Some volatile registers accidentally had defaults added, though there were
no harmful side effects other than increased memory consumption.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Reported by Andrew Morton <akpm@linux-foundation.org>, with
i386 allmodconfig:

ERROR: "pm8606_osc_disable" [drivers/video/backlight/88pm860x_bl.ko] undefined!
ERROR: "pm8606_osc_enable" [drivers/video/backlight/88pm860x_bl.ko] undefined!
ERROR: "pm8606_osc_disable" [drivers/leds/leds-88pm860x.ko] undefined!
ERROR: "pm8606_osc_enable" [drivers/leds/leds-88pm860x.ko] undefined!
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This patch allows to check if the other core is in WFI
mode. It is the last check the idle routine has to do before
entering into the retention state.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This patch allows to check if there are some pending irqs
on the prcmu.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

In the case we go to the retention mode, we decoupled the gic
in order to have the A9 core to reach a stable WFI state.
But we want the prcmu to wake up the A9 when the gic has a pending
irq which is done by copying the gic settings to the to the prcmu.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This patch introduces a routine to check if there are some
irqs pending on the gic. Usually this check is not relevant because
it appears racy (an irq can arrive right after this check), but in
the ux500 it makes sense because the prcmu decouples the gic from
the A9 cores.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

I missed in my previous patch the A9_MASK_REQ[_MASK] were
already defined. Let's remove the duplicate definitions.

The PRCMU_GIC_DELAY macro could be removed as it is not
really useful here.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Also remove a redundant semicolon from return statement.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Since several sub modules such as backlight, leds and vibrator depend
on the refernce group and internal oscillator in pm8606, so modified
the power control of led in pm8606 by unified interface.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Since several sub modules such as backlight, leds and vibrator depend
on the refernce group and internal oscillator in pm8606, so modified
the power control of backlight in pm8606 by unified interface.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Cc: Richard Purdie <rpurdie@rpsys.net>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Collect all the sub-devices of pm8606 initialization into
device_pm8606_init() for code clean up.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The reference group and internal oscillator are shared by sub-devs
like led, backlight and vibrator in PM8606 chip. Now introduce a
voting mechanism to enable/disable it.

Add pm8606_osc_enable() and pm8606_osc_disable() interface and
related defines to support this. This interface will be called by
vibrator led and backlight driver.The refernce group and internal
oscillator are enabled only when at least one of it's clients holds
it on or disabled only all the clients don't use it any more based
on the above mechanism.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Signed-off-by: NHai Dong <hai.dong@polycom.com>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Include linux/mfd/pcf50633/mbc.h instead of duplicating the same defines.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Include linux/mfd/pcf50633/pmic.h instead of duplicating the same defines.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

It should be marked as readable but wasn't, breaking DC servo operation.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Using regmap apis for accessing the device registers and
using RBTREE caching mechanims for caching registers.

Enabling caching of the registers which is used for voltage
controls. By doing this, the modify_bits operation is faster as
it does not involve the i2c register read from device, just read
from cache. This results faster set voltage operation.
Signed-off-by: NLaxman Dewangan <ldewangan@nvidia.com>
Reviewed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

For 88pm860x pmic, it can wake the system from low power mode by irq,
its sub-devs like RTC and onkey can be enabled for this usage.
Signed-off-by: NJett.Zhou <jtzhou@marvell.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

twl6030_irq_set_wake is not used anywhere else in the kernel
except as irq_chip.irq_set_wake. No reason for it to be exported.

Also fixes build warning:
drivers/mfd/twl6030-irq.c:230:5: warning: symbol 'twl6030_irq_set_wake' was not declared. Should it be static?
Signed-off-by: NNishanth Menon <nm@ti.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

TWL6030 family of PMIC use a shadow interrupt status register
while kernel processes the current interrupt event.
However, any write(0 or 1) to register INT_STS_A, INT_STS_B or
INT_STS_C clears all 3 interrupt status registers.

Since clear of the interrupt is done on 32k clk, depending on I2C
bus speed, we could in-adverently clear the status of a interrupt
status pending on shadow register in the current implementation.
This is due to the fact that multi-byte i2c write operation into
three seperate status register could result in multiple load
and clear of status and result in lost interrupts.

Instead, doing a single byte write to INT_STS_A register with 0x0
will clear all three interrupt status registers without the related
risk.
Acked-by: NSantosh Shilimkar <santosh.shilimkar@ti.com>
Signed-off-by: NNishanth Menon <nm@ti.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

da9052 has been converted to use regmap API, so we can remove the unused
io_lock mutex.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This patch allows to decouple and recouple the gic from the PRCMU.
This is needed to put the A9 core in retention mode with the cpuidle
driver.

It is based on top of the "DB8500 PRCMU update" patchset.
Signed-off-by: NDaniel Lezcano <daniel.lezcano@linaro.org>
Acked-by: NRickard Andersson <rickard.andersson@stericsson.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

irq_base of the tps65912 irq platform data should be
initialized with the board provided irq_base data.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The code is currently always checking the first resource of every
device only (several times.) This has been broken since the ACPI check
was added in February 2010 in commit
91fedede.

Fix the check to run on each resource individually, once.
Signed-off-by: NJean Delvare <khali@linux-fr.org>
Cc: stable@vger.kernel.org
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Fixes warnings and needed for correctness.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Acked-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

There could be possible race condition where the host dev's
are alloced and added to platform dev just before usbhs_init.
Just move usbhs_init before adding child dev.

CC: Alan Stern <stern@rowland.harvard.edu>
Signed-off-by: NGovindraj.R <govindraj.raja@ti.com>
Acked-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Remove the hardcoded magic values for dma mask and use the dma
mask api/macro available.
Signed-off-by: NGovindraj.R <govindraj.raja@ti.com>
Reviewed-by: NFelipe Balbi <balbi@ti.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The callers of pcf50633_write_block assume pcf50633_write_block return 0 on
success, thus make it return 0 instead of the number of registers written on
success.

Currently pcf50633_write_block is called in drivers/mfd/pcf50633-irq.c and
drivers/rtc/rtc-pcf50633.c.
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Acked-by: NHarald Welte <laforge@gnumonks.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The val_len parameters for regmap_raw_write should be "count * sizeof(u8)"
(or simply "count") instead of "count * sizeof(u16)".
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The first parameter should be "number of elements" and the second parameter
should be "element size".
Signed-off-by: NAxel Lin <axel.lin@gmail.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The AB9540 variant of the AB8500 is basically close enough
to use the same driver. This adds the new registers and
deviations for this new chip variant.
Reviewed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NMaxime Coquelin <maxime.coquelin@stericsson.com>
Signed-off-by: NAlex Macro <alex.macro@stericsson.com>
Signed-off-by: NMichel Jaouen <michel.jaouen@stericsson.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This makes the AB8500 state struct contain the IRQ mask and
register offsets previously hard-coded so as to make room for
more AB8500 variants.
Reviewed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NMaxime Coquelin <maxime.coquelin@stericsson.com>
Signed-off-by: NAlex Macro <alex.macro@stericsson.com>
Signed-off-by: NMichel Jaouen <michel.jaouen@stericsson.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

There are currently four different versions of the AB8500
around: AB8500, AB8505, AB9540 and AB8540. Unfortunately:

- Some of the chips (AB8500, AB8505, AB9540) cannot read
  the AB8500_REV_REG register but return errors

- Some of them have the same ID value in the hardware
  register AB8500_REV_REV, for example the first versions
  of AB8505 and AB9540 have 0xFF in this register -
  just like the AB8500.

So we need to be able to enforce a certain version from
the platform. We do this by using the id of the platform
device that provides the read/write functions.
Reviewed-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NMaxime Coquelin <maxime.coquelin@stericsson.com>
Signed-off-by: NAlex Macro <alex.macro@stericsson.com>
Signed-off-by: NMichel Jaouen <michel.jaouen@stericsson.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

We were not freeing the irq properly in the error path in
the AB8500 driver.

Cc: Mark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NMaxime Coquelin <maxime.coquelin@stericsson.com>
Signed-off-by: NAlex Macro <alex.macro@stericsson.com>
Signed-off-by: NMichel Jaouen <michel.jaouen@stericsson.com>
Signed-off-by: NLinus Walleij <linus.walleij@linaro.org>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

The jack detection on WM1811 is often required during system suspend, add
it as another check when deciding if we should suspend.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Cc: stable@vger.kernel.org
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

This is the default state that the runtime PM infrastructure expects so
instead just kick the runtime PM core to suspend us if we're not doing
anything (as is default).
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>

Omitted in the regmap conversion.
Signed-off-by: NMark Brown <broonie@opensource.wolfsonmicro.com>
Signed-off-by: NSamuel Ortiz <sameo@linux.intel.com>