Last minute ASoC updates for 3.4

There's a couple of small features here that were added late on but have
been in -next in my tree and some bug fixes.  The wm_hubs stuff is
actually bug fixes - the stuff that's currently in 3.4 is a half way
house between the two solutions that the latest change allows the
machine to select between.

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: Mark Brown <broonie@opensource.wolfsonmicro.com>

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

It can just be enabled all the time with no impact.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

PXA's SSP engine fails to take its current channel phase into account
when enabling a stream while the engine is already running. This
results in randomly swapped left/right channels on either the record
or the playback side, depending on which one was enabled first.

The following patch fixes this by factoring out the bit field
modifications in question to a separate function that pauses the
engine temporarily, modifies the bits and kicks it off again
afterwards. Appearantly, a transition of SSCR0_SSE syncs both
directions properly.

The patch has been rolled out to quite a number of devices over the
last weeks and seems to fix the issue reliably.
Signed-off-by: Daniel Mack <zonque@gmail.com>
Reported-and-tested-by: Sven Neumann <s.neumann@raumfeld.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The WM8776 codec driver requires the machine driver to set one of the
SND_SOC_DAIFMT_CBx_xxx values.  The P1022DS machine driver should be setting
SND_SOC_DAIFMT_CBM_CFM, but since that value was zero, no one noticed.

Commit 75d9ac46 ("ASoC: Allow DAI formats to be specified in the
dai_link"), however, changed the value of SND_SOC_DAIFMT_CBM_CFM from zero
to a non-zero value, which means that it now needs to be specifically set
by the machine driver.

We also set SND_SOC_DAIFMT_NB_NF, for the same reason.
Signed-off-by: Timur Tabi <timur@freescale.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org

The dma size will be changed by requested number of
channel(mono/stereo) from platform. For mono recording,
channels_min value should be 1.
Signed-off-by: Sangsu Park <sangsu4u.park@samsung.com>
Signed-off-by: Sangbeom Kim <sbkim73@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Signed-off-by: Takashi Iwai <tiwai@suse.de>

The irq handler must check whether the MPU401 instance is still alive.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Signed-off-by: Takashi Iwai <tiwai@suse.de>

We're currently not freeing card->rtd in cases where the card is
unregistered before being instantiated - convert it to devm_kzalloc() to
make sure that happens.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Acked-by: Liam Girdwood <lrg@ti.com>

Linus decided to go for another week so here's a few more updates - a
mixed bag here, a few minor diagnostic tweaks, some driver enhancements
and the dmaengine conversion for ep93xx drivers which was tested a while
ago and just waiting for a signoff.

GFP_ATOMIC is used in snd_pcm_link() just because the kmalloc is
called inside a lock.  Since this function isn't too critical for
speed and is rarely called in practice, better to allocate the chunk
at first before spinlock and free it in error paths, so that
GFP_KERNEL can be used.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Allows the constraint lists to be declared const by drivers which seems
reasonable; there's plenty of other constification we could do if we were
being complete but this was easy and quick.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

The WM8996 specification has been updated to specify 44.1kHz as a supported
sample rate. Update the driver to accept this configuration.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Now the mute-LED is controlled without powersave hack, and the ifdefs must
be removed.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Convert mx27vis-aic32x4 to platform driver.
Signed-off-by: Fabio Estevam <fabio.estevam@freescale.com>
Tested-by: Javier Martin <javier.martin@vista-silicon.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

A previous commit af65cbf2 (ALSA: hdmi: fix printout of SAD sampling
rates) fixed the sample rates shown in /proc/asound/cardX/eldY and
kernel log to not be entirely wrong. However, a missing rate from the
array added in the patch causes HDMI rates 88.2 kHz, 96 kHz, 176.4 kHz,
and 192 kHz to be shown as 96 kHz, 176.4 kHz, 192 kHz, and 384 kHz,
respectively.

Fix the reporting by adding the ALSA rate 64 kHz into the conversion
array between 48 kHz and 88.2 kHz.
Signed-off-by: Anssi Hannula <anssi.hannula@iki.fi>
Cc: Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Takashi Iwai <tiwai@suse.de>

We need to resume two legacy registers to recover MIDI/FM functionality
on S3/S4 resume, too.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Fixed a trailing white space error detected
in sound/core/control.c by checkpatch.pl script.
Signed-off-by: Jeffrin Jose <ahiliation@yahoo.co.in>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Since it's not always safe to assume that the vmaster hook is purely the
mute-LED control, add the flag indicating whether to expose the mute-LED
enum control or not.  Currently, conexant codec sets this off for non-HP
laptops where EAPD may be used really as EAPD.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Create snd_hda_add_vmaster_hook() and snd_hda_sync_vmaster_hook()
helper functions to handle the mute-LED in vmaster hook more
commonly.  In the former function, a new enum control "Mute-LED Mode"
is added.  This provides user to choose whether the mute-LED should be
turned on/off explicitly or to follow the master-mute status.
Reviewed-by: David Henningsson <david.henningsson@canonical.com>
Signed-off-by: Takashi Iwai <tiwai@suse.de>

Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Acked-by: Vinod Koul <vinod.koul@linux.intel.com>
Tested-by: Mika Westerberg <mika.westerberg@iki.fi>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Otherwise a wrong position will be reported after restarting a stream and the
first few samples might be skipped.
Signed-off-by: Mika Westerberg <mika.westerberg@iki.fi>
Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
Signed-off-by: Mark 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: Mark Brown <broonie@opensource.wolfsonmicro.com>

Added the vmaster hook for controlling EAPD dynamically to Conexant
auto-parser.  When the Master is muted, EAPDs are turned off as well.
This will fix the missing mute-LED control on some machines in
addition to the more power-saving in the auto-parser mode.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

We've had ugly static handling of the mute-LED with a powersave hook
for ALC269 HP laptops just like done in patch_sigmatel.c.  This is now
rewritten with the new vmaster hook and a fixup code.

For that, the new fixup action, ALC_FIXUP_ACT_BUILD, is introduced.
It's called after build_controls is called.  The reason of this new
action is that vmaster hook must be added at this stage (not in init
or probe).
Signed-off-by: Takashi Iwai <tiwai@suse.de>

The mute-LED is controlled in patch_sigmatel.c by (ab-)using the
powersave hook.  This can be now rewritten with the vmaster hook
instead, which is much simpler and can work even without
CONFIG_SND_HDA_POWER_SAVE kconfig.

A drawback is that the mute-LED corresponds _only_ to the Master mixer
switch instead of checking the whole DACs.  But usually this shouldn't
be a big problem as PA enables the mixer elements accordingly.

Also, this patch changes the code to create vmaster always even on
STAC9200 and STAC925x.  The former "Master" on these chips are renamed
as "PCM" now.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

It'll be used for adding hooks in later patches.
Signed-off-by: Takashi Iwai <tiwai@suse.de>

The McBSP driver stack has been moved, and rewritten resulting a single
driver - selected by CONFIG_SND_OMAP_SOC_MCBSP. There is no longer need to
have CONFIG_OMAP_MCBSP anymore.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonoie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

The McBSP driver stack has been moved to ASoC. The CONFIG_OMAP_MCBSP will
be removed since the CONFIG_SND_OMAP_SOC_MCBSP will trigger to build the
McBSP (audio) drivers.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

We are setting SNDRV_PCM_HW_PARAM_BUFFER_SIZE based on
SNDRV_PCM_HW_PARAM_CHANNELS, not vice versa. This bug didn't
have much impact because the rules are evaluated multiple times
by the core, and intended value got set eventually.
Signed-off-by: Grazvydas Ignotas <notasas@gmail.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

Remove the no longer valid check for McBSP1 regarding to signal mux
selection (on OMAP4 McBSP4 has 6 pin setup).
Only clear the srgr2, pcr0 register configuration if the requested clock
configuration will actually going to touch it. In this way we can avoid
issues if the CLKR/FSR mux has been configured after the clock selection.

We are going to check for the valid McBSP port in the
omap_mcbsp_6pin_src_mux() function based on the validity of the mux_signal
callback (which is only valid for ports having 6 pin setup).
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

Use single function for the CLKR/FSR mux configuration.
OMAP2/3 has 6 pin configuration on McBSP1 instance, while on OMAP4
McBSP4 instance have the 6 pin configuration so the omap2_mcbsp1_mux_* is
not correct name for all support OMAP versions
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Acked-by: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

Clock signal muxing, and functional clock related defines are only needed
in ASoC drivers.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Acked-by: Mark Brown <broonoie@opensource.wolfsonmicro.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>

On OMAP2/3 McBSP1 port has 6 pin setup, while on OMAP4 the port is McBSP4.
Implement the CLKR/FSR clock mux selection for OMAP4, and make sure that
we add the correct callback for the correct port across supported OMAP
versions.
Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Acked-by: Tony Lindgren <tony@atomide.com>
Signed-off-by: Liam Girdwood <lrg@ti.com>