Merge remote-tracking branch 'asoc/topic/wm8955' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8804' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8770' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8753' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8750' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8741' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8510' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8400' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm8350' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm2200' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm2000' into asoc-next

Merge remote-tracking branch 'asoc/topic/wm0010' into asoc-next

Merge remote-tracking branch 'asoc/topic/ux500' into asoc-next

Merge remote-tracking branch 'asoc/topic/tlv320aic32x4' into asoc-next

Merge remote-tracking branch 'asoc/topic/si476x' into asoc-next

Merge remote-tracking branch 'asoc/topic/samsung' into asoc-next

Merge remote-tracking branch 'asoc/topic/rt5631' into asoc-next

Merge remote-tracking branch 'asoc/topic/omap' into asoc-next

Merge remote-tracking branch 'asoc/topic/max98090' into asoc-next

Merge remote-tracking branch 'asoc/topic/max9768' into asoc-next

Merge remote-tracking branch 'asoc/topic/log' into asoc-next

Merge remote-tracking branch 'asoc/topic/lm49453' into asoc-next

Merge remote-tracking branch 'asoc/topic/kirkwood' into asoc-next

Merge remote-tracking branch 'asoc/topic/jz4740' into asoc-next

Merge remote-tracking branch 'asoc/topic/jack' into asoc-next

Merge remote-tracking branch 'asoc/topic/hotplug' into asoc-next

Merge remote-tracking branch 'asoc/topic/fsl' into asoc-next

Merge remote-tracking branch 'asoc/topic/fsi' into asoc-next

Merge remote-tracking branch 'asoc/topic/dmaengine' into asoc-next

Merge remote-tracking branch 'asoc/topic/davinci' into asoc-next

Merge remote-tracking branch 'asoc/topic/da9055' into asoc-next

Merge remote-tracking branch 'asoc/topic/da7210' into asoc-next

Merge remote-tracking branch 'asoc/topic/cs4271' into asoc-next

Merge remote-tracking branch 'asoc/topic/atmel' into asoc-next

Merge remote-tracking branch 'asoc/topic/arizona' into asoc-next

Merge remote-tracking branch 'asoc/topic/ak4642' into asoc-next

Merge remote-tracking branch 'asoc/topic/ak4535' into asoc-next

Merge remote-tracking branch 'asoc/topic/ak4104' into asoc-next

Merge remote-tracking branch 'asoc/topic/adsp' into asoc-next

Merge remote-tracking branch 'asoc/topic/ab8500' into asoc-next

Merge remote-tracking branch 'asoc/fix/omap' into asoc-next

ASoC: wm5110: Enable volume ramp control

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

ASoC: wm5102: Enable volume ramp control

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

ASoC: arizona: Add volume ramp controls

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

ASoC: Samsung: Do not register samsung audio dma device as pdev

Previously, the ASoC 'platform' (PCM/DMA) object was instantiated via a
platform_device. This didn't represent the hardware well, since there
was no separate hardware associated with this platform_device; it was a
virtual device with sole purpose to call snd_soc_register_platform().

This change removes the platform_device completely. Each Samsung DAI now
registers the ASoC 'platform' itself. Machine drivers are adjusted for
the new 'platform' name.

Signed-off-by: Padmavathi Venna <padma.v@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: McASP: implement a way to force BCLK/LRCLK ratios

Depending on the Codec, the the BCLK/LRCLK ratio might not be freely
chosen by the CPU DAI.

For example, some Codec might want to be supplied with 32-bit samples
for both its channels regardless of the actual audio word size the CPU
sends. In such cases, the rest of the bits on the data lines must be
padded with zeros:

          _______________________________
LRCLK    /                               \
      --'                                 `---------- .....

BCLK  ||||||||||||||||||||||||||||||||||||||||||||||| .....

DATA  ____||||||||||||||||_________________|||||||||| .....

          |<--  data  -->|<--   pads  --> |

This patch adds a new clock divider to configure the BCLK/LRCLK ratio.
If the machine code uses that divider, the driver uses the specified
value, instead of deriving that information from the audio word size.

Otherwise, the original behaviour is retained.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: McASP: calculate values for channel size

Change davinci_config_channel_size() to derive the values for XSSZ and
XROT in DAVINCI_MCASP_[RT]XFMT_REG from the configured word length
rather than hard-coding them in a switch/case block.

Also, by directly passing the word length to
davinci_config_channel_size(), we can get rid of the
DAVINCI_AUDIO_WORD_* enum.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: McASP: remove unused variables

codec_fmt and sample_rate variables are unused in both snd_platform_data
and davinci_audio_dev, so drop them.

Signed-off-by: Daniel Mack <zonque@gmail.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: omap-twl4030: Update the header file to support more boards

The common machine driver will be able to support new boards where the voice
port is also in use.
At the same time add possibility to fine tune the connections from twl4030
on the board: jack detection GPIO, input/output selection

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: arizona: Log the clock we're setting the DAI to use

Useful for diagnostics.

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

ASoC: arizona: Store the DAI clock ID when setting

So the code to suppress duplicate changes is effective.

Reported-by: Kyung Kwee Ryu <Kyung-Kwee.Ryu@wolfsonmicro.comyu@wolfsonmicro.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: arizona: Make FLL lock timeout very high

Provide robustness against low quality FLL sync clocks by increasing the
timeout for lock to an absurdly high point; we should never get anywhere
near hitting the timeout in a real system unless it is failing.

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

ASoC: wm5110: Add LHPF coefficient configuration

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

ASoC: wm5110: Add EQ coefficient configuration

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

ASoC: wm5102: Make EQ coefficents configurable

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

ASoC: zoom2: Remove HS mux GPIO handling to avoid kernel crash due to BUG_ON()

The machine driver try to use GPIO15 of twl4030 for HS MUX which supposed to
select between TWL's HSOL/R and tlv320aic3254's HPL/R.
The TWL's GPIO allocated dynamically so the (OMAP_MAX_GPIO_LINES + 15) is no
longer valid GPIO number causing a kernel crash due to BUG_ON()
Also the current machine driver supports only TWL audio currently: there is
no need to control the GPIO.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: da7210: Remove unnecessary regmap_exit call

Use of devm_regmap_init_spi does not require an explicit
regmap_exit call.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm5102: Add support for configuring LHPF coefficients

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

ASoC: adsp: Set DSP clock rate to SYSCLK rate

For simplicity always run the DSP at the SYSCLK rate.

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

ASoC: adsp: Keep ADSP2 memory powered off when not in use

Turn off the ADSP memory when we aren't using it, saving a small amount of
power.

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

ASoC: atmel-pcm: dma support based on pcm dmaengine

Specify overrun bit in interrupt mask
Add dmaengine specific routines

Signed-off-by: Nicolas Ferre <nicolas.ferre@atmel.com>
[voice.shen@atmel.com: adapt to soc dmaengine framework]
Signed-off-by: Bo Shen <voice.shen@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: atmel-pcm: split into two file

This patch is split original atmel-pcm.c into new atmel-pcm.c and
atmel-pcm-pdc.c two files. The new atmel-pcm.c is the share routine
while will be used for pdc or dma transfer.

Using SND_ATMEL_SOC_PDC to select using PDC for audio transfer

Signed-off-by: Bo Shen <voice.shen@atmel.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm5102: Correct base address for Y region

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

ASoC: wm8955: Use devm_regmap_init_i2c()

devm_regmap_init_i2c() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: max9768: Use devm_regmap_init_i2c()

devm_regmap_init_i2c() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: arizona: Add parentheses

Some compiler versions complain.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm_adsp: Set the core enable as well as start bit for ADSP2

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

ASoC: wm2200: Remove empty labels

Removed unnecessary labels probably left out of some previous cleanup.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: Samsung: Get I2S src_clk from clock alias id.

As the I2S src clks are registered with clkdev using generic
connection id, driver can get the clk name using generic id.
So the variable representing the array of rclk src clks is
deleted.

Signed-off-by: Padmavathi Venna <padma.v@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: omap-abe-twl6040: Use "ti,jack-detection" DT property as boolean

Handle the jack detection inforamtion as bool from devicetree.

Signed-off-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm5102: Enable DVFS support for the DSP

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

Merge branch 'topic/adsp' of git://git./linux/kernel/git/broonie/sound into asoc-arizona

ASoC: lm49453: Use devm_regmap_init_i2c()

devm_regmap_init_i2c() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: da7210: Use devm_regmap_init_i2c()

devm_regmap_init_i2c() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: ak4535: Use devm_regmap_init_i2c()

devm_regmap_init_i2c() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm8400: Use devm_regulator_bulk_get

devm_regulator_bulk_get() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm8350: Use devm_regulator_bulk_get

devm_regulator_bulk_get() is device managed and makes error
handling and code cleanup simpler.

Signed-off-by: Sachin Kamat <sachin.kamat@linaro.org>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: wm_adsp: Enable DVFS for ADSP2

Some ADSP devices can make use of DVFS to optimise power consumption
depending on the operating frequency of the DSP core. Implement
support for this in the generic ADSP code.

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

mfd: arizona: Register haptics devices

Both WM5102 and WM5110 support haptics, register the device.

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

ASoC: wm5110: Implement haptics support

Provide a haptics widget for use by the haptics driver and expose the DAPM
context for it.

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

ASoC: wm5102: Implement haptics support

Provide a haptics widget for use by the haptics driver and expose the DAPM
context for it.

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

Input - arizona-haptics: Add driver haptics module on Arizona CODECs

The Arizona CODECs contain a haptics module providing vibration feedback
support. Implement basic support for this, providing simple start/stop and
signal magnitude control.

Since the output path for haptics is routed through the CODEC audio routing
it is modelled as a signal generator within ASoC, the haptics driver calls
DAPM to start and stop the output drivers. An appropriate output path must
be configured via ALSA to connect the haptics source to the correct output.

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

mfd: arizona: Allow the CODEC DAPM context to be accessed elsewhere

Some other device functions need to integrate with signal sources in the
audio portion (primarily for haptics) so allow CODEC to export the DAPM
context by pointing to it from the core driver.

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

ASoC: arizona: Record FLL setting when disabling

Otherwise we skip reenables.

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

Merge tag 'v3.7-rc7' into asoc-wm2200

Linux 3.7-rc7

Merge tag 'v3.7-rc7' into asoc-ux500

Linux 3.7-rc7

ASoC: OMAP: mcbsp fixes for enabling ARM multiplatform support

We cannot include any plat or mach headers for the multiplatform
support.

Fix the issue by defining local mcbsp_omap1().

Signed-off-by: Tony Lindgren <tony@atomide.com>
Acked-by: Jarkko Nikula <jarkko.nikula@bitmer.com>
Acked-by: Peter Ujfalusi <peter.ujfalusi@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: fsl: fix miscompilation of snd-soc-imx-pcm

Hi Mark,

thanks your insisting on a better description for the patch, I found a
more appropriate solution for the problem:

Compiling the SoC Audio driver for Freescale i.MX as a module
(CONFIG_SND_SOC_IMX_PCM=m) results in a non-functional sound driver
indicated by the error message:
| imx-sgtl5000 sound.1: platform imx-pcm-audio not registered
| imx-sgtl5000 sound.1: snd_soc_register_card failed (-517)
| platform sound.1: Driver imx-sgtl5000 requests probe deferral
instead of the message:
| imx-sgtl5000 sound.1:  sgtl5000 <-> 63fcc000.ssi mapping ok
that is to be expected upon loading the snd-soc-imx-pcm.ko module.

The build log reveals, that the file imx-pcm-dma.o (or imx-pcm-fiq.o
depending on the kernel configuration), which should be linked
together with imx-pcm.o into snd-imx-pcm.ko, is not being compiled in
this case.

The make rules for these files shows that the target object imx-pcm.o
is assigned to the variable snd-soc-imx-pcm-y while
imx-pcm-{dma,fiq}.o are added to to
snd-soc-imx-pcm-$(CONFIG_SND_SOC_IMX_PCM_DMA) and
snd-soc-imx-pcm-$(CONFIG_SND_SOC_IMX_PCM_FIQ) which resolve to
snd-soc-imx-pcm-m in this case.

According to Documentation/kbuild/modules.txt:
|When the module is built from multiple sources, an additional line is
|needed listing the files:
|
|        <module_name>-y := <src1>.o <src2>.o ...
Thus the type of the config variables CONFIG_SND_SOC_IMX_PCM_DMA and
CONFIG_SND_SOC_IMX_PCM_FIQ should be 'bool' instead of 'tristate' to
resolve to 'y' when selected.

Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: mxs-saif: export mxs_saif_put_mclk,mxs_saif_get_mclk

Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: fsl: fix miscompilation of snd-soc-imx-pcm

Due to a broken make rule, sound/soc/fsl/imx-pcm-dma.c or
sound/soc/fsl/imx-pcm-fiq.c (whatever is selected via Kconfig) will
not be compiled into imx-pcm.o when building as module, i.e.:
CONFIG_SND_SOC_IMX_PCM=m
CONFIG_SND_SOC_IMX_PCM_DMA=m
resulting in a non-functional sound driver.

This gives the error messages:
| imx-sgtl5000 sound.1: platform imx-pcm-audio not registered
| imx-sgtl5000 sound.1: snd_soc_register_card failed (-517)
| platform sound.1: Driver imx-sgtl5000 requests probe deferral
when loading the driver instead of what's to be expected:
| imx-sgtl5000 sound.1:  sgtl5000 <-> 63fcc000.ssi mapping ok

Signed-off-by: Lothar Waßmann <LW@KARO-electronics.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

ASoC: pcm: Report error code when we fail to init platform

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

ASoC: wm0010: Ignore interrupts in power down mode

The device should not be generating interrupts when it does not have power
so ignore incoming interrupts.

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

ASoC: arizona: Suppress noop FLL updates

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

ASoC: wm5102: Add missing routes for ASRC inputs

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

ASoC: wm2000: Get the CODEC pointer from the widget

Don't rely on the kcontrol for robustness reasons, the widget mechanism
is what the framework uses.

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

ASoC: arizona: Support higher clock rates

Some devices support higher clock rates, allow users to select these.

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

Linux 3.7-rc7

Merge branch 'merge' of git://git./linux/kernel/git/benh/powerpc

Pull powerpc EEH bugfixes from Benjamin Herrenschmidt.

Two one-liner fixes for the new EEH code.

* 'merge' of git://git.kernel.org/pub/scm/linux/kernel/git/benh/powerpc:
  powerpc/eeh: Do not invalidate PE properly
  powerpc/pseries: Fix oops with MSIs when missing EEH PEs

Merge branch 'upstream' of git://git.linux-mips.org/ralf/upstream-linus

Pull MIPS fixes from Ralf Baechle:
 "Three issues fixed accross the field:

   - Some functions that were recently outlined as part of a preemption
     fix were causing problems with function tracing.
   - The recently merged in-kernel MPI library uses very outdated
     headers that contain MIPS-specific code which won't build on with
     gcc 4.4 or newer.
   - The MIPS non-NUMA memory initialization was making only a very
     half-baked attempt at merging adjacent memory ranges.  This kept
     the code simple enough but is now causing issues with kexec."

* 'upstream' of git://git.linux-mips.org/pub/scm/ralf/upstream-linus:
  MPI: Fix compilation on MIPS with GCC 4.4 and newer
  MIPS: Fix crash that occurs when function tracing is enabled
  MIPS: Merge overlapping bootmem ranges

powerpc/eeh: Do not invalidate PE properly

While the EEH does recovery on the specific PE that has PCI errors,
the PCI devices belonging to the PE will be removed and the PE will
be marked as invalid since we still need the information stored in
the PE. We only invalidate the PE when it doesn't have associated
EEH devices and valid child PEs. However, the code used to check
that is wrong. The patch fixes that.

Signed-off-by: Gavin Shan <shangw@linux.vnet.ibm.com>
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>