linux-toradex.git/drivers/rtc/rtc-jz4740.c, branch v6.0-rc3

rtc: rework rtc_register_device() resource management

2020-11-19T11:50:12+00:00

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 
Signed-off-by: Alexandre Belloni 
Link: https://lore.kernel.org/r/20201109163409.24301-8-brgl@bgdev.pl

rtc: ingenic: Reset regulator register in probe

2020-05-11T14:35:05+00:00

The regulator register specifies how many input clock cycles (minus one)
are contained in one tick of the 1 Hz clock.

Since this register can contain bogus values after the system boots, it
needs to be reset in the probe register, otherwise the RTC may count way
to slow or way too fast.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-7-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Fix masking of error code

2020-05-11T14:35:03+00:00

The code was returning -ENOENT on any error of platform_get_irq(), even
if it returned a different error.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-6-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Remove unused fields from private structure

2020-05-11T14:35:01+00:00

The 'clk' and 'irq' fields were only ever used in the probe function.
Therefore they can be moved to be simple local variables of the probe
function.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-5-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Set wakeup params in probe

2020-05-11T14:34:59+00:00

We can write the wakeup timing parameters as soon as the driver probes,
there's no need to wait the very last moment.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-4-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Enable clock in probe

2020-05-11T14:34:58+00:00

It makes no sense to request a clock and not enable it even though the
hardware is being used. So the driver now enables the clock in the
probe. Besides, now we can properly handle errors.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-3-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Use local 'dev' variable in probe

2020-05-11T14:34:55+00:00

Clean a bit the probe function by adding a local struct device *dev
variable.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-2-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: ingenic: Only support probing from devicetree

2020-05-11T14:34:54+00:00

With the recent work on supporting Device Tree on Ingenic SoCs, no
driver ever probes from platform code anymore, so we can clean a bit
this driver by removing the non-devicetree paths.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200505221336.222313-1-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: jz4740: Rename vendor-specific DT properties

2020-03-24T10:03:23+00:00

These properties are never set anywhere within any of the upstream
devicetree files, so I assume I'm not breaking the ABI with this change.

Rename vendor-specific DT properties to have the 'ingenic,' prefix,
which they should have had from the start.

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200311182318.22154-2-paul@crapouillou.net
Signed-off-by: Alexandre Belloni

rtc: jz4740: Add support for JZ4760 SoC

2020-03-24T10:03:22+00:00

The WENR feature (set a magic value to enable RTC registers read-write)
first appeared on the JZ4760; the JZ4780 came much later.

Since it would be dangerous to specify a newer SoC's compatible string as
the fallback of an older SoC's compatible string, we add support for the
"ingenic,jz4760-rtc" compatible string in the driver.

This will permit to support the JZ4770 by having:
compatible = "ingenic,jz4770-rtc", "ingenic,jz4760-rtc";

Instead of doing:
compatible = "ingenic,jz4770-rtc", "ingenic,jz4780-rtc";

Signed-off-by: Paul Cercueil 
Link: https://lore.kernel.org/r/20200311182318.22154-1-paul@crapouillou.net
Signed-off-by: Alexandre Belloni