diff mbox

ARM: architected timers: Mark the clockevent with the C3_STOP feature

Message ID 1341395673-29660-1-git-send-email-santosh.shilimkar@ti.com (mailing list archive)
State New, archived
Headers show

Commit Message

Santosh Shilimkar July 4, 2012, 9:54 a.m. UTC
ARM arch timers stop in low power state and hence can not wakeup CPUs in
deeper idle states when used as cloc event devices. Marking these clock event
devices with C3_STOP so that during lowpower states, the tick is managed by
wakeup capable broadcast timer.

Cc: Marc Zyngier <marc.zyngier@arm.com>
Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
---
 arch/arm/kernel/arch_timer.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Comments

Marc Zyngier July 4, 2012, 9:56 a.m. UTC | #1
On 04/07/12 10:54, Santosh Shilimkar wrote:
> ARM arch timers stop in low power state and hence can not wakeup CPUs in
> deeper idle states when used as cloc event devices. Marking these clock event
> devices with C3_STOP so that during lowpower states, the tick is managed by
> wakeup capable broadcast timer.
> 
> Cc: Marc Zyngier <marc.zyngier@arm.com>
> Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>

Hi Santosh,

I already have queued an identical patch from Lorenzo.
Thanks for reminder anyway!

	M.

> ---
>  arch/arm/kernel/arch_timer.c |    2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/arch/arm/kernel/arch_timer.c b/arch/arm/kernel/arch_timer.c
> index dd58035..df44c8c 100644
> --- a/arch/arm/kernel/arch_timer.c
> +++ b/arch/arm/kernel/arch_timer.c
> @@ -137,7 +137,7 @@ static int __cpuinit arch_timer_setup(struct clock_event_device *clk)
>  	/* Be safe... */
>  	arch_timer_disable();
>  
> -	clk->features = CLOCK_EVT_FEAT_ONESHOT;
> +	clk->features = CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_C3STOP;
>  	clk->name = "arch_sys_timer";
>  	clk->rating = 450;
>  	clk->set_mode = arch_timer_set_mode;
Santosh Shilimkar July 4, 2012, 10 a.m. UTC | #2
On Wed, Jul 4, 2012 at 3:26 PM, Marc Zyngier <marc.zyngier@arm.com> wrote:
>
> On 04/07/12 10:54, Santosh Shilimkar wrote:
> > ARM arch timers stop in low power state and hence can not wakeup CPUs in
> > deeper idle states when used as cloc event devices. Marking these clock
> > event
> > devices with C3_STOP so that during lowpower states, the tick is managed
> > by
> > wakeup capable broadcast timer.
> >
> > Cc: Marc Zyngier <marc.zyngier@arm.com>
> > Signed-off-by: Santosh Shilimkar <santosh.shilimkar@ti.com>
>
> Hi Santosh,
>
> I already have queued an identical patch from Lorenzo.
> Thanks for reminder anyway!
>
OK. Thanks for the information Mark.
Looks like I missed that one on the list.

Regards,
Santosh
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Russell King - ARM Linux July 4, 2012, 10:26 a.m. UTC | #3
On Wed, Jul 04, 2012 at 03:24:33PM +0530, Santosh Shilimkar wrote:
> ARM arch timers stop in low power state and hence can not wakeup CPUs in
> deeper idle states when used as cloc event devices. Marking these clock event
> devices with C3_STOP so that during lowpower states, the tick is managed by
> wakeup capable broadcast timer.

Will tells me that the arch timers don't stop in low power modes, they
just can't produce wakeup events.  Apparantly the spec says:

   The system counter must be implemented in an always-on power domain.
   Use of lower-frequency modes must not affect the implemented accuracy.

Are you sure your above description of the problem is correct?
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Santosh Shilimkar July 4, 2012, 10:41 a.m. UTC | #4
On Wed, Jul 4, 2012 at 3:56 PM, Russell King - ARM Linux
<linux@arm.linux.org.uk> wrote:
>
> On Wed, Jul 04, 2012 at 03:24:33PM +0530, Santosh Shilimkar wrote:
> > ARM arch timers stop in low power state and hence can not wakeup CPUs in
> > deeper idle states when used as cloc event devices. Marking these clock
> > event
> > devices with C3_STOP so that during lowpower states, the tick is managed
> > by
> > wakeup capable broadcast timer.
>
> Will tells me that the arch timers don't stop in low power modes, they
> just can't produce wakeup events.  Apparantly the spec says:
>
>    The system counter must be implemented in an always-on power domain.
>    Use of lower-frequency modes must not affect the implemented accuracy.
>
> Are you sure your above description of the problem is correct?

Yes. Will is right.
The arch timers don't stop with a real time counter implementation.
The issue is they are not wakeup capable hence you need a broad-cast switching
which in turn needs C3_STOP.

Regards
Santosh
--
To unsubscribe from this list: send the line "unsubscribe linux-omap" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
diff mbox

Patch

diff --git a/arch/arm/kernel/arch_timer.c b/arch/arm/kernel/arch_timer.c
index dd58035..df44c8c 100644
--- a/arch/arm/kernel/arch_timer.c
+++ b/arch/arm/kernel/arch_timer.c
@@ -137,7 +137,7 @@  static int __cpuinit arch_timer_setup(struct clock_event_device *clk)
 	/* Be safe... */
 	arch_timer_disable();
 
-	clk->features = CLOCK_EVT_FEAT_ONESHOT;
+	clk->features = CLOCK_EVT_FEAT_ONESHOT | CLOCK_EVT_FEAT_C3STOP;
 	clk->name = "arch_sys_timer";
 	clk->rating = 450;
 	clk->set_mode = arch_timer_set_mode;