Re: [PATCH] rtc: pcf85363: Allow to wake up system without IRQ

[Mike Looijmans]

Met vriendelijke groet / kind regards,

Mike Looijmans
System Expert


TOPIC Embedded Products B.V.
Materiaalweg 4, 5681 RJ Best
The Netherlands

T: +31 (0) 499 33 69 69
E: mike.looijmans@xxxxxxxx
W: www.topic.nl

Please consider the environment before printing this e-mail
On 17-08-2023 10:17, Alexandre Belloni wrote:
> On 17/08/2023 07:42:25+0200, Mike Looijmans wrote:
> > On 16-08-2023 23:40, Alexandre Belloni wrote:
> > > Hello Mike,
> > >
> > > Sorry for the very late review, there is a small change that is needed:
> >
> > Better late than never
> >
> > > On 02/05/2023 07:54:58+0200, Mike Looijmans wrote:
> > > > When wakeup-source is set in the devicetree, set up the device for
> > > > using the output as interrupt instead of clock. This is similar to
> > > > how other RTC devices handle this.
> > > >
> > > > This allows the clock chip to turn on the board when wired to do
> > > > so in hardware.
> > > >
> > > > Signed-off-by: Mike Looijmans <mike.looijmans@xxxxxxxx>
> > > >
> > > > ---
> > > >
> > > >    drivers/rtc/rtc-pcf85363.c | 17 ++++++++++-------
> > > >    1 file changed, 10 insertions(+), 7 deletions(-)
> > > >
> > > > diff --git a/drivers/rtc/rtc-pcf85363.c b/drivers/rtc/rtc-pcf85363.c
> > > > index 8958eadf1c3e..b1543bcdc530 100644
> > > > --- a/drivers/rtc/rtc-pcf85363.c
> > > > +++ b/drivers/rtc/rtc-pcf85363.c
> > > > @@ -434,23 +434,26 @@ static int pcf85363_probe(struct i2c_client *client)
> > > >    	pcf85363->rtc->range_max = RTC_TIMESTAMP_END_2099;
> > > >    	clear_bit(RTC_FEATURE_ALARM, pcf85363->rtc->features);
> > > > +	if (client->irq > 0 || device_property_read_bool(&client->dev,
> > > > +							 "wakeup-source")) {
> > > > +		regmap_write(pcf85363->regmap, CTRL_FLAGS, 0);
> > > > +		regmap_update_bits(pcf85363->regmap, CTRL_PIN_IO,
> > > > +				   PIN_IO_INTA_OUT, PIN_IO_INTAPM);
> > > > +		device_init_wakeup(&client->dev, true);
> > > > +		set_bit(RTC_FEATURE_ALARM, pcf85363->rtc->features);
> > >
> > > This should not be done when devm_request_threaded_irq fails.
> >
> > Yeah, easiest would be to just clear the feature flag when that happens.
>
> My comment was also for device_init_wakeup(&client->dev, true);. I think
> the easiest would be to move this block later on and set client->irq to
> 0 when devm_request_threaded_irq fails.

Ah, clear. That also properly handles things when IRQ fails but wakeup-source 
was present. In table form:

IRQ   wakeup-source   Call "device_init_wakeup" etc.
N/A   No              No
N/A   Yes             Yes
OK    No              Yes
OK    Yes             Yes
FAIL  No              No
FAIL  Yes             Yes

We still have to program the registers before registering the IRQ I think.