Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751651AbZIBLSN (ORCPT ); Wed, 2 Sep 2009 07:18:13 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751390AbZIBLSM (ORCPT ); Wed, 2 Sep 2009 07:18:12 -0400 Received: from e28smtp04.in.ibm.com ([59.145.155.4]:45639 "EHLO e28smtp04.in.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751264AbZIBLSL (ORCPT ); Wed, 2 Sep 2009 07:18:11 -0400 Date: Wed, 2 Sep 2009 16:47:59 +0530 From: Gautham R Shenoy To: Peter Zijlstra Cc: Ingo Molnar , linux-kernel@vger.kernel.org, Andreas Herrmann , Balbir Singh Subject: Re: [RFC][PATCH 3/8] sched: update the cpu_power sum during load-balance Message-ID: <20090902111759.GB3817@in.ibm.com> Reply-To: ego@in.ibm.com References: <20090901083431.748830771@chello.nl> <20090901083825.985050292@chello.nl> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20090901083825.985050292@chello.nl> User-Agent: Mutt/1.5.17 (2007-11-01) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3496 Lines: 101 On Tue, Sep 01, 2009 at 10:34:34AM +0200, Peter Zijlstra wrote: > In order to prepare for a more dynamic cpu_power, update the group sum > while walking the sched domains during load-balance. > > Signed-off-by: Peter Zijlstra > --- > kernel/sched.c | 33 +++++++++++++++++++++++++++++---- > 1 file changed, 29 insertions(+), 4 deletions(-) > > Index: linux-2.6/kernel/sched.c > =================================================================== > --- linux-2.6.orig/kernel/sched.c > +++ linux-2.6/kernel/sched.c > @@ -3699,6 +3699,28 @@ static inline int check_power_save_busie > } > #endif /* CONFIG_SCHED_MC || CONFIG_SCHED_SMT */ > > +static void update_sched_power(struct sched_domain *sd) > +{ > + struct sched_domain *child = sd->child; > + struct sched_group *group, *sdg = sd->groups; > + unsigned long power = sdg->__cpu_power; > + > + if (!child) { > + /* compute cpu power for this cpu */ > + return; > + } > + > + sdg->__cpu_power = 0; > + > + group = child->groups; > + do { > + sdg->__cpu_power += group->__cpu_power; > + group = group->next; > + } while (group != child->groups); > + > + if (power != sdg->__cpu_power) > + sdg->reciprocal_cpu_power = reciprocal_value(sdg->__cpu_power); > +} > > /** > * update_sg_lb_stats - Update sched_group's statistics for load balancing. > @@ -3712,7 +3734,8 @@ static inline int check_power_save_busie > * @balance: Should we balance. > * @sgs: variable to hold the statistics for this group. > */ > -static inline void update_sg_lb_stats(struct sched_group *group, int this_cpu, > +static inline void update_sg_lb_stats(struct sched_domain *sd, > + struct sched_group *group, int this_cpu, > enum cpu_idle_type idle, int load_idx, int *sd_idle, > int local_group, const struct cpumask *cpus, > int *balance, struct sg_lb_stats *sgs) > @@ -3723,8 +3746,11 @@ static inline void update_sg_lb_stats(st > unsigned long sum_avg_load_per_task; > unsigned long avg_load_per_task; > > - if (local_group) > + if (local_group) { > balance_cpu = group_first_cpu(group); > + if (balance_cpu == this_cpu) > + update_sched_power(sd); > + } I guess the intention of this check is to ensure that the cpu_power for the group of sd is updated only by a specific member of the group and that would ideally be the first member of the group. Thus, this check has more to do with this_cpu being the group_first_cpu() than this_cpu being the balance_cpu. Correct ? > > /* Tally up the load of all CPUs in the group */ > sum_avg_load_per_task = avg_load_per_task = 0; > @@ -3828,7 +3854,7 @@ static inline void update_sd_lb_stats(st > local_group = cpumask_test_cpu(this_cpu, > sched_group_cpus(group)); > memset(&sgs, 0, sizeof(sgs)); > - update_sg_lb_stats(group, this_cpu, idle, load_idx, sd_idle, > + update_sg_lb_stats(sd, group, this_cpu, idle, load_idx, sd_idle, > local_group, cpus, balance, &sgs); > > if (local_group && balance && !(*balance)) > @@ -3863,7 +3889,6 @@ static inline void update_sd_lb_stats(st > update_sd_power_savings_stats(group, sds, local_group, &sgs); > group = group->next; > } while (group != sd->groups); > - > } > > /** > > -- -- Thanks and Regards gautham -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/