RTC core never calls rv8803_set_alarm with an invalid alarm time,
so if an invalid alarm time > 0 is set, external factors must have
corrupted the RTC's alarm time and possibly other registers.
Play it safe by marking the date/time invalid, so all registers are
reinitialized on a ->set_time.
This may cause existing setups to lose time if they so far set only
date/time, but ignored that the alarm registers had an invalid date
value, e.g.:
rtc rtc0: invalid alarm value: 2020-3-27 7:82:0
These systems will have their ->get_time return -EINVAL till
->set_time initializes the alarm value (and sets a new time).
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Link: https://lore.kernel.org/r/20221123095527.2771434-3-s.hauer@pengutronix.de
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Allow the IRQ type to be passed from the device tree if available as there
may be components changing the trigger type of the interrupt between the
RTC and the IRQ controller.
Link: https://lore.kernel.org/r/20230123200217.1236011-12-alexandre.belloni@bootlin.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
.probe_new() doesn't get the i2c_device_id * parameter, so determine
that explicitly in .probe(). The device_id array has to move up for that
to work.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Link: https://lore.kernel.org/r/20221021130706.178687-9-u.kleine-koenig@pengutronix.de
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Add the missing unlock before return from function rv8803_set_time()
in the error handling case.
Fixes: f8176e0bb8 ("rtc: rv8803: initialize registers on post-probe voltage loss")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Yang Yingliang <yangyingliang@huawei.com>
Reviewed-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20220627080822.3881072-1-yangyingliang@huawei.com
The reference manuals of both the RX8803 and RV8803 dictate that
"[On V2F/VLF = ] all registers must be initialized".
The RV-8803 application manual (rev. 1.6) further specifies that crossing
V_LOW2 threshold enables flag V2F and triggers a Power-On reset.
According to table 3.11 in the document, all control registers are
defined to sensible values.
However, The Epson RX-8803 doesn't offer the same guarantees.
It explicitly states:
During the initial power-up, the TEST bit is reset to "0" and the VLF
bit is set to "1".
∗ At this point, all other register values are _undefined_, so be sure to
perform a reset before using the module.
Commit d3700b6b64 ("rtc: rv8803: Stop the clock while setting the time")
also had this rationale:
Indeed, all the registers must be initialized if the voltage has been
lower than VLOW2 (triggering V2F), but not low enough to trigger a POR.
We should follow the advice and initialize all applicable registers.
We can group the registers into 3 groups:
A) Already correctly handled registers:
* 0B-0Ch | Timer Counter | unused and disabled by clearing TE in 0Dh
* 0Dh | Extension Reg | already initialized in rv8803_regs_configure
* 0Eh | Flag Reg | handled in IRQ handler, except for VLF, VDET
* 0Eh | VLF, VDET | cleared in ->set_time
* 10h | 100th Seconds | Already reset via RESET bit
* 20-21h | Capture Buffer | holds timestamp unused by driver
* 2Fh | Event Control | resets automatically
B) Registers that are hardware initialized on POR, but not on VLF:
* 0Fh | Control Reg
* 2Ch | OSC Offset
C) RAM that is undefined on voltage loss:
* 00-06h | Date/Time
* 07h | RAM
* 08-0Ah | Alarm
This means we should initialize after VLF the registers in group B
(RV8803_CTRL and RV8803_OSC_OFFSET).
Group C is all-zero after voltage loss on the RV-8803, but undefined on
the RX-8803. This is ok for Date/Time because ->get_time returns an
error code for as long as the voltage loss flag is active. It's cleared
on ->set_time however. Zeroing both RAM and alarm ensures a fixed value
is read afterwards.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20220426071056.1187235-4-s.hauer@pengutronix.de
The driver probe currently initializes some registers to non-POR
values. These values are not reinstated if the RTC experiences voltage
loss later on. Fix this.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20220426071056.1187235-3-s.hauer@pengutronix.de
The driver probe currently initializes some registers to non-POR
values. These values are not reinstated if the RTC experiences voltage
loss later on. Prepare for fixing this by factoring out the
initialization to a separate function.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20220426071056.1187235-2-s.hauer@pengutronix.de
The Epson RX8804 RTC has the same programming model as RV8803.
Add support for it in the driver.
Signed-off-by: Fabio Estevam <festevam@gmail.com>
Reviewed-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20211130125830.1166194-2-festevam@gmail.com
ctrl is set from read_regs(..FLAG, 2, ctrl), so ctrl[0] is FLAG
and ctrl[1] is the CTRL register.
Use ctrl[0] to write back to the FLAG register as appropriate.
Signed-off-by: Dominique Martinet <dominique.martinet@atmark-techno.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20211101013400.325855-1-dominique.martinet@atmark-techno.com
rtc_register_device() is a managed interface but it doesn't use devres
by itself - instead it marks an rtc_device as "registered" and the devres
callback for devm_rtc_allocate_device() takes care of resource release.
This doesn't correspond with the design behind devres where managed
structures should not be aware of being managed. The correct solution
here is to register a separate devres callback for unregistering the
device.
While at it: rename rtc_register_device() to devm_rtc_register_device()
and add it to the list of managed interfaces in devres.rst. This way we
can avoid any potential confusion of driver developers who may expect
there to exist a corresponding unregister function.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20201109163409.24301-8-brgl@bgdev.pl
rtc_nvmem_register() is a managed interface. It doesn't require any
release function to be called at driver detach. To avoid confusing
driver authors, let's rename it to devm_rtc_nvmem_register() and add it
to the list of managed interfaces in Documentation/.
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20201109163409.24301-6-brgl@bgdev.pl
The nvram sysfs attributes have been deprecated at least since v4.13, more
than 3 years ago and nobody ever complained about the deprecation warning.
Remove the sysfs attributes now.
[Bartosz: remove the declaration of rtc_nvmem_unregister()]
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
Link: https://lore.kernel.org/r/20201109163409.24301-5-brgl@bgdev.pl
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.com
Signed-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.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
The RTC core now has error messages in case of registration failure, there
is no need to have other messages in the drivers.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Link: https://lore.kernel.org/r/20190818220041.17833-2-alexandre.belloni@bootlin.com
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
We have a dedicated pointer for that, so use it. Much easier to read and
less computation involved.
Signed-off-by: Wolfram Sang <wsa+renesas@sang-engineering.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
The Epson rx8803 is mostly similar to the Microcrystal RV8803 but the size
of the offset register is 4 bits vs 6 bits but it has a configurable
temperature compensation.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
The probe function is not allowed to fail after registering the RTC. Call
rtc_register_device() at the end.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Call rtc_nvmem_register instead of letting the core do it and stop using
the nvmem_config member of struct rtc_device.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.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: Alexandre Belloni <alexandre.belloni@free-electrons.com>
This removes a possible race condition and allows for further improvement
of the driver.
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The driver doesn't have a struct of_device_id table but supported devices
are registered via Device Trees. This is working on the assumption that a
I2C device registered via OF will always match a legacy I2C device ID and
that the MODALIAS reported will always be of the form i2c:<device>.
But this could change in the future so the correct approach is to have an
OF device ID table if the devices are registered via OF.
Signed-off-by: Javier Martinez Canillas <javier@osg.samsung.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
There might be designs where the power supply circuit is designed
in a way that VDETOFF and SWOFF is required to be set. Otherwise the
RTC detects a power loss. Add a device tree interface for this.
Signed-off-by: Carsten Resch <Carsten.Resch@de.bosch.com>
Signed-off-by: Dirk Behme <dirk.behme@de.bosch.com>
Signed-off-by: Oleksij Rempel <fixed-term.Oleksij.Rempel@de.bosch.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
V1F indicates that the time accuracy may have been compromised because
of a voltage drop (possibly only temporary) below VLOW1, which stops the
temperature compensation. When the time is set, the accuracy is
restored, so V1F should be cleared in order to indicate this and to be
able to detect the next temperature compensation loss. This is the same
principle as for V2F, which is cleared when the time is set to indicate
that the time is no longer invalid and to be able to detect the next
data loss.
Signed-off-by: Benoît Thébaudeau <benoit@wsystem.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
According to the application manual of the RX8900, the RESET bit must be
set to 1 to prevent a timer update while setting the time. This also
resets the subsecond counter. The application manual of the RV-8803 does
not mention such a requirement, and it says that the 100th Seconds
register is cleared when writing to the Seconds register, but using the
RESET bit for the RV-8803 too should not be an issue and is probably
safer.
This change also ensures that the RESET bit is initialized properly in
all cases. Indeed, all the registers must be initialized if the voltage
has been lower than VLOW2 (triggering V2F), but not low enough to
trigger a POR.
Signed-off-by: Benoît Thébaudeau <benoit@wsystem.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The I²C NACK issue of the RV-8803 may occur after any I²C START
condition, depending on the timings. Consequently, the workaround must
be applied for all the I²C transfers.
This commit abstracts the I²C transfer code into register access
functions. This avoids duplicating the I²C workaround everywhere. This
also avoids the duplication of the code handling the return value of
i2c_smbus_read_i2c_block_data(). Error messages are issued in case of
definitive register access failures (if the workaround fails). This
change also makes the I²C transfer return value checks consistent.
Signed-off-by: Benoît Thébaudeau <benoit@wsystem.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The Weekday register is encoded as 2^tm_wday, with tm_wday in 0..6, so
using tm_wday = ffs(reg) to fill tm_wday from the register value is
wrong because this gives the expected value + 1. This could be fixed as
tm_wday = ffs(reg) - 1, but tm_wday = ilog2(reg) works as well and is
more direct.
Signed-off-by: Benoît Thébaudeau <benoit@wsystem.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The RTC core always calls rtc_valid_tm() after ->read_time() in case of
success (in __rtc_read_time()), so do not call it twice.
Signed-off-by: Benoît Thébaudeau <benoit@wsystem.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Since commit d68778b80d ("rtc: initialize output parameter for read
alarm to "uninitialized"") there is no need to explicitly set
unsupported members to -1. So drop the respective assignments from
drivers.
Signed-off-by: Uwe Kleine-König <uwe@kleine-koenig.org>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The previous workaround may still fail as there are actually 4 retries to
be done to ensure the communication succeed. Also, some I2C adapter drivers
may return -EIO instead of -ENXIO.
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The rv8803 has a 60µs window where it will not answer on the i2c bus.
It also means there will be no ack for the communication. Make sure
communication is tried multiple times when this happens (the i2c subsystem
mandates -ENXIO is that case but the number of retries is host specific).
The critical parts are the probe function and the alarm callback so make
sure we handle the failure there.
Cc: stable@vger.kernel.org # v4.4
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Fix a scheduling while atomic issue caused by rv8803_set_time()
holding a spinlock during the call to i2c_smbus_read_byte_data().
Signed-off-by: Oleksij Rempel <fixed-term.Oleksij.Rempel@de.bosch.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The Seiko Epson's RTC RX8900 layout register is compatible with the
RV8803. So let's add its ID in order to reuse the same driver.
Signed-off-by: Gregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
The function can return negative values, so its result should
be assigned to signed variable.
The problem has been detected using proposed semantic patch
scripts/coccinelle/tests/assign_signed_to_unsigned.cocci [1].
[1]: http://permalink.gmane.org/gmane.linux.kernel/2046107
Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
This driver supports the following functions:
- reading and settings time
- alarms when connected to an IRQ
- reading and clearing the voltage low flags
- nvram
Signed-off-by: Alexandre Belloni <alexandre.belloni@free-electrons.com>
Sascha Hauer