diff options
Diffstat (limited to 'kernel/sched')
| -rw-r--r-- | kernel/sched/autogroup.c | 2 | ||||
| -rw-r--r-- | kernel/sched/core.c | 132 | ||||
| -rw-r--r-- | kernel/sched/cpufreq_schedutil.c | 43 | ||||
| -rw-r--r-- | kernel/sched/cputime.c | 16 | ||||
| -rw-r--r-- | kernel/sched/deadline.c | 129 | ||||
| -rw-r--r-- | kernel/sched/debug.c | 24 | ||||
| -rw-r--r-- | kernel/sched/ext.c | 426 | ||||
| -rw-r--r-- | kernel/sched/ext.h | 8 | ||||
| -rw-r--r-- | kernel/sched/fair.c | 630 | ||||
| -rw-r--r-- | kernel/sched/features.h | 9 | ||||
| -rw-r--r-- | kernel/sched/idle.c | 5 | ||||
| -rw-r--r-- | kernel/sched/isolation.c | 22 | ||||
| -rw-r--r-- | kernel/sched/pelt.c | 4 | ||||
| -rw-r--r-- | kernel/sched/psi.c | 7 | ||||
| -rw-r--r-- | kernel/sched/rt.c | 2 | ||||
| -rw-r--r-- | kernel/sched/sched.h | 87 | ||||
| -rw-r--r-- | kernel/sched/stats.c | 11 | ||||
| -rw-r--r-- | kernel/sched/stats.h | 13 | ||||
| -rw-r--r-- | kernel/sched/syscalls.c | 25 | ||||
| -rw-r--r-- | kernel/sched/topology.c | 14 |
20 files changed, 953 insertions, 656 deletions
diff --git a/kernel/sched/autogroup.c b/kernel/sched/autogroup.c index db68a964e34e..83d46b9b8ec8 100644 --- a/kernel/sched/autogroup.c +++ b/kernel/sched/autogroup.c @@ -9,7 +9,7 @@ static struct autogroup autogroup_default; static atomic_t autogroup_seq_nr; #ifdef CONFIG_SYSCTL -static struct ctl_table sched_autogroup_sysctls[] = { +static const struct ctl_table sched_autogroup_sysctls[] = { { .procname = "sched_autogroup_enabled", .data = &sysctl_sched_autogroup_enabled, diff --git a/kernel/sched/core.c b/kernel/sched/core.c index c6d8232ad9ee..165c90ba64ea 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -740,39 +740,43 @@ static void update_rq_clock_task(struct rq *rq, s64 delta) s64 __maybe_unused steal = 0, irq_delta = 0; #ifdef CONFIG_IRQ_TIME_ACCOUNTING - irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; + if (irqtime_enabled()) { + irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; - /* - * Since irq_time is only updated on {soft,}irq_exit, we might run into - * this case when a previous update_rq_clock() happened inside a - * {soft,}IRQ region. - * - * When this happens, we stop ->clock_task and only update the - * prev_irq_time stamp to account for the part that fit, so that a next - * update will consume the rest. This ensures ->clock_task is - * monotonic. - * - * It does however cause some slight miss-attribution of {soft,}IRQ - * time, a more accurate solution would be to update the irq_time using - * the current rq->clock timestamp, except that would require using - * atomic ops. - */ - if (irq_delta > delta) - irq_delta = delta; + /* + * Since irq_time is only updated on {soft,}irq_exit, we might run into + * this case when a previous update_rq_clock() happened inside a + * {soft,}IRQ region. + * + * When this happens, we stop ->clock_task and only update the + * prev_irq_time stamp to account for the part that fit, so that a next + * update will consume the rest. This ensures ->clock_task is + * monotonic. + * + * It does however cause some slight miss-attribution of {soft,}IRQ + * time, a more accurate solution would be to update the irq_time using + * the current rq->clock timestamp, except that would require using + * atomic ops. + */ + if (irq_delta > delta) + irq_delta = delta; - rq->prev_irq_time += irq_delta; - delta -= irq_delta; - delayacct_irq(rq->curr, irq_delta); + rq->prev_irq_time += irq_delta; + delta -= irq_delta; + delayacct_irq(rq->curr, irq_delta); + } #endif #ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING if (static_key_false((¶virt_steal_rq_enabled))) { - steal = paravirt_steal_clock(cpu_of(rq)); + u64 prev_steal; + + steal = prev_steal = paravirt_steal_clock(cpu_of(rq)); steal -= rq->prev_steal_time_rq; if (unlikely(steal > delta)) steal = delta; - rq->prev_steal_time_rq += steal; + rq->prev_steal_time_rq = prev_steal; delta -= steal; } #endif @@ -789,6 +793,7 @@ static void update_rq_clock_task(struct rq *rq, s64 delta) void update_rq_clock(struct rq *rq) { s64 delta; + u64 clock; lockdep_assert_rq_held(rq); @@ -800,11 +805,14 @@ void update_rq_clock(struct rq *rq) SCHED_WARN_ON(rq->clock_update_flags & RQCF_UPDATED); rq->clock_update_flags |= RQCF_UPDATED; #endif + clock = sched_clock_cpu(cpu_of(rq)); + scx_rq_clock_update(rq, clock); - delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; + delta = clock - rq->clock; if (delta < 0) return; rq->clock += delta; + update_rq_clock_task(rq, delta); } @@ -1168,13 +1176,13 @@ int get_nohz_timer_target(void) struct sched_domain *sd; const struct cpumask *hk_mask; - if (housekeeping_cpu(cpu, HK_TYPE_TIMER)) { + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) { if (!idle_cpu(cpu)) return cpu; default_cpu = cpu; } - hk_mask = housekeeping_cpumask(HK_TYPE_TIMER); + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); guard(rcu)(); @@ -1189,7 +1197,7 @@ int get_nohz_timer_target(void) } if (default_cpu == -1) - default_cpu = housekeeping_any_cpu(HK_TYPE_TIMER); + default_cpu = housekeeping_any_cpu(HK_TYPE_KERNEL_NOISE); return default_cpu; } @@ -1341,7 +1349,7 @@ bool sched_can_stop_tick(struct rq *rq) if (scx_enabled() && !scx_can_stop_tick(rq)) return false; - if (rq->cfs.nr_running > 1) + if (rq->cfs.h_nr_queued > 1) return false; /* @@ -3534,7 +3542,7 @@ static int select_fallback_rq(int cpu, struct task_struct *p) * * More yuck to audit. */ - do_set_cpus_allowed(p, task_cpu_possible_mask(p)); + do_set_cpus_allowed(p, task_cpu_fallback_mask(p)); state = fail; break; case fail: @@ -4646,7 +4654,7 @@ static int sysctl_schedstats(const struct ctl_table *table, int write, void *buf #endif /* CONFIG_SCHEDSTATS */ #ifdef CONFIG_SYSCTL -static struct ctl_table sched_core_sysctls[] = { +static const struct ctl_table sched_core_sysctls[] = { #ifdef CONFIG_SCHEDSTATS { .procname = "sched_schedstats", @@ -5632,7 +5640,7 @@ void sched_tick(void) unsigned long hw_pressure; u64 resched_latency; - if (housekeeping_cpu(cpu, HK_TYPE_TICK)) + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) arch_scale_freq_tick(); sched_clock_tick(); @@ -5771,7 +5779,7 @@ static void sched_tick_start(int cpu) int os; struct tick_work *twork; - if (housekeeping_cpu(cpu, HK_TYPE_TICK)) + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) return; WARN_ON_ONCE(!tick_work_cpu); @@ -5792,7 +5800,7 @@ static void sched_tick_stop(int cpu) struct tick_work *twork; int os; - if (housekeeping_cpu(cpu, HK_TYPE_TICK)) + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) return; WARN_ON_ONCE(!tick_work_cpu); @@ -6018,7 +6026,7 @@ __pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) * opportunity to pull in more work from other CPUs. */ if (likely(!sched_class_above(prev->sched_class, &fair_sched_class) && - rq->nr_running == rq->cfs.h_nr_running)) { + rq->nr_running == rq->cfs.h_nr_queued)) { p = pick_next_task_fair(rq, prev, rf); if (unlikely(p == RETRY_TASK)) @@ -6641,7 +6649,6 @@ static void __sched notrace __schedule(int sched_mode) * as a preemption by schedule_debug() and RCU. */ bool preempt = sched_mode > SM_NONE; - bool block = false; unsigned long *switch_count; unsigned long prev_state; struct rq_flags rf; @@ -6702,7 +6709,7 @@ static void __sched notrace __schedule(int sched_mode) goto picked; } } else if (!preempt && prev_state) { - block = try_to_block_task(rq, prev, prev_state); + try_to_block_task(rq, prev, prev_state); switch_count = &prev->nvcsw; } @@ -6748,7 +6755,8 @@ picked: migrate_disable_switch(rq, prev); psi_account_irqtime(rq, prev, next); - psi_sched_switch(prev, next, block); + psi_sched_switch(prev, next, !task_on_rq_queued(prev) || + prev->se.sched_delayed); trace_sched_switch(preempt, prev, next, prev_state); @@ -7701,9 +7709,9 @@ void sched_show_task(struct task_struct *p) if (pid_alive(p)) ppid = task_pid_nr(rcu_dereference(p->real_parent)); rcu_read_unlock(); - pr_cont(" stack:%-5lu pid:%-5d tgid:%-5d ppid:%-6d flags:0x%08lx\n", + pr_cont(" stack:%-5lu pid:%-5d tgid:%-5d ppid:%-6d task_flags:0x%04x flags:0x%08lx\n", free, task_pid_nr(p), task_tgid_nr(p), - ppid, read_task_thread_flags(p)); + ppid, p->flags, read_task_thread_flags(p)); print_worker_info(KERN_INFO, p); print_stop_info(KERN_INFO, p); @@ -7930,19 +7938,26 @@ void sched_setnuma(struct task_struct *p, int nid) #ifdef CONFIG_HOTPLUG_CPU /* - * Ensure that the idle task is using init_mm right before its CPU goes - * offline. + * Invoked on the outgoing CPU in context of the CPU hotplug thread + * after ensuring that there are no user space tasks left on the CPU. + * + * If there is a lazy mm in use on the hotplug thread, drop it and + * switch to init_mm. + * + * The reference count on init_mm is dropped in finish_cpu(). */ -void idle_task_exit(void) +static void sched_force_init_mm(void) { struct mm_struct *mm = current->active_mm; - BUG_ON(cpu_online(smp_processor_id())); - BUG_ON(current != this_rq()->idle); - if (mm != &init_mm) { - switch_mm(mm, &init_mm, current); + mmgrab_lazy_tlb(&init_mm); + local_irq_disable(); + current->active_mm = &init_mm; + switch_mm_irqs_off(mm, &init_mm, current); + local_irq_enable(); finish_arch_post_lock_switch(); + mmdrop_lazy_tlb(mm); } /* finish_cpu(), as ran on the BP, will clean up the active_mm state */ @@ -8180,19 +8195,14 @@ static void cpuset_cpu_active(void) cpuset_update_active_cpus(); } -static int cpuset_cpu_inactive(unsigned int cpu) +static void cpuset_cpu_inactive(unsigned int cpu) { if (!cpuhp_tasks_frozen) { - int ret = dl_bw_check_overflow(cpu); - - if (ret) - return ret; cpuset_update_active_cpus(); } else { num_cpus_frozen++; partition_sched_domains(1, NULL, NULL); } - return 0; } static inline void sched_smt_present_inc(int cpu) @@ -8254,6 +8264,11 @@ int sched_cpu_deactivate(unsigned int cpu) struct rq *rq = cpu_rq(cpu); int ret; + ret = dl_bw_deactivate(cpu); + + if (ret) + return ret; + /* * Remove CPU from nohz.idle_cpus_mask to prevent participating in * load balancing when not active @@ -8299,15 +8314,7 @@ int sched_cpu_deactivate(unsigned int cpu) return 0; sched_update_numa(cpu, false); - ret = cpuset_cpu_inactive(cpu); - if (ret) { - sched_smt_present_inc(cpu); - sched_set_rq_online(rq, cpu); - balance_push_set(cpu, false); - set_cpu_active(cpu, true); - sched_update_numa(cpu, true); - return ret; - } + cpuset_cpu_inactive(cpu); sched_domains_numa_masks_clear(cpu); return 0; } @@ -8344,6 +8351,7 @@ int sched_cpu_starting(unsigned int cpu) int sched_cpu_wait_empty(unsigned int cpu) { balance_hotplug_wait(); + sched_force_init_mm(); return 0; } @@ -10590,7 +10598,7 @@ void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) return; /* No page allocation under rq lock */ - task_work_add(curr, work, TWA_RESUME | TWAF_NO_ALLOC); + task_work_add(curr, work, TWA_RESUME); } void sched_mm_cid_exit_signals(struct task_struct *t) diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 28c77904ea74..1a19d69b91ed 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -83,7 +83,7 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time) if (unlikely(sg_policy->limits_changed)) { sg_policy->limits_changed = false; - sg_policy->need_freq_update = true; + sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS); return true; } @@ -96,7 +96,7 @@ static bool sugov_update_next_freq(struct sugov_policy *sg_policy, u64 time, unsigned int next_freq) { if (sg_policy->need_freq_update) - sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS); + sg_policy->need_freq_update = false; else if (sg_policy->next_freq == next_freq) return false; @@ -604,31 +604,6 @@ static const struct kobj_type sugov_tunables_ktype = { /********************** cpufreq governor interface *********************/ -#ifdef CONFIG_ENERGY_MODEL -static void rebuild_sd_workfn(struct work_struct *work) -{ - rebuild_sched_domains_energy(); -} - -static DECLARE_WORK(rebuild_sd_work, rebuild_sd_workfn); - -/* - * EAS shouldn't be attempted without sugov, so rebuild the sched_domains - * on governor changes to make sure the scheduler knows about it. - */ -static void sugov_eas_rebuild_sd(void) -{ - /* - * When called from the cpufreq_register_driver() path, the - * cpu_hotplug_lock is already held, so use a work item to - * avoid nested locking in rebuild_sched_domains(). - */ - schedule_work(&rebuild_sd_work); -} -#else -static inline void sugov_eas_rebuild_sd(void) { }; -#endif - struct cpufreq_governor schedutil_gov; static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy) @@ -691,7 +666,11 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy) } sg_policy->thread = thread; - kthread_bind_mask(thread, policy->related_cpus); + if (policy->dvfs_possible_from_any_cpu) + set_cpus_allowed_ptr(thread, policy->related_cpus); + else + kthread_bind_mask(thread, policy->related_cpus); + init_irq_work(&sg_policy->irq_work, sugov_irq_work); mutex_init(&sg_policy->work_lock); @@ -784,7 +763,11 @@ static int sugov_init(struct cpufreq_policy *policy) goto fail; out: - sugov_eas_rebuild_sd(); + /* + * Schedutil is the preferred governor for EAS, so rebuild sched domains + * on governor changes to make sure the scheduler knows about them. + */ + em_rebuild_sched_domains(); mutex_unlock(&global_tunables_lock); return 0; @@ -826,7 +809,7 @@ static void sugov_exit(struct cpufreq_policy *policy) sugov_policy_free(sg_policy); cpufreq_disable_fast_switch(policy); - sugov_eas_rebuild_sd(); + em_rebuild_sched_domains(); } static int sugov_start(struct cpufreq_policy *policy) diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 0bed0fa1acd9..5d9143dd0879 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -9,6 +9,8 @@ #ifdef CONFIG_IRQ_TIME_ACCOUNTING +DEFINE_STATIC_KEY_FALSE(sched_clock_irqtime); + /* * There are no locks covering percpu hardirq/softirq time. * They are only modified in vtime_account, on corresponding CPU @@ -22,16 +24,14 @@ */ DEFINE_PER_CPU(struct irqtime, cpu_irqtime); -static int sched_clock_irqtime; - void enable_sched_clock_irqtime(void) { - sched_clock_irqtime = 1; + static_branch_enable(&sched_clock_irqtime); } void disable_sched_clock_irqtime(void) { - sched_clock_irqtime = 0; + static_branch_disable(&sched_clock_irqtime); } static void irqtime_account_delta(struct irqtime *irqtime, u64 delta, @@ -57,7 +57,7 @@ void irqtime_account_irq(struct task_struct *curr, unsigned int offset) s64 delta; int cpu; - if (!sched_clock_irqtime) + if (!irqtime_enabled()) return; cpu = smp_processor_id(); @@ -90,8 +90,6 @@ static u64 irqtime_tick_accounted(u64 maxtime) #else /* CONFIG_IRQ_TIME_ACCOUNTING */ -#define sched_clock_irqtime (0) - static u64 irqtime_tick_accounted(u64 dummy) { return 0; @@ -478,7 +476,7 @@ void account_process_tick(struct task_struct *p, int user_tick) if (vtime_accounting_enabled_this_cpu()) return; - if (sched_clock_irqtime) { + if (irqtime_enabled()) { irqtime_account_process_tick(p, user_tick, 1); return; } @@ -507,7 +505,7 @@ void account_idle_ticks(unsigned long ticks) { u64 cputime, steal; - if (sched_clock_irqtime) { + if (irqtime_enabled()) { irqtime_account_idle_ticks(ticks); return; } diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index db47f33cb7d2..38e4537790af 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -26,7 +26,7 @@ static unsigned int sysctl_sched_dl_period_max = 1 << 22; /* ~4 seconds */ static unsigned int sysctl_sched_dl_period_min = 100; /* 100 us */ #ifdef CONFIG_SYSCTL -static struct ctl_table sched_dl_sysctls[] = { +static const struct ctl_table sched_dl_sysctls[] = { { .procname = "sched_deadline_period_max_us", .data = &sysctl_sched_dl_period_max, @@ -342,6 +342,29 @@ static void dl_rq_change_utilization(struct rq *rq, struct sched_dl_entity *dl_s __add_rq_bw(new_bw, &rq->dl); } +static __always_inline +void cancel_dl_timer(struct sched_dl_entity *dl_se, struct hrtimer *timer) +{ + /* + * If the timer callback was running (hrtimer_try_to_cancel == -1), + * it will eventually call put_task_struct(). + */ + if (hrtimer_try_to_cancel(timer) == 1 && !dl_server(dl_se)) + put_task_struct(dl_task_of(dl_se)); +} + +static __always_inline +void cancel_replenish_timer(struct sched_dl_entity *dl_se) +{ + cancel_dl_timer(dl_se, &dl_se->dl_timer); +} + +static __always_inline +void cancel_inactive_timer(struct sched_dl_entity *dl_se) +{ + cancel_dl_timer(dl_se, &dl_se->inactive_timer); +} + static void dl_change_utilization(struct task_struct *p, u64 new_bw) { WARN_ON_ONCE(p->dl.flags & SCHED_FLAG_SUGOV); @@ -495,10 +518,7 @@ static void task_contending(struct sched_dl_entity *dl_se, int flags) * will not touch the rq's active utilization, * so we are still safe. */ - if (hrtimer_try_to_cancel(&dl_se->inactive_timer) == 1) { - if (!dl_server(dl_se)) - put_task_struct(dl_task_of(dl_se)); - } + cancel_inactive_timer(dl_se); } else { /* * Since "dl_non_contending" is not set, the @@ -1647,6 +1667,7 @@ void dl_server_start(struct sched_dl_entity *dl_se) if (!dl_se->dl_runtime) return; + dl_se->dl_server_active = 1; enqueue_dl_entity(dl_se, ENQUEUE_WAKEUP); if (!dl_task(dl_se->rq->curr) || dl_entity_preempt(dl_se, &rq->curr->dl)) resched_curr(dl_se->rq); @@ -1661,6 +1682,7 @@ void dl_server_stop(struct sched_dl_entity *dl_se) hrtimer_try_to_cancel(&dl_se->dl_timer); dl_se->dl_defer_armed = 0; dl_se->dl_throttled = 0; + dl_se->dl_server_active = 0; } void dl_server_init(struct sched_dl_entity *dl_se, struct rq *rq, @@ -2113,13 +2135,8 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags) * The replenish timer needs to be canceled. No * problem if it fires concurrently: boosted threads * are ignored in dl_task_timer(). - * - * If the timer callback was running (hrtimer_try_to_cancel == -1), - * it will eventually call put_task_struct(). */ - if (hrtimer_try_to_cancel(&p->dl.dl_timer) == 1 && - !dl_server(&p->dl)) - put_task_struct(p); + cancel_replenish_timer(&p->dl); p->dl.dl_throttled = 0; } } else if (!dl_prio(p->normal_prio)) { @@ -2287,8 +2304,7 @@ static void migrate_task_rq_dl(struct task_struct *p, int new_cpu __maybe_unused * will not touch the rq's active utilization, * so we are still safe. */ - if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) - put_task_struct(p); + cancel_inactive_timer(&p->dl); } sub_rq_bw(&p->dl, &rq->dl); rq_unlock(rq, &rf); @@ -2421,8 +2437,10 @@ again: if (dl_server(dl_se)) { p = dl_se->server_pick_task(dl_se); if (!p) { - dl_se->dl_yielded = 1; - update_curr_dl_se(rq, dl_se, 0); + if (dl_server_active(dl_se)) { + dl_se->dl_yielded = 1; + update_curr_dl_se(rq, dl_se, 0); + } goto again; } rq->dl_server = dl_se; @@ -2502,16 +2520,13 @@ static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu return NULL; next_node = rb_first_cached(&rq->dl.pushable_dl_tasks_root); - -next_node: - if (next_node) { + while (next_node) { p = __node_2_pdl(next_node); if (task_is_pushable(rq, p, cpu)) return p; next_node = rb_next(next_node); - goto next_node; } return NULL; @@ -2960,11 +2975,22 @@ void dl_add_task_root_domain(struct task_struct *p) void dl_clear_root_domain(struct root_domain *rd) { - unsigned long flags; + int i; - raw_spin_lock_irqsave(&rd->dl_bw.lock, flags); + guard(raw_spinlock_irqsave)(&rd->dl_bw.lock); rd->dl_bw.total_bw = 0; - raw_spin_unlock_irqrestore(&rd->dl_bw.lock, flags); + + /* + * dl_server bandwidth is only restored when CPUs are attached to root + * domains (after domains are created or CPUs moved back to the + * default root doamin). + */ + for_each_cpu(i, rd->span) { + struct sched_dl_entity *dl_se = &cpu_rq(i)->fair_server; + + if (dl_server(dl_se) && cpu_active(i)) + rd->dl_bw.total_bw += dl_se->dl_bw; + } } #endif /* CONFIG_SMP */ @@ -3025,8 +3051,7 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p) */ static void switched_to_dl(struct rq *rq, struct task_struct *p) { - if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) - put_task_struct(p); + cancel_inactive_timer(&p->dl); /* * In case a task is setscheduled to SCHED_DEADLINE we need to keep @@ -3449,29 +3474,31 @@ int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, } enum dl_bw_request { - dl_bw_req_check_overflow = 0, + dl_bw_req_deactivate = 0, dl_bw_req_alloc, dl_bw_req_free }; static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) { - unsigned long flags; + unsigned long flags, cap; struct dl_bw *dl_b; bool overflow = 0; + u64 fair_server_bw = 0; rcu_read_lock_sched(); dl_b = dl_bw_of(cpu); raw_spin_lock_irqsave(&dl_b->lock, flags); - if (req == dl_bw_req_free) { + cap = dl_bw_capacity(cpu); + switch (req) { + case dl_bw_req_free: __dl_sub(dl_b, dl_bw, dl_bw_cpus(cpu)); - } else { - unsigned long cap = dl_bw_capacity(cpu); - + break; + case dl_bw_req_alloc: overflow = __dl_overflow(dl_b, cap, 0, dl_bw); - if (req == dl_bw_req_alloc && !overflow) { + if (!overflow) { /* * We reserve space in the destination * root_domain, as we can't fail after this point. @@ -3480,6 +3507,42 @@ static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) */ __dl_add(dl_b, dl_bw, dl_bw_cpus(cpu)); } + break; + case dl_bw_req_deactivate: + /* + * cpu is not off yet, but we need to do the math by + * considering it off already (i.e., what would happen if we + * turn cpu off?). + */ + cap -= arch_scale_cpu_capacity(cpu); + + /* + * cpu is going offline and NORMAL tasks will be moved away + * from it. We can thus discount dl_server bandwidth + * contribution as it won't need to be servicing tasks after + * the cpu is off. + */ + if (cpu_rq(cpu)->fair_server.dl_server) + fair_server_bw = cpu_rq(cpu)->fair_server.dl_bw; + + /* + * Not much to check if no DEADLINE bandwidth is present. + * dl_servers we can discount, as tasks will be moved out the + * offlined CPUs anyway. + */ + if (dl_b->total_bw - fair_server_bw > 0) { + /* + * Leaving at least one CPU for DEADLINE tasks seems a + * wise thing to do. As said above, cpu is not offline + * yet, so account for that. + */ + if (dl_bw_cpus(cpu) - 1) + overflow = __dl_overflow(dl_b, cap, fair_server_bw, 0); + else + overflow = 1; + } + + break; } raw_spin_unlock_irqrestore(&dl_b->lock, flags); @@ -3488,9 +3551,9 @@ static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) return overflow ? -EBUSY : 0; } -int dl_bw_check_overflow(int cpu) +int dl_bw_deactivate(int cpu) { - return dl_bw_manage(dl_bw_req_check_overflow, cpu, 0); + return dl_bw_manage(dl_bw_req_deactivate, cpu, 0); } int dl_bw_alloc(int cpu, u64 dl_bw) diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index a48b2a701ec2..fd7e85220715 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -379,7 +379,7 @@ static ssize_t sched_fair_server_write(struct file *filp, const char __user *ubu return -EINVAL; } - if (rq->cfs.h_nr_running) { + if (rq->cfs.h_nr_queued) { update_rq_clock(rq); dl_server_stop(&rq->fair_server); } @@ -392,7 +392,7 @@ static ssize_t sched_fair_server_write(struct file *filp, const char __user *ubu printk_deferred("Fair server disabled in CPU %d, system may crash due to starvation.\n", cpu_of(rq)); - if (rq->cfs.h_nr_running) + if (rq->cfs.h_nr_queued) dl_server_start(&rq->fair_server); } @@ -843,12 +843,10 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) SPLIT_NS(right_vruntime)); spread = right_vruntime - left_vruntime; SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "spread", SPLIT_NS(spread)); - SEQ_printf(m, " .%-30s: %d\n", "nr_running", cfs_rq->nr_running); - SEQ_printf(m, " .%-30s: %d\n", "h_nr_running", cfs_rq->h_nr_running); - SEQ_printf(m, " .%-30s: %d\n", "idle_nr_running", - cfs_rq->idle_nr_running); - SEQ_printf(m, " .%-30s: %d\n", "idle_h_nr_running", - cfs_rq->idle_h_nr_running); + SEQ_printf(m, " .%-30s: %d\n", "nr_queued", cfs_rq->nr_queued); + SEQ_printf(m, " .%-30s: %d\n", "h_nr_runnable", cfs_rq->h_nr_runnable); + SEQ_printf(m, " .%-30s: %d\n", "h_nr_queued", cfs_rq->h_nr_queued); + SEQ_printf(m, " .%-30s: %d\n", "h_nr_idle", cfs_rq->h_nr_idle); SEQ_printf(m, " .%-30s: %ld\n", "load", cfs_rq->load.weight); #ifdef CONFIG_SMP SEQ_printf(m, " .%-30s: %lu\n", "load_avg", @@ -1294,8 +1292,10 @@ void resched_latency_warn(int cpu, u64 latency) { static DEFINE_RATELIMIT_STATE(latency_check_ratelimit, 60 * 60 * HZ, 1); - WARN(__ratelimit(&latency_check_ratelimit), - "sched: CPU %d need_resched set for > %llu ns (%d ticks) " - "without schedule\n", - cpu, latency, cpu_rq(cpu)->ticks_without_resched); + if (likely(!__ratelimit(&latency_check_ratelimit))) + return; + + pr_err("sched: CPU %d need_resched set for > %llu ns (%d ticks) without schedule\n", + cpu, latency, cpu_rq(cpu)->ticks_without_resched); + dump_stack(); } diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 7fff1d045477..8857c0709bdd 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -206,7 +206,7 @@ struct scx_dump_ctx { */ struct sched_ext_ops { /** - * select_cpu - Pick the target CPU for a task which is being woken up + * @select_cpu: Pick the target CPU for a task which is being woken up * @p: task being woken up * @prev_cpu: the cpu @p was on before sleeping * @wake_flags: SCX_WAKE_* @@ -233,7 +233,7 @@ struct sched_ext_ops { s32 (*select_cpu)(struct task_struct *p, s32 prev_cpu, u64 wake_flags); /** - * enqueue - Enqueue a task on the BPF scheduler + * @enqueue: Enqueue a task on the BPF scheduler * @p: task being enqueued * @enq_flags: %SCX_ENQ_* * @@ -248,7 +248,7 @@ struct sched_ext_ops { void (*enqueue)(struct task_struct *p, u64 enq_flags); /** - * dequeue - Remove a task from the BPF scheduler + * @dequeue: Remove a task from the BPF scheduler * @p: task being dequeued * @deq_flags: %SCX_DEQ_* * @@ -264,7 +264,7 @@ struct sched_ext_ops { void (*dequeue)(struct task_struct *p, u64 deq_flags); /** - * dispatch - Dispatch tasks from the BPF scheduler and/or user DSQs + * @dispatch: Dispatch tasks from the BPF scheduler and/or user DSQs * @cpu: CPU to dispatch tasks for * @prev: previous task being switched out * @@ -287,7 +287,7 @@ struct sched_ext_ops { void (*dispatch)(s32 cpu, struct task_struct *prev); /** - * tick - Periodic tick + * @tick: Periodic tick * @p: task running currently * * This operation is called every 1/HZ seconds on CPUs which are @@ -297,7 +297,7 @@ struct sched_ext_ops { void (*tick)(struct task_struct *p); /** - * runnable - A task is becoming runnable on its associated CPU + * @runnable: A task is becoming runnable on its associated CPU * @p: task becoming runnable * @enq_flags: %SCX_ENQ_* * @@ -324,7 +324,7 @@ struct sched_ext_ops { void (*runnable)(struct task_struct *p, u64 enq_flags); /** - * running - A task is starting to run on its associated CPU + * @running: A task is starting to run on its associated CPU * @p: task starting to run * * See ->runnable() for explanation on the task state notifiers. @@ -332,7 +332,7 @@ struct sched_ext_ops { void (*running)(struct task_struct *p); /** - * stopping - A task is stopping execution + * @stopping: A task is stopping execution * @p: task stopping to run * @runnable: is task @p still runnable? * @@ -343,7 +343,7 @@ struct sched_ext_ops { void (*stopping)(struct task_struct *p, bool runnable); /** - * quiescent - A task is becoming not runnable on its associated CPU + * @quiescent: A task is becoming not runnable on its associated CPU * @p: task becoming not runnable * @deq_flags: %SCX_DEQ_* * @@ -363,7 +363,7 @@ struct sched_ext_ops { void (*quiescent)(struct task_struct *p, u64 deq_flags); /** - * yield - Yield CPU + * @yield: Yield CPU * @from: yielding task * @to: optional yield target task * @@ -378,7 +378,7 @@ struct sched_ext_ops { bool (*yield)(struct task_struct *from, struct task_struct *to); /** - * core_sched_before - Task ordering for core-sched + * @core_sched_before: Task ordering for core-sched * @a: task A * @b: task B * @@ -396,7 +396,7 @@ struct sched_ext_ops { bool (*core_sched_before)(struct task_struct *a, struct task_struct *b); /** - * set_weight - Set task weight + * @set_weight: Set task weight * @p: task to set weight for * @weight: new weight [1..10000] * @@ -405,7 +405,7 @@ struct sched_ext_ops { void (*set_weight)(struct task_struct *p, u32 weight); /** - * set_cpumask - Set CPU affinity + * @set_cpumask: Set CPU affinity * @p: task to set CPU affinity for * @cpumask: cpumask of cpus that @p can run on * @@ -415,7 +415,7 @@ struct sched_ext_ops { const struct cpumask *cpumask); /** - * update_idle - Update the idle state of a CPU + * @update_idle: Update the idle state of a CPU * @cpu: CPU to udpate the idle state for * @idle: whether entering or exiting the idle state * @@ -436,7 +436,7 @@ struct sched_ext_ops { void (*update_idle)(s32 cpu, bool idle); /** - * cpu_acquire - A CPU is becoming available to the BPF scheduler + * @cpu_acquire: A CPU is becoming available to the BPF scheduler * @cpu: The CPU being acquired by the BPF scheduler. * @args: Acquire arguments, see the struct definition. * @@ -446,7 +446,7 @@ struct sched_ext_ops { void (*cpu_acquire)(s32 cpu, struct scx_cpu_acquire_args *args); /** - * cpu_release - A CPU is taken away from the BPF scheduler + * @cpu_release: A CPU is taken away from the BPF scheduler * @cpu: The CPU being released by the BPF scheduler. * @args: Release arguments, see the struct definition. * @@ -458,7 +458,7 @@ struct sched_ext_ops { void (*cpu_release)(s32 cpu, struct scx_cpu_release_args *args); /** - * init_task - Initialize a task to run in a BPF scheduler + * @init_task: Initialize a task to run in a BPF scheduler * @p: task to initialize for BPF scheduling * @args: init arguments, see the struct definition * @@ -473,8 +473,9 @@ struct sched_ext_ops { s32 (*init_task)(struct task_struct *p, struct scx_init_task_args *args); /** - * exit_task - Exit a previously-running task from the system + * @exit_task: Exit a previously-running task from the system * @p: task to exit + * @args: exit arguments, see the struct definition * * @p is exiting or the BPF scheduler is being unloaded. Perform any * necessary cleanup for @p. @@ -482,7 +483,7 @@ struct sched_ext_ops { void (*exit_task)(struct task_struct *p, struct scx_exit_task_args *args); /** - * enable - Enable BPF scheduling for a task + * @enable: Enable BPF scheduling for a task * @p: task to enable BPF scheduling for * * Enable @p for BPF scheduling. enable() is called on @p any time it @@ -491,7 +492,7 @@ struct sched_ext_ops { void (*enable)(struct task_struct *p); /** - * disable - Disable BPF scheduling for a task + * @disable: Disable BPF scheduling for a task * @p: task to disable BPF scheduling for * * @p is exiting, leaving SCX or the BPF scheduler is being unloaded. @@ -501,7 +502,7 @@ struct sched_ext_ops { void (*disable)(struct task_struct *p); /** - * dump - Dump BPF scheduler state on error + * @dump: Dump BPF scheduler state on error * @ctx: debug dump context * * Use scx_bpf_dump() to generate BPF scheduler specific debug dump. @@ -509,7 +510,7 @@ struct sched_ext_ops { void (*dump)(struct scx_dump_ctx *ctx); /** - * dump_cpu - Dump BPF scheduler state for a CPU on error + * @dump_cpu: Dump BPF scheduler state for a CPU on error * @ctx: debug dump context * @cpu: CPU to generate debug dump for * @idle: @cpu is currently idle without any runnable tasks @@ -521,7 +522,7 @@ struct sched_ext_ops { void (*dump_cpu)(struct scx_dump_ctx *ctx, s32 cpu, bool idle); /** - * dump_task - Dump BPF scheduler state for a runnable task on error + * @dump_task: Dump BPF scheduler state for a runnable task on error * @ctx: debug dump context * @p: runnable task to generate debug dump for * @@ -532,7 +533,7 @@ struct sched_ext_ops { #ifdef CONFIG_EXT_GROUP_SCHED /** - * cgroup_init - Initialize a cgroup + * @cgroup_init: Initialize a cgroup * @cgrp: cgroup being initialized * @args: init arguments, see the struct definition * @@ -547,7 +548,7 @@ struct sched_ext_ops { struct scx_cgroup_init_args *args); /** - * cgroup_exit - Exit a cgroup + * @cgroup_exit: Exit a cgroup * @cgrp: cgroup being exited * * Either the BPF scheduler is being unloaded or @cgrp destroyed, exit @@ -556,7 +557,7 @@ struct sched_ext_ops { void (*cgroup_exit)(struct cgroup *cgrp); /** - * cgroup_prep_move - Prepare a task to be moved to a different cgroup + * @cgroup_prep_move: Prepare a task to be moved to a different cgroup * @p: task being moved * @from: cgroup @p is being moved from * @to: cgroup @p is being moved to @@ -571,7 +572,7 @@ struct sched_ext_ops { struct cgroup *from, struct cgroup *to); /** - * cgroup_move - Commit cgroup move + * @cgroup_move: Commit cgroup move * @p: task being moved * @from: cgroup @p is being moved from * @to: cgroup @p is being moved to @@ -582,7 +583,7 @@ struct sched_ext_ops { struct cgroup *from, struct cgroup *to); /** - * cgroup_cancel_move - Cancel cgroup move + * @cgroup_cancel_move: Cancel cgroup move * @p: task whose cgroup move is being canceled * @from: cgroup @p was being moved from * @to: cgroup @p was being moved to @@ -594,7 +595,7 @@ struct sched_ext_ops { struct cgroup *from, struct cgroup *to); /** - * cgroup_set_weight - A cgroup's weight is being changed + * @cgroup_set_weight: A cgroup's weight is being changed * @cgrp: cgroup whose weight is being updated * @weight: new weight [1..10000] * @@ -608,7 +609,7 @@ struct sched_ext_ops { */ /** - * cpu_online - A CPU became online + * @cpu_online: A CPU became online * @cpu: CPU which just came up * * @cpu just came online. @cpu will not call ops.enqueue() or @@ -617,7 +618,7 @@ struct sched_ext_ops { void (*cpu_online)(s32 cpu); /** - * cpu_offline - A CPU is going offline + * @cpu_offline: A CPU is going offline * @cpu: CPU which is going offline * * @cpu is going offline. @cpu will not call ops.enqueue() or @@ -630,12 +631,12 @@ struct sched_ext_ops { */ /** - * init - Initialize the BPF scheduler + * @init: Initialize the BPF scheduler */ s32 (*init)(void); /** - * exit - Clean up after the BPF scheduler + * @exit: Clean up after the BPF scheduler * @info: Exit info * * ops.exit() is also called on ops.init() failure, which is a bit @@ -645,17 +646,17 @@ struct sched_ext_ops { void (*exit)(struct scx_exit_info *info); /** - * dispatch_max_batch - Max nr of tasks that dispatch() can dispatch + * @dispatch_max_batch: Max nr of tasks that dispatch() can dispatch */ u32 dispatch_max_batch; /** - * flags - %SCX_OPS_* flags + * @flags: %SCX_OPS_* flags */ u64 flags; /** - * timeout_ms - The maximum amount of time, in milliseconds, that a + * @timeout_ms: The maximum amount of time, in milliseconds, that a * runnable task should be able to wait before being scheduled. The * maximum timeout may not exceed the default timeout of 30 seconds. * @@ -664,13 +665,13 @@ struct sched_ext_ops { u32 timeout_ms; /** - * exit_dump_len - scx_exit_info.dump buffer length. If 0, the default + * @exit_dump_len: scx_exit_info.dump buffer length. If 0, the default * value of 32768 is used. */ u32 exit_dump_len; /** - * hotplug_seq - A sequence number that may be set by the scheduler to + * @hotplug_seq: A sequence number that may be set by the scheduler to * detect when a hotplug event has occurred during the loading process. * If 0, no detection occurs. Otherwise, the scheduler will fail to * load if the sequence number does not match @scx_hotplug_seq on the @@ -679,7 +680,7 @@ struct sched_ext_ops { u64 hotplug_seq; /** - * name - BPF scheduler's name + * @name: BPF scheduler's name * * Must be a non-zero valid BPF object name including only isalnum(), * '_' and '.' chars. Shows up in kernel.sched_ext_ops sysctl while the @@ -960,7 +961,7 @@ static DEFINE_PER_CPU(struct task_struct *, direct_dispatch_task); static struct scx_dispatch_q **global_dsqs; static const struct rhashtable_params dsq_hash_params = { - .key_len = 8, + .key_len = sizeof_field(struct scx_dispatch_q, id), .key_offset = offsetof(struct scx_dispatch_q, id), .head_offset = offsetof(struct scx_dispatch_q, hash_node), }; @@ -1408,7 +1409,6 @@ static struct task_struct *scx_task_iter_next(struct scx_task_iter *iter) /** * scx_task_iter_next_locked - Next non-idle task with its rq locked * @iter: iterator to walk - * @include_dead: Whether we should include dead tasks in the iteration * * Visit the non-idle task with its rq lock held. Allows callers to specify * whether they would like to filter out dead tasks. See scx_task_iter_start() @@ -2747,6 +2747,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) { struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); bool prev_on_scx = prev->sched_class == &ext_sched_class; + bool prev_on_rq = prev->scx.flags & SCX_TASK_QUEUED; int nr_loops = SCX_DSP_MAX_LOOPS; lockdep_assert_rq_held(rq); @@ -2779,8 +2780,7 @@ static int balance_one(struct rq *rq, struct task_struct *prev) * See scx_ops_disable_workfn() for the explanation on the * bypassing test. */ - if ((prev->scx.flags & SCX_TASK_QUEUED) && - prev->scx.slice && !scx_rq_bypassing(rq)) { + if (prev_on_rq && prev->scx.slice && !scx_rq_bypassing(rq)) { rq->scx.flags |= SCX_RQ_BAL_KEEP; goto has_tasks; } @@ -2813,6 +2813,10 @@ static int balance_one(struct rq *rq, struct task_struct *prev) flush_dispatch_buf(rq); + if (prev_on_rq && prev->scx.slice) { + rq->scx.flags |= SCX_RQ_BAL_KEEP; + goto has_tasks; + } if (rq->scx.local_dsq.nr) goto has_tasks; if (consume_global_dsq(rq)) @@ -2838,8 +2842,7 @@ no_tasks: * Didn't find another task to run. Keep running @prev unless * %SCX_OPS_ENQ_LAST is in effect. */ - if ((prev->scx.flags & SCX_TASK_QUEUED) && - (!static_branch_unlikely(&scx_ops_enq_last) || + if (prev_on_rq && (!static_branch_unlikely(&scx_ops_enq_last) || scx_rq_bypassing(rq))) { rq->scx.flags |= SCX_RQ_BAL_KEEP; goto has_tasks; @@ -3034,7 +3037,7 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, */ if (p->scx.slice && !scx_rq_bypassing(rq)) { dispatch_enqueue(&rq->scx.local_dsq, p, SCX_ENQ_HEAD); - return; + goto switch_class; } /* @@ -3051,6 +3054,7 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, } } +switch_class: if (next && next->sched_class != &ext_sched_class) switch_class(rq, next); } @@ -3132,6 +3136,7 @@ static struct task_struct *pick_task_scx(struct rq *rq) * scx_prio_less - Task ordering for core-sched * @a: task A * @b: task B + * @in_fi: in forced idle state * * Core-sched is implemented as an additional scheduling layer on top of the * usual sched_class'es and needs to find out the expected task ordering. For @@ -3180,6 +3185,10 @@ static bool test_and_clear_cpu_idle(int cpu) * scx_pick_idle_cpu() can get caught in an infinite loop as * @cpu is never cleared from idle_masks.smt. Ensure that @cpu * is eventually cleared. + * + * NOTE: Use cpumask_intersects() and cpumask_test_cpu() to + * reduce memory writes, which may help alleviate cache + * coherence pressure. */ if (cpumask_intersects(smt, idle_masks.smt)) cpumask_andnot(idle_masks.smt, idle_masks.smt, smt); @@ -3216,6 +3225,74 @@ found: } /* + * Return the amount of CPUs in the same LLC domain of @cpu (or zero if the LLC + * domain is not defined). + */ +static unsigned int llc_weight(s32 cpu) +{ + struct sched_domain *sd; + + sd = rcu_dereference(per_cpu(sd_llc, cpu)); + if (!sd) + return 0; + + return sd->span_weight; +} + +/* + * Return the cpumask representing the LLC domain of @cpu (or NULL if the LLC + * domain is not defined). + */ +static struct cpumask *llc_span(s32 cpu) +{ + struct sched_domain *sd; + + sd = rcu_dereference(per_cpu(sd_llc, cpu)); + if (!sd) + return 0; + + return sched_domain_span(sd); +} + +/* + * Return the amount of CPUs in the same NUMA domain of @cpu (or zero if the + * NUMA domain is not defined). + */ +static unsigned int numa_weight(s32 cpu) +{ + struct sched_domain *sd; + struct sched_group *sg; + + sd = rcu_dereference(per_cpu(sd_numa, cpu)); + if (!sd) + return 0; + sg = sd->groups; + if (!sg) + return 0; + + return sg->group_weight; +} + +/* + * Return the cpumask representing the NUMA domain of @cpu (or NULL if the NUMA + * domain is not defined). + */ +static struct cpumask *numa_span(s32 cpu) +{ + struct sched_domain *sd; + struct sched_group *sg; + + sd = rcu_dereference(per_cpu(sd_numa, cpu)); + if (!sd) + return NULL; + sg = sd->groups; + if (!sg) + return NULL; + + return sched_group_span(sg); +} + +/* * Return true if the LLC domains do not perfectly overlap with the NUMA * domains, false otherwise. */ @@ -3246,19 +3323,10 @@ static bool llc_numa_mismatch(void) * overlapping, which is incorrect (as NUMA 1 has two distinct LLC * domains). */ - for_each_online_cpu(cpu) { - const struct cpumask *numa_cpus; - struct sched_domain *sd; - - sd = rcu_dereference(per_cpu(sd_llc, cpu)); - if (!sd) + for_each_online_cpu(cpu) + if (llc_weight(cpu) != numa_weight(cpu)) return true; - numa_cpus = cpumask_of_node(cpu_to_node(cpu)); - if (sd->span_weight != cpumask_weight(numa_cpus)) - return true; - } - return false; } @@ -3276,8 +3344,7 @@ static bool llc_numa_mismatch(void) static void update_selcpu_topology(void) { bool enable_llc = false, enable_numa = false; - struct sched_domain *sd; - const struct cpumask *cpus; + unsigned int nr_cpus; s32 cpu = cpumask_first(cpu_online_mask); /* @@ -3291,10 +3358,12 @@ static void update_selcpu_topology(void) * CPUs. */ rcu_read_lock(); - sd = rcu_dereference(per_cpu(sd_llc, cpu)); - if (sd) { - if (sd->span_weight < num_online_cpus()) + nr_cpus = llc_weight(cpu); + if (nr_cpus > 0) { + if (nr_cpus < num_online_cpus()) enable_llc = true; + pr_debug("sched_ext: LLC=%*pb weight=%u\n", + cpumask_pr_args(llc_span(cpu)), llc_weight(cpu)); } /* @@ -3306,15 +3375,19 @@ static void update_selcpu_topology(void) * enabling both NUMA and LLC optimizations is unnecessary, as checking * for an idle CPU in the same domain twice is redundant. */ - cpus = cpumask_of_node(cpu_to_node(cpu)); - if ((cpumask_weight(cpus) < num_online_cpus()) && llc_numa_mismatch()) - enable_numa = true; + nr_cpus = numa_weight(cpu); + if (nr_cpus > 0) { + if (nr_cpus < num_online_cpus() && llc_numa_mismatch()) + enable_numa = true; + pr_debug("sched_ext: NUMA=%*pb weight=%u\n", + cpumask_pr_args(numa_span(cpu)), numa_weight(cpu)); + } rcu_read_unlock(); pr_debug("sched_ext: LLC idle selection %s\n", - enable_llc ? "enabled" : "disabled"); + str_enabled_disabled(enable_llc)); pr_debug("sched_ext: NUMA idle selection %s\n", - enable_numa ? "enabled" : "disabled"); + str_enabled_disabled(enable_numa)); if (enable_llc) static_branch_enable_cpuslocked(&scx_selcpu_topo_llc); @@ -3344,6 +3417,8 @@ static void update_selcpu_topology(void) * 4. Pick a CPU within the same NUMA node, if enabled: * - choose a CPU from the same NUMA node to reduce memory access latency. * + * 5. Pick any idle CPU usable by the task. + * * Step 3 and 4 are performed only if the system has, respectively, multiple * LLC domains / multiple NUMA nodes (see scx_selcpu_topo_llc and * scx_selcpu_topo_numa). @@ -3360,7 +3435,6 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, *found = false; - /* * This is necessary to protect llc_cpus. */ @@ -3379,15 +3453,10 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, */ if (p->nr_cpus_allowed >= num_possible_cpus()) { if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) - numa_cpus = cpumask_of_node(cpu_to_node(prev_cpu)); + numa_cpus = numa_span(prev_cpu); - if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) { - struct sched_domain *sd; - - sd = rcu_dereference(per_cpu(sd_llc, prev_cpu)); - if (sd) - llc_cpus = sched_domain_span(sd); - } + if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) + llc_cpus = llc_span(prev_cpu); } /* @@ -3586,20 +3655,9 @@ static void reset_idle_masks(void) cpumask_copy(idle_masks.smt, cpu_online_mask); } -void __scx_update_idle(struct rq *rq, bool idle) +static void update_builtin_idle(int cpu, bool idle) { - int cpu = cpu_of(rq); - - if (SCX_HAS_OP(update_idle) && !scx_rq_bypassing(rq)) { - SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); - if (!static_branch_unlikely(&scx_builtin_idle_enabled)) - return; - } - - if (idle) - cpumask_set_cpu(cpu, idle_masks.cpu); - else - cpumask_clear_cpu(cpu, idle_masks.cpu); + assign_cpu(cpu, idle_masks.cpu, idle); #ifdef CONFIG_SCHED_SMT if (sched_smt_active()) { @@ -3610,10 +3668,8 @@ void __scx_update_idle(struct rq *rq, bool idle) * idle_masks.smt handling is racy but that's fine as * it's only for optimization and self-correcting. */ - for_each_cpu(cpu, smt) { - if (!cpumask_test_cpu(cpu, idle_masks.cpu)) - return; - } + if (!cpumask_subset(smt, idle_masks.cpu)) + return; cpumask_or(idle_masks.smt, idle_masks.smt, smt); } else { cpumask_andnot(idle_masks.smt, idle_masks.smt, smt); @@ -3622,6 +3678,57 @@ void __scx_update_idle(struct rq *rq, bool idle) #endif } +/* + * Update the idle state of a CPU to @idle. + * + * If @do_notify is true, ops.update_idle() is invoked to notify the scx + * scheduler of an actual idle state transition (idle to busy or vice + * versa). If @do_notify is false, only the idle state in the idle masks is + * refreshed without invoking ops.update_idle(). + * + * This distinction is necessary, because an idle CPU can be "reserved" and + * awakened via scx_bpf_pick_idle_cpu() + scx_bpf_kick_cpu(), marking it as + * busy even if no tasks are dispatched. In this case, the CPU may return + * to idle without a true state transition. Refreshing the idle masks + * without invoking ops.update_idle() ensures accurate idle state tracking + * while avoiding unnecessary updates and maintaining balanced state + * transitions. + */ +void __scx_update_idle(struct rq *rq, bool idle, bool do_notify) +{ + int cpu = cpu_of(rq); + + lockdep_assert_rq_held(rq); + + /* + * Trigger ops.update_idle() only when transitioning from a task to + * the idle thread and vice versa. + * + * Idle transitions are indicated by do_notify being set to true, + * managed by put_prev_task_idle()/set_next_task_idle(). + */ + if (SCX_HAS_OP(update_idle) && do_notify && !scx_rq_bypassing(rq)) + SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); + + /* + * Update the idle masks: + * - for real idle transitions (do_notify == true) + * - for idle-to-idle transitions (indicated by the previous task + * being the idle thread, managed by pick_task_idle()) + * + * Skip updating idle masks if the previous task is not the idle + * thread, since set_next_task_idle() has already handled it when + * transitioning from a task to the idle thread (calling this + * function with do_notify == true). + * + * In this way we can avoid updating the idle masks twice, + * unnecessarily. + */ + if (static_branch_likely(&scx_builtin_idle_enabled)) + if (do_notify || is_idle_task(rq->curr)) + update_builtin_idle(cpu, idle); +} + static void handle_hotplug(struct rq *rq, bool online) { int cpu = cpu_of(rq); @@ -4641,6 +4748,7 @@ bool task_should_scx(int policy) /** * scx_softlockup - sched_ext softlockup handler + * @dur_s: number of seconds of CPU stuck due to soft lockup * * On some multi-socket setups (e.g. 2x Intel 8480c), the BPF scheduler can * live-lock the system by making many CPUs target the same DSQ to the point @@ -4684,6 +4792,7 @@ static void scx_clear_softlockup(void) /** * scx_ops_bypass - [Un]bypass scx_ops and guarantee forward progress + * @bypass: true for bypass, false for unbypass * * Bypassing guarantees that all runnable tasks make forward progress without * trusting the BPF scheduler. We can't grab any mutexes or rwsems as they might @@ -4744,10 +4853,9 @@ static void scx_ops_bypass(bool bypass) */ for_each_possible_cpu(cpu) { struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; struct task_struct *p, *n; - rq_lock(rq, &rf); + raw_spin_rq_lock(rq); if (bypass) { WARN_ON_ONCE(rq->scx.flags & SCX_RQ_BYPASSING); @@ -4763,7 +4871,7 @@ static void scx_ops_bypass(bool bypass) * sees scx_rq_bypassing() before moving tasks to SCX. */ if (!scx_enabled()) { - rq_unlock_irqrestore(rq, &rf); + raw_spin_rq_unlock(rq); continue; } @@ -4783,10 +4891,11 @@ static void scx_ops_bypass(bool bypass) sched_enq_and_set_task(&ctx); } - rq_unlock(rq, &rf); - /* resched to restore ticks and idle state */ - resched_cpu(cpu); + if (cpu_online(cpu) || cpu == smp_processor_id()) + resched_curr(rq); + + raw_spin_rq_unlock(rq); } atomic_dec(&scx_ops_breather_depth); @@ -4852,7 +4961,7 @@ static void scx_ops_disable_workfn(struct kthread_work *work) struct task_struct *p; struct rhashtable_iter rht_iter; struct scx_dispatch_q *dsq; - int i, kind; + int i, kind, cpu; kind = atomic_read(&scx_exit_kind); while (true) { @@ -4935,6 +5044,15 @@ static void scx_ops_disable_workfn(struct kthread_work *work) scx_task_iter_stop(&sti); percpu_up_write(&scx_fork_rwsem); + /* + * Invalidate all the rq clocks to prevent getting outdated + * rq clocks from a previous scx scheduler. + */ + for_each_possible_cpu(cpu) { + struct rq *rq = cpu_rq(cpu); + scx_rq_clock_invalidate(rq); + } + /* no task is on scx, turn off all the switches and flush in-progress calls */ static_branch_disable(&__scx_ops_enabled); for (i = SCX_OPI_BEGIN; i < SCX_OPI_END; i++) @@ -5159,9 +5277,9 @@ static void scx_dump_task(struct seq_buf *s, struct scx_dump_ctx *dctx, scx_get_task_state(p), p->scx.flags & ~SCX_TASK_STATE_MASK, p->scx.dsq_flags, ops_state & SCX_OPSS_STATE_MASK, ops_state >> SCX_OPSS_QSEQ_SHIFT); - dump_line(s, " sticky/holding_cpu=%d/%d dsq_id=%s dsq_vtime=%llu", + dump_line(s, " sticky/holding_cpu=%d/%d dsq_id=%s dsq_vtime=%llu slice=%llu", p->scx.sticky_cpu, p->scx.holding_cpu, dsq_id_buf, - p->scx.dsq_vtime); + p->scx.dsq_vtime, p->scx.slice); dump_line(s, " cpus=%*pb", cpumask_pr_args(p->cpus_ptr)); if (SCX_HAS_OP(dump_task)) { @@ -5352,7 +5470,7 @@ static struct kthread_worker *scx_create_rt_helper(const char *name) { struct kthread_worker *helper; - helper = kthread_create_worker(0, name); + helper = kthread_run_worker(0, name); if (helper) sched_set_fifo(helper->task); return helper; @@ -6236,6 +6354,15 @@ void __init init_sched_ext_class(void) __bpf_kfunc_start_defs(); +static bool check_builtin_idle_enabled(void) +{ + if (static_branch_likely(&scx_builtin_idle_enabled)) + return true; + + scx_ops_error("built-in idle tracking is disabled"); + return false; +} + /** * scx_bpf_select_cpu_dfl - The default implementation of ops.select_cpu() * @p: task_struct to select a CPU for @@ -6253,10 +6380,8 @@ __bpf_kfunc_start_defs(); __bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, bool *is_idle) { - if (!static_branch_likely(&scx_builtin_idle_enabled)) { - scx_ops_error("built-in idle tracking is disabled"); + if (!check_builtin_idle_enabled()) goto prev_cpu; - } if (!scx_kf_allowed(SCX_KF_SELECT_CPU)) goto prev_cpu; @@ -6340,9 +6465,7 @@ __bpf_kfunc_start_defs(); * ops.select_cpu(), and ops.dispatch(). * * When called from ops.select_cpu() or ops.enqueue(), it's for direct dispatch - * and @p must match the task being enqueued. Also, %SCX_DSQ_LOCAL_ON can't be - * used to target the local DSQ of a CPU other than the enqueueing one. Use - * ops.select_cpu() to be on the target CPU in the first place. + * and @p must match the task being enqueued. * * When called from ops.select_cpu(), @enq_flags and @dsp_id are stored, and @p * will be directly inserted into the corresponding dispatch queue after @@ -7013,7 +7136,7 @@ __bpf_kfunc int bpf_iter_scx_dsq_new(struct bpf_iter_scx_dsq *it, u64 dsq_id, return -ENOENT; INIT_LIST_HEAD(&kit->cursor.node); - kit->cursor.flags |= SCX_DSQ_LNODE_ITER_CURSOR | flags; + kit->cursor.flags = SCX_DSQ_LNODE_ITER_CURSOR | flags; kit->cursor.priv = READ_ONCE(kit->dsq->seq); return 0; @@ -7181,7 +7304,7 @@ __bpf_kfunc void scx_bpf_error_bstr(char *fmt, unsigned long long *data, } /** - * scx_bpf_dump - Generate extra debug dump specific to the BPF scheduler + * scx_bpf_dump_bstr - Generate extra debug dump specific to the BPF scheduler * @fmt: format string * @data: format string parameters packaged using ___bpf_fill() macro * @data__sz: @data len, must end in '__sz' for the verifier @@ -7273,7 +7396,6 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu) * scx_bpf_cpuperf_set - Set the relative performance target of a CPU * @cpu: CPU of interest * @perf: target performance level [0, %SCX_CPUPERF_ONE] - * @flags: %SCX_CPUPERF_* flags * * Set the target performance level of @cpu to @perf. @perf is in linear * relative scale between 0 and %SCX_CPUPERF_ONE. This determines how the @@ -7350,10 +7472,8 @@ __bpf_kfunc void scx_bpf_put_cpumask(const struct cpumask *cpumask) */ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void) { - if (!static_branch_likely(&scx_builtin_idle_enabled)) { - scx_ops_error("built-in idle tracking is disabled"); + if (!check_builtin_idle_enabled()) return cpu_none_mask; - } #ifdef CONFIG_SMP return idle_masks.cpu; @@ -7371,10 +7491,8 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void) */ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void) { - if (!static_branch_likely(&scx_builtin_idle_enabled)) { - scx_ops_error("built-in idle tracking is disabled"); + if (!check_builtin_idle_enabled()) return cpu_none_mask; - } #ifdef CONFIG_SMP if (sched_smt_active()) @@ -7389,6 +7507,7 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void) /** * scx_bpf_put_idle_cpumask - Release a previously acquired referenced kptr to * either the percpu, or SMT idle-tracking cpumask. + * @idle_mask: &cpumask to use */ __bpf_kfunc void scx_bpf_put_idle_cpumask(const struct cpumask *idle_mask) { @@ -7412,10 +7531,8 @@ __bpf_kfunc void scx_bpf_put_idle_cpumask(const struct cpumask *idle_mask) */ __bpf_kfunc bool scx_bpf_test_and_clear_cpu_idle(s32 cpu) { - if (!static_branch_likely(&scx_builtin_idle_enabled)) { - scx_ops_error("built-in idle tracking is disabled"); + if (!check_builtin_idle_enabled()) return false; - } if (ops_cpu_valid(cpu, NULL)) return test_and_clear_cpu_idle(cpu); @@ -7445,10 +7562,8 @@ __bpf_kfunc bool scx_bpf_test_and_clear_cpu_idle(s32 cpu) __bpf_kfunc s32 scx_bpf_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags) { - if (!static_branch_likely(&scx_builtin_idle_enabled)) { - scx_ops_error("built-in idle tracking is disabled"); + if (!check_builtin_idle_enabled()) return -EBUSY; - } return scx_pick_idle_cpu(cpus_allowed, flags); } @@ -7543,6 +7658,68 @@ out: } #endif +/** + * scx_bpf_now - Returns a high-performance monotonically non-decreasing + * clock for the current CPU. The clock returned is in nanoseconds. + * + * It provides the following properties: + * + * 1) High performance: Many BPF schedulers call bpf_ktime_get_ns() frequently + * to account for execution time and track tasks' runtime properties. + * Unfortunately, in some hardware platforms, bpf_ktime_get_ns() -- which + * eventually reads a hardware timestamp counter -- is neither performant nor + * scalable. scx_bpf_now() aims to provide a high-performance clock by + * using the rq clock in the scheduler core whenever possible. + * + * 2) High enough resolution for the BPF scheduler use cases: In most BPF + * scheduler use cases, the required clock resolution is lower than the most + * accurate hardware clock (e.g., rdtsc in x86). scx_bpf_now() basically + * uses the rq clock in the scheduler core whenever it is valid. It considers + * that the rq clock is valid from the time the rq clock is updated + * (update_rq_clock) until the rq is unlocked (rq_unpin_lock). + * + * 3) Monotonically non-decreasing clock for the same CPU: scx_bpf_now() + * guarantees the clock never goes backward when comparing them in the same + * CPU. On the other hand, when comparing clocks in different CPUs, there + * is no such guarantee -- the clock can go backward. It provides a + * monotonically *non-decreasing* clock so that it would provide the same + * clock values in two different scx_bpf_now() calls in the same CPU + * during the same period of when the rq clock is valid. + */ +__bpf_kfunc u64 scx_bpf_now(void) +{ + struct rq *rq; + u64 clock; + + preempt_disable(); + + rq = this_rq(); + if (smp_load_acquire(&rq->scx.flags) & SCX_RQ_CLK_VALID) { + /* + * If the rq clock is valid, use the cached rq clock. + * + * Note that scx_bpf_now() is re-entrant between a process + * context and an interrupt context (e.g., timer interrupt). + * However, we don't need to consider the race between them + * because such race is not observable from a caller. + */ + clock = READ_ONCE(rq->scx.clock); + } else { + /* + * Otherwise, return a fresh rq clock. + * + * The rq clock is updated outside of the rq lock. + * In this case, keep the updated rq clock invalid so the next + * kfunc call outside the rq lock gets a fresh rq clock. + */ + clock = sched_clock_cpu(cpu_of(rq)); + } + + preempt_enable(); + + return clock; +} + __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_any) @@ -7574,6 +7751,7 @@ BTF_ID_FLAGS(func, scx_bpf_cpu_rq) #ifdef CONFIG_CGROUP_SCHED BTF_ID_FLAGS(func, scx_bpf_task_cgroup, KF_RCU | KF_ACQUIRE) #endif +BTF_ID_FLAGS(func, scx_bpf_now) BTF_KFUNCS_END(scx_kfunc_ids_any) static const struct btf_kfunc_id_set scx_kfunc_set_any = { diff --git a/kernel/sched/ext.h b/kernel/sched/ext.h index b1675bb59fc4..4d022d17ac7d 100644 --- a/kernel/sched/ext.h +++ b/kernel/sched/ext.h @@ -57,15 +57,15 @@ static inline void init_sched_ext_class(void) {} #endif /* CONFIG_SCHED_CLASS_EXT */ #if defined(CONFIG_SCHED_CLASS_EXT) && defined(CONFIG_SMP) -void __scx_update_idle(struct rq *rq, bool idle); +void __scx_update_idle(struct rq *rq, bool idle, bool do_notify); -static inline void scx_update_idle(struct rq *rq, bool idle) +static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) { if (scx_enabled()) - __scx_update_idle(rq, idle); + __scx_update_idle(rq, idle, do_notify); } #else -static inline void scx_update_idle(struct rq *rq, bool idle) {} +static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) {} #endif #ifdef CONFIG_CGROUP_SCHED diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index aa0238ee4857..ce2e94ccad0c 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -37,6 +37,7 @@ #include <linux/sched/cputime.h> #include <linux/sched/isolation.h> #include <linux/sched/nohz.h> +#include <linux/sched/prio.h> #include <linux/cpuidle.h> #include <linux/interrupt.h> @@ -51,6 +52,8 @@ #include <asm/switch_to.h> +#include <uapi/linux/sched/types.h> + #include "sched.h" #include "stats.h" #include "autogroup.h" @@ -130,7 +133,7 @@ static unsigned int sysctl_numa_balancing_promote_rate_limit = 65536; #endif #ifdef CONFIG_SYSCTL -static struct ctl_table sched_fair_sysctls[] = { +static const struct ctl_table sched_fair_sysctls[] = { #ifdef CONFIG_CFS_BANDWIDTH { .procname = "sched_cfs_bandwidth_slice_us", @@ -523,7 +526,7 @@ void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec); * Scheduling class tree data structure manipulation methods: */ -static inline u64 max_vruntime(u64 max_vruntime, u64 vruntime) +static inline __maybe_unused u64 max_vruntime(u64 max_vruntime, u64 vruntime) { s64 delta = (s64)(vruntime - max_vruntime); if (delta > 0) @@ -532,7 +535,7 @@ static inline u64 max_vruntime(u64 max_vruntime, u64 vruntime) return max_vruntime; } -static inline u64 min_vruntime(u64 min_vruntime, u64 vruntime) +static inline __maybe_unused u64 min_vruntime(u64 min_vruntime, u64 vruntime) { s64 delta = (s64)(vruntime - min_vruntime); if (delta < 0) @@ -689,21 +692,16 @@ u64 avg_vruntime(struct cfs_rq *cfs_rq) * * XXX could add max_slice to the augmented data to track this. */ -static s64 entity_lag(u64 avruntime, struct sched_entity *se) +static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) { s64 vlag, limit; - vlag = avruntime - se->vruntime; - limit = calc_delta_fair(max_t(u64, 2*se->slice, TICK_NSEC), se); - - return clamp(vlag, -limit, limit); -} - -static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ SCHED_WARN_ON(!se->on_rq); - se->vlag = entity_lag(avg_vruntime(cfs_rq), se); + vlag = avg_vruntime(cfs_rq) - se->vruntime; + limit = calc_delta_fair(max_t(u64, 2*se->slice, TICK_NSEC), se); + + se->vlag = clamp(vlag, -limit, limit); } /* @@ -915,7 +913,7 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) * We can safely skip eligibility check if there is only one entity * in this cfs_rq, saving some cycles. */ - if (cfs_rq->nr_running == 1) + if (cfs_rq->nr_queued == 1) return curr && curr->on_rq ? curr : se; if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr))) @@ -1159,8 +1157,6 @@ static inline void update_curr_task(struct task_struct *p, s64 delta_exec) trace_sched_stat_runtime(p, delta_exec); account_group_exec_runtime(p, delta_exec); cgroup_account_cputime(p, delta_exec); - if (p->dl_server) - dl_server_update(p->dl_server, delta_exec); } static inline bool did_preempt_short(struct cfs_rq *cfs_rq, struct sched_entity *curr) @@ -1237,17 +1233,22 @@ static void update_curr(struct cfs_rq *cfs_rq) update_curr_task(p, delta_exec); /* - * Any fair task that runs outside of fair_server should - * account against fair_server such that it can account for - * this time and possibly avoid running this period. + * If the fair_server is active, we need to account for the + * fair_server time whether or not the task is running on + * behalf of fair_server or not: + * - If the task is running on behalf of fair_server, we need + * to limit its time based on the assigned runtime. + * - Fair task that runs outside of fair_server should account + * against fair_server such that it can account for this time + * and possibly avoid running this period. */ - if (p->dl_server != &rq->fair_server) + if (dl_server_active(&rq->fair_server)) dl_server_update(&rq->fair_server, delta_exec); } account_cfs_rq_runtime(cfs_rq, delta_exec); - if (cfs_rq->nr_running == 1) + if (cfs_rq->nr_queued == 1) return; if (resched || did_preempt_short(cfs_rq, curr)) { @@ -2128,7 +2129,7 @@ static void update_numa_stats(struct task_numa_env *env, ns->load += cpu_load(rq); ns->runnable += cpu_runnable(rq); ns->util += cpu_util_cfs(cpu); - ns->nr_running += rq->cfs.h_nr_running; + ns->nr_running += rq->cfs.h_nr_runnable; ns->compute_capacity += capacity_of(cpu); if (find_idle && idle_core < 0 && !rq->nr_running && idle_cpu(cpu)) { @@ -3679,9 +3680,7 @@ account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) list_add(&se->group_node, &rq->cfs_tasks); } #endif - cfs_rq->nr_running++; - if (se_is_idle(se)) - cfs_rq->idle_nr_running++; + cfs_rq->nr_queued++; } static void @@ -3694,9 +3693,7 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) list_del_init(&se->group_node); } #endif - cfs_rq->nr_running--; - if (se_is_idle(se)) - cfs_rq->idle_nr_running--; + cfs_rq->nr_queued--; } /* @@ -3771,137 +3768,32 @@ static inline void dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } #endif -static void reweight_eevdf(struct sched_entity *se, u64 avruntime, - unsigned long weight) -{ - unsigned long old_weight = se->load.weight; - s64 vlag, vslice; - - /* - * VRUNTIME - * -------- - * - * COROLLARY #1: The virtual runtime of the entity needs to be - * adjusted if re-weight at !0-lag point. - * - * Proof: For contradiction assume this is not true, so we can - * re-weight without changing vruntime at !0-lag point. - * - * Weight VRuntime Avg-VRuntime - * before w v V - * after w' v' V' - * - * Since lag needs to be preserved through re-weight: - * - * lag = (V - v)*w = (V'- v')*w', where v = v' - * ==> V' = (V - v)*w/w' + v (1) - * - * Let W be the total weight of the entities before reweight, - * since V' is the new weighted average of entities: - * - * V' = (WV + w'v - wv) / (W + w' - w) (2) - * - * by using (1) & (2) we obtain: - * - * (WV + w'v - wv) / (W + w' - w) = (V - v)*w/w' + v - * ==> (WV-Wv+Wv+w'v-wv)/(W+w'-w) = (V - v)*w/w' + v - * ==> (WV - Wv)/(W + w' - w) + v = (V - v)*w/w' + v - * ==> (V - v)*W/(W + w' - w) = (V - v)*w/w' (3) - * - * Since we are doing at !0-lag point which means V != v, we - * can simplify (3): - * - * ==> W / (W + w' - w) = w / w' - * ==> Ww' = Ww + ww' - ww - * ==> W * (w' - w) = w * (w' - w) - * ==> W = w (re-weight indicates w' != w) - * - * So the cfs_rq contains only one entity, hence vruntime of - * the entity @v should always equal to the cfs_rq's weighted - * average vruntime @V, which means we will always re-weight - * at 0-lag point, thus breach assumption. Proof completed. - * - * - * COROLLARY #2: Re-weight does NOT affect weighted average - * vruntime of all the entities. - * - * Proof: According to corollary #1, Eq. (1) should be: - * - * (V - v)*w = (V' - v')*w' - * ==> v' = V' - (V - v)*w/w' (4) - * - * According to the weighted average formula, we have: - * - * V' = (WV - wv + w'v') / (W - w + w') - * = (WV - wv + w'(V' - (V - v)w/w')) / (W - w + w') - * = (WV - wv + w'V' - Vw + wv) / (W - w + w') - * = (WV + w'V' - Vw) / (W - w + w') - * - * ==> V'*(W - w + w') = WV + w'V' - Vw - * ==> V' * (W - w) = (W - w) * V (5) - * - * If the entity is the only one in the cfs_rq, then reweight - * always occurs at 0-lag point, so V won't change. Or else - * there are other entities, hence W != w, then Eq. (5) turns - * into V' = V. So V won't change in either case, proof done. - * - * - * So according to corollary #1 & #2, the effect of re-weight - * on vruntime should be: - * - * v' = V' - (V - v) * w / w' (4) - * = V - (V - v) * w / w' - * = V - vl * w / w' - * = V - vl' - */ - if (avruntime != se->vruntime) { - vlag = entity_lag(avruntime, se); - vlag = div_s64(vlag * old_weight, weight); - se->vruntime = avruntime - vlag; - } - - /* - * DEADLINE - * -------- - * - * When the weight changes, the virtual time slope changes and - * we should adjust the relative virtual deadline accordingly. - * - * d' = v' + (d - v)*w/w' - * = V' - (V - v)*w/w' + (d - v)*w/w' - * = V - (V - v)*w/w' + (d - v)*w/w' - * = V + (d - V)*w/w' - */ - vslice = (s64)(se->deadline - avruntime); - vslice = div_s64(vslice * old_weight, weight); - se->deadline = avruntime + vslice; -} +static void place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags); static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, unsigned long weight) { bool curr = cfs_rq->curr == se; - u64 avruntime; if (se->on_rq) { /* commit outstanding execution time */ update_curr(cfs_rq); - avruntime = avg_vruntime(cfs_rq); + update_entity_lag(cfs_rq, se); + se->deadline -= se->vruntime; + se->rel_deadline = 1; if (!curr) __dequeue_entity(cfs_rq, se); update_load_sub(&cfs_rq->load, se->load.weight); } dequeue_load_avg(cfs_rq, se); - if (se->on_rq) { - reweight_eevdf(se, avruntime, weight); - } else { - /* - * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), - * we need to scale se->vlag when w_i changes. - */ - se->vlag = div_s64(se->vlag * se->load.weight, weight); - } + /* + * Because we keep se->vlag = V - v_i, while: lag_i = w_i*(V - v_i), + * we need to scale se->vlag when w_i changes. + */ + se->vlag = div_s64(se->vlag * se->load.weight, weight); + if (se->rel_deadline) + se->deadline = div_s64(se->deadline * se->load.weight, weight); update_load_set(&se->load, weight); @@ -3916,6 +3808,7 @@ static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, enqueue_load_avg(cfs_rq, se); if (se->on_rq) { update_load_add(&cfs_rq->load, se->load.weight); + place_entity(cfs_rq, se, 0); if (!curr) __enqueue_entity(cfs_rq, se); @@ -4062,7 +3955,11 @@ static void update_cfs_group(struct sched_entity *se) struct cfs_rq *gcfs_rq = group_cfs_rq(se); long shares; - if (!gcfs_rq) + /* + * When a group becomes empty, preserve its weight. This matters for + * DELAY_DEQUEUE. + */ + if (!gcfs_rq || !gcfs_rq->load.weight) return; if (throttled_hierarchy(gcfs_rq)) @@ -5230,7 +5127,7 @@ static inline void update_misfit_status(struct task_struct *p, struct rq *rq) static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) { - return !cfs_rq->nr_running; + return !cfs_rq->nr_queued; } #define UPDATE_TG 0x0 @@ -5268,6 +5165,22 @@ static inline void update_misfit_status(struct task_struct *p, struct rq *rq) {} #endif /* CONFIG_SMP */ +void __setparam_fair(struct task_struct *p, const struct sched_attr *attr) +{ + struct sched_entity *se = &p->se; + + p->static_prio = NICE_TO_PRIO(attr->sched_nice); + if (attr->sched_runtime) { + se->custom_slice = 1; + se->slice = clamp_t(u64, attr->sched_runtime, + NSEC_PER_MSEC/10, /* HZ=1000 * 10 */ + NSEC_PER_MSEC*100); /* HZ=100 / 10 */ + } else { + se->custom_slice = 0; + se->slice = sysctl_sched_base_slice; + } +} + static void place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { @@ -5286,7 +5199,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * * EEVDF: placement strategy #1 / #2 */ - if (sched_feat(PLACE_LAG) && cfs_rq->nr_running && se->vlag) { + if (sched_feat(PLACE_LAG) && cfs_rq->nr_queued && se->vlag) { struct sched_entity *curr = cfs_rq->curr; unsigned long load; @@ -5356,7 +5269,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) se->vruntime = vruntime - lag; - if (sched_feat(PLACE_REL_DEADLINE) && se->rel_deadline) { + if (se->rel_deadline) { se->deadline += se->vruntime; se->rel_deadline = 0; return; @@ -5379,8 +5292,6 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) static void check_enqueue_throttle(struct cfs_rq *cfs_rq); static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq); -static inline bool cfs_bandwidth_used(void); - static void requeue_delayed_entity(struct sched_entity *se); @@ -5402,7 +5313,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * When enqueuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. * - For group_entity, update its runnable_weight to reflect the new - * h_nr_running of its group cfs_rq. + * h_nr_runnable of its group cfs_rq. * - For group_entity, update its weight to reflect the new share of * its group cfs_rq * - Add its new weight to cfs_rq->load.weight @@ -5435,7 +5346,7 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) __enqueue_entity(cfs_rq, se); se->on_rq = 1; - if (cfs_rq->nr_running == 1) { + if (cfs_rq->nr_queued == 1) { check_enqueue_throttle(cfs_rq); if (!throttled_hierarchy(cfs_rq)) { list_add_leaf_cfs_rq(cfs_rq); @@ -5471,9 +5382,33 @@ static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); -static inline void finish_delayed_dequeue_entity(struct sched_entity *se) +static void set_delayed(struct sched_entity *se) +{ + se->sched_delayed = 1; + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + cfs_rq->h_nr_runnable--; + if (cfs_rq_throttled(cfs_rq)) + break; + } +} + +static void clear_delayed(struct sched_entity *se) { se->sched_delayed = 0; + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + cfs_rq->h_nr_runnable++; + if (cfs_rq_throttled(cfs_rq)) + break; + } +} + +static inline void finish_delayed_dequeue_entity(struct sched_entity *se) +{ + clear_delayed(se); if (sched_feat(DELAY_ZERO) && se->vlag > 0) se->vlag = 0; } @@ -5482,8 +5417,10 @@ static bool dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { bool sleep = flags & DEQUEUE_SLEEP; + int action = UPDATE_TG; update_curr(cfs_rq); + clear_buddies(cfs_rq, se); if (flags & DEQUEUE_DELAYED) { SCHED_WARN_ON(!se->sched_delayed); @@ -5500,15 +5437,12 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (sched_feat(DELAY_DEQUEUE) && delay && !entity_eligible(cfs_rq, se)) { - if (cfs_rq->next == se) - cfs_rq->next = NULL; update_load_avg(cfs_rq, se, 0); - se->sched_delayed = 1; + set_delayed(se); return false; } } - int action = UPDATE_TG; if (entity_is_task(se) && task_on_rq_migrating(task_of(se))) action |= DO_DETACH; @@ -5516,7 +5450,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * When dequeuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. * - For group_entity, update its runnable_weight to reflect the new - * h_nr_running of its group cfs_rq. + * h_nr_runnable of its group cfs_rq. * - Subtract its previous weight from cfs_rq->load.weight. * - For group entity, update its weight to reflect the new share * of its group cfs_rq. @@ -5526,8 +5460,6 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) update_stats_dequeue_fair(cfs_rq, se, flags); - clear_buddies(cfs_rq, se); - update_entity_lag(cfs_rq, se); if (sched_feat(PLACE_REL_DEADLINE) && !sleep) { se->deadline -= se->vruntime; @@ -5556,7 +5488,7 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (flags & DEQUEUE_DELAYED) finish_delayed_dequeue_entity(se); - if (cfs_rq->nr_running == 0) + if (cfs_rq->nr_queued == 0) update_idle_cfs_rq_clock_pelt(cfs_rq); return true; @@ -5618,17 +5550,19 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags); static struct sched_entity * pick_next_entity(struct rq *rq, struct cfs_rq *cfs_rq) { + struct sched_entity *se; + /* - * Enabling NEXT_BUDDY will affect latency but not fairness. + * Picking the ->next buddy will affect latency but not fairness. */ - if (sched_feat(NEXT_BUDDY) && + if (sched_feat(PICK_BUDDY) && cfs_rq->next && entity_eligible(cfs_rq, cfs_rq->next)) { /* ->next will never be delayed */ SCHED_WARN_ON(cfs_rq->next->sched_delayed); return cfs_rq->next; } - struct sched_entity *se = pick_eevdf(cfs_rq); + se = pick_eevdf(cfs_rq); if (se->sched_delayed) { dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); /* @@ -5904,7 +5838,7 @@ static int tg_throttle_down(struct task_group *tg, void *data) list_del_leaf_cfs_rq(cfs_rq); SCHED_WARN_ON(cfs_rq->throttled_clock_self); - if (cfs_rq->nr_running) + if (cfs_rq->nr_queued) cfs_rq->throttled_clock_self = rq_clock(rq); } cfs_rq->throttle_count++; @@ -5917,8 +5851,8 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) struct rq *rq = rq_of(cfs_rq); struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); struct sched_entity *se; - long task_delta, idle_task_delta, dequeue = 1; - long rq_h_nr_running = rq->cfs.h_nr_running; + long queued_delta, runnable_delta, idle_delta, dequeue = 1; + long rq_h_nr_queued = rq->cfs.h_nr_queued; raw_spin_lock(&cfs_b->lock); /* This will start the period timer if necessary */ @@ -5948,8 +5882,9 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq); rcu_read_unlock(); - task_delta = cfs_rq->h_nr_running; - idle_task_delta = cfs_rq->idle_h_nr_running; + queued_delta = cfs_rq->h_nr_queued; + runnable_delta = cfs_rq->h_nr_runnable; + idle_delta = cfs_rq->h_nr_idle; for_each_sched_entity(se) { struct cfs_rq *qcfs_rq = cfs_rq_of(se); int flags; @@ -5969,10 +5904,11 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) dequeue_entity(qcfs_rq, se, flags); if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; + idle_delta = cfs_rq->h_nr_queued; - qcfs_rq->h_nr_running -= task_delta; - qcfs_rq->idle_h_nr_running -= idle_task_delta; + qcfs_rq->h_nr_queued -= queued_delta; + qcfs_rq->h_nr_runnable -= runnable_delta; + qcfs_rq->h_nr_idle -= idle_delta; if (qcfs_rq->load.weight) { /* Avoid re-evaluating load for this entity: */ @@ -5991,17 +5927,18 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) se_update_runnable(se); if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; + idle_delta = cfs_rq->h_nr_queued; - qcfs_rq->h_nr_running -= task_delta; - qcfs_rq->idle_h_nr_running -= idle_task_delta; + qcfs_rq->h_nr_queued -= queued_delta; + qcfs_rq->h_nr_runnable -= runnable_delta; + qcfs_rq->h_nr_idle -= idle_delta; } /* At this point se is NULL and we are at root level*/ - sub_nr_running(rq, task_delta); + sub_nr_running(rq, queued_delta); /* Stop the fair server if throttling resulted in no runnable tasks */ - if (rq_h_nr_running && !rq->cfs.h_nr_running) + if (rq_h_nr_queued && !rq->cfs.h_nr_queued) dl_server_stop(&rq->fair_server); done: /* @@ -6010,7 +5947,7 @@ done: */ cfs_rq->throttled = 1; SCHED_WARN_ON(cfs_rq->throttled_clock); - if (cfs_rq->nr_running) + if (cfs_rq->nr_queued) cfs_rq->throttled_clock = rq_clock(rq); return true; } @@ -6020,8 +5957,8 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) struct rq *rq = rq_of(cfs_rq); struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); struct sched_entity *se; - long task_delta, idle_task_delta; - long rq_h_nr_running = rq->cfs.h_nr_running; + long queued_delta, runnable_delta, idle_delta; + long rq_h_nr_queued = rq->cfs.h_nr_queued; se = cfs_rq->tg->se[cpu_of(rq)]; @@ -6054,8 +5991,9 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) goto unthrottle_throttle; } - task_delta = cfs_rq->h_nr_running; - idle_task_delta = cfs_rq->idle_h_nr_running; + queued_delta = cfs_rq->h_nr_queued; + runnable_delta = cfs_rq->h_nr_runnable; + idle_delta = cfs_rq->h_nr_idle; for_each_sched_entity(se) { struct cfs_rq *qcfs_rq = cfs_rq_of(se); @@ -6069,10 +6007,11 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) enqueue_entity(qcfs_rq, se, ENQUEUE_WAKEUP); if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; + idle_delta = cfs_rq->h_nr_queued; - qcfs_rq->h_nr_running += task_delta; - qcfs_rq->idle_h_nr_running += idle_task_delta; + qcfs_rq->h_nr_queued += queued_delta; + qcfs_rq->h_nr_runnable += runnable_delta; + qcfs_rq->h_nr_idle += idle_delta; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(qcfs_rq)) @@ -6086,10 +6025,11 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) se_update_runnable(se); if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; + idle_delta = cfs_rq->h_nr_queued; - qcfs_rq->h_nr_running += task_delta; - qcfs_rq->idle_h_nr_running += idle_task_delta; + qcfs_rq->h_nr_queued += queued_delta; + qcfs_rq->h_nr_runnable += runnable_delta; + qcfs_rq->h_nr_idle += idle_delta; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(qcfs_rq)) @@ -6097,17 +6037,17 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) } /* Start the fair server if un-throttling resulted in new runnable tasks */ - if (!rq_h_nr_running && rq->cfs.h_nr_running) + if (!rq_h_nr_queued && rq->cfs.h_nr_queued) dl_server_start(&rq->fair_server); /* At this point se is NULL and we are at root level*/ - add_nr_running(rq, task_delta); + add_nr_running(rq, queued_delta); unthrottle_throttle: assert_list_leaf_cfs_rq(rq); /* Determine whether we need to wake up potentially idle CPU: */ - if (rq->curr == rq->idle && rq->cfs.nr_running) + if (rq->curr == rq->idle && rq->cfs.nr_queued) resched_curr(rq); } @@ -6408,7 +6348,7 @@ static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) if (!cfs_bandwidth_used()) return; - if (!cfs_rq->runtime_enabled || cfs_rq->nr_running) + if (!cfs_rq->runtime_enabled || cfs_rq->nr_queued) return; __return_cfs_rq_runtime(cfs_rq); @@ -6679,6 +6619,10 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_rq_held(rq); + // Do not unthrottle for an active CPU + if (cpumask_test_cpu(cpu_of(rq), cpu_active_mask)) + return; + /* * The rq clock has already been updated in the * set_rq_offline(), so we should skip updating @@ -6694,18 +6638,20 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) continue; /* - * clock_task is not advancing so we just need to make sure - * there's some valid quota amount - */ - cfs_rq->runtime_remaining = 1; - /* * Offline rq is schedulable till CPU is completely disabled * in take_cpu_down(), so we prevent new cfs throttling here. */ cfs_rq->runtime_enabled = 0; - if (cfs_rq_throttled(cfs_rq)) - unthrottle_cfs_rq(cfs_rq); + if (!cfs_rq_throttled(cfs_rq)) + continue; + + /* + * clock_task is not advancing so we just need to make sure + * there's some valid quota amount + */ + cfs_rq->runtime_remaining = 1; + unthrottle_cfs_rq(cfs_rq); } rcu_read_unlock(); @@ -6754,11 +6700,6 @@ static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) #else /* CONFIG_CFS_BANDWIDTH */ -static inline bool cfs_bandwidth_used(void) -{ - return false; -} - static void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) {} static bool check_cfs_rq_runtime(struct cfs_rq *cfs_rq) { return false; } static void check_enqueue_throttle(struct cfs_rq *cfs_rq) {} @@ -6816,7 +6757,7 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p) SCHED_WARN_ON(task_rq(p) != rq); - if (rq->cfs.h_nr_running > 1) { + if (rq->cfs.h_nr_queued > 1) { u64 ran = se->sum_exec_runtime - se->prev_sum_exec_runtime; u64 slice = se->slice; s64 delta = slice - ran; @@ -6904,7 +6845,7 @@ static inline void check_update_overutilized_status(struct rq *rq) { } /* Runqueue only has SCHED_IDLE tasks enqueued */ static int sched_idle_rq(struct rq *rq) { - return unlikely(rq->nr_running == rq->cfs.idle_h_nr_running && + return unlikely(rq->nr_running == rq->cfs.h_nr_idle && rq->nr_running); } @@ -6931,19 +6872,19 @@ requeue_delayed_entity(struct sched_entity *se) if (sched_feat(DELAY_ZERO)) { update_entity_lag(cfs_rq, se); if (se->vlag > 0) { - cfs_rq->nr_running--; + cfs_rq->nr_queued--; if (se != cfs_rq->curr) __dequeue_entity(cfs_rq, se); se->vlag = 0; place_entity(cfs_rq, se, 0); if (se != cfs_rq->curr) __enqueue_entity(cfs_rq, se); - cfs_rq->nr_running++; + cfs_rq->nr_queued++; } } update_load_avg(cfs_rq, se, 0); - se->sched_delayed = 0; + clear_delayed(se); } /* @@ -6956,9 +6897,10 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) { struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se; - int idle_h_nr_running = task_has_idle_policy(p); + int h_nr_idle = task_has_idle_policy(p); + int h_nr_runnable = 1; int task_new = !(flags & ENQUEUE_WAKEUP); - int rq_h_nr_running = rq->cfs.h_nr_running; + int rq_h_nr_queued = rq->cfs.h_nr_queued; u64 slice = 0; /* @@ -6983,6 +6925,9 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) if (p->in_iowait) cpufreq_update_util(rq, SCHED_CPUFREQ_IOWAIT); + if (task_new && se->sched_delayed) + h_nr_runnable = 0; + for_each_sched_entity(se) { if (se->on_rq) { if (se->sched_delayed) @@ -7003,11 +6948,12 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) enqueue_entity(cfs_rq, se, flags); slice = cfs_rq_min_slice(cfs_rq); - cfs_rq->h_nr_running++; - cfs_rq->idle_h_nr_running += idle_h_nr_running; + cfs_rq->h_nr_runnable += h_nr_runnable; + cfs_rq->h_nr_queued++; + cfs_rq->h_nr_idle += h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; + h_nr_idle = 1; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(cfs_rq)) @@ -7026,18 +6972,19 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) se->slice = slice; slice = cfs_rq_min_slice(cfs_rq); - cfs_rq->h_nr_running++; - cfs_rq->idle_h_nr_running += idle_h_nr_running; + cfs_rq->h_nr_runnable += h_nr_runnable; + cfs_rq->h_nr_queued++; + cfs_rq->h_nr_idle += h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; + h_nr_idle = 1; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(cfs_rq)) goto enqueue_throttle; } - if (!rq_h_nr_running && rq->cfs.h_nr_running) { + if (!rq_h_nr_queued && rq->cfs.h_nr_queued) { /* Account for idle runtime */ if (!rq->nr_running) dl_server_update_idle_time(rq, rq->curr); @@ -7084,19 +7031,22 @@ static void set_next_buddy(struct sched_entity *se); static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) { bool was_sched_idle = sched_idle_rq(rq); - int rq_h_nr_running = rq->cfs.h_nr_running; + int rq_h_nr_queued = rq->cfs.h_nr_queued; bool task_sleep = flags & DEQUEUE_SLEEP; bool task_delayed = flags & DEQUEUE_DELAYED; struct task_struct *p = NULL; - int idle_h_nr_running = 0; - int h_nr_running = 0; + int h_nr_idle = 0; + int h_nr_queued = 0; + int h_nr_runnable = 0; struct cfs_rq *cfs_rq; u64 slice = 0; if (entity_is_task(se)) { p = task_of(se); - h_nr_running = 1; - idle_h_nr_running = task_has_idle_policy(p); + h_nr_queued = 1; + h_nr_idle = task_has_idle_policy(p); + if (task_sleep || task_delayed || !se->sched_delayed) + h_nr_runnable = 1; } else { cfs_rq = group_cfs_rq(se); slice = cfs_rq_min_slice(cfs_rq); @@ -7112,11 +7062,12 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) break; } - cfs_rq->h_nr_running -= h_nr_running; - cfs_rq->idle_h_nr_running -= idle_h_nr_running; + cfs_rq->h_nr_runnable -= h_nr_runnable; + cfs_rq->h_nr_queued -= h_nr_queued; + cfs_rq->h_nr_idle -= h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = h_nr_running; + h_nr_idle = h_nr_queued; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(cfs_rq)) @@ -7150,20 +7101,21 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) se->slice = slice; slice = cfs_rq_min_slice(cfs_rq); - cfs_rq->h_nr_running -= h_nr_running; - cfs_rq->idle_h_nr_running -= idle_h_nr_running; + cfs_rq->h_nr_runnable -= h_nr_runnable; + cfs_rq->h_nr_queued -= h_nr_queued; + cfs_rq->h_nr_idle -= h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = h_nr_running; + h_nr_idle = h_nr_queued; /* end evaluation on encountering a throttled cfs_rq */ if (cfs_rq_throttled(cfs_rq)) return 0; } - sub_nr_running(rq, h_nr_running); + sub_nr_running(rq, h_nr_queued); - if (rq_h_nr_running && !rq->cfs.h_nr_running) + if (rq_h_nr_queued && !rq->cfs.h_nr_queued) dl_server_stop(&rq->fair_server); /* balance early to pull high priority tasks */ @@ -8780,7 +8732,7 @@ static void check_preempt_wakeup_fair(struct rq *rq, struct task_struct *p, int if (unlikely(throttled_hierarchy(cfs_rq_of(pse)))) return; - if (sched_feat(NEXT_BUDDY) && !(wake_flags & WF_FORK)) { + if (sched_feat(NEXT_BUDDY) && !(wake_flags & WF_FORK) && !pse->sched_delayed) { set_next_buddy(pse); } @@ -8852,7 +8804,7 @@ static struct task_struct *pick_task_fair(struct rq *rq) again: cfs_rq = &rq->cfs; - if (!cfs_rq->nr_running) + if (!cfs_rq->nr_queued) return NULL; do { @@ -8969,7 +8921,7 @@ static struct task_struct *__pick_next_task_fair(struct rq *rq, struct task_stru static bool fair_server_has_tasks(struct sched_dl_entity *dl_se) { - return !!dl_se->rq->cfs.nr_running; + return !!dl_se->rq->cfs.nr_queued; } static struct task_struct *fair_server_pick_task(struct sched_dl_entity *dl_se) @@ -9300,43 +9252,43 @@ static int task_hot(struct task_struct *p, struct lb_env *env) #ifdef CONFIG_NUMA_BALANCING /* - * Returns 1, if task migration degrades locality - * Returns 0, if task migration improves locality i.e migration preferred. - * Returns -1, if task migration is not affected by locality. + * Returns a positive value, if task migration degrades locality. + * Returns 0, if task migration is not affected by locality. + * Returns a negative value, if task migration improves locality i.e migration preferred. */ -static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env) +static long migrate_degrades_locality(struct task_struct *p, struct lb_env *env) { struct numa_group *numa_group = rcu_dereference(p->numa_group); unsigned long src_weight, dst_weight; int src_nid, dst_nid, dist; if (!static_branch_likely(&sched_numa_balancing)) - return -1; + return 0; if (!p->numa_faults || !(env->sd->flags & SD_NUMA)) - return -1; + return 0; src_nid = cpu_to_node(env->src_cpu); dst_nid = cpu_to_node(env->dst_cpu); if (src_nid == dst_nid) - return -1; + return 0; /* Migrating away from the preferred node is always bad. */ if (src_nid == p->numa_preferred_nid) { if (env->src_rq->nr_running > env->src_rq->nr_preferred_running) return 1; else - return -1; + return 0; } /* Encourage migration to the preferred node. */ if (dst_nid == p->numa_preferred_nid) - return 0; + return -1; /* Leaving a core idle is often worse than degrading locality. */ if (env->idle == CPU_IDLE) - return -1; + return 0; dist = node_distance(src_nid, dst_nid); if (numa_group) { @@ -9347,37 +9299,77 @@ static int migrate_degrades_locality(struct task_struct *p, struct lb_env *env) dst_weight = task_weight(p, dst_nid, dist); } - return dst_weight < src_weight; + return src_weight - dst_weight; } #else -static inline int migrate_degrades_locality(struct task_struct *p, +static inline long migrate_degrades_locality(struct task_struct *p, struct lb_env *env) { - return -1; + return 0; } #endif /* + * Check whether the task is ineligible on the destination cpu + * + * When the PLACE_LAG scheduling feature is enabled and + * dst_cfs_rq->nr_queued is greater than 1, if the task + * is ineligible, it will also be ineligible when + * it is migrated to the destination cpu. + */ +static inline int task_is_ineligible_on_dst_cpu(struct task_struct *p, int dest_cpu) +{ + struct cfs_rq *dst_cfs_rq; + +#ifdef CONFIG_FAIR_GROUP_SCHED + dst_cfs_rq = task_group(p)->cfs_rq[dest_cpu]; +#else + dst_cfs_rq = &cpu_rq(dest_cpu)->cfs; +#endif + if (sched_feat(PLACE_LAG) && dst_cfs_rq->nr_queued && + !entity_eligible(task_cfs_rq(p), &p->se)) + return 1; + + return 0; +} + +/* * can_migrate_task - may task p from runqueue rq be migrated to this_cpu? */ static int can_migrate_task(struct task_struct *p, struct lb_env *env) { - int tsk_cache_hot; + long degrades, hot; lockdep_assert_rq_held(env->src_rq); + if (p->sched_task_hot) + p->sched_task_hot = 0; /* * We do not migrate tasks that are: - * 1) throttled_lb_pair, or - * 2) cannot be migrated to this CPU due to cpus_ptr, or - * 3) running (obviously), or - * 4) are cache-hot on their current CPU. + * 1) delayed dequeued unless we migrate load, or + * 2) throttled_lb_pair, or + * 3) cannot be migrated to this CPU due to cpus_ptr, or + * 4) running (obviously), or + * 5) are cache-hot on their current CPU. */ + if ((p->se.sched_delayed) && (env->migration_type != migrate_load)) + return 0; + if (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu)) return 0; + /* + * We want to prioritize the migration of eligible tasks. + * For ineligible tasks we soft-limit them and only allow + * them to migrate when nr_balance_failed is non-zero to + * avoid load-balancing trying very hard to balance the load. + */ + if (!env->sd->nr_balance_failed && + task_is_ineligible_on_dst_cpu(p, env->dst_cpu)) + return 0; + /* Disregard percpu kthreads; they are where they need to be. */ if (kthread_is_per_cpu(p)) return 0; @@ -9433,16 +9425,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) if (env->flags & LBF_ACTIVE_LB) return 1; - tsk_cache_hot = migrate_degrades_locality(p, env); - if (tsk_cache_hot == -1) - tsk_cache_hot = task_hot(p, env); + degrades = migrate_degrades_locality(p, env); + if (!degrades) + hot = task_hot(p, env); + else + hot = degrades > 0; - if (tsk_cache_hot <= 0 || - env->sd->nr_balance_failed > env->sd->cache_nice_tries) { - if (tsk_cache_hot == 1) { - schedstat_inc(env->sd->lb_hot_gained[env->idle]); - schedstat_inc(p->stats.nr_forced_migrations); - } + if (!hot || env->sd->nr_balance_failed > env->sd->cache_nice_tries) { + if (hot) + p->sched_task_hot = 1; return 1; } @@ -9457,6 +9448,12 @@ static void detach_task(struct task_struct *p, struct lb_env *env) { lockdep_assert_rq_held(env->src_rq); + if (p->sched_task_hot) { + p->sched_task_hot = 0; + schedstat_inc(env->sd->lb_hot_gained[env->idle]); + schedstat_inc(p->stats.nr_forced_migrations); + } + deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK); set_task_cpu(p, env->dst_cpu); } @@ -9617,6 +9614,9 @@ static int detach_tasks(struct lb_env *env) continue; next: + if (p->sched_task_hot) + schedstat_inc(p->stats.nr_failed_migrations_hot); + list_move(&p->se.group_node, tasks); } @@ -9759,7 +9759,7 @@ static bool __update_blocked_fair(struct rq *rq, bool *done) if (update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq)) { update_tg_load_avg(cfs_rq); - if (cfs_rq->nr_running == 0) + if (cfs_rq->nr_queued == 0) update_idle_cfs_rq_clock_pelt(cfs_rq); if (cfs_rq == &rq->cfs) @@ -10291,7 +10291,7 @@ sched_reduced_capacity(struct rq *rq, struct sched_domain *sd) * When there is more than 1 task, the group_overloaded case already * takes care of cpu with reduced capacity */ - if (rq->cfs.h_nr_running != 1) + if (rq->cfs.h_nr_runnable != 1) return false; return check_cpu_capacity(rq, sd); @@ -10313,7 +10313,8 @@ static inline void update_sg_lb_stats(struct lb_env *env, bool *sg_overloaded, bool *sg_overutilized) { - int i, nr_running, local_group; + int i, nr_running, local_group, sd_flags = env->sd->flags; + bool balancing_at_rd = !env->sd->parent; memset(sgs, 0, sizeof(*sgs)); @@ -10326,21 +10327,14 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->group_load += load; sgs->group_util += cpu_util_cfs(i); sgs->group_runnable += cpu_runnable(rq); - sgs->sum_h_nr_running += rq->cfs.h_nr_running; + sgs->sum_h_nr_running += rq->cfs.h_nr_runnable; nr_running = rq->nr_running; sgs->sum_nr_running += nr_running; - if (nr_running > 1) - *sg_overloaded = 1; - if (cpu_overutilized(i)) *sg_overutilized = 1; -#ifdef CONFIG_NUMA_BALANCING - sgs->nr_numa_running += rq->nr_numa_running; - sgs->nr_preferred_running += rq->nr_preferred_running; -#endif /* * No need to call idle_cpu() if nr_running is not 0 */ @@ -10350,10 +10344,21 @@ static inline void update_sg_lb_stats(struct lb_env *env, continue; } + /* Overload indicator is only updated at root domain */ + if (balancing_at_rd && nr_running > 1) + *sg_overloaded = 1; + +#ifdef CONFIG_NUMA_BALANCING + /* Only fbq_classify_group() uses this to classify NUMA groups */ + if (sd_flags & SD_NUMA) { + sgs->nr_numa_running += rq->nr_numa_running; + sgs->nr_preferred_running += rq->nr_preferred_running; + } +#endif if (local_group) continue; - if (env->sd->flags & SD_ASYM_CPUCAPACITY) { + if (sd_flags & SD_ASYM_CPUCAPACITY) { /* Check for a misfit task on the cpu */ if (sgs->group_misfit_task_load < rq->misfit_task_load) { sgs->group_misfit_task_load = rq->misfit_task_load; @@ -10641,7 +10646,7 @@ static inline void update_sg_wakeup_stats(struct sched_domain *sd, sgs->group_util += cpu_util_without(i, p); sgs->group_runnable += cpu_runnable_without(rq, p); local = task_running_on_cpu(i, p); - sgs->sum_h_nr_running += rq->cfs.h_nr_running - local; + sgs->sum_h_nr_running += rq->cfs.h_nr_runnable - local; nr_running = rq->nr_running - local; sgs->sum_nr_running += nr_running; @@ -11423,7 +11428,7 @@ static struct rq *sched_balance_find_src_rq(struct lb_env *env, if (rt > env->fbq_type) continue; - nr_running = rq->cfs.h_nr_running; + nr_running = rq->cfs.h_nr_runnable; if (!nr_running) continue; @@ -11582,7 +11587,7 @@ static int need_active_balance(struct lb_env *env) * available on dst_cpu. */ if (env->idle && - (env->src_rq->cfs.h_nr_running == 1)) { + (env->src_rq->cfs.h_nr_runnable == 1)) { if ((check_cpu_capacity(env->src_rq, sd)) && (capacity_of(env->src_cpu)*sd->imbalance_pct < capacity_of(env->dst_cpu)*100)) return 1; @@ -11662,6 +11667,28 @@ static int should_we_balance(struct lb_env *env) return group_balance_cpu(sg) == env->dst_cpu; } +static void update_lb_imbalance_stat(struct lb_env *env, struct sched_domain *sd, + enum cpu_idle_type idle) +{ + if (!schedstat_enabled()) + return; + + switch (env->migration_type) { + case migrate_load: + __schedstat_add(sd->lb_imbalance_load[idle], env->imbalance); + break; + case migrate_util: + __schedstat_add(sd->lb_imbalance_util[idle], env->imbalance); + break; + case migrate_task: + __schedstat_add(sd->lb_imbalance_task[idle], env->imbalance); + break; + case migrate_misfit: + __schedstat_add(sd->lb_imbalance_misfit[idle], env->imbalance); + break; + } +} + /* * Check this_cpu to ensure it is balanced within domain. Attempt to move * tasks if there is an imbalance. @@ -11712,7 +11739,7 @@ redo: WARN_ON_ONCE(busiest == env.dst_rq); - schedstat_add(sd->lb_imbalance[idle], env.imbalance); + update_lb_imbalance_stat(&env, sd, idle); env.src_cpu = busiest->cpu; env.src_rq = busiest; @@ -12210,16 +12237,13 @@ static inline int on_null_domain(struct rq *rq) * - When one of the busy CPUs notices that there may be an idle rebalancing * needed, they will kick the idle load balancer, which then does idle * load balancing for all the idle CPUs. - * - * - HK_TYPE_MISC CPUs are used for this task, because HK_TYPE_SCHED is not set - * anywhere yet. */ static inline int find_new_ilb(void) { const struct cpumask *hk_mask; int ilb_cpu; - hk_mask = housekeeping_cpumask(HK_TYPE_MISC); + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); for_each_cpu_and(ilb_cpu, nohz.idle_cpus_mask, hk_mask) { @@ -12237,7 +12261,8 @@ static inline int find_new_ilb(void) * Kick a CPU to do the NOHZ balancing, if it is time for it, via a cross-CPU * SMP function call (IPI). * - * We pick the first idle CPU in the HK_TYPE_MISC housekeeping set (if there is one). + * We pick the first idle CPU in the HK_TYPE_KERNEL_NOISE housekeeping set + * (if there is one). */ static void kick_ilb(unsigned int flags) { @@ -12325,7 +12350,7 @@ static void nohz_balancer_kick(struct rq *rq) * If there's a runnable CFS task and the current CPU has reduced * capacity, kick the ILB to see if there's a better CPU to run on: */ - if (rq->cfs.h_nr_running >= 1 && check_cpu_capacity(rq, sd)) { + if (rq->cfs.h_nr_runnable >= 1 && check_cpu_capacity(rq, sd)) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } @@ -12457,10 +12482,6 @@ void nohz_balance_enter_idle(int cpu) if (!cpu_active(cpu)) return; - /* Spare idle load balancing on CPUs that don't want to be disturbed: */ - if (!housekeeping_cpu(cpu, HK_TYPE_SCHED)) - return; - /* * Can be set safely without rq->lock held * If a clear happens, it will have evaluated last additions because @@ -12680,13 +12701,6 @@ static void nohz_newidle_balance(struct rq *this_rq) { int this_cpu = this_rq->cpu; - /* - * This CPU doesn't want to be disturbed by scheduler - * housekeeping - */ - if (!housekeeping_cpu(this_cpu, HK_TYPE_SCHED)) - return; - /* Will wake up very soon. No time for doing anything else*/ if (this_rq->avg_idle < sysctl_sched_migration_cost) return; @@ -12823,11 +12837,11 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) * have been enqueued in the meantime. Since we're not going idle, * pretend we pulled a task. */ - if (this_rq->cfs.h_nr_running && !pulled_task) + if (this_rq->cfs.h_nr_queued && !pulled_task) pulled_task = 1; /* Is there a task of a high priority class? */ - if (this_rq->nr_running != this_rq->cfs.h_nr_running) + if (this_rq->nr_running != this_rq->cfs.h_nr_queued) pulled_task = -1; out: @@ -12848,9 +12862,9 @@ out: /* * This softirq handler is triggered via SCHED_SOFTIRQ from two places: * - * - directly from the local scheduler_tick() for periodic load balancing + * - directly from the local sched_tick() for periodic load balancing * - * - indirectly from a remote scheduler_tick() for NOHZ idle balancing + * - indirectly from a remote sched_tick() for NOHZ idle balancing * through the SMP cross-call nohz_csd_func() */ static __latent_entropy void sched_balance_softirq(void) @@ -12941,7 +12955,7 @@ static inline void task_tick_core(struct rq *rq, struct task_struct *curr) * MIN_NR_TASKS_DURING_FORCEIDLE - 1 tasks and use that to check * if we need to give up the CPU. */ - if (rq->core->core_forceidle_count && rq->cfs.nr_running == 1 && + if (rq->core->core_forceidle_count && rq->cfs.nr_queued == 1 && __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE)) resched_curr(rq); } @@ -13085,7 +13099,7 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) if (!task_on_rq_queued(p)) return; - if (rq->cfs.nr_running == 1) + if (rq->cfs.nr_queued == 1) return; /* @@ -13495,7 +13509,7 @@ int sched_group_set_idle(struct task_group *tg, long idle) for_each_possible_cpu(i) { struct rq *rq = cpu_rq(i); struct sched_entity *se = tg->se[i]; - struct cfs_rq *parent_cfs_rq, *grp_cfs_rq = tg->cfs_rq[i]; + struct cfs_rq *grp_cfs_rq = tg->cfs_rq[i]; bool was_idle = cfs_rq_is_idle(grp_cfs_rq); long idle_task_delta; struct rq_flags rf; @@ -13506,16 +13520,8 @@ int sched_group_set_idle(struct task_group *tg, long idle) if (WARN_ON_ONCE(was_idle == cfs_rq_is_idle(grp_cfs_rq))) goto next_cpu; - if (se->on_rq) { - parent_cfs_rq = cfs_rq_of(se); - if (cfs_rq_is_idle(grp_cfs_rq)) - parent_cfs_rq->idle_nr_running++; - else - parent_cfs_rq->idle_nr_running--; - } - - idle_task_delta = grp_cfs_rq->h_nr_running - - grp_cfs_rq->idle_h_nr_running; + idle_task_delta = grp_cfs_rq->h_nr_queued - + grp_cfs_rq->h_nr_idle; if (!cfs_rq_is_idle(grp_cfs_rq)) idle_task_delta *= -1; @@ -13525,7 +13531,7 @@ int sched_group_set_idle(struct task_group *tg, long idle) if (!se->on_rq) break; - cfs_rq->idle_h_nr_running += idle_task_delta; + cfs_rq->h_nr_idle += idle_task_delta; /* Already accounted at parent level and above. */ if (cfs_rq_is_idle(cfs_rq)) diff --git a/kernel/sched/features.h b/kernel/sched/features.h index a3d331dd2d8f..3c12d9f93331 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -32,6 +32,15 @@ SCHED_FEAT(PREEMPT_SHORT, true) SCHED_FEAT(NEXT_BUDDY, false) /* + * Allow completely ignoring cfs_rq->next; which can be set from various + * places: + * - NEXT_BUDDY (wakeup preemption) + * - yield_to_task() + * - cgroup dequeue / pick + */ +SCHED_FEAT(PICK_BUDDY, true) + +/* * Consider buddies to be cache hot, decreases the likeliness of a * cache buddy being migrated away, increases cache locality. */ diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 621696269584..2c85c86b455f 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -452,19 +452,20 @@ static void wakeup_preempt_idle(struct rq *rq, struct task_struct *p, int flags) static void put_prev_task_idle(struct rq *rq, struct task_struct *prev, struct task_struct *next) { dl_server_update_idle_time(rq, prev); - scx_update_idle(rq, false); + scx_update_idle(rq, false, true); } static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool first) { update_idle_core(rq); - scx_update_idle(rq, true); + scx_update_idle(rq, true, true); schedstat_inc(rq->sched_goidle); next->se.exec_start = rq_clock_task(rq); } struct task_struct *pick_task_idle(struct rq *rq) { + scx_update_idle(rq, true, false); return rq->idle; } diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index 5891e715f00d..81bc8b329ef1 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -9,15 +9,9 @@ */ enum hk_flags { - HK_FLAG_TIMER = BIT(HK_TYPE_TIMER), - HK_FLAG_RCU = BIT(HK_TYPE_RCU), - HK_FLAG_MISC = BIT(HK_TYPE_MISC), - HK_FLAG_SCHED = BIT(HK_TYPE_SCHED), - HK_FLAG_TICK = BIT(HK_TYPE_TICK), HK_FLAG_DOMAIN = BIT(HK_TYPE_DOMAIN), - HK_FLAG_WQ = BIT(HK_TYPE_WQ), HK_FLAG_MANAGED_IRQ = BIT(HK_TYPE_MANAGED_IRQ), - HK_FLAG_KTHREAD = BIT(HK_TYPE_KTHREAD), + HK_FLAG_KERNEL_NOISE = BIT(HK_TYPE_KERNEL_NOISE), }; DEFINE_STATIC_KEY_FALSE(housekeeping_overridden); @@ -97,7 +91,7 @@ void __init housekeeping_init(void) static_branch_enable(&housekeeping_overridden); - if (housekeeping.flags & HK_FLAG_TICK) + if (housekeeping.flags & HK_FLAG_KERNEL_NOISE) sched_tick_offload_init(); for_each_set_bit(type, &housekeeping.flags, HK_TYPE_MAX) { @@ -121,7 +115,7 @@ static int __init housekeeping_setup(char *str, unsigned long flags) unsigned int first_cpu; int err = 0; - if ((flags & HK_FLAG_TICK) && !(housekeeping.flags & HK_FLAG_TICK)) { + if ((flags & HK_FLAG_KERNEL_NOISE) && !(housekeeping.flags & HK_FLAG_KERNEL_NOISE)) { if (!IS_ENABLED(CONFIG_NO_HZ_FULL)) { pr_warn("Housekeeping: nohz unsupported." " Build with CONFIG_NO_HZ_FULL\n"); @@ -177,7 +171,7 @@ static int __init housekeeping_setup(char *str, unsigned long flags) housekeeping_setup_type(type, housekeeping_staging); } - if ((flags & HK_FLAG_TICK) && !(housekeeping.flags & HK_FLAG_TICK)) + if ((flags & HK_FLAG_KERNEL_NOISE) && !(housekeeping.flags & HK_FLAG_KERNEL_NOISE)) tick_nohz_full_setup(non_housekeeping_mask); housekeeping.flags |= flags; @@ -195,8 +189,7 @@ static int __init housekeeping_nohz_full_setup(char *str) { unsigned long flags; - flags = HK_FLAG_TICK | HK_FLAG_WQ | HK_FLAG_TIMER | HK_FLAG_RCU | - HK_FLAG_MISC | HK_FLAG_KTHREAD; + flags = HK_FLAG_KERNEL_NOISE; return housekeeping_setup(str, flags); } @@ -210,9 +203,12 @@ static int __init housekeeping_isolcpus_setup(char *str) int len; while (isalpha(*str)) { + /* + * isolcpus=nohz is equivalent to nohz_full. + */ if (!strncmp(str, "nohz,", 5)) { str += 5; - flags |= HK_FLAG_TICK; + flags |= HK_FLAG_KERNEL_NOISE; continue; } diff --git a/kernel/sched/pelt.c b/kernel/sched/pelt.c index fc07382361a8..7a8534a2deff 100644 --- a/kernel/sched/pelt.c +++ b/kernel/sched/pelt.c @@ -275,7 +275,7 @@ ___update_load_avg(struct sched_avg *sa, unsigned long load) * * group: [ see update_cfs_group() ] * se_weight() = tg->weight * grq->load_avg / tg->load_avg - * se_runnable() = grq->h_nr_running + * se_runnable() = grq->h_nr_runnable * * runnable_sum = se_runnable() * runnable = grq->runnable_sum * runnable_avg = runnable_sum @@ -321,7 +321,7 @@ int __update_load_avg_cfs_rq(u64 now, struct cfs_rq *cfs_rq) { if (___update_load_sum(now, &cfs_rq->avg, scale_load_down(cfs_rq->load.weight), - cfs_rq->h_nr_running, + cfs_rq->h_nr_runnable, cfs_rq->curr != NULL)) { ___update_load_avg(&cfs_rq->avg, 1); diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 84dad1511d1e..bb56805e3d47 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -998,7 +998,7 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st s64 delta; u64 irq; - if (static_branch_likely(&psi_disabled)) + if (static_branch_likely(&psi_disabled) || !irqtime_enabled()) return; if (!curr->pid) @@ -1240,6 +1240,11 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res) if (static_branch_likely(&psi_disabled)) return -EOPNOTSUPP; +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + if (!irqtime_enabled() && res == PSI_IRQ) + return -EOPNOTSUPP; +#endif + /* Update averages before reporting them */ mutex_lock(&group->avgs_lock); now = sched_clock(); diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index bd66a46b06ac..4b8e33c615b1 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -26,7 +26,7 @@ static int sched_rt_handler(const struct ctl_table *table, int write, void *buff size_t *lenp, loff_t *ppos); static int sched_rr_handler(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos); -static struct ctl_table sched_rt_sysctls[] = { +static const struct ctl_table sched_rt_sysctls[] = { { .procname = "sched_rt_period_us", .data = &sysctl_sched_rt_period, diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 76f5f53a645f..38e0e323dda2 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -362,7 +362,7 @@ extern void __getparam_dl(struct task_struct *p, struct sched_attr *attr); extern bool __checkparam_dl(const struct sched_attr *attr); extern bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr); extern int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial); -extern int dl_bw_check_overflow(int cpu); +extern int dl_bw_deactivate(int cpu); extern s64 dl_scaled_delta_exec(struct rq *rq, struct sched_dl_entity *dl_se, s64 delta_exec); /* * SCHED_DEADLINE supports servers (nested scheduling) with the following @@ -398,6 +398,11 @@ extern void __dl_server_attach_root(struct sched_dl_entity *dl_se, struct rq *rq extern int dl_server_apply_params(struct sched_dl_entity *dl_se, u64 runtime, u64 period, bool init); +static inline bool dl_server_active(struct sched_dl_entity *dl_se) +{ + return dl_se->dl_server_active; +} + #ifdef CONFIG_CGROUP_SCHED extern struct list_head task_groups; @@ -645,10 +650,10 @@ struct balance_callback { /* CFS-related fields in a runqueue */ struct cfs_rq { struct load_weight load; - unsigned int nr_running; - unsigned int h_nr_running; /* SCHED_{NORMAL,BATCH,IDLE} */ - unsigned int idle_nr_running; /* SCHED_IDLE */ - unsigned int idle_h_nr_running; /* SCHED_IDLE */ + unsigned int nr_queued; + unsigned int h_nr_queued; /* SCHED_{NORMAL,BATCH,IDLE} */ + unsigned int h_nr_runnable; /* SCHED_{NORMAL,BATCH,IDLE} */ + unsigned int h_nr_idle; /* SCHED_IDLE */ s64 avg_vruntime; u64 avg_load; @@ -754,6 +759,7 @@ enum scx_rq_flags { SCX_RQ_BAL_PENDING = 1 << 2, /* balance hasn't run yet */ SCX_RQ_BAL_KEEP = 1 << 3, /* balance decided to keep current */ SCX_RQ_BYPASSING = 1 << 4, + SCX_RQ_CLK_VALID = 1 << 5, /* RQ clock is fresh and valid */ SCX_RQ_IN_WAKEUP = 1 << 16, SCX_RQ_IN_BALANCE = 1 << 17, @@ -766,9 +772,10 @@ struct scx_rq { unsigned long ops_qseq; u64 extra_enq_flags; /* see move_task_to_local_dsq() */ u32 nr_running; - u32 flags; u32 cpuperf_target; /* [0, SCHED_CAPACITY_SCALE] */ bool cpu_released; + u32 flags; + u64 clock; /* current per-rq clock -- see scx_bpf_now() */ cpumask_var_t cpus_to_kick; cpumask_var_t cpus_to_kick_if_idle; cpumask_var_t cpus_to_preempt; @@ -899,7 +906,7 @@ struct dl_rq { static inline void se_update_runnable(struct sched_entity *se) { if (!entity_is_task(se)) - se->runnable_weight = se->my_q->h_nr_running; + se->runnable_weight = se->my_q->h_nr_runnable; } static inline long se_runnable(struct sched_entity *se) @@ -1717,6 +1724,38 @@ struct rq_flags { extern struct balance_callback balance_push_callback; +#ifdef CONFIG_SCHED_CLASS_EXT +extern const struct sched_class ext_sched_class; + +DECLARE_STATIC_KEY_FALSE(__scx_ops_enabled); /* SCX BPF scheduler loaded */ +DECLARE_STATIC_KEY_FALSE(__scx_switched_all); /* all fair class tasks on SCX */ + +#define scx_enabled() static_branch_unlikely(&__scx_ops_enabled) +#define scx_switched_all() static_branch_unlikely(&__scx_switched_all) + +static inline void scx_rq_clock_update(struct rq *rq, u64 clock) +{ + if (!scx_enabled()) + return; + WRITE_ONCE(rq->scx.clock, clock); + smp_store_release(&rq->scx.flags, rq->scx.flags | SCX_RQ_CLK_VALID); +} + +static inline void scx_rq_clock_invalidate(struct rq *rq) +{ + if (!scx_enabled()) + return; + WRITE_ONCE(rq->scx.flags, rq->scx.flags & ~SCX_RQ_CLK_VALID); +} + +#else /* !CONFIG_SCHED_CLASS_EXT */ +#define scx_enabled() false +#define scx_switched_all() false + +static inline void scx_rq_clock_update(struct rq *rq, u64 clock) {} +static inline void scx_rq_clock_invalidate(struct rq *rq) {} +#endif /* !CONFIG_SCHED_CLASS_EXT */ + /* * Lockdep annotation that avoids accidental unlocks; it's like a * sticky/continuous lockdep_assert_held(). @@ -1746,7 +1785,7 @@ static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf) if (rq->clock_update_flags > RQCF_ACT_SKIP) rf->clock_update_flags = RQCF_UPDATED; #endif - + scx_rq_clock_invalidate(rq); lockdep_unpin_lock(__rq_lockp(rq), rf->cookie); } @@ -2271,7 +2310,7 @@ static inline int task_on_cpu(struct rq *rq, struct task_struct *p) static inline int task_on_rq_queued(struct task_struct *p) { - return p->on_rq == TASK_ON_RQ_QUEUED; + return READ_ONCE(p->on_rq) == TASK_ON_RQ_QUEUED; } static inline int task_on_rq_migrating(struct task_struct *p) @@ -2505,19 +2544,6 @@ extern const struct sched_class rt_sched_class; extern const struct sched_class fair_sched_class; extern const struct sched_class idle_sched_class; -#ifdef CONFIG_SCHED_CLASS_EXT -extern const struct sched_class ext_sched_class; - -DECLARE_STATIC_KEY_FALSE(__scx_ops_enabled); /* SCX BPF scheduler loaded */ -DECLARE_STATIC_KEY_FALSE(__scx_switched_all); /* all fair class tasks on SCX */ - -#define scx_enabled() static_branch_unlikely(&__scx_ops_enabled) -#define scx_switched_all() static_branch_unlikely(&__scx_switched_all) -#else /* !CONFIG_SCHED_CLASS_EXT */ -#define scx_enabled() false -#define scx_switched_all() false -#endif /* !CONFIG_SCHED_CLASS_EXT */ - /* * Iterate only active classes. SCX can take over all fair tasks or be * completely disabled. If the former, skip fair. If the latter, skip SCX. @@ -2565,7 +2591,7 @@ static inline bool sched_rt_runnable(struct rq *rq) static inline bool sched_fair_runnable(struct rq *rq) { - return rq->cfs.nr_running > 0; + return rq->cfs.nr_queued > 0; } extern struct task_struct *pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf); @@ -3233,6 +3259,12 @@ struct irqtime { }; DECLARE_PER_CPU(struct irqtime, cpu_irqtime); +DECLARE_STATIC_KEY_FALSE(sched_clock_irqtime); + +static inline int irqtime_enabled(void) +{ + return static_branch_likely(&sched_clock_irqtime); +} /* * Returns the irqtime minus the softirq time computed by ksoftirqd. @@ -3253,6 +3285,13 @@ static inline u64 irq_time_read(int cpu) return total; } +#else + +static inline int irqtime_enabled(void) +{ + return 0; +} + #endif /* CONFIG_IRQ_TIME_ACCOUNTING */ #ifdef CONFIG_CPU_FREQ @@ -3500,6 +3539,8 @@ unsigned long scale_irq_capacity(unsigned long util, unsigned long irq, unsigned #endif /* !CONFIG_HAVE_SCHED_AVG_IRQ */ +extern void __setparam_fair(struct task_struct *p, const struct sched_attr *attr); + #if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) #define perf_domain_span(pd) (to_cpumask(((pd)->em_pd->cpus))) diff --git a/kernel/sched/stats.c b/kernel/sched/stats.c index eb0cdcd4d921..4346fd81c31f 100644 --- a/kernel/sched/stats.c +++ b/kernel/sched/stats.c @@ -103,7 +103,7 @@ void __update_stats_enqueue_sleeper(struct rq *rq, struct task_struct *p, * Bump this up when changing the output format or the meaning of an existing * format, so that tools can adapt (or abort) */ -#define SCHEDSTAT_VERSION 16 +#define SCHEDSTAT_VERSION 17 static int show_schedstat(struct seq_file *seq, void *v) { @@ -138,14 +138,17 @@ static int show_schedstat(struct seq_file *seq, void *v) for_each_domain(cpu, sd) { enum cpu_idle_type itype; - seq_printf(seq, "domain%d %*pb", dcount++, + seq_printf(seq, "domain%d %s %*pb", dcount++, sd->name, cpumask_pr_args(sched_domain_span(sd))); for (itype = 0; itype < CPU_MAX_IDLE_TYPES; itype++) { - seq_printf(seq, " %u %u %u %u %u %u %u %u", + seq_printf(seq, " %u %u %u %u %u %u %u %u %u %u %u", sd->lb_count[itype], sd->lb_balanced[itype], sd->lb_failed[itype], - sd->lb_imbalance[itype], + sd->lb_imbalance_load[itype], + sd->lb_imbalance_util[itype], + sd->lb_imbalance_task[itype], + sd->lb_imbalance_misfit[itype], sd->lb_gained[itype], sd->lb_hot_gained[itype], sd->lb_nobusyq[itype], diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h index 8ee0add5a48a..19cdbe96f93d 100644 --- a/kernel/sched/stats.h +++ b/kernel/sched/stats.h @@ -138,6 +138,10 @@ static inline void psi_enqueue(struct task_struct *p, int flags) if (flags & ENQUEUE_RESTORE) return; + /* psi_sched_switch() will handle the flags */ + if (task_on_cpu(task_rq(p), p)) + return; + if (p->se.sched_delayed) { /* CPU migration of "sleeping" task */ SCHED_WARN_ON(!(flags & ENQUEUE_MIGRATED)); @@ -244,7 +248,10 @@ static inline void sched_info_dequeue(struct rq *rq, struct task_struct *t) delta = rq_clock(rq) - t->sched_info.last_queued; t->sched_info.last_queued = 0; t->sched_info.run_delay += delta; - + if (delta > t->sched_info.max_run_delay) + t->sched_info.max_run_delay = delta; + if (delta && (!t->sched_info.min_run_delay || delta < t->sched_info.min_run_delay)) + t->sched_info.min_run_delay = delta; rq_sched_info_dequeue(rq, delta); } @@ -266,6 +273,10 @@ static void sched_info_arrive(struct rq *rq, struct task_struct *t) t->sched_info.run_delay += delta; t->sched_info.last_arrival = now; t->sched_info.pcount++; + if (delta > t->sched_info.max_run_delay) + t->sched_info.max_run_delay = delta; + if (delta && (!t->sched_info.min_run_delay || delta < t->sched_info.min_run_delay)) + t->sched_info.min_run_delay = delta; rq_sched_info_arrive(rq, delta); } diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c index ff0e5ab4e37c..456d339be98f 100644 --- a/kernel/sched/syscalls.c +++ b/kernel/sched/syscalls.c @@ -300,20 +300,10 @@ static void __setscheduler_params(struct task_struct *p, p->policy = policy; - if (dl_policy(policy)) { + if (dl_policy(policy)) __setparam_dl(p, attr); - } else if (fair_policy(policy)) { - p->static_prio = NICE_TO_PRIO(attr->sched_nice); - if (attr->sched_runtime) { - p->se.custom_slice = 1; - p->se.slice = clamp_t(u64, attr->sched_runtime, - NSEC_PER_MSEC/10, /* HZ=1000 * 10 */ - NSEC_PER_MSEC*100); /* HZ=100 / 10 */ - } else { - p->se.custom_slice = 0; - p->se.slice = sysctl_sched_base_slice; - } - } + else if (fair_policy(policy)) + __setparam_fair(p, attr); /* rt-policy tasks do not have a timerslack */ if (rt_or_dl_task_policy(p)) { @@ -1140,6 +1130,13 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask) return 0; /* + * The special/sugov task isn't part of regular bandwidth/admission + * control so let userspace change affinities. + */ + if (dl_entity_is_special(&p->dl)) + return 0; + + /* * Since bandwidth control happens on root_domain basis, * if admission test is enabled, we only admit -deadline * tasks allowed to run on all the CPUs in the task's @@ -1433,7 +1430,7 @@ int __sched yield_to(struct task_struct *p, bool preempt) struct rq *rq, *p_rq; int yielded = 0; - scoped_guard (irqsave) { + scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { rq = this_rq(); again: diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index 9748a4c8d668..c49aea8c1025 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -312,7 +312,7 @@ static int sched_energy_aware_handler(const struct ctl_table *table, int write, return ret; } -static struct ctl_table sched_energy_aware_sysctls[] = { +static const struct ctl_table sched_energy_aware_sysctls[] = { { .procname = "sched_energy_aware", .data = &sysctl_sched_energy_aware, @@ -1635,9 +1635,7 @@ sd_init(struct sched_domain_topology_level *tl, .max_newidle_lb_cost = 0, .last_decay_max_lb_cost = jiffies, .child = child, -#ifdef CONFIG_SCHED_DEBUG .name = tl->name, -#endif }; sd_span = sched_domain_span(sd); @@ -2338,10 +2336,8 @@ static struct sched_domain *build_sched_domain(struct sched_domain_topology_leve if (!cpumask_subset(sched_domain_span(child), sched_domain_span(sd))) { pr_err("BUG: arch topology borken\n"); -#ifdef CONFIG_SCHED_DEBUG pr_err(" the %s domain not a subset of the %s domain\n", child->name, sd->name); -#endif /* Fixup, ensure @sd has at least @child CPUs. */ cpumask_or(sched_domain_span(sd), sched_domain_span(sd), @@ -2721,9 +2717,11 @@ void partition_sched_domains_locked(int ndoms_new, cpumask_var_t doms_new[], /* * This domain won't be destroyed and as such - * its dl_bw->total_bw needs to be cleared. It - * will be recomputed in function - * update_tasks_root_domain(). + * its dl_bw->total_bw needs to be cleared. + * Tasks contribution will be then recomputed + * in function dl_update_tasks_root_domain(), + * dl_servers contribution in function + * dl_restore_server_root_domain(). */ rd = cpu_rq(cpumask_any(doms_cur[i]))->rd; dl_clear_root_domain(rd); |