Use the BIT macro were possbiel.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

u32 should be used instead of uint32_t
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Use sizeof where possible to ensure we don't read/write more than the
allocated buffer.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

ds1307->regs is never used before being read or initialized locally. There
is no point in keeping a copy in memory.

Also limit the size of the read buffer to what is really used, rename buf
to regs for consistency and use sizeof() where possible.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Based on QNAP's arch/arm/mach-rtk119x/driver/rtk_rtc_drv.c code and
mach-rtk119x/driver/dc2vo/fpga/include/mis_reg.h register definitions.

The base year 2014 was observed on all of Zidoo X9S, ProBox2 Ava and
Beelink Lake I.
Signed-off-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The RTC can output its 32kHz clock outside of the SoC, for example to clock
a WiFi chip.

Create a new clock that other devices will be able to retrieve, while
maintaining the DT stability by providing a default name for that clock if
clock-output-names doesn't list one.
Signed-off-by: NMaxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Add device driver for a virtual RTC device in Android emulator.

The compatible string used by OS for binding the driver is defined
as "google,goldfish-rtc".
Signed-off-by: NMiodrag Dinic <miodrag.dinic@imgtec.com>
Signed-off-by: NGoran Ferenc <goran.ferenc@imgtec.com>
Signed-off-by: NAleksandar Markovic <aleksandar.markovic@imgtec.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

This adds support for reading and writing date/time from/to ds1341 chip.

ds1341 chip has other features - alarms, input clock (can be used instead
of intercal oscillator for better accuracy), output clock ("square wave
generation"). However, not all of that is available at the same time.
Same chip pins, CLKIN/nINTA and SQW/nINTB, can be used either for
input/output clocks, or for alarm interrupts. Role of these pins on
particular board depends on hardware wiring.

We can add device tree properties that describe if each of pins is wired
as clock, or as interrupt, or left unconnected, and enable support for
corresponding functionality based on that. But that is cumbersome, requires
hardware for testing, and has to deal with bit enabling/disabling output
clock also affects which pins alarm interrupts are routed to.

Another factor is that there are hardware setups (i.e. ZII RDU2) that
power DS1341 from SuperCap, which makes power saving critical. For such
setups, kernel driver should leave register bits that control mentioned
pins in the state configured by bootloader.

Given all that, it was decided to limit support to "only date/time" for
now. That is enough for common use case. Full (and cumbersome)
implementation can be added later if ever needed.
Signed-off-by: NNikita Yushchenko <nikita.yoush@cogentembedded.com>
Reviewed-by: NLinus Walleij <linus.walleij@linaro.org>
Tested-by: NAleksander Morgado <aleksander@aleksander.es>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Remove member nvram_offset from struct ds1307 and use the value stored
in struct chip_desc directly.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Factor out offset to struct chip_desc and remove it from struct ds1307.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Factor out rtc_ops to struct chip_desc and use ds13xx_rtc_ops as default.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Factor out irq_handler to struct chip_desc and use ds1307_irq as default.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Change the usage of variable want_irq to reflect its name. Don't set
it to true in case wakeup is enabled but no interrupt number is given.
In addition set variable ds1307_can_wakeup_device if chip->alarm
is set only.
This allows to simplify the code and make it better understandable.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Constify struct chip_desc variables.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Instead of storing the trickle_charger_setup value in struct chip_desc
we can let function ds1307_trickle_init return it because it's used
in the probe function only.
This allows us to constify struct chip_desc variables in a next step.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Factor out the bbsqi bit to struct chip_desc.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The irq number is used in the probe function only, so we don't have
to store it in struct ds1307.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The RK808 and RK805 PMICs are using a similar register map.
We can reuse the rtc driver for the RK805 PMIC. So let's add
the RK805 in the Kconfig description.
Signed-off-by: NElaine Zhang <zhangqing@rock-chips.com>
Signed-off-by: NJoseph Chen <chenjh@rock-chips.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

i2c_device_id are not supposed to change at runtime. All functions
working with i2c_device_id provided by <linux/i2c.h> work with
const i2c_device_id. So mark the non-const structs as const.
Signed-off-by: NArvind Yadav <arvind.yadav.cs@gmail.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

There are no driver left using .open and .release. There is no good use
case for them as there is nothing the character device interface does that
should not be done in the sysfs interface or in-kernel interface.

Remove those callbacks now to avoid future confusion.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Currently, the IRQs are disabled when the rtc character device is closed.
This means that the device needs to stay open to get alarms while the usual
use case will open the device, set the alarm and close the device as is
done in rtcwake.

Keep the alarm functional on character device release so the platform can
actually wakeup.
Reviewed-by: NFabio Estevam <fabio.estevam@nxp.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Currently, the IRQs are disabled when the rtc character device is closed.
This means that the device needs to stay open to get alarms while the usual
use case will open the device, set the alarm and close the device.

Keep the alarms functional on character device release. Note that the PIE
are never enabled and would anyway be disabled by the core.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Currently, the driver unregisters the IRQs when the rtc character device is
closed. This means that the device needs to stay open to get alarms while
the usual use case will open the device, set the alarm and close the
device.

Move the IRQ requests to sa1100_rtc_probe() and use the devm managed
versions so we don't need to free them.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

In the error path of sa1100_rtc_open(), info->clk is disabled which will
happen again in sa1100_rtc_remove() when the module is removed whereas it
is only enabled once in sa1100_rtc_init().

Fixes: 0cc0c38e ("drivers/rtc/rtc-sa1100.c: move clock enable/disable to probe/remove")
Acked-by: NRobert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

pxa_rtc_open() registers the interrupt handler which will access the RTC
registers. However, pxa_rtc_open() is called before the register range is
ioremapped. Instead, call it after devm_ioremap().
Acked-by: NRobert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The last remaining sysfs attribute is undocumented and useless as it can
only be used to debug the driver. Remove it.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Don't require an IRQ if the wakeup-source device-tree property is present.
Signed-off-by: NEric Cooper <ecc@cmu.edu>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Currently, the driver unregisters the IRQs when the rtc character device is
closed. This means that the device needs to stay open to get alarms while
the usual use case will open the device, set the alarm and close the
device.

Move the IRQ requests to puv3_rtc_probe() and use the devm managed versions
so we don't need to free them.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Use managed RTC device allocation as this allows for further cleanup.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

The check on ret < 0 is redundant as the goto destination is the
next statment.  Remove this redudant check and goto.

Detected by CoverityScan, CID#1268785 ("Identical code for different
branches")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

There are two error return paths that do not kfree clk_data and
we end up with a memory leak. Fix these with a kfree error exit
path.

Detected by CoverityScan, CID#1402959 ("Resource Leak")
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Remove double init of spinlock in sun6i_rtc_clk_init()

Fixes: 847b8bf62eb4 ("rtc: sun6i: Expose the 32kHz oscillator")
Signed-off-by: NAlexey Klimov <alexey.klimov@arm.com>
Reviewed-by: NChen-Yu Tsai <wens@csie.org>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Commit 8b44f5be ("ARM: dts: armada: replace isil,irq2-can-wakeup-machine with wakeup-source property")
removed the last usage of "isil,irq2-can-wakeup-machine" almost
two years ago. So I think we can get rid of supporting this
legacy binding.
Signed-off-by: NHeiner Kallweit <hkallweit1@gmail.com>
Acked-by: NRob Herring <robh@kernel.org>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Implements RTC_VL_READ and RTC_VL_CLR ioctls.
Signed-off-by: NFabien Lahoudere <fabien.lahoudere@collabora.co.uk>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

If RTC time have been altered by low voltage, we notify users
that RTC time is invalid by returning -EINVAL.
The RTC time needs to be set correctly to clear the invalid flag.
Signed-off-by: NFabien Lahoudere <fabien.lahoudere@collabora.co.uk>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Make driver use u64 variables and functions to be sure that
it will support dates after year 2038.
Signed-off-by: NBenjamin Gaignard <benjamin.gaignard@linaro.org>
Acked-by: NPatrice Chotard <patrice.chotard@st.com>
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

ds1307_remove() is now empty, remove it
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

Instead of adding a binary sysfs attribute from the driver (which suffers
from a race condition as the attribute appears after the device), use the
core to register an nvmem device.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

This removes a possible race condition and crash and allows for further
improvement of the driver.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>

rv8803_remove() is now empty, remove it.
Signed-off-by: NAlexandre Belloni <alexandre.belloni@free-electrons.com>