In case the initalization of an soc_device failed, there is no codec
associated with it. soc_suspend() will still dereference the pointer
and cause an Ooops when entering the sleep mode.

This happens on our board with a multi-target kernel image when booted
on a machine without audio circuits.

This patch makes the code bail out very early in this special case.
Signed-off-by: NDaniel Mack <daniel@caiaq.de>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Rather than managing the bias level of the system based on if there is
an active audio stream manage it based on there being an active DAPM
widget. This simplifies the code a little, moving the power handling
into one place, and improves audio performance for bypass paths when no
playbacks or captures are active.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

If the card was not instantiated in snd_soc_instantiate_card, calling
soc-remove will crash because some of codec, cpu_dai and card .remove
methods are called twice.
Fix this by returning from soc_remove immediately.
Signed-off-by: NMike Rapoport <mike@compulab.co.il>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

ASoC uses the standard ALSA data format definitions to specify the wire
format used between the CPU and CODEC. Since the ALSA data formats all
include the endianess of the data but this information is not relevant
by the time the data has been encoded onto the serial link to the CODEC
this means that either all the CODEC drivers need to declare both big and
little endian variants or the core needs to fix up the format constraints
specified by CODEC drivers.

For now take the latter approach - this will need to be revisited if any
CODECs are endianness dependant.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Some limited volume controls (mostly simple attenuations) have only two
settings so the ASoC info functions misreport them as booleans. Since
we currently have no better information check for " Volume" in the
control name and always report any controls matching as being integer.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Also make sure we're checking for the right operation while we're here.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Many devices require symmetric configurations of capture and playback
data formats, often due to shared clocking but sometimes also due to
other shared playback and record configuration in the device. Start
providing core support for this by allowing the DAIs or the machine
to specify that the sample rates used should be kept symmetric.

A flag symmetric_rates is provided in the snd_soc_dai and
snd_soc_dai_link structures. If this is set in either of the DAIs or in
the machine then a constraint will be applied when a stream is already
open preventing any changes in sample rate.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Ensure that any AC97 devices that bind to the CODEC are below the
ASoC device in the device tree so the suspend and resume code can
figure out what order to handle them in.
Reported-by: NRussell King <linux@arm.linux.org.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

AC97 devices may have other drivers hanging off them directly so need to
have resumed when the resume function returns meaning that we can't defer
the resume - complete it immediately for them. Non-AC97 devices should
not have other drivers hanging directly off the ASoC devices.

We only really need the deferral for non-AC97 devices - it's there since
some I2C buses are very slow and non-AC97 codecs often have large numbers
of registers to restore and require delays to bring the codec up cleanly
leading to a substantial impact on overall resume time.
Reported-by: NRussell King <linux@arm.linux.org.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The commit 14fa43f5 ("ASoC: Only
register AC97 bus if it's not done already") added a condition for
calling of soc_ac97_dev_register() but not added for calling of
soc_ac97_dev_unregister().  This patch adds same condition for
soc_ac97_dev_unregister().  Without this fix, kernel crashes when
unloading an asoc driver.
Signed-off-by: NAtsushi Nemoto <anemo@mba.ocn.ne.jp>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Considering the fact that most cpu_dai or codec_dai are using a same
'snd_soc_dai_ops' for several similar interfaces, 'ops' would be better
made a pointer instead, to make sharing easier and code a bit cleaner.

The patch below is rather preliminary since the asoc tree is being
actively developed, and this touches almost every piece of code,
(and possibly many others in development need to be changed as
well). Building of all codecs are OK, yet to every SoC, I didn't test
that.
Signed-off-by: NEric Miao <eric.miao@marvell.com>
Acked-by: NTimur Tabi <timur@freescale.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

ASoC supports both explicit codec drivers for AC97 devices and a simple
driver which uses the standard ALSA AC97 framework for codec support.
When used with the generic AC97 codec support that will provide the
ad hoc AC97 device for drivers like touchscreens to attach to so the
core shouldn't do so.
Reported-by: NManuel Lauss <mano@roarinelk.homelinux.net>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This is a further stage on the road to refactoring away the ASoC
platform device.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Convert from snd_card_new() to the new snd_card_create() function
in other sound subdirectories.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

Many codec drivers were implementing cookie-cutter copies of the function
that adds kcontrols to the codec.

This patch moves this code to a common function snd_soc_add_controls() in
soc-core.c and updates all drivers using copies of this function to use the
new common version.

[Edited to raise priority of error log message and document parameters.
 -- broonie]
Signed-off-by: NIan Molton <ian@mnementh.co.uk>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Merge the recently introduced soc_value_enum structure to the soc_enum.
The value based enums are still handled separately from the normal enum types,
but with the merge some of the newly introduced functions can be removed.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@nokia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This patch introduces a new enum type.
In this enum type each enumerated items referred with a value.

This new enum type can handle enums encoded in bitfield, or any other
weird ways. twl4030 codec has several mux selection register, where the
input/output mux is coded in a bitfield. With the normal enum type this type
of mux can not be handled in a clean way.
Signed-off-by: NPeter Ujfalusi <peter.ujfalusi@nokia.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Almost all parameters that have been misnamed in the comments.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Another part of the backporting of Liam's ASoC v2 work. Using this is
more complicated than the other registration types since currently the
codec is instantiated during the probe of the ASoC device so we can't
currently readily wait for the codec to register.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

The module init entries should be __init instead of __devinit.
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

This will allow codec drivers to be refactored to allow them to be
registered out of line with the ASoC device registration.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Use the lists of platforms, platform DAIs and cards to check to see that
everything has registered. Since relationships are still specified by
direct references to the structures in the drivers and the drivers all
register everything at modprobe there should be no practical effect yet.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

ASoC v2 allows platform drivers to instantiate independantly of the
overall ASoC card. This API allows drivers to notify the core when
they are registered.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Add API calls to register and unregister DAIs with the core.  Currently
these APIs are ineffective.  Since multiple DAIs for a given device are
a common case bulk variants are provided.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

ASoC v2 allows cards, codecs and platforms to instantiate separately,
with the overall ASoC device only being instantiated once all the
required components have registered. As part of backporting Liam's work
introduce an initial version of the card registration functions. At
present these do nothing active and are internal only, they will be
exposed to machine drivers after further backporting.  Adding this now
allows the datastructures used for dynamic card instantiation to be
built up gradually.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

None of the DAIs use it except s3c2412-i2s which only uses it for
dev_() printouts.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

None of the platforms are actually using the SoC device so remove it
(only atmel actually has a suspend method).
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

This is in preparation for the removal of struct snd_soc_device.

The pop time configuration should really be a property of the card not
the codec but since DAPM currently uses the codec rather than the card
using the codec is fine for now.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

As part of the deprecation of snd_soc_device push the registration of
the platform down into the card structure.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

ASoC v2 does not use the struct snd_soc_device at runtime, using struct
snd_soc_card as the root of the card.  Begin removing data from
snd_soc_device by pushing the workqueue data into snd_soc_card, using a
backpointer to the snd_soc_device to keep things going for the time
being.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

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

Currently ASoC card initialisation is completed by a function called
snd_soc_register_card().  As part of the work to allow independant
registration of cards, codecs and machines in ASoC v2 a new function of
the same name has been added so rename the existing function to
facilitate the merge of v2.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

There is no argument named @clk_id in snd_soc_dai_set_fmt,
remove its' comment.
Signed-off-by: NQinghuang Feng <qhfeng.kernel@gmail.com>
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Now that the ASoC resume has been punted to a workqueue for a release
cycle without attracting bug reports it should be safe to make the
log messages associated with it debug level, reducing noise and kernel
size in production configurations.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

DAI type information is only ever used within ASoC in order to special
case AC97 and for diagnostic purposes. Since modern CPUs and codecs
support multi function DAIs which can be configured for several modes
it is more trouble than it's worth to maintain anything other than a
flag identifying AC97 DAIs so remove the type field and replace it with
an ac97_control flag.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Liam Girdwood's ASoC v2 work avoids having two different ops structures
for DAIs by merging the members of struct snd_soc_ops into struct
snd_soc_dai_ops, allowing per DAI configuration for everything.
Backport this change.

This paves the way for future work allowing any combination of DAIs to
be connected rather than having fixed purpose CODEC and CPU DAIs and
only allowing CODEC<->CPU interconnections.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

One of the issues with the ASoC v1 API which has been addressed in the
ASoC v2 work that Liam Girdwood has done is that the ALSA card provided
by ASoC is distributed around the ASoC structures. For example, machine
wide data such as the struct snd_card are maintained as part of the
CODEC data structure, preventing the use of multiple codecs. This has
been addressed by refactoring the data structures so that all the data
for the ALSA card is contained in a single structure snd_soc_card which
replaces the existing snd_soc_machine and snd_soc_device.

Begin the process of backporting this by renaming struct snd_soc_machine
to struct snd_soc_card, better reflecting its function and bringing it
closer to standard ALSA terminology.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>

Rather than try to remember to keep the core version number updated
(which hasn't been happening) just remove it.  It was much more useful
when ASoC was out of tree.
Signed-off-by: NMark brown <broonie@opensource.wolfsonmicro.com>

[stripped sound/isa/* changes, replaced with the next patch -- tiwai]
Signed-off-by: NKay Sievers <kay.sievers@vrfy.org>
Signed-off-by: NTakashi Iwai <tiwai@suse.de>

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