Ensures that we don't interact badly with the power management framework,
especially in the cases where we're doing deferred work or we're using a
direct GPIO for these signals.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Otherwise if someone tries to use all four channels on AIF1 with the
device in master mode we won't be able to clock out all the data.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The conversion of the core driver to irqdomains means that we don't need
and irq_base to have working interrupts so use wm8994_request_irq() to
deal with looking up the interrupt number for the micdet IRQ.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Leave it up to the machine driver to disable accessory detection if
desired, the common pattern is to have accessory detection be a wake
source.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

It seems that the code duplication was added at a merge operation.
Signed-off-by: NJaroslav Kysela <perex@perex.cz>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Volume updates may not be acted upon if there is no clock applied when
the volume update is written. Ensure this doesn't happen by writing out
registers with volume updates after we enable each of the clocks.

There are more registers updated than before as previously we were
relying on wm_hubs to set those for controls it manages.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

Ensure that all the actions get taken at appropriate times by calling the
_PRE and _POST events for the aifNclk_ev functions explicitly.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

Make it clearer what context we're operating in.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Since the analogue portions of the checks for class W are the same over
all the devices factor out these checks into wm_hubs and while we're at
it also use wm_hubs_dac_hp_direct() to enable class W optimisations on
more paths.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The optimisations which we can do with caching the headphone DCS result in
wm_hubs have only been enabled in cases where class W is enabled. However,
there are more use cases which can benefit from the cache, especially with
WM8994 series devices with their more advanced digital routing.

Rather than keying off the class W information from the CODECs have a
check in wm_hubs for a suitable path and use that to determine if we can
deploy our headphone optimisations.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This can be helpful to users when tuning their systems.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If a driver using a custom mic detection callback has provided a table
of mic detection rates via platform data then use it.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use a slightly larger debounce when identifying accessory type and a
slightly smaller one when detecting buttons in response to user feedback
from large scale testing.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If FLL bypass is left enabled when we disable the CODEC then the output
clock will be left running which consumes a small amount of additional
current. Only enable bypass when there is an output.
Signed-off-by: NKyung-Kwee Ryu <Kyung-Kwee.Ryu@wolfsonmicro.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

While reading through sound/soc/codecs/wm8994.c I noticed a fair
amount of trailing whitespace. This patch gets rid of it.
Signed-off-by: NJesper Juhl <jj@chaosbits.net>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This ensures a clean startup of the channels, without this change some
use cases could result in issues in a small proportion of cases.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

Since AIF3 shares clock signals with other audio interfaces in order to
ensure it doesn't drive undesirable clocks we need to tristate it. Rather
than forcing the machine driver to do so have the driver do this.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Later WM8994 class devices can bypass the FLL from BCLK. Do this
automatically when the FLL input and output frequencies match up.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

release_firmware() does its own NULL ptr testing, it's redundant to
also test before calling it.
Signed-off-by: NJesper Juhl <jj@chaosbits.net>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If we're not doing jackdet it's not needed.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This increases the chances we'll manage to hit a partially configured
state on restart and the power savings are extremely small.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Rather than trying to work around machine drivers which try to reprogram
the FLL while it is providing SYSCLK just return an error if they try.
This will avoid audio glitches during FLL reconfiguration, or at least
move the introduction of the glitches to the machine driver.

Since disabling the source for an active SYSCLK is not supported in the
first place systems shouldn't be doing this in the first place.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

DAPM now has a DAPM-level lock which it manages itself so we don't need
to take the CODEC mutex to call DAPM any more.  Also remove a redundant
call to snd_soc_dapm_sync(), jack reporting also triggers a DAPM sync.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Based on latest production information.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The optimal management of VMID depends on a number of factors which vary
dynamically at runtime, for example the connection to a system docking
station. In some circumstances it is desirable to keep VMID enabled all
the time, in others it is desirable to aggressively power it up and down.

Provide a callback allowing machine driver to configure either the normal
power up/down mode (WM8994_VMID_MODE_NORMAL) or to maintain VMID even
when idle (WM8994_VMID_MODE_FORCE). This callback, wm8994_vmid_mode(),
should be called with the CODEC lock.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Currently we can the accdet mutex from within DAPM when updating the device
state which means we take accdet then the CODEC mutex but we also do the
locking the other way around when responding to the jackdet IRQ. Move all
the jackdet use of the CODEC mutex out of the accdet lock to avoid this.
Since all the DAPM interactions depend only on a single threaded IRQ this
is still serialised.

The locking improvements in 3.5 allow a better solution there.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

When we are out of system sleep always use audio mode for jack detection
in order to avoid potential performance issues handing off between modes.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Push everything through one function for active use cases, should be
no practical effect.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If we're not doing jackdet it's not needed.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Normally this will have no effect as we set detection up at system startup
before DAPM syncs take effect, this will only be useful if the system
enables and disables detection at runtime.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The WM1811 DC servo is able to run much faster than previous devices so
the benefit of skipping calibration is not useful.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Using snd_soc_write() means we always write to the register even if it
already contains the newly calculated value.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Don't debounce jack detection for inserted jacks, giving improved
responsiveness.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>