On Wed, 6 Mar 2013, Daniel Lezcano wrote:
On 03/05/2013 09:40 PM, Thomas Gleixner wrote:
On Sat, 2 Mar 2013, Daniel Lezcano wrote:
When a cpu goes to a deep idle state where its local timer is shutdown, it notifies the time frame work to use the broadcast timer instead.
Unfortunately, the broadcast device could wake up any CPU, including an idle one which is not concerned by the wake up at all.
This implies, in the worst case, an idle CPU will wake up to send an IPI to another idle cpu.
This patch solves this by setting the irq affinity to the cpu concerned by the nearest timer event, by this way, the CPU which is wake up is guarantee to be the one concerned by the next event and we are safe with unnecessary wakeup for another idle CPU.
As the irq affinity is not supported by all the archs, a flag is needed to specify which clocksource can handle it : CLOCK_EVT_FEAT_DYNIRQ
Signed-off-by: Daniel Lezcano daniel.lezcano@linaro.org
include/linux/clockchips.h | 5 +++++ kernel/time/tick-broadcast.c | 40 +++++++++++++++++++++++++++++++++------- 2 files changed, 38 insertions(+), 7 deletions(-)
diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index 6634652..c93e2a6 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -55,6 +55,11 @@ enum clock_event_nofitiers { #define CLOCK_EVT_FEAT_C3STOP 0x000008 #define CLOCK_EVT_FEAT_DUMMY 0x000010 +/*
- Clock event device can set its irq affinity dynamically
- */
+#define CLOCK_EVT_FEAT_DYNIRQ 0x000020
/**
- struct clock_event_device - clock event device descriptor
- @event_handler: Assigned by the framework to be called by the low
diff --git a/kernel/time/tick-broadcast.c b/kernel/time/tick-broadcast.c index 6197ac0..9ca8ff5 100644 --- a/kernel/time/tick-broadcast.c +++ b/kernel/time/tick-broadcast.c @@ -406,13 +406,37 @@ struct cpumask *tick_get_broadcast_oneshot_mask(void) return to_cpumask(tick_broadcast_oneshot_mask); } -static int tick_broadcast_set_event(struct clock_event_device *bc, +/*
- Set broadcast interrupt affinity
- */
+static void tick_broadcast_set_affinity(struct clock_event_device *bc,
const struct cpumask *cpumask)+{
- if (!(bc->features & CLOCK_EVT_FEAT_DYNIRQ))
return;- if (cpumask_equal(bc->cpumask, cpumask))
return;- bc->cpumask = cpumask;
This breaks with CONFIG_CPUMASK_OFFSTACK=y. cpumask_copy() is your friend!
This instruction copies the pointer, not the cpumask content.
bc->cpumask is defined as a const struct cpumask * and is used to copy a cpumask pointer not the content.
The cpumask parameter is a pointer to a global cpumask provided by the cpumask_of macro.
But to be in the safe side, I compiled tested with CONFIG_CPUMASK_OFFSTACK=y without problem.
Did I missed something ?
No, I misinterpreted the patch. Assigning a pointer is safe.
Thanks,
tglx