Re: System automatically wakes up because of Intel Rapid Start Technology

[Gabriele Mazzotta]

On Monday 22 December 2014 16:50:37 Gabriele Mazzotta wrote:
> On Monday 22 December 2014 14:59:49 Matthew Garrett wrote:
> > Can you try this diff?
> 
> Unfortunately it doesn't work (I made a change, see here below).

I take a better look at the code and noticed a couple of mistakes in
your, but you nailed it.

Here below the working patch. There was an extra ! and another deadlock.

diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index 5b2e761..4035927 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -48,6 +48,7 @@ struct cmos_rtc {
 	struct device		*dev;
 	int			irq;
 	struct resource		*iomem;
+	struct rtc_wkalrm	alm;
 
 	void			(*wake_on)(struct device *);
 	void			(*wake_off)(struct device *);
@@ -59,6 +60,8 @@ struct cmos_rtc {
 	u8			day_alrm;
 	u8			mon_alrm;
 	u8			century;
+
+	bool			valid_alarm;
 };
 
 /* both platform and pnp busses use negative numbers for invalid irqs */
@@ -881,6 +884,8 @@ static int cmos_suspend(struct device *dev)
 	}
 	spin_unlock_irq(&rtc_lock);
 
+	cmos->valid_alarm = !cmos_read_alarm(dev, &cmos->alm);
+
 	if (tmp & RTC_AIE) {
 		cmos->enabled_wake = 1;
 		if (cmos->wake_on)
@@ -951,6 +956,9 @@ static int cmos_resume(struct device *dev)
 	}
 	spin_unlock_irq(&rtc_lock);
 
+	if (cmos->valid_alarm)
+		cmos_set_alarm(dev, &cmos->alm);
+
 	dev_dbg(dev, "resume, ctrl %02x\n", tmp);
 
 	return 0;

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/