Re: [PATCH V3 1/7] sched: Create sched_select_cpu() to give preferredCPU for power saving

From: Viresh Kumar
Date: Tue Mar 19 2013 - 01:15:18 EST


On 18 March 2013 21:27, Frederic Weisbecker <fweisbec@xxxxxxxxx> wrote:
> 2013/3/18 Viresh Kumar <viresh.kumar@xxxxxxxxxx>:
>>> It would be nice to have some more precise naming. sched_select_busy_cpu() ?
>>
>> You are correct that name can be improved but busy may give it a different
>> meaning. Maybe sched_select_non_idle_cpu()?
>
> That works too yeah.

Here is the fixup that i applied (it fixes another stupid mistake in sched.h).
I have pushed this series here again:

http://git.linaro.org/gitweb?p=people/vireshk/linux.git;a=shortlog;h=refs/heads/sched-wq-migration-v3

commit e769ff0d78fd2fb504ae056a44b70bba7c259126
Author: Viresh Kumar <viresh.kumar@xxxxxxxxxx>
Date: Fri Sep 14 15:13:05 2012 +0530

fixup! sched: Create sched_select_cpu() to give preferred CPU for
power saving

Signed-off-by: Viresh Kumar <viresh.kumar@xxxxxxxxxx>
---
include/linux/sched.h | 19 ++++++++++---------
kernel/sched/core.c | 2 +-
2 files changed, 11 insertions(+), 10 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 216fa0d..db6655a 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -231,11 +231,18 @@ extern void init_idle_bootup_task(struct
task_struct *idle);
extern int runqueue_is_locked(int cpu);

#ifdef CONFIG_SMP
-extern int sched_select_cpu(unsigned int sd_flags);
+extern int sched_select_non_idle_cpu(unsigned int sd_flags);
+#else
+static inline int sched_select_non_idle_cpu(unsigned int sd_flags)
+{
+ return smp_processor_id();
+}
+#endif

-#ifdef CONFIG_NO_HZ
+#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
extern void nohz_balance_enter_idle(int cpu);
extern void set_cpu_sd_state_idle(void);
+
/*
* In the semi idle case, use the nearest busy cpu for migrating timers
* from an idle cpu. This is good for power-savings.
@@ -244,17 +251,11 @@ extern void set_cpu_sd_state_idle(void);
* selecting an idle cpu will add more delays to the timers than intended
* (as that cpu's timer base may not be uptodate wrt jiffies etc).
*/
-#define get_nohz_timer_target() sched_select_cpu(0)
+#define get_nohz_timer_target() sched_select_non_idle_cpu(0)
#else
static inline void nohz_balance_enter_idle(int cpu) { }
static inline void set_cpu_sd_state_idle(void) { }
-
-static inline int sched_select_cpu(unsigned int sd_flags)
-{
- return raw_smp_processor_id();
-}
#endif
-#endif /* CONFIG_SMP */

/*
* Only dump TASK_* tasks. (0 for all tasks)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index ccd76d7..0265a5e 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -627,7 +627,7 @@ void sched_avg_update(struct rq *rq)
* returned back. If it is idle, then we must look for another CPU which have
* all the flags passed as argument as set.
*/
-int sched_select_cpu(unsigned int sd_flags)
+int sched_select_non_idle_cpu(unsigned int sd_flags)
{
struct sched_domain *sd;
int cpu = smp_processor_id();
--
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/