When legacy devices are present on x86 machine, the RTC IRQ has
a dedicated pre-defined value. Use it instead of hard coded number.

Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Guilherme G. Piccoli <gpiccoli@canonical.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20200123131437.28157-2-andriy.shevchenko@linux.intel.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

As reported by Guilherme G. Piccoli:

---8<---8<---8<---

The rtc-cmos interrupt setting was changed in the commit 079062b2
("rtc: cmos: prevent kernel warning on IRQ flags mismatch") in order
to allow shared interrupts; according to that commit's description,
some machine got kernel warnings due to the interrupt line being shared
between rtc-cmos and other hardware, and rtc-cmos didn't allow IRQ sharing
that time.

After the aforementioned commit though it was observed a huge increase
in lost HPET interrupts in some systems, observed through the following
kernel message:

[...] hpet1: lost 35 rtc interrupts

After investigation, it was narrowed down to the shared interrupts
usage when having the kernel option "irqpoll" enabled. In this case,
all IRQ handlers are called for non-timer interrupts, if such handlers
are setup in shared IRQ lines. The rtc-cmos IRQ handler could be set to
hpet_rtc_interrupt(), which will produce the kernel "lost interrupts"
message after doing work - lots of readl/writel to HPET registers, which
are known to be slow.

Although "irqpoll" is not a default kernel option, it's used in some contexts,
one being the kdump kernel (which is an already "impaired" kernel usually
running with 1 CPU available), so the performance burden could be considerable.
Also, the same issue would happen (in a shorter extent though) when using
"irqfixup" kernel option.

In a quick experiment, a virtual machine with uptime of 2 minutes produced
>300 calls to hpet_rtc_interrupt() when "irqpoll" was set, whereas without
sharing interrupts this number reduced to 1 interrupt. Machines with more
hardware than a VM should generate even more unnecessary HPET interrupts
in this scenario.

---8<---8<---8<---

After looking into the rtc-cmos driver history and DSDT table from
the Microsoft Surface 3, we may notice that Hans de Goede submitted
a correct fix (see dependency below). Thus, we simply revert
the culprit commit.

Fixes: 079062b2 ("rtc: cmos: prevent kernel warning on IRQ flags mismatch")
Depends-on: a1e23a42 ("rtc: cmos: Do not assume irq 8 for rtc when there are no legacy irqs")
Reported-by: Guilherme G. Piccoli <gpiccoli@canonical.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Tested-by: Guilherme G. Piccoli <gpiccoli@canonical.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20200123131437.28157-1-andriy.shevchenko@linux.intel.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The IRQ_NOAUTOEN flag tells interrupt core that interrupt shall not be
auto-enabled at the time of requesting interrupt. This is a minor clean-up
change that doesn't fix any problems.
Signed-off-by: Dmitry Osipenko <digetx@gmail.com>
Link: https://lore.kernel.org/r/20200106015615.12602-1-digetx@gmail.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Use FIELD_PREP and FIELD_GET instead of hardcoding already defined values.

Link: https://lore.kernel.org/r/20191229204421.337612-6-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The current rtc time is read out in at91_rtc_setalarm but it it immediately
overwritten by the alarm time. Remove the readout.

Link: https://lore.kernel.org/r/20191229204421.337612-5-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The header was simply moved from the arm mach folder to drivers/rtc but
there is not point in having it separated from the driver.

Also remove unused bit definitions and use BIT and GENMASK.

Link: https://lore.kernel.org/r/20191229204421.337612-4-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Both the sama5d4 and sama5d2 RTCs have more features than the previous
RTCs, add a compatible string for them.

Link: https://lore.kernel.org/r/20191229204421.337612-3-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Convert Real Time Clock for Atmel/Microchip SoCs bindings documentation
to json-schema.

Cc: Rob Herring <robh+dt@kernel.org>
Link: https://lore.kernel.org/r/20191229204421.337612-2-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The RTC procfs interface is deprecated and hasn't been used by userspace
for years.

Link: https://lore.kernel.org/r/20191229204421.337612-1-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Add microchip,sam9x60-rtc to DT bindings documentation.
Signed-off-by: Claudiu Beznea <claudiu.beznea@microchip.com>
Acked-by: Rob Herring <robh@kernel.org>
Link: https://lore.kernel.org/r/1579085987-13976-2-git-send-email-claudiu.beznea@microchip.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Replace (1 << ...) with BIT().

CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Link: https://lore.kernel.org/r/20191218082553.3309554-1-iwamatsu@nigauri.orgSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Use SPDX-License-Identifier instead of a verbose license text.

CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Alexandre Belloni <alexandre.belloni@bootlin.com>
CC: Jonas Jensen <jonas.jensen@gmail.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Link: https://lore.kernel.org/r/20191218082836.3309808-1-iwamatsu@nigauri.orgSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

struct spi_device in struct ds1343_priv is not used, remove it.

CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Alexandre Belloni <alexandre.belloni@bootlin.com>
CC: Raghavendra Chandra Ganiga <ravi23ganiga@gmail.com>
CC: Ankur Srivastava <sankurece@gmail.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Link: https://lore.kernel.org/r/20191218081917.3308211-1-iwamatsu@nigauri.orgSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

struct i2c_client can be referenced from the device structure, so this
doesn't need to have it in struct rx8025_data.
Remove struct i2c_client from struct rx8025_data.

CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Alexandre Belloni <alexandre.belloni@bootlin.com>
CC: Wolfgang Grandegger <wg@grandegger.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Link: https://lore.kernel.org/r/20191218081624.3307752-1-iwamatsu@nigauri.orgSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The RTC has a valid bit in the seconds register that indicates whether
power was lost since the pevious time set. This bit is currently read
once at probe time, cached and updated with set_time.

Howeever, caching the bit may prevent detecting power loss at runtime
(which can happen if the RTC's supply is distinct from the the platform's).

Writing the seconds register when setting time will clear the bit,
so there should be no downside in reading the bit directly instead of
caching it.
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Link: https://lore.kernel.org/r/20191212153111.966923-2-paul.kocialkowski@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The current code returns -EPERM when the voltage loss bit is set.
Since the bit indicates that the time value is not valid, return
-EINVAL instead, which is the appropriate error code for this
situation.

Fixes: dcaf0384 ("rtc: add hym8563 rtc-driver")
Signed-off-by: Paul Kocialkowski <paul.kocialkowski@bootlin.com>
Link: https://lore.kernel.org/r/20191212153111.966923-1-paul.kocialkowski@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

In the current code, the return value of devm_request_threaded_irq may be
returned. This fixes it.

CC: Alessandro Zummo <a.zummo@towertech.it>
CC: Alexandre Belloni <alexandre.belloni@bootlin.com>
CC: Akshay Bhat <akshay.bhat@timesys.com>
Signed-off-by: Nobuhiro Iwamatsu <iwamatsu@nigauri.org>
Link: https://lore.kernel.org/r/20191217121231.2698817-2-iwamatsu@nigauri.orgSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Fixes coccicheck warning:

drivers/rtc/rtc-omap.c:619:2-3: Unneeded semicolon
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: zhengbin <zhengbin13@huawei.com>
Link: https://lore.kernel.org/r/1576466353-45384-1-git-send-email-zhengbin13@huawei.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The driver misses a check for devm_clk_get().
Add the check to fix it.
Signed-off-by: Chuhong Yuan <hslester96@gmail.com>
Link: https://lore.kernel.org/r/20191214074528.16806-1-hslester96@gmail.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Remove redundant messages or messages that would not add any value because
the information is already conveyed properly using errno.

Link: https://lore.kernel.org/r/20191223100108.1083078-1-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rv30{2,4}9_register_driver and rv30{2,4}9_unregister_driver are only called
from the init and exit functions of the module. Annotate them properly.

Link: https://lore.kernel.org/r/20191223101430.1091572-1-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The compatibles have been marked obsolete for more that 2 years, drop them
now. Note that this doesn't currently prevent the driver from probing
because the i2c core will still match using the i2c_device_id table.

Link: https://lore.kernel.org/r/20191212230239.65784-1-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Export the 8 byte RAM using nvmem.

Link: https://lore.kernel.org/r/20191214221022.622482-17-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Despite the comment, the RV3029 uses a 7bit BCD register for the year,
making 2079 the last supported year.

Link: https://lore.kernel.org/r/20191214221022.622482-14-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

This allows further improvement of the driver.

Link: https://lore.kernel.org/r/20191214221022.622482-13-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

In case the data is invalid (PON or VLOW2 are set in STATUS, explicitly
tell userspace that the time is invalid. Only remove VLOW2 when setting a
new valid time.

Link: https://lore.kernel.org/r/20191214221022.622482-12-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The RV3029 can report three different conditions: power on, voltage dropped
and data is lost and voltage is low and temperature compensation has
stopped. The first two conditions amount to the same status, the RTC data
is invalid.

VLOW1 has to be cleared manually to resume temperature compensation, this
is achieved using RTC_VL_CLEAR.

Link: https://lore.kernel.org/r/20191214221022.622482-11-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rv3029_read_regs and rv3029_write_regs are simply calling
regmap_bulk_{read,write}. Drop them and call regmap_{,bulk}_{read,write} as
appropriate.

Link: https://lore.kernel.org/r/20191214221022.622482-10-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

It is unecessay to test alarm->enabled before calling
rv3029_alarm_irq_enable.

Link: https://lore.kernel.org/r/20191214221022.622482-9-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Use regmap_update_bits instead of open coding it in
rv3029_alarm_irq_enable.

Link: https://lore.kernel.org/r/20191214221022.622482-8-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

There is no point in having 2 indirections before calling regmap_read,
especially since rv3029_get_sr also changes the return value without any
good reason. Call regmap_read directly.

Link: https://lore.kernel.org/r/20191214221022.622482-7-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

RV3029_STATUS is read in multiple location but its value is never used
afterwards. Avoid this register access when not necessary.

Link: https://lore.kernel.org/r/20191214221022.622482-6-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

There is no lock preventing concurrent access to the status register from
bth the rtc subsystem and the hwmon subsystem. Use regmap_update_bits to
ensure updating RV3029_STATUS is properly locked.

Link: https://lore.kernel.org/r/20191214221022.622482-5-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

rv3029_update_bits open codes regmap_update_bits and forgets to properly
lock the register range while doing so. Use regmap_update_bits instead.

Link: https://lore.kernel.org/r/20191214221022.622482-4-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Instead of trying to validate the accessed registers in custom functions,
let regmap handle that. This allows to defines all the holes in the
register range and gives access to the regmap debugfs register dump.

Link: https://lore.kernel.org/r/20191214221022.622482-3-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

The correct name for the rtc is rv3029. c2 is actually the package form
factor rv3029c3 exists and simply has a smaller form factor.

Link: https://lore.kernel.org/r/20191214221022.622482-2-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

RX8010_FLAG_VLF means the voltage dropped too low and data has been lost.

Link: https://lore.kernel.org/r/20191214220259.621996-18-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Remove RTC_VL_CLR handling because it is a disservice to userspace as it
removes the important information that the RTC data is invalid. This may
lead userspace to set an invalid system time later on.

Link: https://lore.kernel.org/r/20191214220259.621996-17-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

RV8803_FLAG_V1F means the voltage is too low to keep the temperature
compensation running and the accuracy of the RTC is affected.

RV8803_FLAG_V2F means the voltage dropped so low that data is now invalid.

Link: https://lore.kernel.org/r/20191214220259.621996-16-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>

Stop clearing RV8803_FLAG_V2F in RTC_VL_CLR because it is a disservice to
userspace as it removes the important information that the RTC data is
invalid. This may lead userspace to set an invalid system time later on.

Link: https://lore.kernel.org/r/20191214220259.621996-15-alexandre.belloni@bootlin.comSigned-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>