[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220301131220.4011810-1-vincent.whitchurch@axis.com>
Date: Tue, 1 Mar 2022 14:12:20 +0100
From: Vincent Whitchurch <vincent.whitchurch@...s.com>
To: Alessandro Zummo <a.zummo@...ertech.it>,
Alexandre Belloni <alexandre.belloni@...tlin.com>
CC: <kernel@...s.com>,
Vincent Whitchurch <vincent.whitchurch@...s.com>,
<linux-rtc@...r.kernel.org>, <linux-kernel@...r.kernel.org>
Subject: [PATCH] rtc: pcf8563: clear RTC_FEATURE_ALARM if no irq
If there is no IRQ hooked up, clear RTC_FEATURE_ALARM to make the core
ensure that userspace is made aware that alarms are not supported.
Signed-off-by: Vincent Whitchurch <vincent.whitchurch@...s.com>
---
Notes:
Could there be some board which uses the RTC for wakeup using wakealarm but
which doesn't have an IRQ? Then this patch will stop it from working since
wakealarm requires RTC_FEATURE_ALARM too.
It's unclear how that needs to be addressed. There seems to be a lot of
variation in how different RTC drivers handle these flags. Some call
device_set_wakeup_capable(), either conditionally based on the irq or always,
while others don't call it at all. Some call dev_init_wakeup(), others don't.
Some don't clear RTC_FEATURE_ALARM if wakeup-source is set, others don't check
wakeup-source and only clear RTC_FEATURE_ALARM based on the irq. Others never
clear RTC_FEATURE_ALARM.
drivers/rtc/rtc-pcf8563.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/rtc/rtc-pcf8563.c b/drivers/rtc/rtc-pcf8563.c
index c8bddfb94129..ade3a564bd33 100644
--- a/drivers/rtc/rtc-pcf8563.c
+++ b/drivers/rtc/rtc-pcf8563.c
@@ -580,6 +580,8 @@ static int pcf8563_probe(struct i2c_client *client,
client->irq);
return err;
}
+ } else {
+ clear_bit(RTC_FEATURE_ALARM, pcf8563->rtc->features);
}
err = devm_rtc_register_device(pcf8563->rtc);
--
2.34.1
Powered by blists - more mailing lists