by: Mark Brown <broonie@kernel.org>