| From 7c2accd958cf88e35753cd3ac87eca28dfd101fa Mon Sep 17 00:00:00 2001 |
| From: Venkatesh Pallipadi <venki@google.com> |
| Date: Thu, 10 Feb 2011 10:23:27 +0100 |
| Subject: sched: Remove irq time from available CPU power |
| |
| Commit: aa483808516ca5cacfa0e5849691f64fec25828e upstream |
| |
| The idea was suggested by Peter Zijlstra here: |
| |
| http://marc.info/?l=linux-kernel&m=127476934517534&w=2 |
| |
| irq time is technically not available to the tasks running on the CPU. |
| This patch removes irq time from CPU power piggybacking on |
| sched_rt_avg_update(). |
| |
| Tested this by keeping CPU X busy with a network intensive task having 75% |
| oa a single CPU irq processing (hard+soft) on a 4-way system. And start seven |
| cycle soakers on the system. Without this change, there will be two tasks on |
| each CPU. With this change, there is a single task on irq busy CPU X and |
| remaining 7 tasks are spread around among other 3 CPUs. |
| |
| Signed-off-by: Venkatesh Pallipadi <venki@google.com> |
| Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> |
| LKML-Reference: <1286237003-12406-8-git-send-email-venki@google.com> |
| Signed-off-by: Ingo Molnar <mingo@elte.hu> |
| Signed-off-by: Mike Galbraith <efault@gmx.de> |
| Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl> |
| Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de> |
| --- |
| kernel/sched.c | 26 +++++++++++++++++++++++++- |
| kernel/sched_features.h | 5 +++++ |
| 2 files changed, 30 insertions(+), 1 deletion(-) |
| |
| --- a/kernel/sched.c |
| +++ b/kernel/sched.c |
| @@ -553,6 +553,10 @@ struct rq { |
| u64 avg_idle; |
| #endif |
| |
| +#ifdef CONFIG_IRQ_TIME_ACCOUNTING |
| + u64 prev_irq_time; |
| +#endif |
| + |
| /* calc_load related fields */ |
| unsigned long calc_load_update; |
| long calc_load_active; |
| @@ -622,6 +626,7 @@ static inline int cpu_of(struct rq *rq) |
| #define raw_rq() (&__raw_get_cpu_var(runqueues)) |
| |
| static u64 irq_time_cpu(int cpu); |
| +static void sched_irq_time_avg_update(struct rq *rq, u64 irq_time); |
| |
| inline void update_rq_clock(struct rq *rq) |
| { |
| @@ -632,6 +637,8 @@ inline void update_rq_clock(struct rq *r |
| irq_time = irq_time_cpu(cpu); |
| if (rq->clock - irq_time > rq->clock_task) |
| rq->clock_task = rq->clock - irq_time; |
| + |
| + sched_irq_time_avg_update(rq, irq_time); |
| } |
| |
| /* |
| @@ -1883,6 +1890,15 @@ void account_system_vtime(struct task_st |
| local_irq_restore(flags); |
| } |
| |
| +static void sched_irq_time_avg_update(struct rq *rq, u64 curr_irq_time) |
| +{ |
| + if (sched_clock_irqtime && sched_feat(NONIRQ_POWER)) { |
| + u64 delta_irq = curr_irq_time - rq->prev_irq_time; |
| + rq->prev_irq_time = curr_irq_time; |
| + sched_rt_avg_update(rq, delta_irq); |
| + } |
| +} |
| + |
| #else |
| |
| static u64 irq_time_cpu(int cpu) |
| @@ -1890,6 +1906,8 @@ static u64 irq_time_cpu(int cpu) |
| return 0; |
| } |
| |
| +static void sched_irq_time_avg_update(struct rq *rq, u64 curr_irq_time) { } |
| + |
| #endif |
| |
| #include "sched_stats.h" |
| @@ -3755,7 +3773,13 @@ unsigned long scale_rt_power(int cpu) |
| u64 total, available; |
| |
| total = sched_avg_period() + (rq->clock - rq->age_stamp); |
| - available = total - rq->rt_avg; |
| + |
| + if (unlikely(total < rq->rt_avg)) { |
| + /* Ensures that power won't end up being negative */ |
| + available = 0; |
| + } else { |
| + available = total - rq->rt_avg; |
| + } |
| |
| if (unlikely((s64)total < SCHED_LOAD_SCALE)) |
| total = SCHED_LOAD_SCALE; |
| --- a/kernel/sched_features.h |
| +++ b/kernel/sched_features.h |
| @@ -121,3 +121,8 @@ SCHED_FEAT(ASYM_EFF_LOAD, 1) |
| * release the lock. Decreases scheduling overhead. |
| */ |
| SCHED_FEAT(OWNER_SPIN, 1) |
| + |
| +/* |
| + * Decrement CPU power based on irq activity |
| + */ |
| +SCHED_FEAT(NONIRQ_POWER, 1) |