| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <1400860385-14555-12-git-send-email-vincent.guittot@linaro.org>
Date: Fri, 23 May 2014 17:53:05 +0200
From: Vincent Guittot <vincent.guittot@...aro.org>
To: peterz@...radead.org, mingo@...nel.org,
linux-kernel@...r.kernel.org, linux@....linux.org.uk,
linux-arm-kernel@...ts.infradead.org
Cc: preeti@...ux.vnet.ibm.com, Morten.Rasmussen@....com, efault@....de,
nicolas.pitre@...aro.org, linaro-kernel@...ts.linaro.org,
daniel.lezcano@...aro.org,
Vincent Guittot <vincent.guittot@...aro.org>
Subject: [PATCH v2 11/11] sched: replace capacity by activity
The scheduler tries to compute how many tasks a group of CPUs can handle by
assuming that a task's load is SCHED_LOAD_SCALE and a CPU capacity is
SCHED_POWER_SCALE.
We can now have a better idea of the utilization of a group fo CPUs thanks to
group_actitvity and deduct how many capacity is still available.
Signed-off-by: Vincent Guittot <vincent.guittot@...aro.org>
---
kernel/sched/fair.c | 94 ++++++++++++++---------------------------------------
1 file changed, 24 insertions(+), 70 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 2501e49..05b9502 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5538,11 +5538,10 @@ struct sg_lb_stats {
unsigned long group_power;
unsigned long group_activity; /* Total activity of the group */
unsigned int sum_nr_running; /* Nr tasks running in the group */
- unsigned int group_capacity;
+ long group_capacity;
unsigned int idle_cpus;
unsigned int group_weight;
int group_imb; /* Is there an imbalance in the group ? */
- int group_has_capacity; /* Is there extra capacity in the group? */
#ifdef CONFIG_NUMA_BALANCING
unsigned int nr_numa_running;
unsigned int nr_preferred_running;
@@ -5785,31 +5784,6 @@ void update_group_power(struct sched_domain *sd, int cpu)
}
/*
- * Try and fix up capacity for tiny siblings, this is needed when
- * things like SD_ASYM_PACKING need f_b_g to select another sibling
- * which on its own isn't powerful enough.
- *
- * See update_sd_pick_busiest() and check_asym_packing().
- */
-static inline int
-fix_small_capacity(struct sched_domain *sd, struct sched_group *group)
-{
- /*
- * Only siblings can have significantly less than SCHED_POWER_SCALE
- */
- if (!(sd->flags & SD_SHARE_CPUPOWER))
- return 0;
-
- /*
- * If ~90% of the cpu_power is still there, we're good.
- */
- if (group->sgp->power * 32 > group->sgp->power_orig * 29)
- return 1;
-
- return 0;
-}
-
-/*
* Group imbalance indicates (and tries to solve) the problem where balancing
* groups is inadequate due to tsk_cpus_allowed() constraints.
*
@@ -5843,33 +5817,6 @@ static inline int sg_imbalanced(struct sched_group *group)
return group->sgp->imbalance;
}
-/*
- * Compute the group capacity.
- *
- * Avoid the issue where N*frac(smt_power) >= 1 creates 'phantom' cores by
- * first dividing out the smt factor and computing the actual number of cores
- * and limit power unit capacity with that.
- */
-static inline int sg_capacity(struct lb_env *env, struct sched_group *group)
-{
- unsigned int capacity, smt, cpus;
- unsigned int power, power_orig;
-
- power = group->sgp->power;
- power_orig = group->sgp->power_orig;
- cpus = group->group_weight;
-
- /* smt := ceil(cpus / power), assumes: 1 < smt_power < 2 */
- smt = DIV_ROUND_UP(SCHED_POWER_SCALE * cpus, power_orig);
- capacity = cpus / smt; /* cores */
-
- capacity = min_t(unsigned, capacity, DIV_ROUND_CLOSEST(power, SCHED_POWER_SCALE));
- if (!capacity)
- capacity = fix_small_capacity(env->sd, group);
-
- return capacity;
-}
-
/**
* update_sg_lb_stats - Update sched_group's statistics for load balancing.
* @env: The load balancing environment.
@@ -5918,10 +5865,9 @@ static inline void update_sg_lb_stats(struct lb_env *env,
sgs->group_weight = group->group_weight;
sgs->group_imb = sg_imbalanced(group);
- sgs->group_capacity = sg_capacity(env, group);
+ sgs->group_capacity = group->sgp->power_orig - sgs->group_activity;
+
- if (sgs->group_capacity > sgs->sum_nr_running)
- sgs->group_has_capacity = 1;
}
/**
@@ -5945,7 +5891,15 @@ static bool update_sd_pick_busiest(struct lb_env *env,
if (sgs->avg_load <= sds->busiest_stat.avg_load)
return false;
- if (sgs->sum_nr_running > sgs->group_capacity)
+ /* The group has an obvious long run overload */
+ if (sgs->group_capacity < 0)
+ return true;
+
+ /*
+ * The group has a short run overload because more tasks than available
+ * CPUs are running
+ */
+ if (sgs->sum_nr_running > sgs->group_weight)
return true;
/*
@@ -6052,8 +6006,8 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
* with a large weight task outweighs the tasks on the system).
*/
if (prefer_sibling && sds->local &&
- sds->local_stat.group_has_capacity)
- sgs->group_capacity = min(sgs->group_capacity, 1U);
+ sds->local_stat.group_capacity > 0)
+ sgs->group_capacity = min(sgs->group_capacity, 1L);
if (update_sd_pick_busiest(env, sds, sg, sgs)) {
sds->busiest = sg;
@@ -6228,7 +6182,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
* have to drop below capacity to reach cpu-load equilibrium.
*/
load_above_capacity =
- (busiest->sum_nr_running - busiest->group_capacity);
+ (busiest->sum_nr_running - busiest->group_weight);
load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_POWER_SCALE);
load_above_capacity /= busiest->group_power;
@@ -6294,6 +6248,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
local = &sds.local_stat;
busiest = &sds.busiest_stat;
+ /* ASYM feature bypasses nice load balance check */
if ((env->idle == CPU_IDLE || env->idle == CPU_NEWLY_IDLE) &&
check_asym_packing(env, &sds))
return sds.busiest;
@@ -6313,8 +6268,8 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
goto force_balance;
/* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */
- if (env->idle == CPU_NEWLY_IDLE && local->group_has_capacity &&
- !busiest->group_has_capacity)
+ if (env->idle == CPU_NEWLY_IDLE && (local->group_capacity > 0)
+ && (busiest->group_capacity < 0))
goto force_balance;
/*
@@ -6372,7 +6327,7 @@ static struct rq *find_busiest_queue(struct lb_env *env,
int i;
for_each_cpu_and(i, sched_group_cpus(group), env->cpus) {
- unsigned long power, capacity, wl;
+ unsigned long power, wl;
enum fbq_type rt;
rq = cpu_rq(i);
@@ -6400,18 +6355,17 @@ static struct rq *find_busiest_queue(struct lb_env *env,
if (rt > env->fbq_type)
continue;
- power = power_of(i);
- capacity = DIV_ROUND_CLOSEST(power, SCHED_POWER_SCALE);
- if (!capacity)
- capacity = fix_small_capacity(env->sd, group);
-
wl = weighted_cpuload(i);
/*
* When comparing with imbalance, use weighted_cpuload()
* which is not scaled with the cpu power.
*/
- if (capacity && rq->nr_running == 1 && wl > env->imbalance)
+
+ power = power_of(i);
+
+ if (rq->nr_running == 1 && wl > env->imbalance &&
+ ((power * env->sd->imbalance_pct) >= (rq->cpu_power_orig * 100)))
continue;
/*
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists