This patchset consolidates several changes in the capacity and the usage tracking of the CPU. It provides a frequency invariant metric of the usage of CPUs and generally improves the accuracy of load/usage tracking in the scheduler. The frequency invariant metric is the foundation required for the consolidation of cpufreq and implementation of a fully invariant load tracking. These are currently WIP and require several changes to the load balancer (including how it will use and interprets load and capacity metrics) and extensive validation. The frequency invariance is done with arch_scale_freq_capacity and this patchset doesn't provide the backends of the function which are architecture dependent.
As discussed at LPC14, Morten and I have consolidated our changes into a single patchset to make it easier to review and merge.
During load balance, the scheduler evaluates the number of tasks that a group of CPUs can handle. The current method assumes that tasks have a fix load of SCHED_LOAD_SCALE and CPUs have a default capacity of SCHED_CAPACITY_SCALE. This assumption generates wrong decision by creating ghost cores or by removing real ones when the original capacity of CPUs is different from the default SCHED_CAPACITY_SCALE. With this patch set, we don't try anymore to evaluate the number of available cores based on the group_capacity but instead we evaluate the usage of a group and compare it with its capacity.
This patchset mainly replaces the old capacity_factor method by a new one and keeps the general policy almost unchanged. These new metrics will be also used in later patches.
The CPU usage is based on a running time tracking version of the current implementation of the load average tracking. I also have a version that is based on the new implementation proposal [1] but I haven't provide the patches and results as [1] is still under review. I can provide change above [1] to change how CPU usage is computed and to adapt to new mecanism.
Change since V7 - add freq invariance for usage tracking - add freq invariance for scale_rt - update comments and commits' message - fix init of utilization_avg_contrib - fix prefer_sibling
Change since V6 - add group usage tracking - fix some commits' messages - minor fix like comments and argument order
Change since V5 - remove patches that have been merged since v5 : patches 01, 02, 03, 04, 05, 07 - update commit log and add more details on the purpose of the patches - fix/remove useless code with the rebase on patchset [2] - remove capacity_orig in sched_group_capacity as it is not used - move code in the right patch - add some helper function to factorize code
Change since V4 - rebase to manage conflicts with changes in selection of busiest group
Change since V3: - add usage_avg_contrib statistic which sums the running time of tasks on a rq - use usage_avg_contrib instead of runnable_avg_sum for cpu_utilization - fix replacement power by capacity - update some comments
Change since V2: - rebase on top of capacity renaming - fix wake_affine statistic update - rework nohz_kick_needed - optimize the active migration of a task from CPU with reduced capacity - rename group_activity by group_utilization and remove unused total_utilization - repair SD_PREFER_SIBLING and use it for SMT level - reorder patchset to gather patches with same topics
Change since V1: - add 3 fixes - correct some commit messages - replace capacity computation by activity - take into account current cpu capacity
[1] https://lkml.org/lkml/2014/10/10/131 [2] https://lkml.org/lkml/2014/7/25/589
Morten Rasmussen (2): sched: Track group sched_entity usage contributions sched: Make sched entity usage tracking scale-invariant
Vincent Guittot (8): sched: add per rq cpu_capacity_orig sched: remove frequency scaling from cpu_capacity sched: move cfs task on a CPU with higher capacity sched: add utilization_avg_contrib sched: get CPU's usage statistic sched: replace capacity_factor by usage sched: add SD_PREFER_SIBLING for SMT level sched: make scale_rt invariant with frequency
include/linux/sched.h | 21 ++- kernel/sched/core.c | 15 +- kernel/sched/debug.c | 12 +- kernel/sched/fair.c | 369 ++++++++++++++++++++++++++++++++------------------ kernel/sched/sched.h | 15 +- 5 files changed, 276 insertions(+), 156 deletions(-)
This new field cpu_capacity_orig reflects the original capacity of a CPU before being altered by rt tasks and/or IRQ
The cpu_capacity_orig will be used: - to detect when the capacity of a CPU has been noticeably reduced so we can trig load balance to look for a CPU with better capacity. As an example, we can detect when a CPU handles a significant amount of irq (with CONFIG_IRQ_TIME_ACCOUNTING) but this CPU is seen as an idle CPU by scheduler whereas CPUs, which are really idle, are available. - evaluate the available capacity for CFS tasks
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org Reviewed-by: Kamalesh Babulal kamalesh@linux.vnet.ibm.com --- kernel/sched/core.c | 2 +- kernel/sched/fair.c | 8 +++++++- kernel/sched/sched.h | 1 + 3 files changed, 9 insertions(+), 2 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c84bdc0..45ae52d 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7087,7 +7087,7 @@ void __init sched_init(void) #ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; - rq->cpu_capacity = SCHED_CAPACITY_SCALE; + rq->cpu_capacity = rq->cpu_capacity_orig = SCHED_CAPACITY_SCALE; rq->post_schedule = 0; rq->active_balance = 0; rq->next_balance = jiffies; diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd61cff..c3674da 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4089,6 +4089,11 @@ static unsigned long capacity_of(int cpu) return cpu_rq(cpu)->cpu_capacity; }
+static unsigned long capacity_orig_of(int cpu) +{ + return cpu_rq(cpu)->cpu_capacity_orig; +} + static unsigned long cpu_avg_load_per_task(int cpu) { struct rq *rq = cpu_rq(cpu); @@ -5776,6 +5781,7 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu)
capacity >>= SCHED_CAPACITY_SHIFT;
+ cpu_rq(cpu)->cpu_capacity_orig = capacity; sdg->sgc->capacity_orig = capacity;
if (sched_feat(ARCH_CAPACITY)) @@ -5837,7 +5843,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) * Runtime updates will correct capacity_orig. */ if (unlikely(!rq->sd)) { - capacity_orig += capacity_of(cpu); + capacity_orig += capacity_orig_of(cpu); capacity += capacity_of(cpu); continue; } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 6130251..12483d8 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -579,6 +579,7 @@ struct rq { struct sched_domain *sd;
unsigned long cpu_capacity; + unsigned long cpu_capacity_orig;
unsigned char idle_balance; /* For active balancing */
Now that arch_scale_cpu_capacity has been introduced to scale the original capacity, the arch_scale_freq_capacity is no longer used (it was previously used by ARM arch). Remove arch_scale_freq_capacity from the computation of cpu_capacity. The frequency invariance will be handled in the load tracking and not in the CPU capacity. arch_scale_freq_capacity will be revisited for scaling load with the current frequency of the CPUs in a later patch.
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 7 ------- 1 file changed, 7 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c3674da..bd214d2 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5784,13 +5784,6 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu) cpu_rq(cpu)->cpu_capacity_orig = capacity; sdg->sgc->capacity_orig = capacity;
- if (sched_feat(ARCH_CAPACITY)) - capacity *= arch_scale_freq_capacity(sd, cpu); - else - capacity *= default_scale_capacity(sd, cpu); - - capacity >>= SCHED_CAPACITY_SHIFT; - capacity *= scale_rt_capacity(cpu); capacity >>= SCHED_CAPACITY_SHIFT;
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases : - there is only 1 task on the busiest CPU. - we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) }
/* + * Check whether the capacity of the rq has been noticeably reduced by side + * activity. The imbalance_pct is used for the threshold. + * Return true is the capacity is reduced + */ +static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{ + return ((rq->cpu_capacity * sd->imbalance_pct) < + (rq->cpu_capacity_orig * 100)); +} + +/* * Group imbalance indicates (and tries to solve) the problem where balancing * groups is inadequate due to tsk_cpus_allowed() constraints. * @@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
+ /* + * The dst_cpu is idle and the src_cpu CPU has only 1 CFS task. + * It's worth migrating the task if the src_cpu's capacity is reduced + * because of other sched_class or IRQs whereas capacity stays + * available on dst_cpu. + */ + if ((env->idle != CPU_NOT_IDLE) && + (env->src_rq->cfs.h_nr_running == 1)) { + unsigned long src_eff_capacity, dst_eff_capacity; + + dst_eff_capacity = 100; + dst_eff_capacity *= capacity_of(env->dst_cpu); + dst_eff_capacity *= capacity_orig_of(env->src_cpu); + + src_eff_capacity = sd->imbalance_pct; + src_eff_capacity *= capacity_of(env->src_cpu); + src_eff_capacity *= capacity_orig_of(env->dst_cpu); + + if (src_eff_capacity < dst_eff_capacity) + return 1; + } + return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2); }
@@ -6661,6 +6695,9 @@ static int load_balance(int this_cpu, struct rq *this_rq,
schedstat_add(sd, lb_imbalance[idle], env.imbalance);
+ env.src_cpu = busiest->cpu; + env.src_rq = busiest; + ld_moved = 0; if (busiest->nr_running > 1) { /* @@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED; - env.src_cpu = busiest->cpu; - env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running);
more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle)
/* * Current heuristic for kicking the idle load balancer in the presence - * of an idle cpu is the system. + * of an idle cpu in the system. * - This rq has more than one task. - * - At any scheduler domain level, this cpu's scheduler group has multiple - * busy cpu's exceeding the group's capacity. + * - This rq has at least one CFS task and the capacity of the CPU is + * significantly reduced because of RT tasks or IRQs. + * - At parent of LLC scheduler domain level, this cpu's scheduler group has + * multiple busy cpu. * - For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler * domain span are idle. */ -static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu; + bool kick = false;
if (unlikely(rq->idle_balance)) - return 0; + return false;
/* * We may be recently in ticked or tickless idle mode. At the first @@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus))) - return 0; + return false;
if (time_before(now, nohz.next_balance)) - return 0; + return false;
if (rq->nr_running >= 2) - goto need_kick; + return true;
rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu)); - if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
- if (nr_busy > 1) - goto need_kick_unlock; + if (nr_busy > 1) { + kick = true; + goto unlock; + } + }
- sd = rcu_dereference(per_cpu(sd_asym, cpu)); + sd = rcu_dereference(rq->sd); + if (sd) { + if ((rq->cfs.h_nr_running >= 1) && + check_cpu_capacity(rq, sd)) { + kick = true; + goto unlock; + } + }
+ sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu)) - goto need_kick_unlock; + kick = true;
+unlock: rcu_read_unlock(); - return 0; - -need_kick_unlock: - rcu_read_unlock(); -need_kick: - return 1; + return kick; } #else static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { }
On 14/10/31 下午4:47, Vincent Guittot wrote:
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases :
- there is only 1 task on the busiest CPU.
- we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) } /*
- Check whether the capacity of the rq has been noticeably reduced by side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
- */
+static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{
- return ((rq->cpu_capacity * sd->imbalance_pct) <
(rq->cpu_capacity_orig * 100));+}
+/*
- Group imbalance indicates (and tries to solve) the problem where balancing
- groups is inadequate due to tsk_cpus_allowed() constraints.
@@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
- /*
* The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.
Why specify one task instead of not less than one?
Regards, Wanpeng Li
* It's worth migrating the task if the src_cpu's capacity is reduced* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/- if ((env->idle != CPU_NOT_IDLE) &&
(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;- }
- return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2); }
@@ -6661,6 +6695,9 @@ static int load_balance(int this_cpu, struct rq *this_rq, schedstat_add(sd, lb_imbalance[idle], env.imbalance);
- env.src_cpu = busiest->cpu;
- env.src_rq = busiest;
- ld_moved = 0; if (busiest->nr_running > 1) { /*
@@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED;
env.src_cpu = busiest->cpu;env.src_rq = busiest;more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) /*
- Current heuristic for kicking the idle load balancer in the presence
- of an idle cpu is the system.
- of an idle cpu in the system.
- This rq has more than one task.
- At any scheduler domain level, this cpu's scheduler group has multiple
busy cpu's exceeding the group's capacity.
- This rq has at least one CFS task and the capacity of the CPU is
significantly reduced because of RT tasks or IRQs.
- At parent of LLC scheduler domain level, this cpu's scheduler group has
*/
multiple busy cpu.
- For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
domain span are idle.-static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu;
- bool kick = false;
if (unlikely(rq->idle_balance))
return 0;
return false;/*
- We may be recently in ticked or tickless idle mode. At the first
@@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus)))
return 0;
return false;if (time_before(now, nohz.next_balance))
return 0;
return false;if (rq->nr_running >= 2)
goto need_kick;
return true;rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu));
- if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)goto need_kick_unlock;
if (nr_busy > 1) {kick = true;goto unlock;}- }
- sd = rcu_dereference(per_cpu(sd_asym, cpu));
- sd = rcu_dereference(rq->sd);
- if (sd) {
if ((rq->cfs.h_nr_running >= 1) &&check_cpu_capacity(rq, sd)) {kick = true;goto unlock;}- }
- sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu))
goto need_kick_unlock;
kick = true;+unlock: rcu_read_unlock();
- return 0;
-need_kick_unlock:
- rcu_read_unlock();
-need_kick:
- return 1;
- return kick; } #else static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { }
On 4 November 2014 09:30, Wanpeng Li kernellwp@gmail.com wrote:
On 14/10/31 下午4:47, Vincent Guittot wrote:
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases :
- there is only 1 task on the busiest CPU.
- we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) } /*
- Check whether the capacity of the rq has been noticeably reduced by
side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
- */
+static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{
return ((rq->cpu_capacity * sd->imbalance_pct) <(rq->cpu_capacity_orig * 100));+}
+/*
- Group imbalance indicates (and tries to solve) the problem where
balancing
- groups is inadequate due to tsk_cpus_allowed() constraints.
@@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
/** The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.Why specify one task instead of not less than one?
if cfs.h_nr_running == 0 (which should not occurs at that point), we don't need to do more test to check if it's worth moving the task because there is no task to move.
Regards, Vincent
Regards, Wanpeng Li
* It's worth migrating the task if the src_cpu's capacity isreduced
* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/if ((env->idle != CPU_NOT_IDLE) &&(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;}return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);*this_rq, schedstat_add(sd, lb_imbalance[idle], env.imbalance);
env.src_cpu = busiest->cpu;env.src_rq = busiest;ld_moved = 0; if (busiest->nr_running > 1) { /*@@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED;
env.src_cpu = busiest->cpu;env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate,busiest->nr_running); more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) /*
- Current heuristic for kicking the idle load balancer in the presence
- of an idle cpu is the system.
- of an idle cpu in the system.
- This rq has more than one task.
- At any scheduler domain level, this cpu's scheduler group has
multiple
busy cpu's exceeding the group's capacity.
- This rq has at least one CFS task and the capacity of the CPU is
significantly reduced because of RT tasks or IRQs.
- At parent of LLC scheduler domain level, this cpu's scheduler
group has
*/
multiple busy cpu.
- For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
domain span are idle.-static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu;
bool kick = false; if (unlikely(rq->idle_balance))
return 0;
return false; /* * We may be recently in ticked or tickless idle mode. At the first@@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus)))
return 0;
return false; if (time_before(now, nohz.next_balance))
return 0;
return false; if (rq->nr_running >= 2)
goto need_kick;
return true; rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu));
if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)goto need_kick_unlock;
if (nr_busy > 1) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu));sd = rcu_dereference(rq->sd);if (sd) {if ((rq->cfs.h_nr_running >= 1) &&check_cpu_capacity(rq, sd)) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu))
goto need_kick_unlock;
kick = true;
return 0;-need_kick_unlock:
rcu_read_unlock();-need_kick:
return 1;
return kick;idle) { }
Hi Vincent, On 14/11/4 下午5:41, Vincent Guittot wrote:
On 4 November 2014 09:30, Wanpeng Li kernellwp@gmail.com wrote:
On 14/10/31 下午4:47, Vincent Guittot wrote:
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases :
- there is only 1 task on the busiest CPU.
- we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) } /*
- Check whether the capacity of the rq has been noticeably reduced by
side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
- */
+static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{
return ((rq->cpu_capacity * sd->imbalance_pct) <(rq->cpu_capacity_orig * 100));+}
+/* * Group imbalance indicates (and tries to solve) the problem where balancing * groups is inadequate due to tsk_cpus_allowed() constraints. * @@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
/** The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.Why specify one task instead of not less than one?
if cfs.h_nr_running == 0 (which should not occurs at that point), we don't need to do more test to check if it's worth moving the task because there is no task to move.
Sorry for my confusing statement. I mean cfs.h_nr_running >= 1.
Regards, Wanpeng Li
Regards, Vincent
Regards, Wanpeng Li
* It's worth migrating the task if the src_cpu's capacity isreduced
* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/if ((env->idle != CPU_NOT_IDLE) &&(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;}return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);*this_rq, schedstat_add(sd, lb_imbalance[idle], env.imbalance);
env.src_cpu = busiest->cpu;env.src_rq = busiest;ld_moved = 0; if (busiest->nr_running > 1) { /*@@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED;
env.src_cpu = busiest->cpu;env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate,busiest->nr_running); more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) /* * Current heuristic for kicking the idle load balancer in the presence
- of an idle cpu is the system.
- of an idle cpu in the system.
- This rq has more than one task.
- At any scheduler domain level, this cpu's scheduler group has
multiple
busy cpu's exceeding the group's capacity.
- This rq has at least one CFS task and the capacity of the CPU is
significantly reduced because of RT tasks or IRQs.
- At parent of LLC scheduler domain level, this cpu's scheduler
group has
multiple busy cpu.*/
- For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
domain span are idle.-static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu;
bool kick = false; if (unlikely(rq->idle_balance))
return 0;
return false; /* * We may be recently in ticked or tickless idle mode. At the first@@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus)))
return 0;
return false; if (time_before(now, nohz.next_balance))
return 0;
return false; if (rq->nr_running >= 2)
goto need_kick;
return true; rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu));
if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)goto need_kick_unlock;
if (nr_busy > 1) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu));sd = rcu_dereference(rq->sd);if (sd) {if ((rq->cfs.h_nr_running >= 1) &&check_cpu_capacity(rq, sd)) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu))
goto need_kick_unlock;
kick = true;
return 0;-need_kick_unlock:
rcu_read_unlock();-need_kick:
return 1;
return kick;idle) { }
On 4 November 2014 11:42, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent,
+/* * Group imbalance indicates (and tries to solve) the problem where balancing * groups is inadequate due to tsk_cpus_allowed() constraints. * @@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
/** The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.Why specify one task instead of not less than one?
if cfs.h_nr_running == 0 (which should not occurs at that point), we don't need to do more test to check if it's worth moving the task because there is no task to move.
Sorry for my confusing statement. I mean cfs.h_nr_running >= 1.
ok.
If we have more than 1 task, we fall back into the default balancing policy than should move a not running task or that will use the imbalance field in case of the presence of pinned tasks
Regards, Vincent
Regards, Wanpeng Li
Regards, Vincent
Regards, Wanpeng Li
* It's worth migrating the task if the src_cpu's capacity isreduced
* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/if ((env->idle != CPU_NOT_IDLE) &&(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;}
Hi Vincent, On 10/31/14, 4:47 PM, Vincent Guittot wrote:
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such
I see the cpu capacity will be reduced if RT tasks are running in scale_rt_capacity(), could you point out where cpu capacity reduced due to IRQs?
Regards, Wanpeng Li
situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases :
- there is only 1 task on the busiest CPU.
- we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) } /*
- Check whether the capacity of the rq has been noticeably reduced by side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
- */
+static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{
- return ((rq->cpu_capacity * sd->imbalance_pct) <
(rq->cpu_capacity_orig * 100));+}
+/*
- Group imbalance indicates (and tries to solve) the problem where balancing
- groups is inadequate due to tsk_cpus_allowed() constraints.
@@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
- /*
* The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.* It's worth migrating the task if the src_cpu's capacity is reduced* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/- if ((env->idle != CPU_NOT_IDLE) &&
(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;- }
- return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2); }
@@ -6661,6 +6695,9 @@ static int load_balance(int this_cpu, struct rq *this_rq, schedstat_add(sd, lb_imbalance[idle], env.imbalance);
- env.src_cpu = busiest->cpu;
- env.src_rq = busiest;
- ld_moved = 0; if (busiest->nr_running > 1) { /*
@@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED;
env.src_cpu = busiest->cpu;env.src_rq = busiest;more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) /*
- Current heuristic for kicking the idle load balancer in the presence
- of an idle cpu is the system.
- of an idle cpu in the system.
- This rq has more than one task.
- At any scheduler domain level, this cpu's scheduler group has multiple
busy cpu's exceeding the group's capacity.
- This rq has at least one CFS task and the capacity of the CPU is
significantly reduced because of RT tasks or IRQs.
- At parent of LLC scheduler domain level, this cpu's scheduler group has
*/
multiple busy cpu.
- For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
domain span are idle.-static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu;
- bool kick = false;
if (unlikely(rq->idle_balance))
return 0;
return false;/*
- We may be recently in ticked or tickless idle mode. At the first
@@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus)))
return 0;
return false;if (time_before(now, nohz.next_balance))
return 0;
return false;if (rq->nr_running >= 2)
goto need_kick;
return true;rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu));
- if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)goto need_kick_unlock;
if (nr_busy > 1) {kick = true;goto unlock;}- }
- sd = rcu_dereference(per_cpu(sd_asym, cpu));
- sd = rcu_dereference(rq->sd);
- if (sd) {
if ((rq->cfs.h_nr_running >= 1) &&check_cpu_capacity(rq, sd)) {kick = true;goto unlock;}- }
- sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu))
goto need_kick_unlock;
kick = true;+unlock: rcu_read_unlock();
- return 0;
-need_kick_unlock:
- rcu_read_unlock();
-need_kick:
- return 1;
- return kick; } #else static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { }
On 18 November 2014 11:47, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent, On 10/31/14, 4:47 PM, Vincent Guittot wrote:
When a CPU is used to handle a lot of IRQs or some RT tasks, the remaining capacity for CFS tasks can be significantly reduced. Once we detect such
I see the cpu capacity will be reduced if RT tasks are running in scale_rt_capacity(), could you point out where cpu capacity reduced due to IRQs?
with CONFIG_IRQ_TIME_ACCOUNTING, the time spent in interrupt context will be added in rq->rt_avg
Regards, Vincent
Regards, Wanpeng Li
situation by comparing cpu_capacity_orig and cpu_capacity, we trig an idle load balance to check if it's worth moving its tasks on an idle CPU.
Once the idle load_balance has selected the busiest CPU, it will look for an active load balance for only two cases :
- there is only 1 task on the busiest CPU.
- we haven't been able to move a task of the busiest rq.
A CPU with a reduced capacity is included in the 1st case, and it's worth to actively migrate its task if the idle CPU has got full capacity. This test has been added in need_active_balance.
As a sidenote, this will note generate more spurious ilb because we already trig an ilb if there is more than 1 busy cpu. If this cpu is the only one that has a task, we will trig the ilb once for migrating the task.
The nohz_kick_needed function has been cleaned up a bit while adding the new test
env.src_cpu and env.src_rq must be set unconditionnally because they are used in need_active_balance which is called even if busiest->nr_running equals 1
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/fair.c | 86 ++++++++++++++++++++++++++++++++++++++++------------- 1 file changed, 65 insertions(+), 21 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index bd214d2..54468f3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5889,6 +5889,18 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group) } /*
- Check whether the capacity of the rq has been noticeably reduced by
side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
- */
+static inline int +check_cpu_capacity(struct rq *rq, struct sched_domain *sd) +{
return ((rq->cpu_capacity * sd->imbalance_pct) <(rq->cpu_capacity_orig * 100));+}
+/*
- Group imbalance indicates (and tries to solve) the problem where
balancing
- groups is inadequate due to tsk_cpus_allowed() constraints.
@@ -6562,6 +6574,28 @@ static int need_active_balance(struct lb_env *env) return 1; }
/** The dst_cpu is idle and the src_cpu CPU has only 1 CFS task.* It's worth migrating the task if the src_cpu's capacity isreduced
* because of other sched_class or IRQs whereas capacity stays* available on dst_cpu.*/if ((env->idle != CPU_NOT_IDLE) &&(env->src_rq->cfs.h_nr_running == 1)) {unsigned long src_eff_capacity, dst_eff_capacity;dst_eff_capacity = 100;dst_eff_capacity *= capacity_of(env->dst_cpu);dst_eff_capacity *= capacity_orig_of(env->src_cpu);src_eff_capacity = sd->imbalance_pct;src_eff_capacity *= capacity_of(env->src_cpu);src_eff_capacity *= capacity_orig_of(env->dst_cpu);if (src_eff_capacity < dst_eff_capacity)return 1;}return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2);*this_rq, schedstat_add(sd, lb_imbalance[idle], env.imbalance);
env.src_cpu = busiest->cpu;env.src_rq = busiest;ld_moved = 0; if (busiest->nr_running > 1) { /*@@ -6670,8 +6707,6 @@ static int load_balance(int this_cpu, struct rq *this_rq, * correctly treated as an imbalance. */ env.flags |= LBF_ALL_PINNED;
env.src_cpu = busiest->cpu;env.src_rq = busiest; env.loop_max = min(sysctl_sched_nr_migrate,busiest->nr_running); more_balance: @@ -7371,22 +7406,25 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) /*
- Current heuristic for kicking the idle load balancer in the presence
- of an idle cpu is the system.
- of an idle cpu in the system.
- This rq has more than one task.
- At any scheduler domain level, this cpu's scheduler group has
multiple
busy cpu's exceeding the group's capacity.
- This rq has at least one CFS task and the capacity of the CPU is
significantly reduced because of RT tasks or IRQs.
- At parent of LLC scheduler domain level, this cpu's scheduler
group has
*/
multiple busy cpu.
- For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
domain span are idle.-static inline int nohz_kick_needed(struct rq *rq) +static inline bool nohz_kick_needed(struct rq *rq) { unsigned long now = jiffies; struct sched_domain *sd; struct sched_group_capacity *sgc; int nr_busy, cpu = rq->cpu;
bool kick = false; if (unlikely(rq->idle_balance))
return 0;
return false; /* * We may be recently in ticked or tickless idle mode. At the first@@ -7400,38 +7438,44 @@ static inline int nohz_kick_needed(struct rq *rq) * balancing. */ if (likely(!atomic_read(&nohz.nr_cpus)))
return 0;
return false; if (time_before(now, nohz.next_balance))
return 0;
return false; if (rq->nr_running >= 2)
goto need_kick;
return true; rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_busy, cpu));
if (sd) { sgc = sd->groups->sgc; nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)goto need_kick_unlock;
if (nr_busy > 1) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu));sd = rcu_dereference(rq->sd);if (sd) {if ((rq->cfs.h_nr_running >= 1) &&check_cpu_capacity(rq, sd)) {kick = true;goto unlock;}}
sd = rcu_dereference(per_cpu(sd_asym, cpu)); if (sd && (cpumask_first_and(nohz.idle_cpus_mask, sched_domain_span(sd)) < cpu))
goto need_kick_unlock;
kick = true;
return 0;-need_kick_unlock:
rcu_read_unlock();-need_kick:
return 1;
return kick;idle) { }
Add new statistics which reflect the average time a task is running on the CPU and the sum of these running time of the tasks on a runqueue. The latter is named utilization_load_avg.
This patch is based on the usage metric that was proposed in the 1st versions of the per-entity load tracking patchset by Paul Turner pjt@google.com but that has be removed afterwards. This version differs from the original one in the sense that it's not linked to task_group.
The rq's utilization_load_avg will be used to check if a rq is overloaded or not instead of trying to compute how many tasks a group of CPUs can handle.
Rename runnable_avg_period into avg_period as it is now used with both runnable_avg_sum and running_avg_sum
Add some descriptions of the variables to explain their differences
cc: Paul Turner pjt@google.com cc: Ben Segall bsegall@google.com
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- include/linux/sched.h | 21 ++++++++++++--- kernel/sched/debug.c | 10 ++++--- kernel/sched/fair.c | 74 ++++++++++++++++++++++++++++++++++++++++----------- kernel/sched/sched.h | 8 +++++- 4 files changed, 89 insertions(+), 24 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h index 18f5262..b576b29 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1071,15 +1071,28 @@ struct load_weight { };
struct sched_avg { + u64 last_runnable_update; + s64 decay_count; + /* + * utilization_avg_contrib describes the amount of time that a + * sched_entity is running on a CPU. It is based on running_avg_sum + * and is scaled in the range [0..SCHED_LOAD_SCALE]. + * load_avg_contrib described the amount of time that a sched_entity + * is runnable on a rq. It is based on both runnable_avg_sum and the + * weight of the task. + */ + unsigned long load_avg_contrib, utilization_avg_contrib; /* * These sums represent an infinite geometric series and so are bound * above by 1024/(1-y). Thus we only need a u32 to store them for all * choices of y < 1-2^(-32)*1024. + * running_avg_sum reflects the time that the sched_entity is + * effectively running on the CPU. + * runnable_avg_sum represents the amount of time a sched_entity is on + * a runqueue which includes the running time that is monitored by + * running_avg_sum. */ - u32 runnable_avg_sum, runnable_avg_period; - u64 last_runnable_update; - s64 decay_count; - unsigned long load_avg_contrib; + u32 runnable_avg_sum, avg_period, running_avg_sum; };
#ifdef CONFIG_SCHEDSTATS diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index ce33780..f384452 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -71,7 +71,7 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group if (!se) { struct sched_avg *avg = &cpu_rq(cpu)->avg; P(avg->runnable_avg_sum); - P(avg->runnable_avg_period); + P(avg->avg_period); return; }
@@ -94,7 +94,7 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group P(se->load.weight); #ifdef CONFIG_SMP P(se->avg.runnable_avg_sum); - P(se->avg.runnable_avg_period); + P(se->avg.avg_period); P(se->avg.load_avg_contrib); P(se->avg.decay_count); #endif @@ -214,6 +214,8 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) cfs_rq->runnable_load_avg); SEQ_printf(m, " .%-30s: %ld\n", "blocked_load_avg", cfs_rq->blocked_load_avg); + SEQ_printf(m, " .%-30s: %ld\n", "utilization_load_avg", + cfs_rq->utilization_load_avg); #ifdef CONFIG_FAIR_GROUP_SCHED SEQ_printf(m, " .%-30s: %ld\n", "tg_load_contrib", cfs_rq->tg_load_contrib); @@ -628,8 +630,10 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) P(se.load.weight); #ifdef CONFIG_SMP P(se.avg.runnable_avg_sum); - P(se.avg.runnable_avg_period); + P(se.avg.running_avg_sum); + P(se.avg.avg_period); P(se.avg.load_avg_contrib); + P(se.avg.utilization_avg_contrib); P(se.avg.decay_count); #endif P(policy); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 54468f3..36a882f 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -670,6 +670,7 @@ static int select_idle_sibling(struct task_struct *p, int cpu); static unsigned long task_h_load(struct task_struct *p);
static inline void __update_task_entity_contrib(struct sched_entity *se); +static inline void __update_task_entity_utilization(struct sched_entity *se);
/* Give new task start runnable values to heavy its load in infant time */ void init_task_runnable_average(struct task_struct *p) @@ -678,9 +679,10 @@ void init_task_runnable_average(struct task_struct *p)
p->se.avg.decay_count = 0; slice = sched_slice(task_cfs_rq(p), &p->se) >> 10; - p->se.avg.runnable_avg_sum = slice; - p->se.avg.runnable_avg_period = slice; + p->se.avg.runnable_avg_sum = p->se.avg.running_avg_sum = slice; + p->se.avg.avg_period = slice; __update_task_entity_contrib(&p->se); + __update_task_entity_utilization(&p->se); } #else void init_task_runnable_average(struct task_struct *p) @@ -1548,7 +1550,7 @@ static u64 numa_get_avg_runtime(struct task_struct *p, u64 *period) *period = now - p->last_task_numa_placement; } else { delta = p->se.avg.runnable_avg_sum; - *period = p->se.avg.runnable_avg_period; + *period = p->se.avg.avg_period; }
p->last_sum_exec_runtime = runtime; @@ -2294,7 +2296,8 @@ static u32 __compute_runnable_contrib(u64 n) */ static __always_inline int __update_entity_runnable_avg(u64 now, struct sched_avg *sa, - int runnable) + int runnable, + int running) { u64 delta, periods; u32 runnable_contrib; @@ -2320,7 +2323,7 @@ static __always_inline int __update_entity_runnable_avg(u64 now, sa->last_runnable_update = now;
/* delta_w is the amount already accumulated against our next period */ - delta_w = sa->runnable_avg_period % 1024; + delta_w = sa->avg_period % 1024; if (delta + delta_w >= 1024) { /* period roll-over */ decayed = 1; @@ -2333,7 +2336,9 @@ static __always_inline int __update_entity_runnable_avg(u64 now, delta_w = 1024 - delta_w; if (runnable) sa->runnable_avg_sum += delta_w; - sa->runnable_avg_period += delta_w; + if (running) + sa->running_avg_sum += delta_w; + sa->avg_period += delta_w;
delta -= delta_w;
@@ -2343,20 +2348,26 @@ static __always_inline int __update_entity_runnable_avg(u64 now,
sa->runnable_avg_sum = decay_load(sa->runnable_avg_sum, periods + 1); - sa->runnable_avg_period = decay_load(sa->runnable_avg_period, + sa->running_avg_sum = decay_load(sa->running_avg_sum, + periods + 1); + sa->avg_period = decay_load(sa->avg_period, periods + 1);
/* Efficiently calculate \sum (1..n_period) 1024*y^i */ runnable_contrib = __compute_runnable_contrib(periods); if (runnable) sa->runnable_avg_sum += runnable_contrib; - sa->runnable_avg_period += runnable_contrib; + if (running) + sa->running_avg_sum += runnable_contrib; + sa->avg_period += runnable_contrib; }
/* Remainder of delta accrued against u_0` */ if (runnable) sa->runnable_avg_sum += delta; - sa->runnable_avg_period += delta; + if (running) + sa->running_avg_sum += delta; + sa->avg_period += delta;
return decayed; } @@ -2372,6 +2383,8 @@ static inline u64 __synchronize_entity_decay(struct sched_entity *se) return 0;
se->avg.load_avg_contrib = decay_load(se->avg.load_avg_contrib, decays); + se->avg.utilization_avg_contrib = + decay_load(se->avg.utilization_avg_contrib, decays); se->avg.decay_count = 0;
return decays; @@ -2408,7 +2421,7 @@ static inline void __update_tg_runnable_avg(struct sched_avg *sa,
/* The fraction of a cpu used by this cfs_rq */ contrib = div_u64((u64)sa->runnable_avg_sum << NICE_0_SHIFT, - sa->runnable_avg_period + 1); + sa->avg_period + 1); contrib -= cfs_rq->tg_runnable_contrib;
if (abs(contrib) > cfs_rq->tg_runnable_contrib / 64) { @@ -2461,7 +2474,8 @@ static inline void __update_group_entity_contrib(struct sched_entity *se)
static inline void update_rq_runnable_avg(struct rq *rq, int runnable) { - __update_entity_runnable_avg(rq_clock_task(rq), &rq->avg, runnable); + __update_entity_runnable_avg(rq_clock_task(rq), &rq->avg, runnable, + runnable); __update_tg_runnable_avg(&rq->avg, &rq->cfs); } #else /* CONFIG_FAIR_GROUP_SCHED */ @@ -2479,7 +2493,7 @@ static inline void __update_task_entity_contrib(struct sched_entity *se)
/* avoid overflowing a 32-bit type w/ SCHED_LOAD_SCALE */ contrib = se->avg.runnable_avg_sum * scale_load_down(se->load.weight); - contrib /= (se->avg.runnable_avg_period + 1); + contrib /= (se->avg.avg_period + 1); se->avg.load_avg_contrib = scale_load(contrib); }
@@ -2498,6 +2512,27 @@ static long __update_entity_load_avg_contrib(struct sched_entity *se) return se->avg.load_avg_contrib - old_contrib; }
+ +static inline void __update_task_entity_utilization(struct sched_entity *se) +{ + u32 contrib; + + /* avoid overflowing a 32-bit type w/ SCHED_LOAD_SCALE */ + contrib = se->avg.running_avg_sum * scale_load_down(SCHED_LOAD_SCALE); + contrib /= (se->avg.avg_period + 1); + se->avg.utilization_avg_contrib = scale_load(contrib); +} + +static long __update_entity_utilization_avg_contrib(struct sched_entity *se) +{ + long old_contrib = se->avg.utilization_avg_contrib; + + if (entity_is_task(se)) + __update_task_entity_utilization(se); + + return se->avg.utilization_avg_contrib - old_contrib; +} + static inline void subtract_blocked_load_contrib(struct cfs_rq *cfs_rq, long load_contrib) { @@ -2514,7 +2549,7 @@ static inline void update_entity_load_avg(struct sched_entity *se, int update_cfs_rq) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - long contrib_delta; + long contrib_delta, utilization_delta; u64 now;
/* @@ -2526,18 +2561,22 @@ static inline void update_entity_load_avg(struct sched_entity *se, else now = cfs_rq_clock_task(group_cfs_rq(se));
- if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq)) + if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq, + cfs_rq->curr == se)) return;
contrib_delta = __update_entity_load_avg_contrib(se); + utilization_delta = __update_entity_utilization_avg_contrib(se);
if (!update_cfs_rq) return;
- if (se->on_rq) + if (se->on_rq) { cfs_rq->runnable_load_avg += contrib_delta; - else + cfs_rq->utilization_load_avg += utilization_delta; + } else { subtract_blocked_load_contrib(cfs_rq, -contrib_delta); + } }
/* @@ -2612,6 +2651,7 @@ static inline void enqueue_entity_load_avg(struct cfs_rq *cfs_rq, }
cfs_rq->runnable_load_avg += se->avg.load_avg_contrib; + cfs_rq->utilization_load_avg += se->avg.utilization_avg_contrib; /* we force update consideration on load-balancer moves */ update_cfs_rq_blocked_load(cfs_rq, !wakeup); } @@ -2630,6 +2670,7 @@ static inline void dequeue_entity_load_avg(struct cfs_rq *cfs_rq, update_cfs_rq_blocked_load(cfs_rq, !sleep);
cfs_rq->runnable_load_avg -= se->avg.load_avg_contrib; + cfs_rq->utilization_load_avg -= se->avg.utilization_avg_contrib; if (sleep) { cfs_rq->blocked_load_avg += se->avg.load_avg_contrib; se->avg.decay_count = atomic64_read(&cfs_rq->decay_counter); @@ -2967,6 +3008,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) */ update_stats_wait_end(cfs_rq, se); __dequeue_entity(cfs_rq, se); + update_entity_load_avg(se, 1); }
update_stats_curr_start(cfs_rq, se); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 12483d8..35eca7d 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -343,8 +343,14 @@ struct cfs_rq { * Under CFS, load is tracked on a per-entity basis and aggregated up. * This allows for the description of both thread and group usage (in * the FAIR_GROUP_SCHED case). + * runnable_load_avg is the sum of the load_avg_contrib of the + * sched_entities on the rq. + * blocked_load_avg is similar to runnable_load_avg except that its + * the blocked sched_entities on the rq. + * utilization_load_avg is the sum of the average running time of the + * sched_entities on the rq. */ - unsigned long runnable_load_avg, blocked_load_avg; + unsigned long runnable_load_avg, blocked_load_avg, utilization_load_avg; atomic64_t decay_counter; u64 last_decay; atomic_long_t removed_load;
From: Morten Rasmussen morten.rasmussen@arm.com
Adds usage contribution tracking for group entities. Unlike se->avg.load_avg_contrib, se->avg.utilization_avg_contrib for group entities is the sum of se->avg.utilization_avg_contrib for all entities on the group runqueue. It is _not_ influenced in any way by the task group h_load. Hence it is representing the actual cpu usage of the group, not its intended load contribution which may differ significantly from the utilization on lightly utilized systems.
cc: Paul Turner pjt@google.com cc: Ben Segall bsegall@google.com
Signed-off-by: Morten Rasmussen morten.rasmussen@arm.com Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/debug.c | 2 ++ kernel/sched/fair.c | 3 +++ 2 files changed, 5 insertions(+)
diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index f384452..efb47ed 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -94,8 +94,10 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group P(se->load.weight); #ifdef CONFIG_SMP P(se->avg.runnable_avg_sum); + P(se->avg.running_avg_sum); P(se->avg.avg_period); P(se->avg.load_avg_contrib); + P(se->avg.utilization_avg_contrib); P(se->avg.decay_count); #endif #undef PN diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 36a882f..9ab5233 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2529,6 +2529,9 @@ static long __update_entity_utilization_avg_contrib(struct sched_entity *se)
if (entity_is_task(se)) __update_task_entity_utilization(se); + else + se->avg.utilization_avg_contrib = + group_cfs_rq(se)->utilization_load_avg;
return se->avg.utilization_avg_contrib - old_contrib; }
Monitor the usage level of each group of each sched_domain level. The usage is the portion of cpu_capacity_orig that is currently used on a CPU or group of CPUs. We use the utilization_load_avg to evaluate the usage level of each group.
The utilization_load_avg only takes into account the running time of the CFS tasks on a CPU with a maximum value of SCHED_LOAD_SCALE when the CPU is fully utilized. Nevertheless, we must cap utilization_load_avg which can be temporaly greater than SCHED_LOAD_SCALE after the migration of a task on this CPU and until the metrics are stabilized.
The utilization_load_avg is in the range [0..SCHED_LOAD_SCALE] to reflect the running load on the CPU whereas the available capacity for the CFS task is in the range [0..cpu_capacity_orig]. In order to test if a CPU is fully utilized by CFS tasks, we have to scale the utilization in the cpu_capacity_orig range of the CPU to get the usage of the latter. The usage can then be compared with the available capacity (ie cpu_capacity) to deduct the usage level of a CPU.
The frequency scaling invariance of the usage is not taken into account in this patch, it will be solved in another patch which will deal with frequency scaling invariance on the running_load_avg.
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 29 +++++++++++++++++++++++++++++ 1 file changed, 29 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 9ab5233..7ca5656 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -4552,6 +4552,33 @@ static int select_idle_sibling(struct task_struct *p, int target) done: return target; } +/* + * get_cpu_usage returns the amount of capacity of a CPU that is used by CFS + * tasks. The unit of the return value must capacity so we can compare the + * usage with the capacity of the CPU that is available for CFS task (ie + * cpu_capacity). + * cfs.utilization_load_avg is the sum of running time of runnable tasks on a + * CPU. It represents the amount of utilization of a CPU in the range + * [0..SCHED_LOAD_SCALE]. The usage of a CPU can't be higher than the full + * capacity of the CPU because it's about the running time on this CPU. + * Nevertheless, cfs.utilization_load_avg can be higher than SCHED_LOAD_SCALE + * because of unfortunate rounding in avg_period and running_load_avg or just + * after migrating tasks until the average stabilizes with the new running + * time. So we need to check that the usage stays into the range + * [0..cpu_capacity_orig] and cap if necessary. + * Without capping the usage, a group could be seen as overloaded (CPU0 usage + * at 121% + CPU1 usage at 80%) whereas CPU1 has 20% of available capacity/ + */ +static int get_cpu_usage(int cpu) +{ + unsigned long usage = cpu_rq(cpu)->cfs.utilization_load_avg; + unsigned long capacity = capacity_orig_of(cpu); + + if (usage >= SCHED_LOAD_SCALE) + return capacity; + + return (usage * capacity) >> SCHED_LOAD_SHIFT; +}
/* * select_task_rq_fair: Select target runqueue for the waking task in domains @@ -5681,6 +5708,7 @@ struct sg_lb_stats { unsigned long sum_weighted_load; /* Weighted load of group's tasks */ unsigned long load_per_task; unsigned long group_capacity; + unsigned long group_usage; /* Total usage of the group */ unsigned int sum_nr_running; /* Nr tasks running in the group */ unsigned int group_capacity_factor; unsigned int idle_cpus; @@ -6048,6 +6076,7 @@ static inline void update_sg_lb_stats(struct lb_env *env, load = source_load(i, load_idx);
sgs->group_load += load; + sgs->group_usage += get_cpu_usage(i); sgs->sum_nr_running += rq->cfs.h_nr_running;
if (rq->nr_running > 1)
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's capacity is SCHED_CAPACITY_SCALE. group_capacity_factor divides the capacity of the group by SCHED_LOAD_SCALE to estimate how many task can run in the group. Then, it compares this value with the sum of nr_running to decide if the group is overloaded or not. But the group_capacity_factor is hardly working for SMT system, it sometimes works for big cores but fails to do the right thing for little cores.
Below are two examples to illustrate the problem that this patch solves:
1- If the original capacity of a CPU is less than SCHED_CAPACITY_SCALE (640 as an example), a group of 3 CPUS will have a max capacity_factor of 2 (div_round_closest(3x640/1024) = 2) which means that it will be seen as overloaded even if we have only one task per CPU.
2 - If the original capacity of a CPU is greater than SCHED_CAPACITY_SCALE (1512 as an example), a group of 4 CPUs will have a capacity_factor of 4 (at max and thanks to the fix [0] for SMT system that prevent the apparition of ghost CPUs) but if one CPU is fully used by rt tasks (and its capacity is reduced to nearly nothing), the capacity factor of the group will still be 4 (div_round_closest(3*1512/1024) = 5 which is cap to 4 with [0]).
So, this patch tries to solve this issue by removing capacity_factor and replacing it with the 2 following metrics : -The available CPU's capacity for CFS tasks which is already used by load_balance. -The usage of the CPU by the CFS tasks. For the latter, utilization_avg_contrib has been re-introduced to compute the usage of a CPU by CFS tasks.
group_capacity_factor and group_has_free_capacity has been removed and replaced by group_no_capacity. We compare the number of task with the number of CPUs and we evaluate the level of utilization of the CPUs to define if a group is overloaded or if a group has capacity to handle more tasks.
For SD_PREFER_SIBLING, a group is tagged overloaded if it has more than 1 task so it will be selected in priority (among the overloaded groups). Since [1], SD_PREFER_SIBLING is no more concerned by the computation of load_above_capacity because local is not overloaded.
Finally, the sched_group->sched_group_capacity->capacity_orig has been removed because it's no more used during load balance.
[1] https://lkml.org/lkml/2014/8/12/295
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/core.c | 12 ----- kernel/sched/fair.c | 150 +++++++++++++++++++++++---------------------------- kernel/sched/sched.h | 2 +- 3 files changed, 69 insertions(+), 95 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 45ae52d..37fb92c 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5373,17 +5373,6 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; }
- /* - * Even though we initialize ->capacity to something semi-sane, - * we leave capacity_orig unset. This allows us to detect if - * domain iteration is still funny without causing /0 traps. - */ - if (!group->sgc->capacity_orig) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: domain->cpu_capacity not set\n"); - break; - } - if (!cpumask_weight(sched_group_cpus(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: empty group\n"); @@ -5868,7 +5857,6 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu) * die on a /0 trap. */ sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span); - sg->sgc->capacity_orig = sg->sgc->capacity;
/* * Make sure the first group of this domain contains the diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7ca5656..c04f2b2 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5710,11 +5710,10 @@ struct sg_lb_stats { unsigned long group_capacity; unsigned long group_usage; /* Total usage of the group */ unsigned int sum_nr_running; /* Nr tasks running in the group */ - unsigned int group_capacity_factor; unsigned int idle_cpus; unsigned int group_weight; enum group_type group_type; - int group_has_free_capacity; + int group_no_capacity; #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running; @@ -5855,7 +5854,6 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu) capacity >>= SCHED_CAPACITY_SHIFT;
cpu_rq(cpu)->cpu_capacity_orig = capacity; - sdg->sgc->capacity_orig = capacity;
capacity *= scale_rt_capacity(cpu); capacity >>= SCHED_CAPACITY_SHIFT; @@ -5871,7 +5869,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) { struct sched_domain *child = sd->child; struct sched_group *group, *sdg = sd->groups; - unsigned long capacity, capacity_orig; + unsigned long capacity; unsigned long interval;
interval = msecs_to_jiffies(sd->balance_interval); @@ -5883,7 +5881,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) return; }
- capacity_orig = capacity = 0; + capacity = 0;
if (child->flags & SD_OVERLAP) { /* @@ -5903,19 +5901,15 @@ void update_group_capacity(struct sched_domain *sd, int cpu) * Use capacity_of(), which is set irrespective of domains * in update_cpu_capacity(). * - * This avoids capacity/capacity_orig from being 0 and + * This avoids capacity from being 0 and * causing divide-by-zero issues on boot. - * - * Runtime updates will correct capacity_orig. */ if (unlikely(!rq->sd)) { - capacity_orig += capacity_orig_of(cpu); capacity += capacity_of(cpu); continue; }
sgc = rq->sd->groups->sgc; - capacity_orig += sgc->capacity_orig; capacity += sgc->capacity; } } else { @@ -5926,42 +5920,15 @@ void update_group_capacity(struct sched_domain *sd, int cpu)
group = child->groups; do { - capacity_orig += group->sgc->capacity_orig; capacity += group->sgc->capacity; group = group->next; } while (group != child->groups); }
- sdg->sgc->capacity_orig = capacity_orig; sdg->sgc->capacity = capacity; }
/* - * 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_CAPACITY_SCALE - */ - if (!(sd->flags & SD_SHARE_CPUCAPACITY)) - return 0; - - /* - * If ~90% of the cpu_capacity is still there, we're good. - */ - if (group->sgc->capacity * 32 > group->sgc->capacity_orig * 29) - return 1; - - return 0; -} - -/* * Check whether the capacity of the rq has been noticeably reduced by side * activity. The imbalance_pct is used for the threshold. * Return true is the capacity is reduced @@ -6008,37 +5975,54 @@ static inline int sg_imbalanced(struct sched_group *group) }
/* - * Compute the group capacity factor. - * - * Avoid the issue where N*frac(smt_capacity) >= 1 creates 'phantom' cores by - * first dividing out the smt factor and computing the actual number of cores - * and limit unit capacity with that. + * group_has_capacity returns true if the group has spare capacity that could + * be used by some tasks. We consider that a group has spare capacity if the + * number of task is smaller than the number of CPUs or if the usage is lower + * than the available capacity for CFS tasks. For the latter, we use a + * threshold to stabilize the state, to take into account the variance of the + * tasks' load and to return true if the available capacity in meaningful for + * the load balancer. As an example, an available capacity of 1% can appear + * but it doesn't make any benefit for the load balance. */ -static inline int sg_capacity_factor(struct lb_env *env, struct sched_group *group) +static inline bool +group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs) { - unsigned int capacity_factor, smt, cpus; - unsigned int capacity, capacity_orig; + if ((sgs->group_capacity * 100) > + (sgs->group_usage * env->sd->imbalance_pct)) + return true;
- capacity = group->sgc->capacity; - capacity_orig = group->sgc->capacity_orig; - cpus = group->group_weight; + if (sgs->sum_nr_running < sgs->group_weight) + return true;
- /* smt := ceil(cpus / capacity), assumes: 1 < smt_capacity < 2 */ - smt = DIV_ROUND_UP(SCHED_CAPACITY_SCALE * cpus, capacity_orig); - capacity_factor = cpus / smt; /* cores */ + return false; +}
- capacity_factor = min_t(unsigned, - capacity_factor, DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE)); - if (!capacity_factor) - capacity_factor = fix_small_capacity(env->sd, group); +/* + * group_is_overloaded returns true if the group has more tasks than it can + * handle. We consider that a group is overloaded if the number of tasks is + * greater than the number of CPUs and the tasks already use all available + * capacity for CFS tasks. For the latter, we use a threshold to stabilize + * the state, to take into account the variance of tasks' load and to return + * true if available capacity is no more meaningful for load balancer + */ +static inline bool +group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs) +{ + if (sgs->sum_nr_running <= sgs->group_weight) + return false; + + if ((sgs->group_capacity * 100) < + (sgs->group_usage * env->sd->imbalance_pct)) + return true;
- return capacity_factor; + return false; }
-static enum group_type -group_classify(struct sched_group *group, struct sg_lb_stats *sgs) +static enum group_type group_classify(struct lb_env *env, + struct sched_group *group, + struct sg_lb_stats *sgs) { - if (sgs->sum_nr_running > sgs->group_capacity_factor) + if (sgs->group_no_capacity) return group_overloaded;
if (sg_imbalanced(group)) @@ -6099,11 +6083,9 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running;
sgs->group_weight = group->group_weight; - sgs->group_capacity_factor = sg_capacity_factor(env, group); - sgs->group_type = group_classify(group, sgs);
- if (sgs->group_capacity_factor > sgs->sum_nr_running) - sgs->group_has_free_capacity = 1; + sgs->group_no_capacity = group_is_overloaded(env, sgs); + sgs->group_type = group_classify(env, group, sgs); }
/** @@ -6225,17 +6207,20 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
/* * In case the child domain prefers tasks go to siblings - * first, lower the sg capacity factor to one so that we'll try + * first, lower the sg capacity so that we'll try * and move all the excess tasks away. We lower the capacity * of a group only if the local group has the capacity to fit - * these excess tasks, i.e. nr_running < group_capacity_factor. The - * extra check prevents the case where you always pull from the - * heaviest group when it is already under-utilized (possible - * with a large weight task outweighs the tasks on the system). + * these excess tasks. The extra check prevents the case where + * you always pull from the heaviest group when it is already + * under-utilized (possible with a large weight task outweighs + * the tasks on the system). */ if (prefer_sibling && sds->local && - sds->local_stat.group_has_free_capacity) - sgs->group_capacity_factor = min(sgs->group_capacity_factor, 1U); + group_has_capacity(env, &sds->local_stat) && + (sgs->sum_nr_running > 1)) { + sgs->group_no_capacity = 1; + sgs->group_type = group_overloaded; + }
if (update_sd_pick_busiest(env, sds, sg, sgs)) { sds->busiest = sg; @@ -6414,11 +6399,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ if (busiest->group_type == group_overloaded && local->group_type == group_overloaded) { - load_above_capacity = - (busiest->sum_nr_running - busiest->group_capacity_factor); - - load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_CAPACITY_SCALE); - load_above_capacity /= busiest->group_capacity; + load_above_capacity = busiest->sum_nr_running * + SCHED_LOAD_SCALE; + if (load_above_capacity > busiest->group_capacity) + load_above_capacity -= busiest->group_capacity; + else + load_above_capacity = ~0UL; }
/* @@ -6481,6 +6467,7 @@ 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; @@ -6501,8 +6488,8 @@ 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_free_capacity && - !busiest->group_has_free_capacity) + if (env->idle == CPU_NEWLY_IDLE && group_has_capacity(env, local) && + busiest->group_no_capacity) goto force_balance;
/* @@ -6561,7 +6548,7 @@ 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 capacity, capacity_factor, wl; + unsigned long capacity, wl; enum fbq_type rt;
rq = cpu_rq(i); @@ -6590,9 +6577,6 @@ static struct rq *find_busiest_queue(struct lb_env *env, continue;
capacity = capacity_of(i); - capacity_factor = DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE); - if (!capacity_factor) - capacity_factor = fix_small_capacity(env->sd, group);
wl = weighted_cpuload(i);
@@ -6600,7 +6584,9 @@ static struct rq *find_busiest_queue(struct lb_env *env, * When comparing with imbalance, use weighted_cpuload() * which is not scaled with the cpu capacity. */ - if (capacity_factor && rq->nr_running == 1 && wl > env->imbalance) + + if (rq->nr_running == 1 && wl > env->imbalance && + !check_cpu_capacity(rq, env->sd)) continue;
/* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 35eca7d..7963981 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -759,7 +759,7 @@ struct sched_group_capacity { * CPU capacity of this group, SCHED_LOAD_SCALE being max capacity * for a single CPU. */ - unsigned int capacity, capacity_orig; + unsigned int capacity; unsigned long next_update; int imbalance; /* XXX unrelated to capacity but shared group state */ /*
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
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's capacity is SCHED_CAPACITY_SCALE. group_capacity_factor divides the capacity of the group by SCHED_LOAD_SCALE to estimate how many task can run in the group. Then, it compares this value with the sum of nr_running to decide if the group is overloaded or not. But the group_capacity_factor is hardly working for SMT system, it sometimes works for big cores but fails to do the right thing for little cores.
Below are two examples to illustrate the problem that this patch solves:
1- If the original capacity of a CPU is less than SCHED_CAPACITY_SCALE (640 as an example), a group of 3 CPUS will have a max capacity_factor of 2 (div_round_closest(3x640/1024) = 2) which means that it will be seen as overloaded even if we have only one task per CPU.
2 - If the original capacity of a CPU is greater than SCHED_CAPACITY_SCALE (1512 as an example), a group of 4 CPUs will have a capacity_factor of 4 (at max and thanks to the fix [0] for SMT system that prevent the apparition
What's the story of 'fix [0] for SMT system' here?
Regards, Wanpeng Li
of ghost CPUs) but if one CPU is fully used by rt tasks (and its capacity is reduced to nearly nothing), the capacity factor of the group will still be 4 (div_round_closest(3*1512/1024) = 5 which is cap to 4 with [0]).
So, this patch tries to solve this issue by removing capacity_factor and replacing it with the 2 following metrics : -The available CPU's capacity for CFS tasks which is already used by load_balance. -The usage of the CPU by the CFS tasks. For the latter, utilization_avg_contrib has been re-introduced to compute the usage of a CPU by CFS tasks.
group_capacity_factor and group_has_free_capacity has been removed and replaced by group_no_capacity. We compare the number of task with the number of CPUs and we evaluate the level of utilization of the CPUs to define if a group is overloaded or if a group has capacity to handle more tasks.
For SD_PREFER_SIBLING, a group is tagged overloaded if it has more than 1 task so it will be selected in priority (among the overloaded groups). Since [1], SD_PREFER_SIBLING is no more concerned by the computation of load_above_capacity because local is not overloaded.
Finally, the sched_group->sched_group_capacity->capacity_orig has been removed because it's no more used during load balance.
[1] https://lkml.org/lkml/2014/8/12/295
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org
kernel/sched/core.c | 12 ----- kernel/sched/fair.c | 150 +++++++++++++++++++++++---------------------------- kernel/sched/sched.h | 2 +- 3 files changed, 69 insertions(+), 95 deletions(-)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 45ae52d..37fb92c 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5373,17 +5373,6 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; }
/** Even though we initialize ->capacity to something semi-sane,* we leave capacity_orig unset. This allows us to detect if* domain iteration is still funny without causing /0 traps.*/if (!group->sgc->capacity_orig) {printk(KERN_CONT "\n");printk(KERN_ERR "ERROR: domain->cpu_capacity not set\n");break;}- if (!cpumask_weight(sched_group_cpus(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: empty group\n");
@@ -5868,7 +5857,6 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu) * die on a /0 trap. */ sg->sgc->capacity = SCHED_CAPACITY_SCALE * cpumask_weight(sg_span);
sg->sgc->capacity_orig = sg->sgc->capacity;/* * Make sure the first group of this domain contains the diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7ca5656..c04f2b2 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5710,11 +5710,10 @@ struct sg_lb_stats { unsigned long group_capacity; unsigned long group_usage; /* Total usage of the group */ unsigned int sum_nr_running; /* Nr tasks running in the group */
- unsigned int group_capacity_factor; unsigned int idle_cpus; unsigned int group_weight; enum group_type group_type;
- int group_has_free_capacity;
- int group_no_capacity; #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running;
@@ -5855,7 +5854,6 @@ static void update_cpu_capacity(struct sched_domain *sd, int cpu) capacity >>= SCHED_CAPACITY_SHIFT; cpu_rq(cpu)->cpu_capacity_orig = capacity;
- sdg->sgc->capacity_orig = capacity;
capacity *= scale_rt_capacity(cpu); capacity >>= SCHED_CAPACITY_SHIFT; @@ -5871,7 +5869,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) { struct sched_domain *child = sd->child; struct sched_group *group, *sdg = sd->groups;
- unsigned long capacity, capacity_orig;
- unsigned long capacity; unsigned long interval;
interval = msecs_to_jiffies(sd->balance_interval); @@ -5883,7 +5881,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) return; }
- capacity_orig = capacity = 0;
- capacity = 0;
if (child->flags & SD_OVERLAP) { /* @@ -5903,19 +5901,15 @@ void update_group_capacity(struct sched_domain *sd, int cpu) * Use capacity_of(), which is set irrespective of domains * in update_cpu_capacity(). *
* This avoids capacity/capacity_orig from being 0 and
* This avoids capacity from being 0 and * causing divide-by-zero issues on boot.
** Runtime updates will correct capacity_orig. */ if (unlikely(!rq->sd)) {capacity_orig += capacity_orig_of(cpu); capacity += capacity_of(cpu); continue; }sgc = rq->sd->groups->sgc;
capacity_orig += sgc->capacity_orig; capacity += sgc->capacity;@@ -5926,42 +5920,15 @@ void update_group_capacity(struct sched_domain *sd, int cpu) group = child->groups; do {
capacity_orig += group->sgc->capacity_orig; capacity += group->sgc->capacity; group = group->next;
- sdg->sgc->capacity_orig = capacity_orig; sdg->sgc->capacity = capacity; }
/*
- 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_CAPACITY_SCALE*/- if (!(sd->flags & SD_SHARE_CPUCAPACITY))
return 0;- /*
* If ~90% of the cpu_capacity is still there, we're good.*/- if (group->sgc->capacity * 32 > group->sgc->capacity_orig * 29)
return 1;- return 0;
-}
-/*
- Check whether the capacity of the rq has been noticeably reduced by side
- activity. The imbalance_pct is used for the threshold.
- Return true is the capacity is reduced
@@ -6008,37 +5975,54 @@ static inline int sg_imbalanced(struct sched_group *group) } /*
- Compute the group capacity factor.
- Avoid the issue where N*frac(smt_capacity) >= 1 creates 'phantom' cores by
- first dividing out the smt factor and computing the actual number of cores
- and limit unit capacity with that.
- group_has_capacity returns true if the group has spare capacity that could
- be used by some tasks. We consider that a group has spare capacity if the
- number of task is smaller than the number of CPUs or if the usage is lower
- than the available capacity for CFS tasks. For the latter, we use a
- threshold to stabilize the state, to take into account the variance of the
- tasks' load and to return true if the available capacity in meaningful for
- the load balancer. As an example, an available capacity of 1% can appear
*/
- but it doesn't make any benefit for the load balance.
-static inline int sg_capacity_factor(struct lb_env *env, struct sched_group *group) +static inline bool +group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs) {
- unsigned int capacity_factor, smt, cpus;
- unsigned int capacity, capacity_orig;
- if ((sgs->group_capacity * 100) >
(sgs->group_usage * env->sd->imbalance_pct))return true;
- capacity = group->sgc->capacity;
- capacity_orig = group->sgc->capacity_orig;
- cpus = group->group_weight;
- if (sgs->sum_nr_running < sgs->group_weight)
return true;
- /* smt := ceil(cpus / capacity), assumes: 1 < smt_capacity < 2 */
- smt = DIV_ROUND_UP(SCHED_CAPACITY_SCALE * cpus, capacity_orig);
- capacity_factor = cpus / smt; /* cores */
- return false;
+}
- capacity_factor = min_t(unsigned,
capacity_factor, DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE));- if (!capacity_factor)
capacity_factor = fix_small_capacity(env->sd, group);+/*
- group_is_overloaded returns true if the group has more tasks than it can
- handle. We consider that a group is overloaded if the number of tasks is
- greater than the number of CPUs and the tasks already use all available
- capacity for CFS tasks. For the latter, we use a threshold to stabilize
- the state, to take into account the variance of tasks' load and to return
- true if available capacity is no more meaningful for load balancer
- */
+static inline bool +group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs) +{
- if (sgs->sum_nr_running <= sgs->group_weight)
return false;- if ((sgs->group_capacity * 100) <
(sgs->group_usage * env->sd->imbalance_pct))return true;
- return capacity_factor;
- return false; }
-static enum group_type -group_classify(struct sched_group *group, struct sg_lb_stats *sgs) +static enum group_type group_classify(struct lb_env *env,
struct sched_group *group,struct sg_lb_stats *sgs)
- if (sgs->sum_nr_running > sgs->group_capacity_factor)
- if (sgs->group_no_capacity) return group_overloaded;
if (sg_imbalanced(group)) @@ -6099,11 +6083,9 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running; sgs->group_weight = group->group_weight;
- sgs->group_capacity_factor = sg_capacity_factor(env, group);
- sgs->group_type = group_classify(group, sgs);
- if (sgs->group_capacity_factor > sgs->sum_nr_running)
sgs->group_has_free_capacity = 1;
- sgs->group_no_capacity = group_is_overloaded(env, sgs);
- sgs->group_type = group_classify(env, group, sgs); }
/** @@ -6225,17 +6207,20 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd /* * In case the child domain prefers tasks go to siblings
* first, lower the sg capacity factor to one so that we'll try
* first, lower the sg capacity so that we'll try
- and move all the excess tasks away. We lower the capacity
- of a group only if the local group has the capacity to fit
* these excess tasks, i.e. nr_running < group_capacity_factor. The* extra check prevents the case where you always pull from the* heaviest group when it is already under-utilized (possible* with a large weight task outweighs the tasks on the system).
* these excess tasks. The extra check prevents the case where* you always pull from the heaviest group when it is already* under-utilized (possible with a large weight task outweighs* the tasks on the system).
sds->local_stat.group_has_free_capacity)sgs->group_capacity_factor = min(sgs->group_capacity_factor, 1U);
group_has_capacity(env, &sds->local_stat) &&(sgs->sum_nr_running > 1)) {sgs->group_no_capacity = 1;sgs->group_type = group_overloaded;}if (update_sd_pick_busiest(env, sds, sg, sgs)) { sds->busiest = sg; @@ -6414,11 +6399,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ if (busiest->group_type == group_overloaded && local->group_type == group_overloaded) {
load_above_capacity =(busiest->sum_nr_running - busiest->group_capacity_factor);load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_CAPACITY_SCALE);load_above_capacity /= busiest->group_capacity;
load_above_capacity = busiest->sum_nr_running *SCHED_LOAD_SCALE;if (load_above_capacity > busiest->group_capacity)load_above_capacity -= busiest->group_capacity;elseload_above_capacity = ~0UL;/* @@ -6481,6 +6467,7 @@ 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;
@@ -6501,8 +6488,8 @@ 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_free_capacity &&
!busiest->group_has_free_capacity)
- if (env->idle == CPU_NEWLY_IDLE && group_has_capacity(env, local) &&
busiest->group_no_capacity)/* @@ -6561,7 +6548,7 @@ 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 capacity, capacity_factor, wl;
unsigned long capacity, wl;rq = cpu_rq(i); @@ -6590,9 +6577,6 @@ static struct rq *find_busiest_queue(struct lb_env *env, continue; capacity = capacity_of(i);
capacity_factor = DIV_ROUND_CLOSEST(capacity, SCHED_CAPACITY_SCALE);if (!capacity_factor)capacity_factor = fix_small_capacity(env->sd, group);wl = weighted_cpuload(i); @@ -6600,7 +6584,9 @@ static struct rq *find_busiest_queue(struct lb_env *env, * When comparing with imbalance, use weighted_cpuload() * which is not scaled with the cpu capacity. */
if (capacity_factor && rq->nr_running == 1 && wl > env->imbalance)
if (rq->nr_running == 1 && wl > env->imbalance &&!check_cpu_capacity(rq, env->sd)) continue;/* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 35eca7d..7963981 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -759,7 +759,7 @@ struct sched_group_capacity { * CPU capacity of this group, SCHED_LOAD_SCALE being max capacity * for a single CPU. */
- unsigned int capacity, capacity_orig;
- unsigned int capacity; unsigned long next_update; int imbalance; /* XXX unrelated to capacity but shared group state */ /*
On 3 November 2014 08:01, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
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's capacity is SCHED_CAPACITY_SCALE. group_capacity_factor divides the capacity of the group by SCHED_LOAD_SCALE to estimate how many task can run in the group. Then, it compares this value with the sum of nr_running to decide if the group is overloaded or not. But the group_capacity_factor is hardly working for SMT system, it sometimes works for big cores but fails to do the right thing for little cores.
Below are two examples to illustrate the problem that this patch solves:
1- If the original capacity of a CPU is less than SCHED_CAPACITY_SCALE (640 as an example), a group of 3 CPUS will have a max capacity_factor of 2 (div_round_closest(3x640/1024) = 2) which means that it will be seen as overloaded even if we have only one task per CPU.
2 - If the original capacity of a CPU is greater than SCHED_CAPACITY_SCALE (1512 as an example), a group of 4 CPUs will have a capacity_factor of 4 (at max and thanks to the fix [0] for SMT system that prevent the apparition
What's the story of 'fix [0] for SMT system' here?
The [0] url has disappeared. please find it below: [0] https://lkml.org/lkml/2013/8/28/194
Let me know if your need more explanation than my answer of your previous question for the cover letter
Regards, Vincent
Regards, Wanpeng Li
of ghost CPUs) but if one CPU is fully used by rt tasks (and its capacity is reduced to nearly nothing), the capacity factor of the group will still be 4 (div_round_closest(3*1512/1024) = 5 which is cap to 4 with [0]).
[snip]
On Fri, Oct 31, 2014 at 09:47:29AM +0100, Vincent Guittot wrote:
@@ -6414,11 +6399,12 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ if (busiest->group_type == group_overloaded && local->group_type == group_overloaded) {
load_above_capacity =(busiest->sum_nr_running - busiest->group_capacity_factor);load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_CAPACITY_SCALE);load_above_capacity /= busiest->group_capacity;
load_above_capacity = busiest->sum_nr_running *SCHED_LOAD_SCALE;if (load_above_capacity > busiest->group_capacity)load_above_capacity -= busiest->group_capacity;elseload_above_capacity = ~0UL;/*
It seems to me we no longer have need to assume each task contributes SCHED_LOAD_SCALE, do we?
But as it stands I tihnk this patch already does too much -- it could do with a splitting, but let me stare at is a wee bit more.
Add the SD_PREFER_SIBLING flag for SMT level in order to ensure that the scheduler will put at least 1 task per core.
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org Reviewed-by: Preeti U. Murthy preeti@linux.vnet.ibm.com --- kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 37fb92c..731f2ad 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6165,6 +6165,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu) */
if (sd->flags & SD_SHARE_CPUCAPACITY) { + sd->flags |= SD_PREFER_SIBLING; sd->imbalance_pct = 110; sd->smt_gain = 1178; /* ~15% */
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
Add the SD_PREFER_SIBLING flag for SMT level in order to ensure that the scheduler will put at least 1 task per core.
What's the behavior before this patch?
Regards, Wanpeng Li
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org Reviewed-by: Preeti U. Murthy preeti@linux.vnet.ibm.com
kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 37fb92c..731f2ad 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6165,6 +6165,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu) */ if (sd->flags & SD_SHARE_CPUCAPACITY) {
sd->flags |= SD_PREFER_SIBLING;
On 4 November 2014 04:21, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
Add the SD_PREFER_SIBLING flag for SMT level in order to ensure that the scheduler will put at least 1 task per core.
What's the behavior before this patch?
Before this patch, the capacity_factor was 1 at core level (whatever the number of CPUs in the core) so the group that represents this core was overloaded as soon as more than 1 task was running on this group. So if a core has 2 tasks running in its group, it is seen as overloaded and the scheduler will try to migrate 1 task on a group that is idle
Regards, Vincent
Regards, Wanpeng Li
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org Reviewed-by: Preeti U. Murthy preeti@linux.vnet.ibm.com
kernel/sched/core.c | 1 + 1 file changed, 1 insertion(+)
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 37fb92c..731f2ad 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -6165,6 +6165,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu) */ if (sd->flags & SD_SHARE_CPUCAPACITY) {
sd->flags |= SD_PREFER_SIBLING; sd->imbalance_pct = 110; sd->smt_gain = 1178; /* ~15% */
From: Morten Rasmussen morten.rasmussen@arm.com
Apply frequency scale-invariance correction factor to usage tracking. Each segment of the running_load_avg geometric series is now scaled by the current frequency so the utilization_avg_contrib of each entity will be invariant with frequency scaling. As a result, utilization_load_avg which is the sum of utilization_avg_contrib, becomes invariant too. So the usage level that is returned by get_cpu_usage, stays relative to the max frequency as the cpu_capacity which is is compared against. Then, we want the keep the load tracking values in a 32bits type, which implies that the max value of {runnable|running}_avg_sum must be lower than 2^32/88761=48388 (88761 is the max weigth of a task). As LOAD_AVG_MAX = 47742, arch_scale_freq_capacity must return a value less than (48388/47742) << SCHED_CAPACITY_SHIFT = 1037 (SCHED_SCALE_CAPACITY = 1024). So we define the range to [0..SCHED_SCALE_CAPACITY] in order to avoid overflow.
cc: Paul Turner pjt@google.com cc: Ben Segall bsegall@google.com
Signed-off-by: Morten Rasmussen morten.rasmussen@arm.com Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 21 ++++++++++++++------- 1 file changed, 14 insertions(+), 7 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c04f2b2..31c3ec2 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -2266,6 +2266,8 @@ static u32 __compute_runnable_contrib(u64 n) return contrib + runnable_avg_yN_sum[n]; }
+unsigned long __weak arch_scale_freq_capacity(struct sched_domain *sd, int cpu); + /* * We can represent the historical contribution to runnable average as the * coefficients of a geometric series. To do this we sub-divide our runnable @@ -2294,7 +2296,7 @@ static u32 __compute_runnable_contrib(u64 n) * load_avg = u_0` + y*(u_0 + u_1*y + u_2*y^2 + ... ) * = u_0 + u_1*y + u_2*y^2 + ... [re-labeling u_i --> u_{i+1}] */ -static __always_inline int __update_entity_runnable_avg(u64 now, +static __always_inline int __update_entity_runnable_avg(u64 now, int cpu, struct sched_avg *sa, int runnable, int running) @@ -2302,6 +2304,7 @@ static __always_inline int __update_entity_runnable_avg(u64 now, u64 delta, periods; u32 runnable_contrib; int delta_w, decayed = 0; + unsigned long scale_freq = arch_scale_freq_capacity(NULL, cpu);
delta = now - sa->last_runnable_update; /* @@ -2337,7 +2340,8 @@ static __always_inline int __update_entity_runnable_avg(u64 now, if (runnable) sa->runnable_avg_sum += delta_w; if (running) - sa->running_avg_sum += delta_w; + sa->running_avg_sum += delta_w * scale_freq + >> SCHED_CAPACITY_SHIFT; sa->avg_period += delta_w;
delta -= delta_w; @@ -2358,7 +2362,8 @@ static __always_inline int __update_entity_runnable_avg(u64 now, if (runnable) sa->runnable_avg_sum += runnable_contrib; if (running) - sa->running_avg_sum += runnable_contrib; + sa->running_avg_sum += runnable_contrib * scale_freq + >> SCHED_CAPACITY_SHIFT; sa->avg_period += runnable_contrib; }
@@ -2366,7 +2371,8 @@ static __always_inline int __update_entity_runnable_avg(u64 now, if (runnable) sa->runnable_avg_sum += delta; if (running) - sa->running_avg_sum += delta; + sa->running_avg_sum += delta * scale_freq + >> SCHED_CAPACITY_SHIFT; sa->avg_period += delta;
return decayed; @@ -2474,8 +2480,8 @@ static inline void __update_group_entity_contrib(struct sched_entity *se)
static inline void update_rq_runnable_avg(struct rq *rq, int runnable) { - __update_entity_runnable_avg(rq_clock_task(rq), &rq->avg, runnable, - runnable); + __update_entity_runnable_avg(rq_clock_task(rq), cpu_of(rq), &rq->avg, + runnable, runnable); __update_tg_runnable_avg(&rq->avg, &rq->cfs); } #else /* CONFIG_FAIR_GROUP_SCHED */ @@ -2553,6 +2559,7 @@ static inline void update_entity_load_avg(struct sched_entity *se, { struct cfs_rq *cfs_rq = cfs_rq_of(se); long contrib_delta, utilization_delta; + int cpu = cpu_of(rq_of(cfs_rq)); u64 now;
/* @@ -2564,7 +2571,7 @@ static inline void update_entity_load_avg(struct sched_entity *se, else now = cfs_rq_clock_task(group_cfs_rq(se));
- if (!__update_entity_runnable_avg(now, &se->avg, se->on_rq, + if (!__update_entity_runnable_avg(now, cpu, &se->avg, se->on_rq, cfs_rq->curr == se)) return;
The average running time of RT tasks is used to estimate the remaining compute capacity for CFS tasks. This remaining capacity is the original capacity scaled down by a factor (aka scale_rt_capacity). This estimation of available capacity must also be invariant with frequency scaling.
A frequency scaling factor is applied on the running time of the RT tasks for computing scale_rt_capacity.
In sched_rt_avg_update, we scale the RT execution time like below: rq->rt_avg += rt_delta * arch_scale_freq_capacity() >> SCHED_CAPACITY_SHIFT
Then, scale_rt_capacity can be summarized by: scale_rt_capacity = SCHED_CAPACITY_SCALE - ((rq->rt_avg << SCHED_CAPACITY_SHIFT) / period)
We can optimize by removing right and left shift in the computation of rq->rt_avg and scale_rt_capacity
The call to arch_scale_frequency_capacity in the rt scheduling path might be a concern for RT folks because I'm not sure whether we can rely on arch_scale_freq_capacity to be short and efficient ?
Signed-off-by: Vincent Guittot vincent.guittot@linaro.org --- kernel/sched/fair.c | 17 +++++------------ kernel/sched/sched.h | 4 +++- 2 files changed, 8 insertions(+), 13 deletions(-)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 31c3ec2..02e8f7f 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5817,7 +5817,7 @@ unsigned long __weak arch_scale_cpu_capacity(struct sched_domain *sd, int cpu) static unsigned long scale_rt_capacity(int cpu) { struct rq *rq = cpu_rq(cpu); - u64 total, available, age_stamp, avg; + u64 total, used, age_stamp, avg; s64 delta;
/* @@ -5833,19 +5833,12 @@ static unsigned long scale_rt_capacity(int cpu)
total = sched_avg_period() + delta;
- if (unlikely(total < avg)) { - /* Ensures that capacity won't end up being negative */ - available = 0; - } else { - available = total - avg; - } + used = div_u64(avg, total);
- if (unlikely((s64)total < SCHED_CAPACITY_SCALE)) - total = SCHED_CAPACITY_SCALE; + if (likely(used < SCHED_CAPACITY_SCALE)) + return SCHED_CAPACITY_SCALE - used;
- total >>= SCHED_CAPACITY_SHIFT; - - return div_u64(available, total); + return 1; }
static void update_cpu_capacity(struct sched_domain *sd, int cpu) diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 7963981..8fd30c1 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1313,9 +1313,11 @@ static inline int hrtick_enabled(struct rq *rq)
#ifdef CONFIG_SMP extern void sched_avg_update(struct rq *rq); +extern unsigned long arch_scale_freq_capacity(struct sched_domain *sd, int cpu); + static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { - rq->rt_avg += rt_delta; + rq->rt_avg += rt_delta * arch_scale_freq_capacity(NULL, cpu_of(rq)); sched_avg_update(rq); } #else
On Fri, Oct 31, 2014 at 09:47:32AM +0100, Vincent Guittot wrote:
The call to arch_scale_frequency_capacity in the rt scheduling path might be a concern for RT folks because I'm not sure whether we can rely on arch_scale_freq_capacity to be short and efficient ?
Well, you put it in quite a hot path on the CFS side too, so it had better be blazing fast anyhow.
That said, can you move all the running + invariance bits at the beginning of the patch set, these appear to be the least controversial bits and 'desired' by quite a few people, so we'll try and get those agreed upon and merged first.
On 3 November 2014 16:51, Peter Zijlstra peterz@infradead.org wrote:
On Fri, Oct 31, 2014 at 09:47:32AM +0100, Vincent Guittot wrote:
The call to arch_scale_frequency_capacity in the rt scheduling path might be a concern for RT folks because I'm not sure whether we can rely on arch_scale_freq_capacity to be short and efficient ?
Well, you put it in quite a hot path on the CFS side too, so it had better be blazing fast anyhow.
That said, can you move all the running + invariance bits at the beginning of the patch set, these appear to be the least controversial bits and 'desired' by quite a few people, so we'll try and get those agreed upon and merged first.
ok
I'm going to resend with the following order
sched: add utilization_avg_contrib sched: Track group sched_entity usage contributions sched: remove frequency scaling from cpu_capacity sched: Make sched entity usage tracking scale-invariant sched: make scale_rt invariant with frequency sched: add per rq cpu_capacity_orig sched: get CPU's usage statistic sched: replace capacity_factor by usage sched: add SD_PREFER_SIBLING for SMT level sched: move cfs task on a CPU with higher capacity
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
This patchset consolidates several changes in the capacity and the usage tracking of the CPU. It provides a frequency invariant metric of the usage of CPUs and generally improves the accuracy of load/usage tracking in the scheduler. The frequency invariant metric is the foundation required for the consolidation of cpufreq and implementation of a fully invariant load tracking. These are currently WIP and require several changes to the load balancer (including how it will use and interprets load and capacity metrics) and extensive validation. The frequency invariance is done with arch_scale_freq_capacity and this patchset doesn't provide the backends of the function which are architecture dependent.
As discussed at LPC14, Morten and I have consolidated our changes into a single patchset to make it easier to review and merge.
During load balance, the scheduler evaluates the number of tasks that a group of CPUs can handle. The current method assumes that tasks have a fix load of SCHED_LOAD_SCALE and CPUs have a default capacity of SCHED_CAPACITY_SCALE. This assumption generates wrong decision by creating ghost cores or by
I don't know the history, could you explain what's the meaning of 'ghost cores' ?
Regards, Wanpeng Li
removing real ones when the original capacity of CPUs is different from the default SCHED_CAPACITY_SCALE. With this patch set, we don't try anymore to evaluate the number of available cores based on the group_capacity but instead we evaluate the usage of a group and compare it with its capacity.
This patchset mainly replaces the old capacity_factor method by a new one and keeps the general policy almost unchanged. These new metrics will be also used in later patches.
The CPU usage is based on a running time tracking version of the current implementation of the load average tracking. I also have a version that is based on the new implementation proposal [1] but I haven't provide the patches and results as [1] is still under review. I can provide change above [1] to change how CPU usage is computed and to adapt to new mecanism.
Change since V7
- add freq invariance for usage tracking
- add freq invariance for scale_rt
- update comments and commits' message
- fix init of utilization_avg_contrib
- fix prefer_sibling
Change since V6
- add group usage tracking
- fix some commits' messages
- minor fix like comments and argument order
Change since V5
- remove patches that have been merged since v5 : patches 01, 02, 03, 04, 05, 07
- update commit log and add more details on the purpose of the patches
- fix/remove useless code with the rebase on patchset [2]
- remove capacity_orig in sched_group_capacity as it is not used
- move code in the right patch
- add some helper function to factorize code
Change since V4
- rebase to manage conflicts with changes in selection of busiest group
Change since V3:
- add usage_avg_contrib statistic which sums the running time of tasks on a rq
- use usage_avg_contrib instead of runnable_avg_sum for cpu_utilization
- fix replacement power by capacity
- update some comments
Change since V2:
- rebase on top of capacity renaming
- fix wake_affine statistic update
- rework nohz_kick_needed
- optimize the active migration of a task from CPU with reduced capacity
- rename group_activity by group_utilization and remove unused total_utilization
- repair SD_PREFER_SIBLING and use it for SMT level
- reorder patchset to gather patches with same topics
Change since V1:
- add 3 fixes
- correct some commit messages
- replace capacity computation by activity
- take into account current cpu capacity
[1] https://lkml.org/lkml/2014/10/10/131 [2] https://lkml.org/lkml/2014/7/25/589
Morten Rasmussen (2): sched: Track group sched_entity usage contributions sched: Make sched entity usage tracking scale-invariant
Vincent Guittot (8): sched: add per rq cpu_capacity_orig sched: remove frequency scaling from cpu_capacity sched: move cfs task on a CPU with higher capacity sched: add utilization_avg_contrib sched: get CPU's usage statistic sched: replace capacity_factor by usage sched: add SD_PREFER_SIBLING for SMT level sched: make scale_rt invariant with frequency
include/linux/sched.h | 21 ++- kernel/sched/core.c | 15 +- kernel/sched/debug.c | 12 +- kernel/sched/fair.c | 369 ++++++++++++++++++++++++++++++++------------------ kernel/sched/sched.h | 15 +- 5 files changed, 276 insertions(+), 156 deletions(-)
On 3 November 2014 03:12, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
This patchset consolidates several changes in the capacity and the usage tracking of the CPU. It provides a frequency invariant metric of the usage of CPUs and generally improves the accuracy of load/usage tracking in the scheduler. The frequency invariant metric is the foundation required for the consolidation of cpufreq and implementation of a fully invariant load tracking. These are currently WIP and require several changes to the load balancer (including how it will use and interprets load and capacity metrics) and extensive validation. The frequency invariance is done with arch_scale_freq_capacity and this patchset doesn't provide the backends of the function which are architecture dependent.
As discussed at LPC14, Morten and I have consolidated our changes into a single patchset to make it easier to review and merge.
During load balance, the scheduler evaluates the number of tasks that a group of CPUs can handle. The current method assumes that tasks have a fix load of SCHED_LOAD_SCALE and CPUs have a default capacity of SCHED_CAPACITY_SCALE. This assumption generates wrong decision by creating ghost cores or by
I don't know the history, could you explain what's the meaning of 'ghost cores' ?
The capacity_factor gives the number of tasks that can be handled by a group of CPUs by dividing the group's capacity by SCHED_CAPACITY_SCALE
For a system with SMT, the default capacity of a core is 1178 so the capacity of each CPU for a dual threads per core is 589.
At CPU level we have a capacity_factor of 1 = div_round_closest(589, 1024) At core level we still have a capacity_factor of 1 = div_round_closest(1178, 1024). This is a intended behavior to promote 1 task per core Then, if we have 4 cores in a node, the capacity_factor is 5 = div_round_closest(4712, 1024) whereas we should have 4. So a 5th ghost core has appeared in the group and the load balancer will not considered the group as overloaded if there is 5 tasks whereas it should in order to try to move this 5th task on an idle core (if there is one) Patch [0] solves some use cases by ensuring that we will not have more cores than possible so we can't have more than 4 core for the previous example. Now, if some RT tasks are running and using almost 1 core (1024 as an example), the capacity_factor is still 4 = div_round_closest(3688, 1024) whereas a core is nearly fully used and the capacity_factor should be 3
[0] https://lkml.org/lkml/2013/8/28/194
Regards, Vincent
Regards, Wanpeng Li
removing real ones when the original capacity of CPUs is different from the default SCHED_CAPACITY_SCALE. With this patch set, we don't try anymore to evaluate the number of available cores based on the group_capacity but instead we evaluate the usage of a group and compare it with its capacity.
This patchset mainly replaces the old capacity_factor method by a new one and keeps the general policy almost unchanged. These new metrics will be also used in later patches.
[snip]
On 14/11/3 下午6:55, Vincent Guittot wrote:
On 3 November 2014 03:12, Wanpeng Li kernellwp@gmail.com wrote:
Hi Vincent, On 14/10/31 下午4:47, Vincent Guittot wrote:
This patchset consolidates several changes in the capacity and the usage tracking of the CPU. It provides a frequency invariant metric of the usage of CPUs and generally improves the accuracy of load/usage tracking in the scheduler. The frequency invariant metric is the foundation required for the consolidation of cpufreq and implementation of a fully invariant load tracking. These are currently WIP and require several changes to the load balancer (including how it will use and interprets load and capacity metrics) and extensive validation. The frequency invariance is done with arch_scale_freq_capacity and this patchset doesn't provide the backends of the function which are architecture dependent.
As discussed at LPC14, Morten and I have consolidated our changes into a single patchset to make it easier to review and merge.
During load balance, the scheduler evaluates the number of tasks that a group of CPUs can handle. The current method assumes that tasks have a fix load of SCHED_LOAD_SCALE and CPUs have a default capacity of SCHED_CAPACITY_SCALE. This assumption generates wrong decision by creating ghost cores or by
I don't know the history, could you explain what's the meaning of 'ghost cores' ?
The capacity_factor gives the number of tasks that can be handled by a group of CPUs by dividing the group's capacity by SCHED_CAPACITY_SCALE
For a system with SMT, the default capacity of a core is 1178 so the capacity of each CPU for a dual threads per core is 589.
At CPU level we have a capacity_factor of 1 = div_round_closest(589, 1024) At core level we still have a capacity_factor of 1 = div_round_closest(1178, 1024). This is a intended behavior to promote 1 task per core Then, if we have 4 cores in a node, the capacity_factor is 5 = div_round_closest(4712, 1024) whereas we should have 4. So a 5th ghost core has appeared in the group and the load balancer will not considered the group as overloaded if there is 5 tasks whereas it should in order to try to move this 5th task on an idle core (if there is one) Patch [0] solves some use cases by ensuring that we will not have more cores than possible so we can't have more than 4 core for the previous example. Now, if some RT tasks are running and using almost 1 core (1024 as an example), the capacity_factor is still 4 = div_round_closest(3688, 1024) whereas a core is nearly fully used and the capacity_factor should be 3
Got it, thanks for your great explanation.
Regards, Wanpeng Li
Regards, Vincent
Regards, Wanpeng Li
removing real ones when the original capacity of CPUs is different from the default SCHED_CAPACITY_SCALE. With this patch set, we don't try anymore to evaluate the number of available cores based on the group_capacity but instead we evaluate the usage of a group and compare it with its capacity.
This patchset mainly replaces the old capacity_factor method by a new one and keeps the general policy almost unchanged. These new metrics will be also used in later patches.
[snip]
linaro-kernel@lists.linaro.org
-
Peter Zijlstra -
Vincent Guittot -
Wanpeng Li