| Age | Commit message (Collapse) | Author | Files | Lines |
|---|
|
Add PCI id for the AlderLake-P.
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Link: https://lore.kernel.org/r/20210114115558.52699-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The remote-endpoint may not be available if it is part of some
pluggable module. One such example would be an audio card, the
Codec endpoint will not be available until it is plugged in.
Hence drop 'remote-endpoint' as a required property.
Cc: Rob Herring <robh@kernel.org>
Cc: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Signed-off-by: Sameer Pujar <spujar@nvidia.com>
Link: https://lore.kernel.org/r/1610948585-16286-1-git-send-email-spujar@nvidia.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Fix devicetree binding errors caused by newly added parameters
Signed-off-by: Michael Sit Wei Hong <michael.wei.hong.sit@intel.com>
Link: https://lore.kernel.org/r/20210118102706.6125-1-michael.wei.hong.sit@intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The resume operation relies on multiple transactions to synchronize
the regmap state, make sure the timeout is one order of magnitude
larger than an individual transaction, so that timeouts of failed
transactions are detected first.
Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Signed-off-by: Bard Liao <yung-chuan.liao@linux.intel.com>
Link: https://lore.kernel.org/r/20210115061651.9740-2-yung-chuan.liao@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
We don't need to print an error message when these ioremap operations
fail. The function that returns an error already prints an error message
and properly attributes it to the device. Drop them to save some code.
Cc: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Cc: Srinivasa Rao <srivasam@codeaurora.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210115034327.617223-5-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
We don't need to cast away __iomem when testing with IS_ERR() or
converting with PTR_ERR(). Modern sparse can handle this just fine.
Drop it.
Cc: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Cc: Srinivasa Rao <srivasam@codeaurora.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210115034327.617223-4-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
This device can sometimes have multiple regmaps. Let's add a name so
that we can differentiate in debugfs more easily.
Cc: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Cc: Srinivasa Rao <srivasam@codeaurora.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210115034327.617223-3-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
This looks like a left over debug print that tells us that HDMI is
enabled. Let's remove it as that's definitely not an error to have HDMI
enabled.
Cc: V Sujith Kumar Reddy <vsujithk@codeaurora.org>
Cc: Srinivasa Rao <srivasam@codeaurora.org>
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Cc: Cheng-Yi Chiang <cychiang@chromium.org>
Fixes: 7cb37b7bd0d3 ("ASoC: qcom: Add support for lpass hdmi driver")
Signed-off-by: Stephen Boyd <swboyd@chromium.org>
Reviewed-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Link: https://lore.kernel.org/r/20210115034327.617223-2-swboyd@chromium.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Invoke hda_dsp_remove() as the .shutdown() callback. This will help to
perform shutdown of the DSP safely on TGL platforms before shutting down
or rebooting the system.
BugLink: https://github.com/thesofproject/linux/issues/2571
Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210113152617.4048541-4-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add the .shutdown() callback to the sof-pci-dev driver, to help to
handle shutting down specific tasks for SOF PCI platforms.
Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210113152617.4048541-3-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add helper snd_sof_device_shutdown() to wrap the platform specific
.shutdown callbacks for SOF platforms.
Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210113152617.4048541-2-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add .shutdown() callback to the struct snd_sof_dsp_ops, for
doing platform specific actions at shutdown.
Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210113152617.4048541-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Instead of manually managing its DMA buffers using
dma_{alloc,free}_coherent() lets the sound core take care of this using
managed buffers.
On one hand this reduces the amount of boiler plate code, but the main
motivation for the change is to use the shared code where possible. This
makes it easier to argue about correctness and that the code does not
contain subtle bugs like data leakage or similar.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20210106133650.13509-3-lars@metafoo.de
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Instead of manually managing its DMA buffers using
dma_{alloc,free}_coherent() lets the sound core take care of this using
managed buffers.
On one hand this reduces the amount of boiler plate code, but the main
motivation for the change is to use the shared code where possible. This
makes it easier to argue about correctness and that the code does not
contain subtle bugs like data leakage or similar.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Link: https://lore.kernel.org/r/20210106133650.13509-2-lars@metafoo.de
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Instead of manually managing its DMA buffers using
dma_{alloc,free}_coherent() lets the sound core take care of this using
managed buffers.
On one hand this reduces the amount of boiler plate code, but the main
motivation for the change is to use the shared code where possible. This
makes it easier to argue about correctness and that the code does not
contain subtle bugs like data leakage or similar.
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Reviewed-by: Codrin Ciubotariu <codrin.ciubotariu@microchip.com>
Link: https://lore.kernel.org/r/20210106133650.13509-1-lars@metafoo.de
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
At the moment it is necessary to set up the DAPM routes between
front-end AIF<->DAI explicitly in the device tree, e.g. using
audio-routing =
"MM_DL1", "MultiMedia1 Playback",
"MM_DL3", "MultiMedia3 Playback",
"MM_DL4", "MultiMedia4 Playback",
"MultiMedia2 Capture", "MM_UL2";
This is prone to mistakes and (sadly) there is no clear error if one
of these routes is missing. :(
Actually, this should not be necessary because the ASoC core normally
automatically links AIF<->DAI within snd_soc_dapm_link_dai_widgets().
This is done using the "stname" parameter of SND_SOC_DAPM_AIF_IN/OUT.
For SND_SOC_DAPM_AIF_IN("MM_DL1", "MultiMedia1 Playback", 0, 0, 0, 0),
it should create the route from above: MM_DL1 <-> MultiMedia1 Playback.
This does not work at the moment because the AIF widget (MM_DL1)
and the DAI widget (MultiMedia1 Playback) belong to different
DAPM contexts (q6routing / q6asm-dai).
Fix this by declaring the AIF widgets in the same driver as the DAIs
(q6asm-dai). Now the routes above are created automatically
and no longer need to be specified in the device tree.
This is also more consistent with the back-end AIFs which are already
declared in q6afe-dais instead of q6routing. q6routing should only link
the components together using mixers.
Cc: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Fixes: 2a9e92d371db ("ASoC: qdsp6: q6asm: Add q6asm dai driver")
Signed-off-by: Stephan Gerhold <stephan@gerhold.net>
Link: https://lore.kernel.org/r/20201211203255.148246-1-stephan@gerhold.net
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
There are two issues with this code. The first error path forgot to set
the error code and instead returns success. The second error path
doesn't clean up.
Fixes: 272b5edd3b8f ("ASoC: Add support for CS42L56 CODEC")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/X9NE/9nK9/TuxuL+@mwanda
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
soc_pcm_apply_symmetry() want to call snd_pcm_hw_constraint_single()
for rate/channel/sample_bits, but, it needs many condition check.
These are very similar but different, thus, it needs to have very
verbose code.
This patch use macro for it and make code more simple.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87wnxo7uyq.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Getting rate/channels/sample_bits from param needs fixed method.
This patch adds new soc_pcm_set_dai_params() and replace existing code.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87y2i47uyw.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
soc-pcm has dpcm_set_fe_update_state() to update FE's runtime_update
(except dpcm_fe_dai_do_trigger() which needs to update it without it).
OTOH, it doesn't have BE's update function.
O: dpcm_set_fe_update_state()
X: dpcm_set_be_update_state()
This patch add BE's dpcm_set_fe_update_state()
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/87zh2k7uz1.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
This patch moves dpcm_set_fe_update_state() to top side.
This is prepare for cleanup soc-pcm.c
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>
Link: https://lore.kernel.org/r/871rfw99jn.wl-kuninori.morimoto.gx@renesas.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
There could be more than one thread read/write the dsp_power_state
simultaneously (e.g. hda_dsp_d0i3_work and sof_ipc_tx_message), add a
mutex power_state_access to make sure the access to it is mutually
exclusive.
Signed-off-by: Keyon Jie <yang.jie@linux.intel.com>
Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com>
Reviewed-by: Bard Liao <bard.liao@intel.com>
Reviewed-by: Guennadi Liakhovetski <guennadi.liakhovetski@linux.intel.com>
Signed-off-by: Kai Vehmanen <kai.vehmanen@linux.intel.com>
Link: https://lore.kernel.org/r/20210105155640.3725238-1-kai.vehmanen@linux.intel.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
sound/soc/soc-core.c: soc_remove_component() unconditionally calls
snd_soc_component_set_jack(component, NULL, NULL); on any components
being removed.
This means that on machines where the machine-driver does not provide
a jack through snd_soc_component_set_jack() es8316_disable_jack_detect()
will still get called and at this time es8316->jack will be NULL and
the es8316->jack->status check in es8316_disable_jack_detect() will
lead to a NULL pointer deref.
Fix this by checking for es8316->jack bein NULL at the start of
es8316_disable_jack_detect() and turn the function into a no-op in
that case.
Cc: russianneuromancer <russianneuromancer@ya.ru>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20210112101725.44200-1-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Because clk_disable_unprepare() already checked NULL clock parameter,
so the additional check is unnecessary, just remove it.
Signed-off-by: Xu Wang <vulab@iscas.ac.cn>
Link: https://lore.kernel.org/r/20210108084456.6603-1-vulab@iscas.ac.cn
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Move the snd_soc_dai_set_tdm_slot() call from cht_codec_init() to
cht_codec_fixup(). There are 2 reasons for doing this:
1. This aligns the cht_bsw_nau8824 with all the other BYT/CHT machine
drivers which also do this from their codec_fixup function.
2. When using the SOF driver, things like the TDM info is set from the
topology file. Moving the call to the codec_fixup function, which gets
skipped when using the SOF driver avoids the call interfering with the
settings when using the SOF driver.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210107115324.11602-1-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add a quirk for the Mele PCG03 Mini PC, being a Mini PC this device
has no speakers and no internal microphone.
To make matters worse the speaker output pins are shorted (to gnd or
to each other?) and SPKVDD is provided. So trying to output sound on the
speakers leads to shorting SPKVDD, this leads to a power dip after
which the codec is an unknown state. Sometimes it drops of the i2c
bus, sometimes it does still respond to i2c transfers, but is otherwise
not functional. TL;DR: trying to use the speaker outputs on this model
is BAD.
Besides not having speakers / an internal mic, this is a Bay Trail CR
device without a CHAN package in ACPI, so we default to SSP0-AIF2 as
codec connection. But the device is actually using SSP0-AIF1, so we
need to quirk that too.
Cc: Rasmus Porsager <rasmus@beat.dk>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210109210119.159032-4-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Some devices, like mini PCs/media/top-set boxes do not have an internal
microphone at all, an example of the is the Mele PCG03 Mini PC.
Add a new BYT_RT5640_NO_INTERNAL_MIC_MAP input-mapping for this,
which does not add any internal-mic routes and modifies the components
and the (optional) long_name strings to reflect this.
Cc: Rasmus Porsager <rasmus@beat.dk>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210109210119.159032-3-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Some devices, like mini PCs/media/top-set boxes do not have any speakers
at all, an example of the is the Mele PCG03 Mini PC.
Add a new BYT_RT5640_NO_SPEAKERS quirk-flag which when sets does not add
speaker routes and modifies the components and the (optional) long_name
strings to reflect that there are no speakers.
Cc: Rasmus Porsager <rasmus@beat.dk>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Acked-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Link: https://lore.kernel.org/r/20210109210119.159032-2-hdegoede@redhat.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The error path here doesn't set "ret" so it returns uninitialized data
instead of a negative error code.
Fixes: 2c1382840c19 ("ASoC: soc-pcm: disconnect BEs if the FE is not ready")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Link: https://lore.kernel.org/r/X/wfXQFxeMLvpO+1@mwanda
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
wm_adsp_read_data_word() used if (ret) to check for an error from
wm_adsp_read_raw_data_block(). While this is perfectly valid,
wm_adsp_read_raw_data_block() itself uses if (ret < 0) and three
calls to wm_adsp_read_data_word() also use if (ret < 0).
This creates an error check chain like this:
1st) if (ret < 0) return ret;
2nd) if (ret) return ret;
3rd) if (ret < 0) ...
This can confuse the compiler into thinking that there are possible
returns > 0 from the middle if() that are not handled by the final
if(). If this was true it would lead to using uninitialized variables
later in the outer function.
Fix this by changing the test in wm_adsp_read_data_word() to be
if (ret < 0).
Signed-off-by: Richard Fitzgerald <rf@opensource.cirrus.com>
Link: https://lore.kernel.org/r/20210111133825.8758-1-rf@opensource.cirrus.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Kernel test robot throws below error ->
sound/soc/soc-pcm.c:2523 dpcm_run_update_startup() error: uninitialized
symbol 'ret'.
Initializing ret = 0 and returning correct -ERRNO in failure path.
Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Souptick Joarder <jrdr.linux@gmail.com>
Link: https://lore.kernel.org/r/1610163901-5523-1-git-send-email-jrdr.linux@gmail.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
To avoid the pop noise, this patch enables the feature of fast discharge.
Signed-off-by: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20210111092544.9064-1-shumingf@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
This patch will be the workaround to fix getting the wrong device ID on the rare chance.
It seems like something unstable when the system resumes. e.g. the bus clock
This patch tries to read the device ID to check several times.
After the test, the driver will get the correct device ID the second time.
Signed-off-by: Shuming Fan <shumingf@realtek.com>
Link: https://lore.kernel.org/r/20210111092740.9128-1-shumingf@realtek.com
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
On ECS EF20 series laptops, the internal mic is on DMIC2/IN2P.
And they need the inv_hp_det to make jack detection to work as
exoected.
Signed-off-by: Chris Chiu <chiu@endlessos.org>
Link: https://lore.kernel.org/r/20210111054141.4668-5-chiu@endlessos.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
The ECS EF20EA laptop use gpio for jack detection instead of rt5645
rt5645 JD. However, the GPIO polarity is inverse for hp-detect based
on the _DSD property of the RTK2 device.
Name (_DSD, Package () {
ToUUID("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),
Package () {
Package () {"hp-detect-gpio", Package() {^RTK2, 0, 0, 1 }},
}
})
This flag will invert the hp-detect gpio polarity.
Signed-off-by: Chris Chiu <chiu@endlessos.org>
Link: https://lore.kernel.org/r/20210111054141.4668-4-chiu@endlessos.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
Add the hp-detect gpio for ECS EF20 series laptops based on the
_CRS defined in DSDT table.
Method (_CRS, 0, NotSerialized)
{
Name (SBUF, ResourceTemplate ()
{
I2cSerialBusV2 (0x001A, ControllerInitiated, 0x00061A80,
AddressingMode7Bit, "\\_SB.PCI0.I2C2",
0x00, ResourceConsumer, , Exclusive,
)
GpioInt (Edge, ActiveBoth, SharedAndWake, PullNone, 0x0000,
"\\_SB.GPO3", 0x00, ResourceConsumer, ,
)
{ // Pin list
0x004F
}
GpioIo (Shared, PullDefault, 0x0000, 0x0000, IoRestrictionInputOnly,
"\\_SB.GPO3", 0x00, ResourceConsumer, ,
)
{ // Pin list
0x004F
}
})
Return (SBUF) /* \_SB_.PCI0.I2C2.RTK2._CRS.SBUF */
}
Signed-off-by: Chris Chiu <chiu@endlessos.org>
Link: https://lore.kernel.org/r/20210111054141.4668-3-chiu@endlessos.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
On at least one laptop (ECS EF20EA) the 'hp-detect' GPIO is defined in
the DSDT table by the ACPI GpioIo resources in _CRS. The GPIO related
information should be mapped to the rt5645 driver to enable the jack
detection also on non-DT platforms.
Signed-off-by: Chris Chiu <chiu@endlessos.org>
Link: https://lore.kernel.org/r/20210111054141.4668-2-chiu@endlessos.org
Signed-off-by: Mark Brown <broonie@kernel.org>
|
|
|
|
Section 3.11 was incorrectly called 3.9, fix it.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: Masahiro Yamada <masahiroy@kernel.org>
|
|
Change my email contact ahead of a likely painful eleven-month migration
to a certain cobalt enteprisey groupware cloud product that will totally
break my workflow. Some day I may get used to having to email being
sequestered behind both claret and cerulean oath2+sms 2fa layers, but
for now I'll stick with keying in one password to receive an email vs.
the required four.
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
|
|
When the creator of SQPOLL io_uring dies (i.e. sqo_task), we don't want
its internals like ->files and ->mm to be poked by the SQPOLL task, it
have never been nice and recently got racy. That can happen when the
owner undergoes destruction and SQPOLL tasks tries to submit new
requests in parallel, and so calls io_sq_thread_acquire*().
That patch halts SQPOLL submissions when sqo_task dies by introducing
sqo_dead flag. Once set, the SQPOLL task must not do any submission,
which is synchronised by uring_lock as well as the new flag.
The tricky part is to make sure that disabling always happens, that
means either the ring is discovered by creator's do_exit() -> cancel,
or if the final close() happens before it's done by the creator. The
last is guaranteed by the fact that for SQPOLL the creator task and only
it holds exactly one file note, so either it pins up to do_exit() or
removed by the creator on the final put in flush. (see comments in
uring_flush() around file->f_count == 2).
One more place that can trigger io_sq_thread_acquire_*() is
__io_req_task_submit(). Shoot off requests on sqo_dead there, even
though actually we don't need to. That's because cancellation of
sqo_task should wait for the request before going any further.
note 1: io_disable_sqo_submit() does io_ring_set_wakeup_flag() so the
caller would enter the ring to get an error, but it still doesn't
guarantee that the flag won't be cleared.
note 2: if final __userspace__ close happens not from the creator
task, the file note will pin the ring until the task dies.
Fixed: b1b6b5a30dce8 ("kernel/io_uring: cancel io_uring before task works")
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
files_cancel() should cancel all relevant requests and drop file notes,
so we should never have file notes after that, including on-exit fput
and flush. Add a WARN_ONCE to be sure.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
A simple preparation change inlining io_uring_attempt_task_drop() into
io_uring_flush().
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
We expect io_rw_reissue() to take place only during submission with
uring_lock held. Add a lockdep annotation to check that invariant.
Signed-off-by: Pavel Begunkov <asml.silence@gmail.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
If BCH_FEATURE_INCOMPAT_OBSO_LARGE_BUCKET is set in incompat feature
set, it means the cache device is created with obsoleted layout with
obso_bucket_site_hi. Now bcache does not support this feature bit, a new
BCH_FEATURE_INCOMPAT_LOG_LARGE_BUCKET_SIZE incompat feature bit is added
for a better layout to support large bucket size.
For the legacy compatibility purpose, if a cache device created with
obsoleted BCH_FEATURE_INCOMPAT_OBSO_LARGE_BUCKET feature bit, all bcache
devices attached to this cache set should be set to read-only. Then the
dirty data can be written back to backing device before re-create the
cache device with BCH_FEATURE_INCOMPAT_LOG_LARGE_BUCKET_SIZE feature bit
by the latest bcache-tools.
This patch checks BCH_FEATURE_INCOMPAT_OBSO_LARGE_BUCKET feature bit
when running a cache set and attach a bcache device to the cache set. If
this bit is set,
- When run a cache set, print an error kernel message to indicate all
following attached bcache device will be read-only.
- When attach a bcache device, print an error kernel message to indicate
the attached bcache device will be read-only, and ask users to update
to latest bcache-tools.
Such change is only for cache device whose bucket size >= 32MB, this is
for the zoned SSD and almost nobody uses such large bucket size at this
moment. If you don't explicit set a large bucket size for a zoned SSD,
such change is totally transparent to your bcache device.
Fixes: ffa470327572 ("bcache: add bucket_size_hi into struct cache_sb_disk for large bucket")
Signed-off-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
When large bucket feature was added, BCH_FEATURE_INCOMPAT_LARGE_BUCKET
was introduced into the incompat feature set. It used bucket_size_hi
(which was added at the tail of struct cache_sb_disk) to extend current
16bit bucket size to 32bit with existing bucket_size in struct
cache_sb_disk.
This is not a good idea, there are two obvious problems,
- Bucket size is always value power of 2, if store log2(bucket size) in
existing bucket_size of struct cache_sb_disk, it is unnecessary to add
bucket_size_hi.
- Macro csum_set() assumes d[SB_JOURNAL_BUCKETS] is the last member in
struct cache_sb_disk, bucket_size_hi was added after d[] which makes
csum_set calculate an unexpected super block checksum.
To fix the above problems, this patch introduces a new incompat feature
bit BCH_FEATURE_INCOMPAT_LOG_LARGE_BUCKET_SIZE, when this bit is set, it
means bucket_size in struct cache_sb_disk stores the order of power-of-2
bucket size value. When user specifies a bucket size larger than 32768
sectors, BCH_FEATURE_INCOMPAT_LOG_LARGE_BUCKET_SIZE will be set to
incompat feature set, and bucket_size stores log2(bucket size) more
than store the real bucket size value.
The obsoleted BCH_FEATURE_INCOMPAT_LARGE_BUCKET won't be used anymore,
it is renamed to BCH_FEATURE_INCOMPAT_OBSO_LARGE_BUCKET and still only
recognized by kernel driver for legacy compatible purpose. The previous
bucket_size_hi is renmaed to obso_bucket_size_hi in struct cache_sb_disk
and not used in bcache-tools anymore.
For cache device created with BCH_FEATURE_INCOMPAT_LARGE_BUCKET feature,
bcache-tools and kernel driver still recognize the feature string and
display it as "obso_large_bucket".
With this change, the unnecessary extra space extend of bcache on-disk
super block can be avoided, and csum_set() may generate expected check
sum as well.
Fixes: ffa470327572 ("bcache: add bucket_size_hi into struct cache_sb_disk for large bucket")
Signed-off-by: Coly Li <colyli@suse.de>
Cc: stable@vger.kernel.org # 5.9+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
This patch adds the check for features which is incompatible for
current supported feature sets.
Now if the bcache device created by bcache-tools has features that
current kernel doesn't support, read_super() will fail with error
messoage. E.g. if an unsupported incompatible feature detected,
bcache register will fail with dmesg "bcache: register_bcache() error :
Unsupported incompatible feature found".
Fixes: d721a43ff69c ("bcache: increase super block version for cache device and backing device")
Fixes: ffa470327572 ("bcache: add bucket_size_hi into struct cache_sb_disk for large bucket")
Signed-off-by: Coly Li <colyli@suse.de>
Cc: stable@vger.kernel.org # 5.9+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
This patch fixes the following typos,
from BCH_FEATURE_COMPAT_SUUP to BCH_FEATURE_COMPAT_SUPP
from BCH_FEATURE_INCOMPAT_SUUP to BCH_FEATURE_INCOMPAT_SUPP
from BCH_FEATURE_INCOMPAT_SUUP to BCH_FEATURE_RO_COMPAT_SUPP
Fixes: d721a43ff69c ("bcache: increase super block version for cache device and backing device")
Fixes: ffa470327572 ("bcache: add bucket_size_hi into struct cache_sb_disk for large bucket")
Signed-off-by: Coly Li <colyli@suse.de>
Cc: stable@vger.kernel.org # 5.9+
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
There is no need to reassign pdev_set_uuid in the second loop iteration,
so move it to the place before second loop.
Signed-off-by: Yi Li <yili@winhong.com>
Signed-off-by: Coly Li <colyli@suse.de>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
|
|
HSDK has hardware floating point and the common use case is with
glibc+hf so enable that as default.
Signed-off-by: Vineet Gupta <vgupta@synopsys.com>
|
Kai Vehmanen
Sameer Pujar
Michael Sit Wei Hong
Pierre-Louis Bossart
Stephen Boyd
Keyon Jie
Lars-Peter Clausen
Stephan Gerhold
Dan Carpenter
Kuninori Morimoto
Hans de Goede
Xu Wang
Richard Fitzgerald
Souptick Joarder
Shuming Fan
Chris Chiu
Linus Torvalds
Viresh Kumar
Darrick J. Wong
Pavel Begunkov
Coly Li
Yi Li
Vineet Gupta