2013-04-10 00:36:29

by dbasehore .

[permalink] [raw]
Subject: [PATCH] Don't disable hpet emulation on suspend

There's a bug where rtc alarms are ignored after the rtc cmos suspends but
before the system finishes suspend. Since hpet emulation is disabled and it
still handles the interrupts, a wake event is never registered which is done
from the rtc layer. This reverts an earlier commit which disables hpet
emulation. To fix the problem mentioned in that commit, the hpet_rtc_timer_init
function is called directly on resume.

This reverts commit d1b2efa83fbf7b33919238fa29ef6ab935820103.

Signed-off-by: Derek Basehore <[email protected]>
---
drivers/rtc/rtc-cmos.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/drivers/rtc/rtc-cmos.c b/drivers/rtc/rtc-cmos.c
index af97c94..cc5bea9 100644
--- a/drivers/rtc/rtc-cmos.c
+++ b/drivers/rtc/rtc-cmos.c
@@ -804,9 +804,8 @@ static int cmos_suspend(struct device *dev)
mask = RTC_IRQMASK;
tmp &= ~mask;
CMOS_WRITE(tmp, RTC_CONTROL);
+ hpet_mask_rtc_irq_bit(mask);

- /* shut down hpet emulation - we don't need it for alarm */
- hpet_mask_rtc_irq_bit(RTC_PIE|RTC_AIE|RTC_UIE);
cmos_checkintr(cmos, tmp);
}
spin_unlock_irq(&rtc_lock);
@@ -870,6 +869,7 @@ static int cmos_resume(struct device *dev)
rtc_update_irq(cmos->rtc, 1, mask);
tmp &= ~RTC_AIE;
hpet_mask_rtc_irq_bit(RTC_AIE);
+ hpet_rtc_timer_init();
} while (mask & RTC_AIE);
spin_unlock_irq(&rtc_lock);
}
--
1.8.1.3


2013-04-10 23:58:04

by Andrew Morton

[permalink] [raw]
Subject: Re: [PATCH] Don't disable hpet emulation on suspend

On Tue, 9 Apr 2013 17:36:13 -0700 Derek Basehore <[email protected]> wrote:

> There's a bug where rtc alarms are ignored after the rtc cmos suspends but
> before the system finishes suspend. Since hpet emulation is disabled and it
> still handles the interrupts, a wake event is never registered which is done
> from the rtc layer. This reverts an earlier commit which disables hpet
> emulation. To fix the problem mentioned in that commit, the hpet_rtc_timer_init
> function is called directly on resume.
>
> This reverts commit d1b2efa83fbf7b33919238fa29ef6ab935820103.
>
> ...
>
> --- a/drivers/rtc/rtc-cmos.c
> +++ b/drivers/rtc/rtc-cmos.c
> @@ -804,9 +804,8 @@ static int cmos_suspend(struct device *dev)
> mask = RTC_IRQMASK;
> tmp &= ~mask;
> CMOS_WRITE(tmp, RTC_CONTROL);
> + hpet_mask_rtc_irq_bit(mask);
>
> - /* shut down hpet emulation - we don't need it for alarm */
> - hpet_mask_rtc_irq_bit(RTC_PIE|RTC_AIE|RTC_UIE);
> cmos_checkintr(cmos, tmp);
> }
> spin_unlock_irq(&rtc_lock);
> @@ -870,6 +869,7 @@ static int cmos_resume(struct device *dev)
> rtc_update_irq(cmos->rtc, 1, mask);
> tmp &= ~RTC_AIE;
> hpet_mask_rtc_irq_bit(RTC_AIE);
> + hpet_rtc_timer_init();
> } while (mask & RTC_AIE);
> spin_unlock_irq(&rtc_lock);
> }

huh, so it's been broken for three years.

This calls hpet_rtc_timer_init() under spin_lock_irq() which is OK with
the current implementation. And things should be OK when
CONFIG_HPET_EMULATE_RTC=n.

Maxim, can you please review/test/etc?