This is required since commit f7df20ec
"regulator: core: Use list_voltage() to read single voltage regulators",
otherwise _regulator_get_voltage returns rdev->desc->ops->list_voltage(rdev, 0).

The Maxim 1586 controls V3 and V6 voltages, but offers no way of reading back
the set up value. Thus this patch caches the setting when setting new voltage.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Simply use devm_gpio_request_one() instead.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

To clear the mask bit, setting data argument to be 0 with proper mask setting
for lp8788_update_bits. We don't need the var array here.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Milo(Woogyom) Kim <milo.kim@ti.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Use of_match_ptr and add ifdef CONFIG_OF guard for regulator_gpio_of_match.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Introduce a regulator_can_change_voltage() function for the subsytems or
drivers which might check if applying voltage change is possible and use
special workaround code when the driver is used with fixed regulators or
regulators with disabled ability to change the voltage.
Signed-off-by: Marek Szyprowski <m.szyprowski@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

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

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

Helps with figuring out when things went wrong.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Some Arizona device support a 1.8V output mode. Enable this in the driver.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The number of voltage slots supported by the LDO is 7, not 6.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Some Arizona devices such as the WM5102 can use DVFS on their digital core,
for these devices allow the voltage range to vary in the default setup.
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

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

Below equation means the "voltage" is the "smallest" voltage within specific
range.

ret = DIV_ROUND_UP(min - bottom) / step;
voltage = ret * step + bottom;

If we do try 1 down when (voltage > max), new voltage is then less than min
voltage. Which means the new voltage is not in the requested voltage range.

This patch also includes below cleanups:
- Use DIV_ROUND_UP
- rename variable 'ret' to 'sel' for better readability because as3711_sel_check
  returns the selector.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Below cases are supposed to be valid:

min_uV == max_uV == info->max_uV
min_uV == max_uV == rdev->desc->min_uV

Don't return -EINVAL for above cases.

This patch also includes below cleanups:
- Use rdev_get_drvdata(rdev) instead of rdev->reg_data.
- Remove unnecessary WARN_ON, it looks pointless.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Acked-by: Guennadi Liakhovetski <g.liakhovetski@gmx.de>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Checking da9052_current_limits[row][i] <= max_uA is not enough, it is possible
da9052_current_limits[row][i] may less than the requested min_uA.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Selecting the minimal value is only true for voltage regulators.
For current regulators the maximum in the given range should be
selected instead.

This issue was reported by Heiko Stuebner for gpio-regulator driver [1],
and the conclusion is to select the max current for current regulators [2].

[1] https://lkml.org/lkml/2012/8/5/162
[2] https://lkml.org/lkml/2012/8/6/183

This patch also ensures da9055_buck_set_current_limit return -EINVAL when the
supported current limit does not meet the request range.
Signed-off-by: Axel Lin <axel.lin@ingics.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

Merge branch 'topic/hotplug' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/regulator into regulator-max8997

Conflicts:
	drivers/regulator/max8997.c

Add device tree based discovery support for max8997.
Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
Acked-by: MyungJoo Ham <myungjoo.ham@samsung.com>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

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

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

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>

Pull sound build error fix from Takashi Iwai:
 "Only a single commit for fixing the build error without CONFIG_PM in
  hda driver."

* tag 'sound-3.7' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound:
  ALSA: hda - Fix build without CONFIG_PM

In case the gpio based volatage selection mode is not used for either of
buck 1/2/5, then only the BUCKxDVS1 register need to be programmed. So
determine whether dvs mode is used and limit the loop count appropriately.
Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

The BUCKxDVSx register programming is now moved prior to setting up of the
gpio based dvs mode. This will ensure that all the BUCKxDVSx registers
are programmed with appropriate voltage values before the gpio based dvs
mode is selected for buck1/2/5.
Signed-off-by: Thomas Abraham <thomas.abraham@linaro.org>
Reviewed-by: Tomasz Figa <t.figa@samsung.com>
Signed-off-by: Mark Brown <broonie@opensource.wolfsonmicro.com>

I forgot this again...  codec->in_pm is in #ifdef CONFIG_PM
Reported-by: Markus Trippelsdorf <markus@trippelsdorf.de>
Signed-off-by: Takashi Iwai <tiwai@suse.de>