Merge tag 'asoc-v6.4' of https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into for-next
ASoC: Updates for v6.4
The bulk of the commits here are for the conversion of drivers to use
void remove callbacks but there's a reasonable amount of other stuff
going on, the pace of development with the SOF code continues to be high
and there's a bunch of new drivers too:
- More core cleanups from Morimto-san.
- Update drivers to have remove() callbacks returning void, mostly
mechanical with some substantial changes.
- Continued feature and simplification work on SOF, including addition
of a no-DSP mode for bringup, HDA MLink and extensions to the IPC4
protocol.
- Hibernation support for CS35L45.
- More DT binding conversions.
- Support for Cirrus Logic CS35L56, Freescale QMC, Maxim MAX98363,
nVidia systems with MAX9809x and RT5631, Realtek RT712, Renesas R-Car
Gen4, Rockchip RK3588 and TI TAS5733.
ALSA: emu10k1: don't stop DSP in _snd_emu10k1_{,audigy_}init_efx()
These functions don't actually touch the DSP until they poke the code
into it, at which point it's temporarily stopped anyway. And fx8010.dbg
is already zero anyway.
- Use correct address limit for Audigy
- Use the right constant to actually make a step on Audigy
- Don't store *_DBG_STEP and the address in emu->fx8010.dbg, as
otherwise unrelated operations would make steps, too
This is untested. as10k1 was never ported to Audigy anyway.
ALSA: emu10k1: skip Sound Blaster-specific hacks for E-MU cards
The rev2 cards use CA0108, but the embedded P17V goes entirely unused.
Also, A_IOCTL (which is really the GPIO port) is actually the FPGA
control port, so messing with it is no good idea.
The hacks are actually mutually exclusive, so make that explicit while
we're at it.
Firstly, fix the distribution between public and private headers.
Otherwise, some of the already public macros wouldn't actually work, and
the SNDRV_EMU10K1_IOCTL_DBG_READ result for Audigy would be useless.
Secondly, add condition code registers for Audigy. These are just
aliases for selected constant registers, and thus are generation-
specific. At least A_CC_REG_ZERO is actually correct ...
Finally, shuffle around some defines to more logical places while at it,
and fix up some more comments.
Firstly, remove the FXWC_* defines - the comment on FXWC implies that
the relevant defines are the (A_)EXTOUT_* ones. It's unclear where this
came from - it was in the initial ALSA import, but neither the driver
from Creative nor kX-project have these defines.
Secondly, remove A_HR, which made plain no sense (was unused, and
clashed with FXRT). Amends commit cbb7d8f9b7b ("emu10k1: Update
registers defines for the Audigy 2/emu10k2.5").
ALSA: emu10k1: fix snd_emu1010_fpga_read() input masking for rev2 cards
Unlike the Alice2 chips used on 1st generation E-MU cards, the
Tina/Tina2 chips used on the 2nd gen cards have only six GPIN pins,
which means that we need to use a smaller mask. Failure to do so would
falsify the read data if the FPGA tried to raise an IRQ right at that
moment. This wasn't a problem so far, as we didn't actually enable FPGA
IRQs, but that's going to change soon.
- In snd_emu10k1_look_for_ctl(), evaluate the cheap condition first
- In _snd_emu10k1_{audigy_,}init_efx(), don't use expensive bit setting
loops to fill arrays
- In snd_emu_proc_ptr_reg_read(), remove useless condition - iobase can
be only 0 or 0x20
These seem to be another instance of cargo-culting from the Windows
driver. It presumably queries the register to decide about the followup
actions, but we don't do that.
These seem to be simply cargo-culted from the Windows driver's behavior.
However, the original reason were presumably read-modify-write cycles,
which we don't do.
ALSA: emu10k1: stop doing weird things with HCFG in snd_emu10k1_emu1010_init()
This doesn't do anything snd_emu10k1_init() wouldn't do later, and none
of the things it does seem relevant for the function itself (which is
pretty much about setting up the FPGA). It was probably a Windows
driver behavior cargo-culting artifact.
As the register definition clearly states, this is a 16-bit register,
yet we did all accesses as 32-bit. The writes in particular would have
the potential to clear the TIMER register (depending on how the bus/card
actually handles the too long writes).
This commit also introduces a separate define A_GPIO which aliases
A_IOCFG, which better reflects the distinct usage on E-MU cards.
This is done in the same commit to keep the churn down, as we're
touching all involved lines anyway.
Assert the validity of the registers and values, as them being out of
range would indicate an error in the driver. Consequently, don't bother
returning error codes; they were ignored everywhere anyway.
Detach it better from the main PCM driver, which it really doesn't have
much in common with.
In particular, this moves the interrupt handler implementation into
p16v.c, and makes it access the substream runtime status more directly,
so it doesn't need to abuse structs snd_emu10k1_pcm and
snd_emu10k1_voice any more.
We don't need private pcm runtime data at all, as the only thing it was
used for (except the back-link to the substream) was the `running` flag.
So store that directly in runtime->private_data.
This somewhat radical strip-down shows that this driver contains some
complexity that was never actually utilized. I suppose the right way to
fully utilize the hardware in a simple way would be introducing more
substreams. This wouldn't require any of the removed code.
ALSA: emu10k1: remove obsolete card type variable and defines
The use of the variable was removed in commit 2b637da5a1b ("clean up
card features"). That commit also broke user space (the ioctl
structure), at which point the defines became meaningless, so I don't
think purging them is a problem.
ALSA: emu10k1: clarify various fx8010.*_mask fields
extin_mask and extout_mask are used only by the SbLive! microcode, so
they have no effect on Audigy.
Eliminate fxbus_mask entirely, as it wasn't actually used for anything.
As a drive-by, remove the pointless pad1 field from struct
snd_emu10k1_fx8010 - it is not visible to user space, so it has no
binary compatibility constraints.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com> Tested-by: Shengjiu Wang <shengjiu.wang@gmail.com> Link: https://lore.kernel.org/r/877cu6f619.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: fsl: imx-audmix: cleanup platform which is using Generic DMA
If CPU is using soc-generic-dmaengine-pcm, Platform Component will be
same as CPU Component. In this case, we can use CPU dlc for Platform dlc.
This patch shares CPU dlc with Platform, and add comment.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com> Tested-by: Shengjiu Wang <shengjiu.wang@gmail.com> Link: https://lore.kernel.org/r/878remf61j.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: fsl: imx-spdif: cleanup platform which is using Generic DMA
If CPU is using soc-generic-dmaengine-pcm, Platform Component will be
same as CPU Component. In this case, we can use CPU dlc for Platform dlc.
This patch shares CPU dlc with Platform, and add comment.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com> Tested-by: Shengjiu Wang <shengjiu.wang@gmail.com> Link: https://lore.kernel.org/r/87a5z2f61w.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC: fsl: imx-es8328: cleanup platform which is using Generic DMA
If CPU is using soc-generic-dmaengine-pcm, Platform Component will be
same as CPU Component. In this case, we can use CPU dlc for Platform dlc.
This patch shares CPU dlc with Platform, and add comment.
Signed-off-by: Kuninori Morimoto <kuninori.morimoto.gx@renesas.com> Acked-by: Shengjiu Wang <shengjiu.wang@gmail.com> Tested-by: Shengjiu Wang <shengjiu.wang@gmail.com> Link: https://lore.kernel.org/r/87bkjif628.wl-kuninori.morimoto.gx@renesas.com Signed-off-by: Mark Brown <broonie@kernel.org>
The auto-silencer supports two modes: "thresholded" to fill up "just
enough", and "top-up" to fill up "as much as possible". The two modes
used rather distinct code paths, which this patch unifies. The only
remaining distinction is how much we actually want to fill.
This fixes a bug in thresholded mode, where we failed to use new_hw_ptr,
resulting in under-fill.
Top-up mode is now more well-behaved and much easier to understand in
corner cases.
This also updates comments in the proximity of silencing-related data
structures.
Mark Brown [Thu, 20 Apr 2023 13:49:35 +0000 (14:49 +0100)]
ASoC: SOF: pcm/Intel: Pause-resume improvements for
Merge series from Peter Ujfalusi <peter.ujfalusi@linux.intel.com>:
Last minute patch for correct the pasue/resume operation with IPC4. The
issues are hardto reproduce and needs extended stress testing to be hit,
in which case the audio breaks due to DMA errors.
Mark Brown [Thu, 20 Apr 2023 13:49:22 +0000 (14:49 +0100)]
ASoC: Intel: boards: updates for 6.4
Merge series from Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>:
Preparation of ALC712 support with different types of SoundWire
devices per link, new RaptorLake SoundWire device, better error
handling for Cirrus devices and cosmetic changes for Max98373.
Bard Liao (3):
ASoC: Intel: sof_sdw: set codec_num = 1 if the device is not
aggregated
ASoC: Intel: sof_sdw: support different devices on the same sdw link
ASoC: Intel: sof_sdw: append codec type to dai link name
Curtis Malainey (1):
ASoC: Intel: sof_cirrus_common: Guard against missing buses
Yong Zhi (2):
ASoC: Intel: sof_sdw: remove late_probe flag in struct
sof_sdw_codec_info
ASoC: Intel: sof_sdw_max98373: change sof_sdw_mx8373_late_probe to
static call
apoorv (1):
ASoC: Intel: soc-acpi: Add entry for rt711-sdca-sdw at link 2 in RPL
match table
Mark Brown [Wed, 19 Apr 2023 18:45:06 +0000 (19:45 +0100)]
ASoC: es8316: Don't use ranges based register lookup for a single register
The es8316 driver uses a register range to specify the single volatile
register it has. While the cost will be in the noise this is a bunch of
overhead compared to just having a volatile_reg() callback so switch to
the callback.
ASoC: SOF: Intel: Split the set_power_op for IPC3 and IPC4
Suspending to S0iX with IPC3 requires the PM_GATE IPC to be sent again
to stop the DMA trace. But with IPC4, this is not needed as the trace is
stopped with the LARGE_CONFIG_SET IPC. Also, sending the MOD_D0IX IPC to
set the D0I3 state again when the DSP is in D0I3 already results in an
imbalance in PM runtime states in the firmware. So split the
set_power_state ops for IPC3 and IPC4 to avoid sending the MOD_D0IX IPC
when the DSP is already in D0I3 with IPC4.
ASoC: SOF: pcm: Add an option to skip platform trigger during stop
In the case of IPC4, a pipeline is only paused during STOP/PAUSE/SUSPEND
triggers and the FW keeps the host DMA running when a pipeline is
paused. The start/stop tests iterate through STOP/START triggers without
involving a hw_free. This means that the pipeline state will only toggle
between PAUSED (during the STOP trigger) and RUNNING (during the START
trigger). So this test should be treated in the same way as a
PAUSE_PUSH/PAUSE_RELEASE test and the DMA should be kept running when
toggling the pipeline states between PAUSED and RUNNING.
Since there is no way to tell if a STOP trigger will be followed by hw_free
or not, this patch proposes to always skip DMA stop during the STOP trigger
and handle it later during hw_free. Introduce a new flag in struct
sof_ipc_pcm_ops, delayed_platform_trigger, that will be used to ensure that
the host DMA will not be stopped during the STOP/PAUSE/RELEASE triggers
and set it for IPC4. The platform_trigger call to stop the DMA will be
invoked during PCM hw_free instead when the pipeline is reset.
ASoC: SOF: Intel: hda: Do not stop/start DMA during pause/release
The FW does not pause/stop the host DMA during pause and stopping the
host DMA from the driver could result in an unknown behaviour. So, skip
triggering the HD-Audio host DMA during pause/release.
ASoC: Intel: sof_sdw: append codec type to dai link name
The existing sdw_sof machine driver constructs two SoundWire interfaces
by direction and sdw link id. It means that we will have exactly the
same dai link name if two dai links are on the same sdw link with the
same direction.
The new Realtek codec has two SoundWire interfaces for jack and DMIC
functions and they are treated as different codecs. To create two dai
links for jack and DMIC, we need to have different dai link names.
This patch suggests to append codec type if there are two or more
different types of devices on the same sdw bus.
ASoC: Intel: sof_sdw: support different devices on the same sdw link
The existing code assumes all devices on the same soundwire link
are the same devices. eg. all rt1316. This commit removes the
assumption and supports different devices on the same soundwire link.
ASoC: Intel: sof_sdw: set codec_num = 1 if the device is not aggregated
We assume adr_link->num_adr = 1 if a device is not aggregated. However,
the assumption is not valid if there are different type devices on the
same soundwire link.
Convert all existing calls that pass "NULL, 0" for reply data to the new
no_reply calls. Also convert any calls that pass in data but don't
actually parse the result.
95% of the calls inside SOF to TX an IPC don't care about a reply. Yet
the previous commit cleaned up a bunch of replies that were being
populated and then thrown away. This adds some functions so users who do
not need replies don't feel obligated to provide the space to the API.
Rob Herring [Fri, 10 Mar 2023 14:47:33 +0000 (08:47 -0600)]
ALSA: Use of_property_read_bool() for boolean properties
It is preferred to use typed property access functions (i.e.
of_property_read_<type> functions) rather than low-level
of_get_property/of_find_property functions for reading properties.
Convert reading boolean properties to to of_property_read_bool().
Rob Herring [Fri, 10 Mar 2023 14:47:33 +0000 (08:47 -0600)]
ALSA: ppc/tumbler: Use of_property_present() for testing DT property presence
It is preferred to use typed property access functions (i.e.
of_property_read_<type> functions) rather than low-level
of_get_property/of_find_property functions for reading properties. As
part of this, convert of_get_property/of_find_property calls to the
recently added of_property_present() helper when we just want to test
for presence of a property and nothing more.
Richard Fitzgerald [Tue, 18 Apr 2023 14:43:08 +0000 (15:43 +0100)]
ASoC: cs35l56: Remove SDW2RX1 mixer source
The mixer source index value for SDW2RX1 is different between
A1 and B0 silicon. As the driver doesn't provide a DAI for SDW2
just remove it as a mixer source option.
Richard Fitzgerald [Tue, 18 Apr 2023 14:43:06 +0000 (15:43 +0100)]
ASoC: cs35l56: Update comment on masking of EINT20 interrupts
EINT20 contains wake-source interrupts and also interface-blocked
interrupts, which all default to unmasked after reset or wake.
The comment in cs35l56_init() only mentioned the wake interrupts.
Update the comment so it's clear that it's intentional to also
mask the *_BLOCKED interrupts.
This is a bug, a logical-or should have been used.
Fixes: fc0b096c9291 ("ASoC: nau8825: Add delay control for input path") Signed-off-by: Tom Rix <trix@redhat.com> Acked-by: David Lin <CTLIN0@nuvoton.com> Link: https://lore.kernel.org/r/20230418120955.3230705-1-trix@redhat.com Signed-off-by: Mark Brown <broonie@kernel.org>
Liliang Ye [Mon, 3 Apr 2023 15:26:47 +0000 (23:26 +0800)]
ASoC: fsl_mqs: move of_node_put() to the correct location
of_node_put() should have been done directly after
mqs_priv->regmap = syscon_node_to_regmap(gpr_np);
otherwise it creates a reference leak on the success path.
To fix this, of_node_put() is moved to the correct location, and change
all the gotos to direct returns.
Fixes: a9d273671440 ("ASoC: fsl_mqs: Fix error handling in probe") Signed-off-by: Liliang Ye <yll@hust.edu.cn> Reviewed-by: Dan Carpenter <error27@gmail.com> Link: https://lore.kernel.org/r/20230403152647.17638-1-yll@hust.edu.cn Signed-off-by: Mark Brown <broonie@kernel.org>
ASoC need to use card->mutex with _INIT or _RUNTIME,
but there is no helper function for it.
This patch adds its helper function and use it.
Because people might misunderstand that _init() is mutex initialization,
this patch renames _INIT to _ROOT and adds new
snd_soc_card_mutex_lock_root() for it.
soc.h has snd_soc_dapm_mutex_lock/unlock() definition and
many drivers are using it, but soc-dapm.c is not.
1st reason is snd_soc_dapm_mutex_lock/unlock() requests
snd_soc_dapm_context pointer as parameter (A), but sometimes soc-dapm.c
needs to use snd_soc_card (B).
2nd reason is it want to use SND_SOC_DAPM_CLASS_INIT for mutex_lock_nested(),
but helper is using _RUNTIME (A).
The conclusion is we want to use "dapm vs card" and "_RUNTIME vs _INIT"
for dapm lock/unlock. To enable this selfish request, this patch uses
_Generic macro. We can use snd_soc_dapm_mutex_lock/unlock() for both
dapm and card case.
Current soc-dapm.c is using both mutex_lock() and mutex_lock_nested().
This patch handles mutex_lock() as mutex_lock_nested(..., 0),
in other words, handles below as same.
Because people might misunderstand that _init() is mutex initialization,
this patch renames _INIT to _ROOT and adds new
snd_soc_dapm_mutex_lock_root() for it.
This patch also moves snd_soc_dapm_subclass definition from soc-dapm.h
to soc.h to keep related code together.
Because very complex soc.h vs soc-dapm.h relationship,
it is difficult/impossible to define these helper into soc-dapm.h.
Convert the WM8753 audio codec bindings to DT schema.
Signed-off-by: Saalim Quadri <danascape@gmail.com> Acked-by: Charles Keepax <ckeepax@opensource.cirrus.com> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Link: https://lore.kernel.org/r/20230414223801.1106550-1-danascape@gmail.com Signed-off-by: Mark Brown <broonie@kernel.org>
Simon Trimmer [Fri, 14 Apr 2023 13:37:53 +0000 (14:37 +0100)]
ASoC: cs35l56: Don't return a value from cs35l56_remove()
cs35l56_remove() always returns 0. Two of the functions that call
it are void and the other one should only return 0. So there's no
point returning anything from cs35l56_remove().
dsp_ready_completion is redundant and can be replaced by a call
flush_work() to wait for cs35l56_dsp_work() to complete.
As the dsp_work is queued by component_probe() it must run before other
ASoC component callbacks and therefore there is no risk of calling
flush_work() before the dsp_work() has been queued.
Simon Trimmer [Fri, 14 Apr 2023 13:37:51 +0000 (14:37 +0100)]
ASoC: cs35l56: Wait for init_complete in cs35l56_component_probe()
Moving the wait from the beginning of the cs35l56_dsp_work() into
cs35l56_component_probe() will prevent the limbo situation that is an
artifact of the two stage SoundWire driver probe and initialisation
where the card is all registered and shows in ALSA but doesn't actually
work because the hardware didn't enumerate.
The other bus drivers perform the probe and init sequentially and are
not susceptible to this issue.
Simon Trimmer [Fri, 14 Apr 2023 13:37:49 +0000 (14:37 +0100)]
ASoC: cs35l56: Rework IRQ allocation
The irq member was being set before calling the init function and then
cs35l56_irq_request() was called only when the init was successful.
However cs35l56_release() calls devm_free_irq() when the irq member is
set and therefore if init() fails then this will cause an attempted free
of an unallocated IRQ.
Instead pass the desired IRQ number to the cs35l56_irq_request()
function and set cs35l56->irq only when it has been successfully
allocated.
David Lin [Fri, 14 Apr 2023 10:39:44 +0000 (18:39 +0800)]
ASoC: nau8825: Add delay control for input path
Change the original fixed delay to the assignment from property. It will make
more flexible to different platforms for avoiding pop noise at the beginning
of recording.
David Lin [Fri, 14 Apr 2023 10:39:42 +0000 (18:39 +0800)]
ASoC: dt-bindings: nau8825: Add delay control for input path
Change the original fixed delay to the assignment from property. It will make
more flexible to different platforms for avoiding pop noise at the beginning
of recording.
David Rau [Thu, 13 Apr 2023 02:41:34 +0000 (02:41 +0000)]
ASoC: da7219: Improve the relability of AAD IRQ process
- Configure the default gound switch delay time before enabling IRQ
to avoid the unexpected delay time is set up
- Apply DA7219 AAD own work queue to handle AAD events
- Replace msleep with queue_delayed_work to have better relability
This commit improves the control of ground switches in AAD IRQ
Mark Brown [Wed, 12 Apr 2023 18:59:27 +0000 (19:59 +0100)]
ASoC: cs35l56: Add system suspend handling
Merge series from Richard Fitzgerald <rf@opensource.cirrus.com>:
This set of patches adds handling for system suspend.
Patches 1..4 make some code changes that simplify the
suspend implementation, mainly to avoid race conditions.
There are two seperate aspects to suspend, and these have
been done as two patches:
- the main suspend-resume handling,
- re-loading the firmware if necessary after resume.
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:28 +0000 (16:25 +0100)]
ASoC: cs35l56: Re-patch firmware after system suspend
Check during cs35l56_system_resume() whether the firmware patch must
be applied again.
The FIRMWARE_MISSING flag in the PROTECTION_STATUS register indicates
whether the firmware has been patched.
In non-secure mode the FIRMWARE_MISSING flag is cleared at the end of
dsp_work(). If it is set after system-resume we know that dsp_work()
must be run again.
In secure mode the pre-OS loader will have done the secure patching
and cleared the FIRMWARE_MISSING flag. So this flag does not tell us
whether firmware memory was lost. But the driver could only be
downloading non-secure tunings, which is always safe to do.
If the driver has control of RESET we will have asserted it during
suspend so the firmware patch will have been lost. The driver would only
have control of RESET in non-secure mode.
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:23 +0000 (16:25 +0100)]
ASoC: cs35l56: Remove quick-cancelling of dsp_work()
Delete the 'removing' flag and don't kick init_completion to make a
quick cancel of dsp_work(). Just let it timeout on the wait for the
completion.
Simplify the code to standard cancelling or flushing of the work.
This avoids introducing corner cases from a layer of custom signalling.
It also avoids potential race conditions when system-suspend handling
is added.
Unless the hardware is broken, the dsp_work() will already have started
and passed the completion before the driver would want to cancel it.
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:27 +0000 (16:25 +0100)]
ASoC: cs35l56: Add basic system suspend handling
This adds the main handling for system suspend but does not handle
re-patching the firmware after system resume.
This is a multi-stage suspend and resume because if there is a
RESET line it is almost certain that it will be shared by all the
amps. So every amp must have done its suspend before we can
assert RESET. Likewise we must de-assert RESET before the amps
can resume.
It's preferable to assert RESET before we turning off regulators, and
while they power up.
The actual suspend and resume is done by using the pair
pm_runtime_force_suspend() and pm_runtime_force_resume() to
re-use our runtime suspend/resume sequences.
pm_runtime_force_suspend() will disable our pm_runtime. If we were
runtime-resumed it calls our runtime_suspend().
pm_runtime_force_resume() re-enables pm_runtime and if we were
originally runtime-resumed before the pm_runtime_force_suspend()
it calls our runtime_resume(). Otherwise it leaves us
runtime-suspended.
The general process is therefore:
suspend() -> finish dsp_work and then run our runtime_suspend
suspend_late() -> assert RESET and turn off supplies
resume_early() -> enable supplies and de-assert RESET
resume() -> pm_runtime_force_resume()
In addition, to prevent the IRQ handler running in the period
between pm_runtime_force_suspend() and pm_runtime_force_resume()
the parent IRQ is temporarily disabled:
- from suspend until suspend_noirq
- from resume_noirq until resume
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:26 +0000 (16:25 +0100)]
ASoC: cs35l56: Always wait for firmware boot in runtime-resume
When we are resuming from a system suspend the CS35L56 has probably
been hard reset (usually a power-on reset). So we must wait for the
firmware to boot. On SoundWire we also need it to re-initialize before
we can read the registers to check the CS35L56 state.
The simplest way to handle this is for runtime-resume to always wait
for firmware boot. If the firmware is already booted the overhead is
only one register read.
The system-resume will have to runtime-resume the driver anyway before
attempting any register access. So this will automatically include the
wait for initialization on SoundWire.
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:25 +0000 (16:25 +0100)]
ASoC: cs35l56: Skip first init_completion wait in dsp_work if init_done
At the start of dsp_work() only wait for init_completion if !init_done.
This allows system suspend to re-queue dsp_work() without having to
do a dummy complete() of init_completion.
A dummy completion in system suspend would have to be conditional on
init_done. But that would create a possible race condition between our
system resume and cs35l56_init() in the corner case that we suspend right
after the SoundWire core has enumerated and reported ATTACHED.
It is safer and simpler to have cs35l56_init() as the only place that
init_completion is completed, and dsp_work() as the only place that
it is consumed.
Richard Fitzgerald [Tue, 11 Apr 2023 15:25:24 +0000 (16:25 +0100)]
ASoC: cs35l56: Use DAPM widget for firmware PLAY/PAUSE
If we use a DAPM widget instead of mute_stream() to send the
PLAY command we can issue the plays to multiple amps in parallel.
With mute_stream each codec driver instance is called one at a
time so we get N * PS0 delay time.
DAPM does each stage on every widget in a card before moving to
the next stage. So all amps will do the PRE_PMU then all will do
the POST_PMU. The PLAY is sent in the PRE_PMU so that they all
power-up in parallel. After the PS0 wait in the first POST_PMU
all the other amps will also be ready so there won't be any extra
delay, or it will be negligible.
There's also no point waiting for the MBOX ack in the PRE_PMU.
We won't see a PS0 state in POST_PMU if it didn't ack the PLAY
command. So we can save a little extra time.
We have SOF and generic ACP support enabled for Rembrandt and
pheonix platforms on some machines. Since we have same PCI id
used for probing, add check for machine configuration flag to
avoid conflict with newer pci drivers. Such machine flag has
been initialized via dmi match on few Chrome machines. If no
flag is specified probe and register older platform device.
Ying Liu [Tue, 11 Apr 2023 17:09:12 +0000 (01:09 +0800)]
ASoC: tas5720: add missing unwind goto in tas5720_codec_probe
Smatch complains that missing unwind goto in tas5720_codec_probe.
When tas5720 has an invalid devtype, it is expected to invoke
regulator_bulk_disable to handle the failure. But the default
option return an error code directly. Fix it by reusing the
probe_fail label.
Fred Oh [Wed, 12 Apr 2023 06:14:57 +0000 (09:14 +0300)]
ASoC: SOF: Intel: mtl: fix page fault in dspless mode when DSP is disabled
The patch to make DSPless mode work even if the DSP is
disabled in BIOS missed to touch the MTL code to add
the needed checks.
If the DSP is disabled this can lead to page fault due to not
accesible registers.
Fixes: 9fc6786f549c ("ASoC: SOF: Intel: hda: make DSPless mode work with DSP disabled in BIOS") Signed-off-by: Fred Oh <fred.oh@linux.intel.com> Reviewed-by: Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Signed-off-by: Peter Ujfalusi <peter.ujfalusi@linux.intel.com> Link: https://lore.kernel.org/r/20230412061457.27937-1-peter.ujfalusi@linux.intel.com Signed-off-by: Mark Brown <broonie@kernel.org>
projects / users / hch / misc.git / log