Commit b2e09f63 authored by Linus Torvalds's avatar Linus Torvalds

Merge branch 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

Pull more scheduler updates from Ingo Molnar:
 "Second round of scheduler changes:
   - try-to-wakeup and IPI reduction speedups, from Andy Lutomirski
   - continued power scheduling cleanups and refactorings, from Nicolas
     Pitre
   - misc fixes and enhancements"

* 'sched-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  sched/deadline: Delete extraneous extern for to_ratio()
  sched/idle: Optimize try-to-wake-up IPI
  sched/idle: Simplify wake_up_idle_cpu()
  sched/idle: Clear polling before descheduling the idle thread
  sched, trace: Add a tracepoint for IPI-less remote wakeups
  cpuidle: Set polling in poll_idle
  sched: Remove redundant assignment to "rt_rq" in update_curr_rt(...)
  sched: Rename capacity related flags
  sched: Final power vs. capacity cleanups
  sched: Remove remaining dubious usage of "power"
  sched: Let 'struct sched_group_power' care about CPU capacity
  sched/fair: Disambiguate existing/remaining "capacity" usage
  sched/fair: Change "has_capacity" to "has_free_capacity"
  sched/fair: Remove "power" from 'struct numa_stats'
  sched: Fix signedness bug in yield_to()
  sched/fair: Use time_after() in record_wakee()
  sched/balancing: Reduce the rate of needless idle load balancing
  sched/fair: Fix unlocked reads of some cfs_b->quota/period
parents 3737a127 535560d8
......@@ -26,30 +26,30 @@
#include <asm/topology.h>
/*
* cpu power scale management
* cpu capacity scale management
*/
/*
* cpu power table
* cpu capacity table
* This per cpu data structure describes the relative capacity of each core.
* On a heteregenous system, cores don't have the same computation capacity
* and we reflect that difference in the cpu_power field so the scheduler can
* take this difference into account during load balance. A per cpu structure
* is preferred because each CPU updates its own cpu_power field during the
* load balance except for idle cores. One idle core is selected to run the
* rebalance_domains for all idle cores and the cpu_power can be updated
* during this sequence.
* and we reflect that difference in the cpu_capacity field so the scheduler
* can take this difference into account during load balance. A per cpu
* structure is preferred because each CPU updates its own cpu_capacity field
* during the load balance except for idle cores. One idle core is selected
* to run the rebalance_domains for all idle cores and the cpu_capacity can be
* updated during this sequence.
*/
static DEFINE_PER_CPU(unsigned long, cpu_scale);
unsigned long arch_scale_freq_power(struct sched_domain *sd, int cpu)
unsigned long arch_scale_freq_capacity(struct sched_domain *sd, int cpu)
{
return per_cpu(cpu_scale, cpu);
}
static void set_power_scale(unsigned int cpu, unsigned long power)
static void set_capacity_scale(unsigned int cpu, unsigned long capacity)
{
per_cpu(cpu_scale, cpu) = power;
per_cpu(cpu_scale, cpu) = capacity;
}
#ifdef CONFIG_OF
......@@ -62,11 +62,11 @@ struct cpu_efficiency {
* Table of relative efficiency of each processors
* The efficiency value must fit in 20bit and the final
* cpu_scale value must be in the range
* 0 < cpu_scale < 3*SCHED_POWER_SCALE/2
* 0 < cpu_scale < 3*SCHED_CAPACITY_SCALE/2
* in order to return at most 1 when DIV_ROUND_CLOSEST
* is used to compute the capacity of a CPU.
* Processors that are not defined in the table,
* use the default SCHED_POWER_SCALE value for cpu_scale.
* use the default SCHED_CAPACITY_SCALE value for cpu_scale.
*/
static const struct cpu_efficiency table_efficiency[] = {
{"arm,cortex-a15", 3891},
......@@ -83,9 +83,9 @@ static unsigned long middle_capacity = 1;
* Iterate all CPUs' descriptor in DT and compute the efficiency
* (as per table_efficiency). Also calculate a middle efficiency
* as close as possible to (max{eff_i} - min{eff_i}) / 2
* This is later used to scale the cpu_power field such that an
* 'average' CPU is of middle power. Also see the comments near
* table_efficiency[] and update_cpu_power().
* This is later used to scale the cpu_capacity field such that an
* 'average' CPU is of middle capacity. Also see the comments near
* table_efficiency[] and update_cpu_capacity().
*/
static void __init parse_dt_topology(void)
{
......@@ -141,15 +141,15 @@ static void __init parse_dt_topology(void)
* cpu_scale because all CPUs have the same capacity. Otherwise, we
* compute a middle_capacity factor that will ensure that the capacity
* of an 'average' CPU of the system will be as close as possible to
* SCHED_POWER_SCALE, which is the default value, but with the
* SCHED_CAPACITY_SCALE, which is the default value, but with the
* constraint explained near table_efficiency[].
*/
if (4*max_capacity < (3*(max_capacity + min_capacity)))
middle_capacity = (min_capacity + max_capacity)
>> (SCHED_POWER_SHIFT+1);
>> (SCHED_CAPACITY_SHIFT+1);
else
middle_capacity = ((max_capacity / 3)
>> (SCHED_POWER_SHIFT-1)) + 1;
>> (SCHED_CAPACITY_SHIFT-1)) + 1;
}
......@@ -158,20 +158,20 @@ static void __init parse_dt_topology(void)
* boot. The update of all CPUs is in O(n^2) for heteregeneous system but the
* function returns directly for SMP system.
*/
static void update_cpu_power(unsigned int cpu)
static void update_cpu_capacity(unsigned int cpu)
{
if (!cpu_capacity(cpu))
return;
set_power_scale(cpu, cpu_capacity(cpu) / middle_capacity);
set_capacity_scale(cpu, cpu_capacity(cpu) / middle_capacity);
printk(KERN_INFO "CPU%u: update cpu_power %lu\n",
cpu, arch_scale_freq_power(NULL, cpu));
printk(KERN_INFO "CPU%u: update cpu_capacity %lu\n",
cpu, arch_scale_freq_capacity(NULL, cpu));
}
#else
static inline void parse_dt_topology(void) {}
static inline void update_cpu_power(unsigned int cpuid) {}
static inline void update_cpu_capacity(unsigned int cpuid) {}
#endif
/*
......@@ -267,7 +267,7 @@ void store_cpu_topology(unsigned int cpuid)
update_siblings_masks(cpuid);
update_cpu_power(cpuid);
update_cpu_capacity(cpuid);
printk(KERN_INFO "CPU%u: thread %d, cpu %d, socket %d, mpidr %x\n",
cpuid, cpu_topology[cpuid].thread_id,
......@@ -297,7 +297,7 @@ void __init init_cpu_topology(void)
{
unsigned int cpu;
/* init core mask and power*/
/* init core mask and capacity */
for_each_possible_cpu(cpu) {
struct cputopo_arm *cpu_topo = &(cpu_topology[cpu]);
......@@ -307,7 +307,7 @@ void __init init_cpu_topology(void)
cpumask_clear(&cpu_topo->core_sibling);
cpumask_clear(&cpu_topo->thread_sibling);
set_power_scale(cpu, SCHED_POWER_SCALE);
set_capacity_scale(cpu, SCHED_CAPACITY_SCALE);
}
smp_wmb();
......
......@@ -749,7 +749,7 @@ int setup_profiling_timer(unsigned int multiplier)
/* cpumask of CPUs with asymetric SMT dependancy */
static const int powerpc_smt_flags(void)
{
int flags = SD_SHARE_CPUPOWER | SD_SHARE_PKG_RESOURCES;
int flags = SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES;
if (cpu_has_feature(CPU_FTR_ASYM_SMT)) {
printk_once(KERN_INFO "Enabling Asymmetric SMT scheduling\n");
......
......@@ -187,8 +187,11 @@ static int poll_idle(struct cpuidle_device *dev,
t1 = ktime_get();
local_irq_enable();
while (!need_resched())
cpu_relax();
if (!current_set_polling_and_test()) {
while (!need_resched())
cpu_relax();
}
current_clr_polling();
t2 = ktime_get();
diff = ktime_to_us(ktime_sub(t2, t1));
......
......@@ -586,7 +586,7 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn);
void kvm_vcpu_block(struct kvm_vcpu *vcpu);
void kvm_vcpu_kick(struct kvm_vcpu *vcpu);
bool kvm_vcpu_yield_to(struct kvm_vcpu *target);
int kvm_vcpu_yield_to(struct kvm_vcpu *target);
void kvm_vcpu_on_spin(struct kvm_vcpu *vcpu);
void kvm_load_guest_fpu(struct kvm_vcpu *vcpu);
void kvm_put_guest_fpu(struct kvm_vcpu *vcpu);
......
......@@ -847,10 +847,10 @@ enum cpu_idle_type {
};
/*
* Increase resolution of cpu_power calculations
* Increase resolution of cpu_capacity calculations
*/
#define SCHED_POWER_SHIFT 10
#define SCHED_POWER_SCALE (1L << SCHED_POWER_SHIFT)
#define SCHED_CAPACITY_SHIFT 10
#define SCHED_CAPACITY_SCALE (1L << SCHED_CAPACITY_SHIFT)
/*
* sched-domains (multiprocessor balancing) declarations:
......@@ -862,7 +862,7 @@ enum cpu_idle_type {
#define SD_BALANCE_FORK 0x0008 /* Balance on fork, clone */
#define SD_BALANCE_WAKE 0x0010 /* Balance on wakeup */
#define SD_WAKE_AFFINE 0x0020 /* Wake task to waking CPU */
#define SD_SHARE_CPUPOWER 0x0080 /* Domain members share cpu power */
#define SD_SHARE_CPUCAPACITY 0x0080 /* Domain members share cpu power */
#define SD_SHARE_POWERDOMAIN 0x0100 /* Domain members share power domain */
#define SD_SHARE_PKG_RESOURCES 0x0200 /* Domain members share cpu pkg resources */
#define SD_SERIALIZE 0x0400 /* Only a single load balancing instance */
......@@ -874,7 +874,7 @@ enum cpu_idle_type {
#ifdef CONFIG_SCHED_SMT
static inline const int cpu_smt_flags(void)
{
return SD_SHARE_CPUPOWER | SD_SHARE_PKG_RESOURCES;
return SD_SHARE_CPUCAPACITY | SD_SHARE_PKG_RESOURCES;
}
#endif
......@@ -1006,7 +1006,7 @@ typedef const int (*sched_domain_flags_f)(void);
struct sd_data {
struct sched_domain **__percpu sd;
struct sched_group **__percpu sg;
struct sched_group_power **__percpu sgp;
struct sched_group_capacity **__percpu sgc;
};
struct sched_domain_topology_level {
......@@ -2173,7 +2173,7 @@ static inline void sched_autogroup_fork(struct signal_struct *sig) { }
static inline void sched_autogroup_exit(struct signal_struct *sig) { }
#endif
extern bool yield_to(struct task_struct *p, bool preempt);
extern int yield_to(struct task_struct *p, bool preempt);
extern void set_user_nice(struct task_struct *p, long nice);
extern int task_prio(const struct task_struct *p);
/**
......
......@@ -530,6 +530,26 @@ TRACE_EVENT(sched_swap_numa,
__entry->dst_pid, __entry->dst_tgid, __entry->dst_ngid,
__entry->dst_cpu, __entry->dst_nid)
);
/*
* Tracepoint for waking a polling cpu without an IPI.
*/
TRACE_EVENT(sched_wake_idle_without_ipi,
TP_PROTO(int cpu),
TP_ARGS(cpu),
TP_STRUCT__entry(
__field( int, cpu )
),
TP_fast_assign(
__entry->cpu = cpu;
),
TP_printk("cpu=%d", __entry->cpu)
);
#endif /* _TRACE_SCHED_H */
/* This part must be outside protection */
......
......@@ -535,7 +535,7 @@ static inline void init_hrtick(void)
__old; \
})
#ifdef TIF_POLLING_NRFLAG
#if defined(CONFIG_SMP) && defined(TIF_POLLING_NRFLAG)
/*
* Atomically set TIF_NEED_RESCHED and test for TIF_POLLING_NRFLAG,
* this avoids any races wrt polling state changes and thereby avoids
......@@ -546,12 +546,44 @@ static bool set_nr_and_not_polling(struct task_struct *p)
struct thread_info *ti = task_thread_info(p);
return !(fetch_or(&ti->flags, _TIF_NEED_RESCHED) & _TIF_POLLING_NRFLAG);
}
/*
* Atomically set TIF_NEED_RESCHED if TIF_POLLING_NRFLAG is set.
*
* If this returns true, then the idle task promises to call
* sched_ttwu_pending() and reschedule soon.
*/
static bool set_nr_if_polling(struct task_struct *p)
{
struct thread_info *ti = task_thread_info(p);
typeof(ti->flags) old, val = ACCESS_ONCE(ti->flags);
for (;;) {
if (!(val & _TIF_POLLING_NRFLAG))
return false;
if (val & _TIF_NEED_RESCHED)
return true;
old = cmpxchg(&ti->flags, val, val | _TIF_NEED_RESCHED);
if (old == val)
break;
val = old;
}
return true;
}
#else
static bool set_nr_and_not_polling(struct task_struct *p)
{
set_tsk_need_resched(p);
return true;
}
#ifdef CONFIG_SMP
static bool set_nr_if_polling(struct task_struct *p)
{
return false;
}
#endif
#endif
/*
......@@ -580,6 +612,8 @@ void resched_task(struct task_struct *p)
if (set_nr_and_not_polling(p))
smp_send_reschedule(cpu);
else
trace_sched_wake_idle_without_ipi(cpu);
}
void resched_cpu(int cpu)
......@@ -642,27 +676,10 @@ static void wake_up_idle_cpu(int cpu)
if (cpu == smp_processor_id())
return;
/*
* This is safe, as this function is called with the timer
* wheel base lock of (cpu) held. When the CPU is on the way
* to idle and has not yet set rq->curr to idle then it will
* be serialized on the timer wheel base lock and take the new
* timer into account automatically.
*/
if (rq->curr != rq->idle)
return;
/*
* We can set TIF_RESCHED on the idle task of the other CPU
* lockless. The worst case is that the other CPU runs the
* idle task through an additional NOOP schedule()
*/
set_tsk_need_resched(rq->idle);
/* NEED_RESCHED must be visible before we test polling */
smp_mb();
if (!tsk_is_polling(rq->idle))
if (set_nr_and_not_polling(rq->idle))
smp_send_reschedule(cpu);
else
trace_sched_wake_idle_without_ipi(cpu);
}
static bool wake_up_full_nohz_cpu(int cpu)
......@@ -888,7 +905,7 @@ static void update_rq_clock_task(struct rq *rq, s64 delta)
rq->clock_task += delta;
#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING)
if ((irq_delta + steal) && sched_feat(NONTASK_POWER))
if ((irq_delta + steal) && sched_feat(NONTASK_CAPACITY))
sched_rt_avg_update(rq, irq_delta + steal);
#endif
}
......@@ -1521,13 +1538,17 @@ static int ttwu_remote(struct task_struct *p, int wake_flags)
}
#ifdef CONFIG_SMP
static void sched_ttwu_pending(void)
void sched_ttwu_pending(void)
{
struct rq *rq = this_rq();
struct llist_node *llist = llist_del_all(&rq->wake_list);
struct task_struct *p;
unsigned long flags;
raw_spin_lock(&rq->lock);
if (!llist)
return;
raw_spin_lock_irqsave(&rq->lock, flags);
while (llist) {
p = llist_entry(llist, struct task_struct, wake_entry);
......@@ -1535,7 +1556,7 @@ static void sched_ttwu_pending(void)
ttwu_do_activate(rq, p, 0);
}
raw_spin_unlock(&rq->lock);
raw_spin_unlock_irqrestore(&rq->lock, flags);
}
void scheduler_ipi(void)
......@@ -1581,8 +1602,14 @@ void scheduler_ipi(void)
static void ttwu_queue_remote(struct task_struct *p, int cpu)
{
if (llist_add(&p->wake_entry, &cpu_rq(cpu)->wake_list))
smp_send_reschedule(cpu);
struct rq *rq = cpu_rq(cpu);
if (llist_add(&p->wake_entry, &cpu_rq(cpu)->wake_list)) {
if (!set_nr_if_polling(rq->idle))
smp_send_reschedule(cpu);
else
trace_sched_wake_idle_without_ipi(cpu);
}
}
bool cpus_share_cache(int this_cpu, int that_cpu)
......@@ -4219,7 +4246,7 @@ EXPORT_SYMBOL(yield);
* false (0) if we failed to boost the target.
* -ESRCH if there's no task to yield to.
*/
bool __sched yield_to(struct task_struct *p, bool preempt)
int __sched yield_to(struct task_struct *p, bool preempt)
{
struct task_struct *curr = current;
struct rq *rq, *p_rq;
......@@ -5245,14 +5272,13 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
}
/*
* Even though we initialize ->power to something semi-sane,
* we leave power_orig unset. This allows us to detect if
* 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->sgp->power_orig) {
if (!group->sgc->capacity_orig) {
printk(KERN_CONT "\n");
printk(KERN_ERR "ERROR: domain->cpu_power not "
"set\n");
printk(KERN_ERR "ERROR: domain->cpu_capacity not set\n");
break;
}
......@@ -5274,9 +5300,9 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level,
cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group));
printk(KERN_CONT " %s", str);
if (group->sgp->power != SCHED_POWER_SCALE) {
printk(KERN_CONT " (cpu_power = %d)",
group->sgp->power);
if (group->sgc->capacity != SCHED_CAPACITY_SCALE) {
printk(KERN_CONT " (cpu_capacity = %d)",
group->sgc->capacity);
}
group = group->next;
......@@ -5334,7 +5360,7 @@ static int sd_degenerate(struct sched_domain *sd)
SD_BALANCE_NEWIDLE |
SD_BALANCE_FORK |
SD_BALANCE_EXEC |
SD_SHARE_CPUPOWER |
SD_SHARE_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
SD_SHARE_POWERDOMAIN)) {
if (sd->groups != sd->groups->next)
......@@ -5365,7 +5391,7 @@ sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent)
SD_BALANCE_NEWIDLE |
SD_BALANCE_FORK |
SD_BALANCE_EXEC |
SD_SHARE_CPUPOWER |
SD_SHARE_CPUCAPACITY |
SD_SHARE_PKG_RESOURCES |
SD_PREFER_SIBLING |
SD_SHARE_POWERDOMAIN);
......@@ -5490,7 +5516,7 @@ static struct root_domain *alloc_rootdomain(void)
return rd;
}
static void free_sched_groups(struct sched_group *sg, int free_sgp)
static void free_sched_groups(struct sched_group *sg, int free_sgc)
{
struct sched_group *tmp, *first;
......@@ -5501,8 +5527,8 @@ static void free_sched_groups(struct sched_group *sg, int free_sgp)
do {
tmp = sg->next;
if (free_sgp && atomic_dec_and_test(&sg->sgp->ref))
kfree(sg->sgp);
if (free_sgc && atomic_dec_and_test(&sg->sgc->ref))
kfree(sg->sgc);
kfree(sg);
sg = tmp;
......@@ -5520,7 +5546,7 @@ static void free_sched_domain(struct rcu_head *rcu)
if (sd->flags & SD_OVERLAP) {
free_sched_groups(sd->groups, 1);
} else if (atomic_dec_and_test(&sd->groups->ref)) {
kfree(sd->groups->sgp);
kfree(sd->groups->sgc);
kfree(sd->groups);
}
kfree(sd);
......@@ -5731,17 +5757,17 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
cpumask_or(covered, covered, sg_span);
sg->sgp = *per_cpu_ptr(sdd->sgp, i);
if (atomic_inc_return(&sg->sgp->ref) == 1)
sg->sgc = *per_cpu_ptr(sdd->sgc, i);
if (atomic_inc_return(&sg->sgc->ref) == 1)
build_group_mask(sd, sg);
/*
* Initialize sgp->power such that even if we mess up the
* Initialize sgc->capacity such that even if we mess up the
* domains and no possible iteration will get us here, we won't
* die on a /0 trap.
*/
sg->sgp->power = SCHED_POWER_SCALE * cpumask_weight(sg_span);
sg->sgp->power_orig = sg->sgp->power;
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
......@@ -5779,8 +5805,8 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
if (sg) {
*sg = *per_cpu_ptr(sdd->sg, cpu);
(*sg)->sgp = *per_cpu_ptr(sdd->sgp, cpu);
atomic_set(&(*sg)->sgp->ref, 1); /* for claim_allocations */
(*sg)->sgc = *per_cpu_ptr(sdd->sgc, cpu);
atomic_set(&(*sg)->sgc->ref, 1); /* for claim_allocations */
}
return cpu;
......@@ -5789,7 +5815,7 @@ static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg)
/*
* build_sched_groups will build a circular linked list of the groups
* covered by the given span, and will set each group's ->cpumask correctly,
* and ->cpu_power to 0.
* and ->cpu_capacity to 0.
*
* Assumes the sched_domain tree is fully constructed
*/
......@@ -5843,16 +5869,16 @@ build_sched_groups(struct sched_domain *sd, int cpu)
}
/*
* Initialize sched groups cpu_power.
* Initialize sched groups cpu_capacity.
*
* cpu_power indicates the capacity of sched group, which is used while
* cpu_capacity indicates the capacity of sched group, which is used while
* distributing the load between different sched groups in a sched domain.
* Typically cpu_power for all the groups in a sched domain will be same unless
* there are asymmetries in the topology. If there are asymmetries, group
* having more cpu_power will pickup more load compared to the group having
* less cpu_power.
* Typically cpu_capacity for all the groups in a sched domain will be same
* unless there are asymmetries in the topology. If there are asymmetries,
* group having more cpu_capacity will pickup more load compared to the
* group having less cpu_capacity.
*/
static void init_sched_groups_power(int cpu, struct sched_domain *sd)
static void init_sched_groups_capacity(int cpu, struct sched_domain *sd)
{
struct sched_group *sg = sd->groups;
......@@ -5866,8 +5892,8 @@ static void init_sched_groups_power(int cpu, struct sched_domain *sd)
if (cpu != group_balance_cpu(sg))
return;
update_group_power(sd, cpu);
atomic_set(&sg->sgp->nr_busy_cpus, sg->group_weight);
update_group_capacity(sd, cpu);
atomic_set(&sg->sgc->nr_busy_cpus, sg->group_weight);
}
/*
......@@ -5958,8 +5984,8 @@ static void claim_allocations(int cpu, struct sched_domain *sd)
if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref))
*per_cpu_ptr(sdd->sg, cpu) = NULL;
if (atomic_read(&(*per_cpu_ptr(sdd->sgp, cpu))->ref))
*per_cpu_ptr(sdd->sgp, cpu) = NULL;
if (atomic_read(&(*per_cpu_ptr(sdd->sgc, cpu))->ref))
*per_cpu_ptr(sdd->sgc, cpu) = NULL;
}
#ifdef CONFIG_NUMA
......@@ -5972,7 +5998,7 @@ static int sched_domains_curr_level;
/*
* SD_flags allowed in topology descriptions.
*
* SD_SHARE_CPUPOWER - describes SMT topologies
* SD_SHARE_CPUCAPACITY - describes SMT topologies
* SD_SHARE_PKG_RESOURCES - describes shared caches
* SD_NUMA - describes NUMA topologies
* SD_SHARE_POWERDOMAIN - describes shared power domain
......@@ -5981,7 +6007,7 @@ static int sched_domains_curr_level;
* SD_ASYM_PACKING - describes SMT quirks
*/
#define TOPOLOGY_SD_FLAGS \
(SD_SHARE_CPUPOWER | \
(SD_SHARE_CPUCAPACITY | \
SD_SHARE_PKG_RESOURCES | \
SD_NUMA | \
SD_ASYM_PACKING | \
......@@ -6027,7 +6053,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu)
| 1*SD_BALANCE_FORK
| 0*SD_BALANCE_WAKE
| 1*SD_WAKE_AFFINE
| 0*SD_SHARE_CPUPOWER
| 0*SD_SHARE_CPUCAPACITY
| 0*SD_SHARE_PKG_RESOURCES
| 0*SD_SERIALIZE
| 0*SD_PREFER_SIBLING
......@@ -6049,7 +6075,7 @@ sd_init(struct sched_domain_topology_level *tl, int cpu)
* Convert topological properties into behaviour.
*/
if (sd->flags & SD_SHARE_CPUPOWER) {
if (sd->flags & SD_SHARE_CPUCAPACITY) {
sd->imbalance_pct = 110;
sd->smt_gain = 1178; /* ~15% */
......@@ -6361,14 +6387,14 @@ static int __sdt_alloc(const struct cpumask *cpu_map)
if (!sdd->sg)
return -ENOMEM;
sdd->sgp = alloc_percpu(struct sched_group_power *);
if (!sdd->sgp)
sdd->sgc = alloc_percpu(struct sched_group_capacity *);
if (!sdd->sgc)
return -ENOMEM;
for_each_cpu(j, cpu_map) {
struct sched_domain *sd;
struct sched_group *sg;
struct sched_group_power *sgp;
struct sched_group_capacity *sgc;
sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(),
GFP_KERNEL, cpu_to_node(j));
......@@ -6386,12 +6412,12 @@ static int __sdt_alloc(const struct cpumask *cpu_map)
*per_cpu_ptr(sdd->sg, j) = sg;
sgp = kzalloc_node(sizeof(struct sched_group_power) + cpumask_size(),
sgc = kzalloc_node(sizeof(struct sched_group_capacity) + cpumask_size(),
GFP_KERNEL, cpu_to_node(j));
if (!sgp)
if (!sgc)
return -ENOMEM;
*per_cpu_ptr(sdd->sgp, j) = sgp;
*per_cpu_ptr(sdd->sgc, j) = sgc;
}
}
......@@ -6418,15 +6444,15 @@ static void __sdt_free(const struct cpumask *cpu_map)
if (sdd->sg)
kfree(*per_cpu_ptr(sdd->sg, j));
if (sdd->sgp)
kfree(*per_cpu_ptr(sdd->sgp, j));
if (sdd->sgc)
kfree(*per_cpu_ptr(sdd->sgc, j));
}
free_percpu(sdd->sd);
sdd->sd = NULL;
free_percpu(sdd->sg);
sdd->sg = NULL;
free_percpu(sdd->sgp);
sdd->sgp = NULL;
free_percpu(sdd->sgc);
sdd->sgc = NULL;
}
}
......@@ -6496,14 +6522,14 @@ static int build_sched_domains(const struct cpumask *cpu_map,
}
}
/* Calculate CPU power for physical packages and nodes */
/* Calculate CPU capacity for physical packages and nodes */
for (i = nr_cpumask_bits-1; i >= 0; i--) {
if (!cpumask_test_cpu(i, cpu_map))
continue;
for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) {
claim_allocations(i, sd);
init_sched_groups_power(i, sd);
init_sched_groups_capacity(i, sd);
}
}
......@@ -6946,7 +6972,7 @@ void __init sched_init(void)
#ifdef CONFIG_SMP
rq->sd = NULL;
rq->rd = NULL;
rq->cpu_power = SCHED_POWER_SCALE;
rq->cpu_capacity = SCHED_CAPACITY_SCALE;
rq->post_schedule = 0;
rq->active_balance = 0;
rq->next_balance = jiffies;
......
......@@ -57,8 +57,6 @@ void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime)
dl_b->dl_runtime = runtime;
}
extern unsigned long to_ratio(u64 period, u64 runtime);
void init_dl_bw(struct dl_bw *dl_b)
{
raw_spin_lock_init(&dl_b->lock);
......
......@@ -1017,7 +1017,7 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page,
static unsigned long weighted_cpuload(const int cpu);
static unsigned long source_load(int cpu, int type);
static unsigned long target_load(int cpu, int type);
static unsigned long power_of(int cpu);
static unsigned long capacity_of(int cpu);
static long effective_load(struct task_group *tg, int cpu, long wl, long wg);
/* Cached statistics for all CPUs within a node */
......@@ -1026,11 +1026,11 @@ struct numa_stats {
unsigned long load;
/* Total compute capacity of CPUs on a node */
unsigned long power;
unsigned long compute_capacity;
/* Approximate capacity in terms of runnable tasks on a node */
unsigned long capacity;
int has_capacity;
unsigned long task_capacity;
int has_free_capacity;
};
/*
......@@ -1046,7 +1046,7 @@ static void update_numa_stats(struct numa_stats *ns, int nid)
ns->nr_running += rq->nr_running;
ns->load += weighted_cpuload(cpu);
ns->power += power_of(cpu);
ns->compute_capacity += capacity_of(cpu);
cpus++;
}
......@@ -1056,15 +1056,16 @@ static void update_numa_stats(struct numa_stats *ns, int nid)
* the @ns structure is NULL'ed and task_numa_compare() will
* not find this node attractive.
*
* We'll either bail at !has_capacity, or we'll detect a huge imbalance
* and bail there.
* We'll either bail at !has_free_capacity, or we'll detect a huge
* imbalance and bail there.
*/
if (!cpus)
return;
ns->load = (ns->load * SCHED_POWER_SCALE) / ns->power;
ns->capacity = DIV_ROUND_CLOSEST(ns->power, SCHED_POWER_SCALE);
ns->has_capacity = (ns->nr_running < ns->capacity);
ns->load = (ns->load * SCHED_CAPACITY_SCALE) / ns->compute_capacity;
ns->task_capacity =
DIV_ROUND_CLOSEST(ns->compute_capacity, SCHED_CAPACITY_SCALE);
ns->has_free_capacity = (ns->nr_running < ns->task_capacity);
}
struct task_numa_env {
......@@ -1195,8 +1196,8 @@ static void task_numa_compare(struct task_numa_env *env,
if (!cur) {
/* Is there capacity at our destination? */
if (env->src_stats.has_capacity &&
!env->dst_stats.has_capacity)
if (env->src_stats.has_free_capacity &&
!env->dst_stats.has_free_capacity)
goto unlock;
goto balance;
......@@ -1213,7 +1214,7 @@ static void task_numa_compare(struct task_numa_env *env,
orig_dst_load = env->dst_stats.load;
orig_src_load = env->src_stats.load;
/* XXX missing power terms */
/* XXX missing capacity terms */
load = task_h_load(env->p);
dst_load = orig_dst_load + load;
src_load = orig_src_load - load;
......@@ -1301,8 +1302,8 @@ static int task_numa_migrate(struct task_struct *p)
groupimp = group_weight(p, env.dst_nid) - groupweight;
update_numa_stats(&env.dst_stats, env.dst_nid);
/* If the preferred nid has capacity, try to use it. */
if (env.dst_stats.has_capacity)
/* If the preferred nid has free capacity, try to use it. */
if (env.dst_stats.has_free_capacity)
task_numa_find_cpu(&env, taskimp, groupimp);
/* No space available on the preferred nid. Look elsewhere. */
......@@ -3225,10 +3226,12 @@ static void expire_cfs_rq_runtime(struct cfs_rq *cfs_rq)
* has not truly expired.
*
* Fortunately we can check determine whether this the case by checking
* whether the global deadline has advanced.
* whether the global deadline has advanced. It is valid to compare
* cfs_b->runtime_expires without any locks since we only care about
* exact equality, so a partial write will still work.
*/
if ((s64)(cfs_rq->runtime_expires - cfs_b->runtime_expires) >= 0) {
if (cfs_rq->runtime_expires != cfs_b->runtime_expires) {
/* extend local deadline, drift is bounded above by 2 ticks */
cfs_rq->runtime_expires += TICK_NSEC;
} else {
......@@ -3457,21 +3460,21 @@ static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b,
static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun)
{
u64 runtime, runtime_expires;
int idle = 1, throttled;
int throttled;
raw_spin_lock(&cfs_b->lock);
/* no need to continue the timer with no bandwidth constraint */
if (cfs_b->quota == RUNTIME_INF)
goto out_unlock;
goto out_deactivate;
throttled = !list_empty(&cfs_b->throttled_cfs_rq);
/* idle depends on !throttled (for the case of a large deficit) */
idle = cfs_b->idle && !throttled;
cfs_b->nr_periods += overrun;
/* if we're going inactive then everything else can be deferred */
if (idle)
goto out_unlock;
/*
* idle depends on !throttled (for the case of a large deficit), and if
* we're going inactive then everything else can be deferred
*/
if (cfs_b->idle && !throttled)
goto out_deactivate;
/*
* if we have relooped after returning idle once, we need to update our
......@@ -3485,7 +3488,7 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun)
if (!throttled) {
/* mark as potentially idle for the upcoming period */
cfs_b->idle = 1;
goto out_unlock;
return 0;
}
/* account preceding periods in which throttling occurred */
......@@ -3525,12 +3528,12 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun)
* timer to remain active while there are any throttled entities.)
*/
cfs_b->idle = 0;
out_unlock:
if (idle)
cfs_b->timer_active = 0;
raw_spin_unlock(&cfs_b->lock);
return idle;
return 0;
out_deactivate:
cfs_b->timer_active = 0;
return 1;
}
/* a cfs_rq won't donate quota below this amount */
......@@ -3707,6 +3710,7 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer)
int overrun;
int idle = 0;
raw_spin_lock(&cfs_b->lock);
for (;;) {
now = hrtimer_cb_get_time(timer);
overrun = hrtimer_forward(timer, now, cfs_b->period);
......@@ -3716,6 +3720,7 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer)
idle = do_sched_cfs_period_timer(cfs_b, overrun);
}
raw_spin_unlock(&cfs_b->lock);
return idle ? HRTIMER_NORESTART : HRTIMER_RESTART;
}
......@@ -3775,8 +3780,6 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq)
struct cfs_rq *cfs_rq;
for_each_leaf_cfs_rq(rq, cfs_rq) {
struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg);
if (!cfs_rq->runtime_enabled)
continue;
......@@ -3784,7 +3787,7 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq)
* clock_task is not advancing so we just need to make sure
* there's some valid quota amount
*/
cfs_rq->runtime_remaining = cfs_b->quota;
cfs_rq->runtime_remaining = 1;
if (cfs_rq_throttled(cfs_rq))
unthrottle_cfs_rq(cfs_rq);
}
......@@ -4041,9 +4044,9 @@ static unsigned long target_load(int cpu, int type)
return max(rq->cpu_load[type-1], total);
}
static unsigned long power_of(int cpu)
static unsigned long capacity_of(int cpu)
{
return cpu_rq(cpu)->cpu_power;
return cpu_rq(cpu)->cpu_capacity;
}
static unsigned long cpu_avg_load_per_task(int cpu)
......@@ -4065,7 +4068,7 @@ static void record_wakee(struct task_struct *p)
* about the boundary, really active task won't care
* about the loss.
*/
if (jiffies > current->wakee_flip_decay_ts + HZ) {
if (time_after(jiffies, current->wakee_flip_decay_ts + HZ)) {
current->wakee_flips >>= 1;
current->wakee_flip_decay_ts = jiffies;
}
......@@ -4286,12 +4289,12 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)
s64 this_eff_load, prev_eff_load;
this_eff_load = 100;
this_eff_load *= power_of(prev_cpu);
this_eff_load *= capacity_of(prev_cpu);
this_eff_load *= this_load +
effective_load(tg, this_cpu, weight, weight);
prev_eff_load = 100 + (sd->imbalance_pct - 100) / 2;
prev_eff_load *= power_of(this_cpu);
prev_eff_load *= capacity_of(this_cpu);
prev_eff_load *= load + effective_load(tg, prev_cpu, 0, weight);
balanced = this_eff_load <= prev_eff_load;
......@@ -4367,8 +4370,8 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p,
avg_load += load;
}
/* Adjust by relative CPU power of the group */
avg_load = (avg_load * SCHED_POWER_SCALE) / group->sgp->power;
/* Adjust by relative CPU capacity of the group */
avg_load = (avg_load * SCHED_CAPACITY_SCALE) / group->sgc->capacity;
if (local_group) {
this_load = avg_load;
......@@ -4948,14 +4951,14 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp
*
* W'_i,n = (2^n - 1) / 2^n * W_i,n + 1 / 2^n * W_i,0 (3)
*
* P_i is the cpu power (or compute capacity) of cpu i, typically it is the
* C_i is the compute capacity of cpu i, typically it is the
* fraction of 'recent' time available for SCHED_OTHER task execution. But it
* can also include other factors [XXX].
*
* To achieve this balance we define a measure of imbalance which follows
* directly from (1):
*
* imb_i,j = max{ avg(W/P), W_i/P_i } - min{ avg(W/P), W_j/P_j } (4)
* imb_i,j = max{ avg(W/C), W_i/C_i } - min{ avg(W/C), W_j/C_j } (4)
*
* We them move tasks around to minimize the imbalance. In the continuous
* function space it is obvious this converges, in the discrete case we get
......@@ -5530,13 +5533,13 @@ struct sg_lb_stats {
unsigned long group_load; /* Total load over the CPUs of the group */
unsigned long sum_weighted_load; /* Weighted load of group's tasks */
unsigned long load_per_task;
unsigned long group_power;
unsigned long group_capacity;
unsigned int sum_nr_running; /* Nr tasks running in the group */
unsigned int group_capacity;
unsigned int group_capacity_factor;
unsigned int idle_cpus;
unsigned int group_weight;
int group_imb; /* Is there an imbalance in the group ? */
int group_has_capacity; /* Is there extra capacity in the group? */
int group_has_free_capacity;
#ifdef CONFIG_NUMA_BALANCING
unsigned int nr_numa_running;
unsigned int nr_preferred_running;
......@@ -5551,7 +5554,7 @@ struct sd_lb_stats {
struct sched_group *busiest; /* Busiest group in this sd */
struct sched_group *local; /* Local group in this sd */
unsigned long total_load; /* Total load of all groups in sd */
unsigned long total_pwr; /* Total power of all groups in sd */
unsigned long total_capacity; /* Total capacity of all groups in sd */
unsigned long avg_load; /* Average load across all groups in sd */
struct sg_lb_stats busiest_stat;/* Statistics of the busiest group */
......@@ -5570,7 +5573,7 @@ static inline void init_sd_lb_stats(struct sd_lb_stats *sds)
.busiest = NULL,
.local = NULL,
.total_load = 0UL,
.total_pwr = 0UL,
.total_capacity = 0UL,
.busiest_stat = {
.avg_load = 0UL,
},
......@@ -5605,17 +5608,17 @@ static inline int get_sd_load_idx(struct sched_domain *sd,
return load_idx;
}
static unsigned long default_scale_freq_power(struct sched_domain *sd, int cpu)
static unsigned long default_scale_capacity(struct sched_domain *sd, int cpu)
{
return SCHED_POWER_SCALE;
return SCHED_CAPACITY_SCALE;
}
unsigned long __weak arch_scale_freq_power(struct sched_domain *sd, int cpu)
unsigned long __weak arch_scale_freq_capacity(struct sched_domain *sd, int cpu)
{
return default_scale_freq_power(sd, cpu);
return default_scale_capacity(sd, cpu);
}
static unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu)
static unsigned long default_scale_smt_capacity(struct sched_domain *sd, int cpu)
{
unsigned long weight = sd->span_weight;
unsigned long smt_gain = sd->smt_gain;
......@@ -5625,12 +5628,12 @@ static unsigned long default_scale_smt_power(struct sched_domain *sd, int cpu)
return smt_gain;
}
unsigned long __weak arch_scale_smt_power(struct sched_domain *sd, int cpu)
unsigned long __weak arch_scale_smt_capacity(struct sched_domain *sd, int cpu)
{
return default_scale_smt_power(sd, cpu);
return default_scale_smt_capacity(sd, cpu);
}
static unsigned long scale_rt_power(int cpu)
static unsigned long scale_rt_capacity(int cpu)
{
struct rq *rq = cpu_rq(cpu);
u64 total, available, age_stamp, avg;
......@@ -5650,71 +5653,71 @@ static unsigned long scale_rt_power(int cpu)
total = sched_avg_period() + delta;
if (unlikely(total < avg)) {
/* Ensures that power won't end up being negative */
/* Ensures that capacity won't end up being negative */
available = 0;
} else {
available = total - avg;
}
if (unlikely((s64)total < SCHED_POWER_SCALE))
total = SCHED_POWER_SCALE;
if (unlikely((s64)total < SCHED_CAPACITY_SCALE))
total = SCHED_CAPACITY_SCALE;
total >>= SCHED_POWER_SHIFT;
total >>= SCHED_CAPACITY_SHIFT;
return div_u64(available, total);
}
static void update_cpu_power(struct sched_domain *sd, int cpu)
static void update_cpu_capacity(struct sched_domain *sd, int cpu)
{
unsigned long weight = sd->span_weight;
unsigned long power = SCHED_POWER_SCALE;
unsigned long capacity = SCHED_CAPACITY_SCALE;
struct sched_group *sdg = sd->groups;
if ((sd->flags & SD_SHARE_CPUPOWER) && weight > 1) {
if (sched_feat(ARCH_POWER))
power *= arch_scale_smt_power(sd, cpu);
if ((sd->flags & SD_SHARE_CPUCAPACITY) && weight > 1) {
if (sched_feat(ARCH_CAPACITY))
capacity *= arch_scale_smt_capacity(sd, cpu);
else
power *= default_scale_smt_power(sd, cpu);
capacity *= default_scale_smt_capacity(sd, cpu);
power >>= SCHED_POWER_SHIFT;
capacity >>= SCHED_CAPACITY_SHIFT;
}
sdg->sgp->power_orig = power;
sdg->sgc->capacity_orig = capacity;
if (sched_feat(ARCH_POWER))
power *= arch_scale_freq_power(sd, cpu);
if (sched_feat(ARCH_CAPACITY))
capacity *= arch_scale_freq_capacity(sd, cpu);
else
power *= default_scale_freq_power(sd, cpu);
capacity *= default_scale_capacity(sd, cpu);
power >>= SCHED_POWER_SHIFT;
capacity >>= SCHED_CAPACITY_SHIFT;
power *= scale_rt_power(cpu);
power >>= SCHED_POWER_SHIFT;
capacity *= scale_rt_capacity(cpu);
capacity >>= SCHED_CAPACITY_SHIFT;
if (!power)
power = 1;
if (!capacity)
capacity = 1;
cpu_rq(cpu)->cpu_power = power;
sdg->sgp->power = power;
cpu_rq(cpu)->cpu_capacity = capacity;
sdg->sgc->capacity = capacity;
}
void update_group_power(struct sched_domain *sd, int cpu)
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 power, power_orig;
unsigned long capacity, capacity_orig;
unsigned long interval;
interval = msecs_to_jiffies(sd->balance_interval);
interval = clamp(interval, 1UL, max_load_balance_interval);
sdg->sgp->next_update = jiffies + interval;
sdg->sgc->next_update = jiffies + interval;
if (!child) {
update_cpu_power(sd, cpu);
update_cpu_capacity(sd, cpu);
return;
}
power_orig = power = 0;
capacity_orig = capacity = 0;
if (child->flags & SD_OVERLAP) {
/*
......@@ -5723,31 +5726,31 @@ void update_group_power(struct sched_domain *sd, int cpu)
*/
for_each_cpu(cpu, sched_group_cpus(sdg)) {
struct sched_group_power *sgp;
struct sched_group_capacity *sgc;
struct rq *rq = cpu_rq(cpu);
/*
* build_sched_domains() -> init_sched_groups_power()
* build_sched_domains() -> init_sched_groups_capacity()
* gets here before we've attached the domains to the
* runqueues.
*
* Use power_of(), which is set irrespective of domains
* in update_cpu_power().
* Use capacity_of(), which is set irrespective of domains
* in update_cpu_capacity().
*
* This avoids power/power_orig from being 0 and
* This avoids capacity/capacity_orig from being 0 and
* causing divide-by-zero issues on boot.
*
* Runtime updates will correct power_orig.
* Runtime updates will correct capacity_orig.
*/
if (unlikely(!rq->sd)) {
power_orig += power_of(cpu);
power += power_of(cpu);
capacity_orig += capacity_of(cpu);
capacity += capacity_of(cpu);
continue;
}
sgp = rq->sd->groups->sgp;
power_orig += sgp->power_orig;
power += sgp->power;
sgc = rq->sd->groups->sgc;
capacity_orig += sgc->capacity_orig;
capacity += sgc->capacity;
}
} else {
/*
......@@ -5757,14 +5760,14 @@ void update_group_power(struct sched_domain *sd, int cpu)
group = child->groups;
do {
power_orig += group->sgp->power_orig;
power += group->sgp->power;
capacity_orig += group->sgc->capacity_orig;
capacity += group->sgc->capacity;
group = group->next;
} while (group != child->groups);
}
sdg->sgp->power_orig = power_orig;
sdg->sgp->power = power;
sdg->sgc->capacity_orig = capacity_orig;
sdg->sgc->capacity = capacity;
}
/*
......@@ -5778,15 +5781,15 @@ static inline int
fix_small_capacity(struct sched_domain *sd, struct sched_group *group)
{
/*
* Only siblings can have significantly less than SCHED_POWER_SCALE
* Only siblings can have significantly less than SCHED_CAPACITY_SCALE
*/
if (!(sd->flags & SD_SHARE_CPUPOWER))
if (!(sd->flags & SD_SHARE_CPUCAPACITY))
return 0;
/*
* If ~90% of the cpu_power is still there, we're good.
* If ~90% of the cpu_capacity is still there, we're good.
*/
if (group->sgp->power * 32 > group->sgp->power_orig * 29)
if (group->sgc->capacity * 32 > group->sgc->capacity_orig * 29)
return 1;
return 0;
......@@ -5823,34 +5826,35 @@ fix_small_capacity(struct sched_domain *sd, struct sched_group *group)
static inline int sg_imbalanced(struct sched_group *group)
{
return group->sgp->imbalance;
return group->sgc->imbalance;
}
/*
* Compute the group capacity.
* Compute the group capacity factor.
*
* Avoid the issue where N*frac(smt_power) >= 1 creates 'phantom' cores by
* 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 power unit capacity with that.
* and limit unit capacity with that.
*/
static inline int sg_capacity(struct lb_env *env, struct sched_group *group)
static inline int sg_capacity_factor(struct lb_env *env, struct sched_group *group)
{
unsigned int capacity, smt, cpus;
unsigned int power, power_orig;
unsigned int capacity_factor, smt, cpus;
unsigned int capacity, capacity_orig;
power = group->sgp->power;
power_orig = group->sgp->power_orig;
capacity = group->sgc->capacity;
capacity_orig = group->sgc->capacity_orig;
cpus = group->group_weight;
/* smt := ceil(cpus / power), assumes: 1 < smt_power < 2 */
smt = DIV_ROUND_UP(SCHED_POWER_SCALE * cpus, power_orig);
capacity = cpus / smt; /* cores */
/* smt := ceil(cpus / capacity), assumes: 1 < smt_capacity < 2 */
smt = DIV_ROUND_UP(SCHED_CAPACITY_SCALE * cpus, capacity_orig);
capacity_factor = cpus / smt; /* cores */
capacity = min_t(unsigned, capacity, DIV_ROUND_CLOSEST(power, SCHED_POWER_SCALE));
if (!capacity)
capacity = fix_small_capacity(env->sd, group);
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);
return capacity;
return capacity_factor;
}
/**
......@@ -5890,9 +5894,9 @@ static inline void update_sg_lb_stats(struct lb_env *env,
sgs->idle_cpus++;
}
/* Adjust by relative CPU power of the group */
sgs->group_power = group->sgp->power;
sgs->avg_load = (sgs->group_load*SCHED_POWER_SCALE) / sgs->group_power;
/* Adjust by relative CPU capacity of the group */
sgs->group_capacity = group->sgc->capacity;
sgs->avg_load = (sgs->group_load*SCHED_CAPACITY_SCALE) / sgs->group_capacity;
if (sgs->sum_nr_running)
sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running;
......@@ -5900,10 +5904,10 @@ static inline void update_sg_lb_stats(struct lb_env *env,
sgs->group_weight = group->group_weight;
sgs->group_imb = sg_imbalanced(group);
sgs->group_capacity = sg_capacity(env, group);
sgs->group_capacity_factor = sg_capacity_factor(env, group);
if (sgs->group_capacity > sgs->sum_nr_running)
sgs->group_has_capacity = 1;
if (sgs->group_capacity_factor > sgs->sum_nr_running)
sgs->group_has_free_capacity = 1;
}
/**
......@@ -5927,7 +5931,7 @@ static bool update_sd_pick_busiest(struct lb_env *env,
if (sgs->avg_load <= sds->busiest_stat.avg_load)
return false;
if (sgs->sum_nr_running > sgs->group_capacity)
if (sgs->sum_nr_running > sgs->group_capacity_factor)
return true;
if (sgs->group_imb)
......@@ -6007,8 +6011,8 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
sgs = &sds->local_stat;
if (env->idle != CPU_NEWLY_IDLE ||
time_after_eq(jiffies, sg->sgp->next_update))
update_group_power(env->sd, env->dst_cpu);
time_after_eq(jiffies, sg->sgc->next_update))
update_group_capacity(env->sd, env->dst_cpu);
}
update_sg_lb_stats(env, sg, load_idx, local_group, sgs);
......@@ -6018,17 +6022,17 @@ 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 to one so that we'll try
* first, lower the sg capacity factor to one 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. The
* 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).
*/
if (prefer_sibling && sds->local &&
sds->local_stat.group_has_capacity)
sgs->group_capacity = min(sgs->group_capacity, 1U);
sds->local_stat.group_has_free_capacity)
sgs->group_capacity_factor = min(sgs->group_capacity_factor, 1U);
if (update_sd_pick_busiest(env, sds, sg, sgs)) {
sds->busiest = sg;
......@@ -6038,7 +6042,7 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
next_group:
/* Now, start updating sd_lb_stats */
sds->total_load += sgs->group_load;
sds->total_pwr += sgs->group_power;
sds->total_capacity += sgs->group_capacity;
sg = sg->next;
} while (sg != env->sd->groups);
......@@ -6085,8 +6089,8 @@ static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds)
return 0;
env->imbalance = DIV_ROUND_CLOSEST(
sds->busiest_stat.avg_load * sds->busiest_stat.group_power,
SCHED_POWER_SCALE);
sds->busiest_stat.avg_load * sds->busiest_stat.group_capacity,
SCHED_CAPACITY_SCALE);
return 1;
}
......@@ -6101,7 +6105,7 @@ static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds)
static inline
void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds)
{
unsigned long tmp, pwr_now = 0, pwr_move = 0;
unsigned long tmp, capa_now = 0, capa_move = 0;
unsigned int imbn = 2;
unsigned long scaled_busy_load_per_task;
struct sg_lb_stats *local, *busiest;
......@@ -6115,8 +6119,8 @@ void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds)
imbn = 1;
scaled_busy_load_per_task =
(busiest->load_per_task * SCHED_POWER_SCALE) /
busiest->group_power;
(busiest->load_per_task * SCHED_CAPACITY_SCALE) /
busiest->group_capacity;
if (busiest->avg_load + scaled_busy_load_per_task >=
local->avg_load + (scaled_busy_load_per_task * imbn)) {
......@@ -6126,38 +6130,38 @@ void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds)
/*
* OK, we don't have enough imbalance to justify moving tasks,
* however we may be able to increase total CPU power used by
* however we may be able to increase total CPU capacity used by
* moving them.
*/
pwr_now += busiest->group_power *
capa_now += busiest->group_capacity *
min(busiest->load_per_task, busiest->avg_load);
pwr_now += local->group_power *
capa_now += local->group_capacity *
min(local->load_per_task, local->avg_load);
pwr_now /= SCHED_POWER_SCALE;
capa_now /= SCHED_CAPACITY_SCALE;
/* Amount of load we'd subtract */
if (busiest->avg_load > scaled_busy_load_per_task) {
pwr_move += busiest->group_power *
capa_move += busiest->group_capacity *
min(busiest->load_per_task,
busiest->avg_load - scaled_busy_load_per_task);
}
/* Amount of load we'd add */
if (busiest->avg_load * busiest->group_power <
busiest->load_per_task * SCHED_POWER_SCALE) {
tmp = (busiest->avg_load * busiest->group_power) /
local->group_power;
if (busiest->avg_load * busiest->group_capacity <
busiest->load_per_task * SCHED_CAPACITY_SCALE) {
tmp = (busiest->avg_load * busiest->group_capacity) /
local->group_capacity;
} else {
tmp = (busiest->load_per_task * SCHED_POWER_SCALE) /
local->group_power;
tmp = (busiest->load_per_task * SCHED_CAPACITY_SCALE) /
local->group_capacity;
}
pwr_move += local->group_power *
capa_move += local->group_capacity *
min(local->load_per_task, local->avg_load + tmp);
pwr_move /= SCHED_POWER_SCALE;
capa_move /= SCHED_CAPACITY_SCALE;
/* Move if we gain throughput */
if (pwr_move > pwr_now)
if (capa_move > capa_now)
env->imbalance = busiest->load_per_task;
}
......@@ -6187,7 +6191,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
/*
* In the presence of smp nice balancing, certain scenarios can have
* max load less than avg load(as we skip the groups at or below
* its cpu_power, while calculating max_load..)
* its cpu_capacity, while calculating max_load..)
*/
if (busiest->avg_load <= sds->avg_load ||
local->avg_load >= sds->avg_load) {
......@@ -6202,10 +6206,10 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
* have to drop below capacity to reach cpu-load equilibrium.
*/
load_above_capacity =
(busiest->sum_nr_running - busiest->group_capacity);
(busiest->sum_nr_running - busiest->group_capacity_factor);
load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_POWER_SCALE);
load_above_capacity /= busiest->group_power;
load_above_capacity *= (SCHED_LOAD_SCALE * SCHED_CAPACITY_SCALE);
load_above_capacity /= busiest->group_capacity;
}
/*
......@@ -6220,9 +6224,9 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s
/* How much load to actually move to equalise the imbalance */
env->imbalance = min(
max_pull * busiest->group_power,
(sds->avg_load - local->avg_load) * local->group_power
) / SCHED_POWER_SCALE;
max_pull * busiest->group_capacity,
(sds->avg_load - local->avg_load) * local->group_capacity
) / SCHED_CAPACITY_SCALE;
/*
* if *imbalance is less than the average load per runnable task
......@@ -6276,7 +6280,8 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
if (!sds.busiest || busiest->sum_nr_running == 0)
goto out_balanced;
sds.avg_load = (SCHED_POWER_SCALE * sds.total_load) / sds.total_pwr;
sds.avg_load = (SCHED_CAPACITY_SCALE * sds.total_load)
/ sds.total_capacity;
/*
* If the busiest group is imbalanced the below checks don't
......@@ -6287,8 +6292,8 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
goto force_balance;
/* SD_BALANCE_NEWIDLE trumps SMP nice when underutilized */
if (env->idle == CPU_NEWLY_IDLE && local->group_has_capacity &&
!busiest->group_has_capacity)
if (env->idle == CPU_NEWLY_IDLE && local->group_has_free_capacity &&
!busiest->group_has_free_capacity)
goto force_balance;
/*
......@@ -6342,11 +6347,11 @@ static struct rq *find_busiest_queue(struct lb_env *env,
struct sched_group *group)
{
struct rq *busiest = NULL, *rq;
unsigned long busiest_load = 0, busiest_power = 1;
unsigned long busiest_load = 0, busiest_capacity = 1;
int i;
for_each_cpu_and(i, sched_group_cpus(group), env->cpus) {
unsigned long power, capacity, wl;
unsigned long capacity, capacity_factor, wl;
enum fbq_type rt;
rq = cpu_rq(i);
......@@ -6374,34 +6379,34 @@ static struct rq *find_busiest_queue(struct lb_env *env,
if (rt > env->fbq_type)
continue;
power = power_of(i);
capacity = DIV_ROUND_CLOSEST(power, SCHED_POWER_SCALE);
if (!capacity)
capacity = fix_small_capacity(env->sd, group);
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);
/*
* When comparing with imbalance, use weighted_cpuload()
* which is not scaled with the cpu power.
* which is not scaled with the cpu capacity.
*/
if (capacity && rq->nr_running == 1 && wl > env->imbalance)
if (capacity_factor && rq->nr_running == 1 && wl > env->imbalance)
continue;
/*
* For the load comparisons with the other cpu's, consider
* the weighted_cpuload() scaled with the cpu power, so that
* the load can be moved away from the cpu that is potentially
* running at a lower capacity.
* the weighted_cpuload() scaled with the cpu capacity, so
* that the load can be moved away from the cpu that is
* potentially running at a lower capacity.
*
* Thus we're looking for max(wl_i / power_i), crosswise
* Thus we're looking for max(wl_i / capacity_i), crosswise
* multiplication to rid ourselves of the division works out
* to: wl_i * power_j > wl_j * power_i; where j is our
* previous maximum.
* to: wl_i * capacity_j > wl_j * capacity_i; where j is
* our previous maximum.
*/
if (wl * busiest_power > busiest_load * power) {
if (wl * busiest_capacity > busiest_load * capacity) {
busiest_load = wl;
busiest_power = power;
busiest_capacity = capacity;
busiest = rq;
}
}
......@@ -6609,7 +6614,7 @@ static int load_balance(int this_cpu, struct rq *this_rq,
* We failed to reach balance because of affinity.
*/
if (sd_parent) {
int *group_imbalance = &sd_parent->groups->sgp->imbalance;
int *group_imbalance = &sd_parent->groups->sgc->imbalance;
if ((env.flags & LBF_SOME_PINNED) && env.imbalance > 0) {
*group_imbalance = 1;
......@@ -6996,7 +7001,7 @@ static inline void set_cpu_sd_state_busy(void)
goto unlock;
sd->nohz_idle = 0;
atomic_inc(&sd->groups->sgp->nr_busy_cpus);
atomic_inc(&sd->groups->sgc->nr_busy_cpus);
unlock:
rcu_read_unlock();
}
......@@ -7013,7 +7018,7 @@ void set_cpu_sd_state_idle(void)
goto unlock;
sd->nohz_idle = 1;
atomic_dec(&sd->groups->sgp->nr_busy_cpus);
atomic_dec(&sd->groups->sgc->nr_busy_cpus);
unlock:
rcu_read_unlock();
}
......@@ -7192,12 +7197,17 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle)
rq = cpu_rq(balance_cpu);
raw_spin_lock_irq(&rq->lock);
update_rq_clock(rq);
update_idle_cpu_load(rq);
raw_spin_unlock_irq(&rq->lock);
rebalance_domains(rq, CPU_IDLE);
/*
* If time for next balance is due,
* do the balance.
*/
if (time_after_eq(jiffies, rq->next_balance)) {
raw_spin_lock_irq(&rq->lock);
update_rq_clock(rq);
update_idle_cpu_load(rq);
raw_spin_unlock_irq(&rq->lock);
rebalance_domains(rq, CPU_IDLE);
}
if (time_after(this_rq->next_balance, rq->next_balance))
this_rq->next_balance = rq->next_balance;
......@@ -7212,7 +7222,7 @@ static void nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle)
* of an idle cpu is 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 power.
* busy cpu's exceeding the group's capacity.
* - For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler
* domain span are idle.
*/
......@@ -7220,7 +7230,7 @@ static inline int nohz_kick_needed(struct rq *rq)
{
unsigned long now = jiffies;
struct sched_domain *sd;
struct sched_group_power *sgp;
struct sched_group_capacity *sgc;
int nr_busy, cpu = rq->cpu;
if (unlikely(rq->idle_balance))
......@@ -7250,8 +7260,8 @@ static inline int nohz_kick_needed(struct rq *rq)
sd = rcu_dereference(per_cpu(sd_busy, cpu));
if (sd) {
sgp = sd->groups->sgp;
nr_busy = atomic_read(&sgp->nr_busy_cpus);
sgc = sd->groups->sgc;
nr_busy = atomic_read(&sgc->nr_busy_cpus);
if (nr_busy > 1)
goto need_kick_unlock;
......
......@@ -37,18 +37,18 @@ SCHED_FEAT(CACHE_HOT_BUDDY, true)
SCHED_FEAT(WAKEUP_PREEMPTION, true)
/*
* Use arch dependent cpu power functions
* Use arch dependent cpu capacity functions
*/
SCHED_FEAT(ARCH_POWER, true)
SCHED_FEAT(ARCH_CAPACITY, true)
SCHED_FEAT(HRTICK, false)
SCHED_FEAT(DOUBLE_TICK, false)
SCHED_FEAT(LB_BIAS, true)
/*
* Decrement CPU power based on time not spent running tasks
* Decrement CPU capacity based on time not spent running tasks
*/
SCHED_FEAT(NONTASK_POWER, true)
SCHED_FEAT(NONTASK_CAPACITY, true)
/*
* Queue remote wakeups on the target CPU and process them
......
......@@ -12,6 +12,8 @@
#include <trace/events/power.h>
#include "sched.h"
static int __read_mostly cpu_idle_force_poll;
void cpu_idle_poll_ctrl(bool enable)
......@@ -67,6 +69,10 @@ void __weak arch_cpu_idle(void)
* cpuidle_idle_call - the main idle function
*
* NOTE: no locks or semaphores should be used here
*
* On archs that support TIF_POLLING_NRFLAG, is called with polling
* set, and it returns with polling set. If it ever stops polling, it
* must clear the polling bit.
*/
static void cpuidle_idle_call(void)
{
......@@ -175,10 +181,22 @@ static void cpuidle_idle_call(void)
/*
* Generic idle loop implementation
*
* Called with polling cleared.
*/
static void cpu_idle_loop(void)
{
while (1) {
/*
* If the arch has a polling bit, we maintain an invariant:
*
* Our polling bit is clear if we're not scheduled (i.e. if
* rq->curr != rq->idle). This means that, if rq->idle has
* the polling bit set, then setting need_resched is
* guaranteed to cause the cpu to reschedule.
*/
__current_set_polling();
tick_nohz_idle_enter();
while (!need_resched()) {
......@@ -218,6 +236,17 @@ static void cpu_idle_loop(void)
*/
preempt_set_need_resched();
tick_nohz_idle_exit();
__current_clr_polling();
/*
* We promise to call sched_ttwu_pending and reschedule
* if need_resched is set while polling is set. That
* means that clearing polling needs to be visible
* before doing these things.
*/
smp_mb__after_atomic();
sched_ttwu_pending();
schedule_preempt_disabled();
}
}
......@@ -239,7 +268,6 @@ void cpu_startup_entry(enum cpuhp_state state)
*/
boot_init_stack_canary();
#endif
__current_set_polling();
arch_cpu_idle_prepare();
cpu_idle_loop();
}
......@@ -918,7 +918,6 @@ static void update_curr_rt(struct rq *rq)
{
struct task_struct *curr = rq->curr;
struct sched_rt_entity *rt_se = &curr->rt;
struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
u64 delta_exec;
if (curr->sched_class != &rt_sched_class)
......@@ -943,7 +942,7 @@ static void update_curr_rt(struct rq *rq)
return;
for_each_sched_rt_entity(rt_se) {
rt_rq = rt_rq_of_se(rt_se);
struct rt_rq *rt_rq = rt_rq_of_se(rt_se);
if (sched_rt_runtime(rt_rq) != RUNTIME_INF) {
raw_spin_lock(&rt_rq->rt_runtime_lock);
......
......@@ -567,7 +567,7 @@ struct rq {
struct root_domain *rd;
struct sched_domain *sd;
unsigned long cpu_power;
unsigned long cpu_capacity;
unsigned char idle_balance;
/* For active balancing */
......@@ -670,6 +670,8 @@ extern int migrate_swap(struct task_struct *, struct task_struct *);
#ifdef CONFIG_SMP
extern void sched_ttwu_pending(void);
#define rcu_dereference_check_sched_domain(p) \
rcu_dereference_check((p), \
lockdep_is_held(&sched_domains_mutex))
......@@ -728,15 +730,15 @@ DECLARE_PER_CPU(struct sched_domain *, sd_numa);
DECLARE_PER_CPU(struct sched_domain *, sd_busy);
DECLARE_PER_CPU(struct sched_domain *, sd_asym);
struct sched_group_power {
struct sched_group_capacity {
atomic_t ref;
/*
* CPU power of this group, SCHED_LOAD_SCALE being max power for a
* single CPU.
* CPU capacity of this group, SCHED_LOAD_SCALE being max capacity
* for a single CPU.
*/
unsigned int power, power_orig;
unsigned int capacity, capacity_orig;
unsigned long next_update;
int imbalance; /* XXX unrelated to power but shared group state */
int imbalance; /* XXX unrelated to capacity but shared group state */
/*
* Number of busy cpus in this group.
*/
......@@ -750,7 +752,7 @@ struct sched_group {
atomic_t ref;
unsigned int group_weight;
struct sched_group_power *sgp;
struct sched_group_capacity *sgc;
/*
* The CPUs this group covers.
......@@ -773,7 +775,7 @@ static inline struct cpumask *sched_group_cpus(struct sched_group *sg)
*/
static inline struct cpumask *sched_group_mask(struct sched_group *sg)
{
return to_cpumask(sg->sgp->cpumask);
return to_cpumask(sg->sgc->cpumask);
}
/**
......@@ -787,6 +789,10 @@ static inline unsigned int group_first_cpu(struct sched_group *group)
extern int group_balance_cpu(struct sched_group *sg);
#else
static inline void sched_ttwu_pending(void) { }
#endif /* CONFIG_SMP */
#include "stats.h"
......@@ -1167,7 +1173,7 @@ extern const struct sched_class idle_sched_class;
#ifdef CONFIG_SMP
extern void update_group_power(struct sched_domain *sd, int cpu);
extern void update_group_capacity(struct sched_domain *sd, int cpu);
extern void trigger_load_balance(struct rq *rq);
......
......@@ -1714,11 +1714,11 @@ void kvm_vcpu_kick(struct kvm_vcpu *vcpu)
EXPORT_SYMBOL_GPL(kvm_vcpu_kick);
#endif /* !CONFIG_S390 */
bool kvm_vcpu_yield_to(struct kvm_vcpu *target)
int kvm_vcpu_yield_to(struct kvm_vcpu *target)
{
struct pid *pid;
struct task_struct *task = NULL;
bool ret = false;
int ret = 0;
rcu_read_lock();
pid = rcu_dereference(target->pid);
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment