diff options
Diffstat (limited to 'kernel/sched/fair.c')
| -rw-r--r-- | kernel/sched/fair.c | 10885 |
1 files changed, 7144 insertions, 3741 deletions
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 50aa2aba69bd..da46c3164537 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -20,25 +20,43 @@ * Adaptive scheduling granularity, math enhancements by Peter Zijlstra * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra */ -#include "sched.h" - -#include <trace/events/sched.h> +#include <linux/energy_model.h> +#include <linux/mmap_lock.h> +#include <linux/hugetlb_inline.h> +#include <linux/jiffies.h> +#include <linux/mm_api.h> +#include <linux/highmem.h> +#include <linux/spinlock_api.h> +#include <linux/cpumask_api.h> +#include <linux/lockdep_api.h> +#include <linux/softirq.h> +#include <linux/refcount_api.h> +#include <linux/topology.h> +#include <linux/sched/clock.h> +#include <linux/sched/cond_resched.h> +#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> +#include <linux/memory-tiers.h> +#include <linux/mempolicy.h> +#include <linux/mutex_api.h> +#include <linux/profile.h> +#include <linux/psi.h> +#include <linux/ratelimit.h> +#include <linux/task_work.h> +#include <linux/rbtree_augmented.h> + +#include <asm/switch_to.h> + +#include <uapi/linux/sched/types.h> -/* - * Targeted preemption latency for CPU-bound tasks: - * - * NOTE: this latency value is not the same as the concept of - * 'timeslice length' - timeslices in CFS are of variable length - * and have no persistent notion like in traditional, time-slice - * based scheduling concepts. - * - * (to see the precise effective timeslice length of your workload, - * run vmstat and monitor the context-switches (cs) field) - * - * (default: 6ms * (1 + ilog(ncpus)), units: nanoseconds) - */ -unsigned int sysctl_sched_latency = 6000000ULL; -static unsigned int normalized_sysctl_sched_latency = 6000000ULL; +#include "sched.h" +#include "stats.h" +#include "autogroup.h" /* * The initial- and re-scaling of tunables is configurable @@ -46,47 +64,30 @@ static unsigned int normalized_sysctl_sched_latency = 6000000ULL; * Options are: * * SCHED_TUNABLESCALING_NONE - unscaled, always *1 - * SCHED_TUNABLESCALING_LOG - scaled logarithmical, *1+ilog(ncpus) + * SCHED_TUNABLESCALING_LOG - scaled logarithmically, *1+ilog(ncpus) * SCHED_TUNABLESCALING_LINEAR - scaled linear, *ncpus * * (default SCHED_TUNABLESCALING_LOG = *(1+ilog(ncpus)) */ -enum sched_tunable_scaling sysctl_sched_tunable_scaling = SCHED_TUNABLESCALING_LOG; +unsigned int sysctl_sched_tunable_scaling = SCHED_TUNABLESCALING_LOG; /* * Minimal preemption granularity for CPU-bound tasks: * - * (default: 0.75 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 0.70 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_min_granularity = 750000ULL; -static unsigned int normalized_sysctl_sched_min_granularity = 750000ULL; +unsigned int sysctl_sched_base_slice = 700000ULL; +static unsigned int normalized_sysctl_sched_base_slice = 700000ULL; -/* - * This value is kept at sysctl_sched_latency/sysctl_sched_min_granularity - */ -static unsigned int sched_nr_latency = 8; - -/* - * After fork, child runs first. If set to 0 (default) then - * parent will (try to) run first. - */ -unsigned int sysctl_sched_child_runs_first __read_mostly; +__read_mostly unsigned int sysctl_sched_migration_cost = 500000UL; -/* - * SCHED_OTHER wake-up granularity. - * - * This option delays the preemption effects of decoupled workloads - * and reduces their over-scheduling. Synchronous workloads will still - * have immediate wakeup/sleep latencies. - * - * (default: 1 msec * (1 + ilog(ncpus)), units: nanoseconds) - */ -unsigned int sysctl_sched_wakeup_granularity = 1000000UL; -static unsigned int normalized_sysctl_sched_wakeup_granularity = 1000000UL; - -const_debug unsigned int sysctl_sched_migration_cost = 500000UL; +static int __init setup_sched_thermal_decay_shift(char *str) +{ + pr_warn("Ignoring the deprecated sched_thermal_decay_shift= option\n"); + return 1; +} +__setup("sched_thermal_decay_shift=", setup_sched_thermal_decay_shift); -#ifdef CONFIG_SMP /* * For asym packing, by default the lower numbered CPU has higher priority. */ @@ -96,13 +97,19 @@ int __weak arch_asym_cpu_priority(int cpu) } /* - * The margin used when comparing utilization with CPU capacity: - * util * margin < capacity * 1024 + * The margin used when comparing utilization with CPU capacity. * * (default: ~20%) */ -static unsigned int capacity_margin = 1280; -#endif +#define fits_capacity(cap, max) ((cap) * 1280 < (max) * 1024) + +/* + * The margin used when comparing CPU capacities. + * is 'cap1' noticeably greater than 'cap2' + * + * (default: ~5%) + */ +#define capacity_greater(cap1, cap2) ((cap1) * 1024 > (cap2) * 1078) #ifdef CONFIG_CFS_BANDWIDTH /* @@ -115,9 +122,46 @@ static unsigned int capacity_margin = 1280; * * (default: 5 msec, units: microseconds) */ -unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL; +static unsigned int sysctl_sched_cfs_bandwidth_slice = 5000UL; #endif +#ifdef CONFIG_NUMA_BALANCING +/* Restrict the NUMA promotion throughput (MB/s) for each target node. */ +static unsigned int sysctl_numa_balancing_promote_rate_limit = 65536; +#endif + +#ifdef CONFIG_SYSCTL +static const struct ctl_table sched_fair_sysctls[] = { +#ifdef CONFIG_CFS_BANDWIDTH + { + .procname = "sched_cfs_bandwidth_slice_us", + .data = &sysctl_sched_cfs_bandwidth_slice, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ONE, + }, +#endif +#ifdef CONFIG_NUMA_BALANCING + { + .procname = "numa_balancing_promote_rate_limit_MBps", + .data = &sysctl_numa_balancing_promote_rate_limit, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + }, +#endif /* CONFIG_NUMA_BALANCING */ +}; + +static int __init sched_fair_sysctl_init(void) +{ + register_sysctl_init("kernel", sched_fair_sysctls); + return 0; +} +late_initcall(sched_fair_sysctl_init); +#endif /* CONFIG_SYSCTL */ + static inline void update_load_add(struct load_weight *lw, unsigned long inc) { lw->weight += inc; @@ -172,13 +216,11 @@ static void update_sysctl(void) #define SET_SYSCTL(name) \ (sysctl_##name = (factor) * normalized_sysctl_##name) - SET_SYSCTL(sched_min_granularity); - SET_SYSCTL(sched_latency); - SET_SYSCTL(sched_wakeup_granularity); + SET_SYSCTL(sched_base_slice); #undef SET_SYSCTL } -void sched_init_granularity(void) +void __init sched_init_granularity(void) { update_sysctl(); } @@ -218,28 +260,40 @@ static void __update_inv_weight(struct load_weight *lw) static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight *lw) { u64 fact = scale_load_down(weight); + u32 fact_hi = (u32)(fact >> 32); int shift = WMULT_SHIFT; + int fs; __update_inv_weight(lw); - if (unlikely(fact >> 32)) { - while (fact >> 32) { - fact >>= 1; - shift--; - } + if (unlikely(fact_hi)) { + fs = fls(fact_hi); + shift -= fs; + fact >>= fs; } - /* hint to use a 32x32->64 mul */ - fact = (u64)(u32)fact * lw->inv_weight; + fact = mul_u32_u32(fact, lw->inv_weight); - while (fact >> 32) { - fact >>= 1; - shift--; + fact_hi = (u32)(fact >> 32); + if (fact_hi) { + fs = fls(fact_hi); + shift -= fs; + fact >>= fs; } return mul_u64_u32_shr(delta_exec, fact, shift); } +/* + * delta /= w + */ +static inline u64 calc_delta_fair(u64 delta, struct sched_entity *se) +{ + if (unlikely(se->load.weight != NICE_0_LOAD)) + delta = __calc_delta(delta, NICE_0_LOAD, &se->load); + + return delta; +} const struct sched_class fair_sched_class; @@ -249,112 +303,107 @@ const struct sched_class fair_sched_class; #ifdef CONFIG_FAIR_GROUP_SCHED -/* cpu runqueue to which this cfs_rq is attached */ -static inline struct rq *rq_of(struct cfs_rq *cfs_rq) -{ - return cfs_rq->rq; -} - -static inline struct task_struct *task_of(struct sched_entity *se) -{ - SCHED_WARN_ON(!entity_is_task(se)); - return container_of(se, struct task_struct, se); -} - /* Walk up scheduling entities hierarchy */ #define for_each_sched_entity(se) \ for (; se; se = se->parent) -static inline struct cfs_rq *task_cfs_rq(struct task_struct *p) +static inline bool list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) { - return p->se.cfs_rq; -} + struct rq *rq = rq_of(cfs_rq); + int cpu = cpu_of(rq); -/* runqueue on which this entity is (to be) queued */ -static inline struct cfs_rq *cfs_rq_of(struct sched_entity *se) -{ - return se->cfs_rq; -} + if (cfs_rq->on_list) + return rq->tmp_alone_branch == &rq->leaf_cfs_rq_list; -/* runqueue "owned" by this group */ -static inline struct cfs_rq *group_cfs_rq(struct sched_entity *grp) -{ - return grp->my_q; -} + cfs_rq->on_list = 1; -static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) -{ - if (!cfs_rq->on_list) { - struct rq *rq = rq_of(cfs_rq); - int cpu = cpu_of(rq); + /* + * Ensure we either appear before our parent (if already + * enqueued) or force our parent to appear after us when it is + * enqueued. The fact that we always enqueue bottom-up + * reduces this to two cases and a special case for the root + * cfs_rq. Furthermore, it also means that we will always reset + * tmp_alone_branch either when the branch is connected + * to a tree or when we reach the top of the tree + */ + if (cfs_rq->tg->parent && + cfs_rq->tg->parent->cfs_rq[cpu]->on_list) { /* - * Ensure we either appear before our parent (if already - * enqueued) or force our parent to appear after us when it is - * enqueued. The fact that we always enqueue bottom-up - * reduces this to two cases and a special case for the root - * cfs_rq. Furthermore, it also means that we will always reset - * tmp_alone_branch either when the branch is connected - * to a tree or when we reach the beg of the tree + * If parent is already on the list, we add the child + * just before. Thanks to circular linked property of + * the list, this means to put the child at the tail + * of the list that starts by parent. */ - if (cfs_rq->tg->parent && - cfs_rq->tg->parent->cfs_rq[cpu]->on_list) { - /* - * If parent is already on the list, we add the child - * just before. Thanks to circular linked property of - * the list, this means to put the child at the tail - * of the list that starts by parent. - */ - list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list, - &(cfs_rq->tg->parent->cfs_rq[cpu]->leaf_cfs_rq_list)); - /* - * The branch is now connected to its tree so we can - * reset tmp_alone_branch to the beginning of the - * list. - */ - rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; - } else if (!cfs_rq->tg->parent) { - /* - * cfs rq without parent should be put - * at the tail of the list. - */ - list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list, - &rq->leaf_cfs_rq_list); - /* - * We have reach the beg of a tree so we can reset - * tmp_alone_branch to the beginning of the list. - */ - rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; - } else { - /* - * The parent has not already been added so we want to - * make sure that it will be put after us. - * tmp_alone_branch points to the beg of the branch - * where we will add parent. - */ - list_add_rcu(&cfs_rq->leaf_cfs_rq_list, - rq->tmp_alone_branch); - /* - * update tmp_alone_branch to points to the new beg - * of the branch - */ - rq->tmp_alone_branch = &cfs_rq->leaf_cfs_rq_list; - } + list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list, + &(cfs_rq->tg->parent->cfs_rq[cpu]->leaf_cfs_rq_list)); + /* + * The branch is now connected to its tree so we can + * reset tmp_alone_branch to the beginning of the + * list. + */ + rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; + return true; + } - cfs_rq->on_list = 1; + if (!cfs_rq->tg->parent) { + /* + * cfs rq without parent should be put + * at the tail of the list. + */ + list_add_tail_rcu(&cfs_rq->leaf_cfs_rq_list, + &rq->leaf_cfs_rq_list); + /* + * We have reach the top of a tree so we can reset + * tmp_alone_branch to the beginning of the list. + */ + rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; + return true; } + + /* + * The parent has not already been added so we want to + * make sure that it will be put after us. + * tmp_alone_branch points to the begin of the branch + * where we will add parent. + */ + list_add_rcu(&cfs_rq->leaf_cfs_rq_list, rq->tmp_alone_branch); + /* + * update tmp_alone_branch to points to the new begin + * of the branch + */ + rq->tmp_alone_branch = &cfs_rq->leaf_cfs_rq_list; + return false; } static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) { if (cfs_rq->on_list) { + struct rq *rq = rq_of(cfs_rq); + + /* + * With cfs_rq being unthrottled/throttled during an enqueue, + * it can happen the tmp_alone_branch points to the leaf that + * we finally want to delete. In this case, tmp_alone_branch moves + * to the prev element but it will point to rq->leaf_cfs_rq_list + * at the end of the enqueue. + */ + if (rq->tmp_alone_branch == &cfs_rq->leaf_cfs_rq_list) + rq->tmp_alone_branch = cfs_rq->leaf_cfs_rq_list.prev; + list_del_rcu(&cfs_rq->leaf_cfs_rq_list); cfs_rq->on_list = 0; } } -/* Iterate through all leaf cfs_rq's on a runqueue: */ -#define for_each_leaf_cfs_rq(rq, cfs_rq) \ - list_for_each_entry_rcu(cfs_rq, &rq->leaf_cfs_rq_list, leaf_cfs_rq_list) +static inline void assert_list_leaf_cfs_rq(struct rq *rq) +{ + WARN_ON_ONCE(rq->tmp_alone_branch != &rq->leaf_cfs_rq_list); +} + +/* Iterate through all leaf cfs_rq's on a runqueue */ +#define for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) \ + list_for_each_entry_safe(cfs_rq, pos, &rq->leaf_cfs_rq_list, \ + leaf_cfs_rq_list) /* Do the two (enqueued) entities belong to the same group ? */ static inline struct cfs_rq * @@ -366,7 +415,7 @@ is_same_group(struct sched_entity *se, struct sched_entity *pse) return NULL; } -static inline struct sched_entity *parent_entity(struct sched_entity *se) +static inline struct sched_entity *parent_entity(const struct sched_entity *se) { return se->parent; } @@ -403,63 +452,70 @@ find_matching_se(struct sched_entity **se, struct sched_entity **pse) } } -#else /* !CONFIG_FAIR_GROUP_SCHED */ +static int tg_is_idle(struct task_group *tg) +{ + return tg->idle > 0; +} -static inline struct task_struct *task_of(struct sched_entity *se) +static int cfs_rq_is_idle(struct cfs_rq *cfs_rq) { - return container_of(se, struct task_struct, se); + return cfs_rq->idle > 0; } -static inline struct rq *rq_of(struct cfs_rq *cfs_rq) +static int se_is_idle(struct sched_entity *se) { - return container_of(cfs_rq, struct rq, cfs); + if (entity_is_task(se)) + return task_has_idle_policy(task_of(se)); + return cfs_rq_is_idle(group_cfs_rq(se)); } +#else /* !CONFIG_FAIR_GROUP_SCHED: */ #define for_each_sched_entity(se) \ for (; se; se = NULL) -static inline struct cfs_rq *task_cfs_rq(struct task_struct *p) +static inline bool list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) { - return &task_rq(p)->cfs; + return true; } -static inline struct cfs_rq *cfs_rq_of(struct sched_entity *se) +static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) { - struct task_struct *p = task_of(se); - struct rq *rq = task_rq(p); +} - return &rq->cfs; +static inline void assert_list_leaf_cfs_rq(struct rq *rq) +{ } -/* runqueue "owned" by this group */ -static inline struct cfs_rq *group_cfs_rq(struct sched_entity *grp) +#define for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) \ + for (cfs_rq = &rq->cfs, pos = NULL; cfs_rq; cfs_rq = pos) + +static inline struct sched_entity *parent_entity(struct sched_entity *se) { return NULL; } -static inline void list_add_leaf_cfs_rq(struct cfs_rq *cfs_rq) +static inline void +find_matching_se(struct sched_entity **se, struct sched_entity **pse) { } -static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) +static inline int tg_is_idle(struct task_group *tg) { + return 0; } -#define for_each_leaf_cfs_rq(rq, cfs_rq) \ - for (cfs_rq = &rq->cfs; cfs_rq; cfs_rq = NULL) - -static inline struct sched_entity *parent_entity(struct sched_entity *se) +static int cfs_rq_is_idle(struct cfs_rq *cfs_rq) { - return NULL; + return 0; } -static inline void -find_matching_se(struct sched_entity **se, struct sched_entity **pse) +static int se_is_idle(struct sched_entity *se) { + return task_has_idle_policy(task_of(se)); } -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* !CONFIG_FAIR_GROUP_SCHED */ static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec); @@ -468,7 +524,7 @@ 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) @@ -477,7 +533,7 @@ 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) @@ -486,209 +542,511 @@ static inline u64 min_vruntime(u64 min_vruntime, u64 vruntime) return min_vruntime; } -static inline int entity_before(struct sched_entity *a, - struct sched_entity *b) +static inline bool entity_before(const struct sched_entity *a, + const struct sched_entity *b) { - return (s64)(a->vruntime - b->vruntime) < 0; + /* + * Tiebreak on vruntime seems unnecessary since it can + * hardly happen. + */ + return (s64)(a->deadline - b->deadline) < 0; } -static void update_min_vruntime(struct cfs_rq *cfs_rq) +static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + return (s64)(se->vruntime - cfs_rq->zero_vruntime); +} + +#define __node_2_se(node) \ + rb_entry((node), struct sched_entity, run_node) + +/* + * Compute virtual time from the per-task service numbers: + * + * Fair schedulers conserve lag: + * + * \Sum lag_i = 0 + * + * Where lag_i is given by: + * + * lag_i = S - s_i = w_i * (V - v_i) + * + * Where S is the ideal service time and V is it's virtual time counterpart. + * Therefore: + * + * \Sum lag_i = 0 + * \Sum w_i * (V - v_i) = 0 + * \Sum w_i * V - w_i * v_i = 0 + * + * From which we can solve an expression for V in v_i (which we have in + * se->vruntime): + * + * \Sum v_i * w_i \Sum v_i * w_i + * V = -------------- = -------------- + * \Sum w_i W + * + * Specifically, this is the weighted average of all entity virtual runtimes. + * + * [[ NOTE: this is only equal to the ideal scheduler under the condition + * that join/leave operations happen at lag_i = 0, otherwise the + * virtual time has non-contiguous motion equivalent to: + * + * V +-= lag_i / W + * + * Also see the comment in place_entity() that deals with this. ]] + * + * However, since v_i is u64, and the multiplication could easily overflow + * transform it into a relative form that uses smaller quantities: + * + * Substitute: v_i == (v_i - v0) + v0 + * + * \Sum ((v_i - v0) + v0) * w_i \Sum (v_i - v0) * w_i + * V = ---------------------------- = --------------------- + v0 + * W W + * + * Which we track using: + * + * v0 := cfs_rq->zero_vruntime + * \Sum (v_i - v0) * w_i := cfs_rq->avg_vruntime + * \Sum w_i := cfs_rq->avg_load + * + * Since zero_vruntime closely tracks the per-task service, these + * deltas: (v_i - v), will be in the order of the maximal (virtual) lag + * induced in the system due to quantisation. + * + * Also, we use scale_load_down() to reduce the size. + * + * As measured, the max (key * weight) value was ~44 bits for a kernel build. + */ +static void +avg_vruntime_add(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + unsigned long weight = scale_load_down(se->load.weight); + s64 key = entity_key(cfs_rq, se); + + cfs_rq->avg_vruntime += key * weight; + cfs_rq->avg_load += weight; +} + +static void +avg_vruntime_sub(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + unsigned long weight = scale_load_down(se->load.weight); + s64 key = entity_key(cfs_rq, se); + + cfs_rq->avg_vruntime -= key * weight; + cfs_rq->avg_load -= weight; +} + +static inline +void avg_vruntime_update(struct cfs_rq *cfs_rq, s64 delta) +{ + /* + * v' = v + d ==> avg_vruntime' = avg_runtime - d*avg_load + */ + cfs_rq->avg_vruntime -= cfs_rq->avg_load * delta; +} + +/* + * Specifically: avg_runtime() + 0 must result in entity_eligible() := true + * For this to be so, the result of this function must have a left bias. + */ +u64 avg_vruntime(struct cfs_rq *cfs_rq) { struct sched_entity *curr = cfs_rq->curr; - struct rb_node *leftmost = rb_first_cached(&cfs_rq->tasks_timeline); + s64 avg = cfs_rq->avg_vruntime; + long load = cfs_rq->avg_load; - u64 vruntime = cfs_rq->min_vruntime; + if (curr && curr->on_rq) { + unsigned long weight = scale_load_down(curr->load.weight); - if (curr) { - if (curr->on_rq) - vruntime = curr->vruntime; - else - curr = NULL; + avg += entity_key(cfs_rq, curr) * weight; + load += weight; } - if (leftmost) { /* non-empty tree */ - struct sched_entity *se; - se = rb_entry(leftmost, struct sched_entity, run_node); - - if (!curr) - vruntime = se->vruntime; - else - vruntime = min_vruntime(vruntime, se->vruntime); + if (load) { + /* sign flips effective floor / ceiling */ + if (avg < 0) + avg -= (load - 1); + avg = div_s64(avg, load); } - /* ensure we never gain time by being placed backwards. */ - cfs_rq->min_vruntime = max_vruntime(cfs_rq->min_vruntime, vruntime); -#ifndef CONFIG_64BIT - smp_wmb(); - cfs_rq->min_vruntime_copy = cfs_rq->min_vruntime; -#endif + return cfs_rq->zero_vruntime + avg; } /* - * Enqueue an entity into the rb-tree: + * lag_i = S - s_i = w_i * (V - v_i) + * + * However, since V is approximated by the weighted average of all entities it + * is possible -- by addition/removal/reweight to the tree -- to move V around + * and end up with a larger lag than we started with. + * + * Limit this to either double the slice length with a minimum of TICK_NSEC + * since that is the timing granularity. + * + * EEVDF gives the following limit for a steady state system: + * + * -r_max < lag < max(r_max, q) + * + * XXX could add max_slice to the augmented data to track this. */ -static void __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) +static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) { - struct rb_node **link = &cfs_rq->tasks_timeline.rb_root.rb_node; - struct rb_node *parent = NULL; - struct sched_entity *entry; - bool leftmost = true; + s64 vlag, limit; - /* - * Find the right place in the rbtree: - */ - while (*link) { - parent = *link; - entry = rb_entry(parent, struct sched_entity, run_node); - /* - * We dont care about collisions. Nodes with - * the same key stay together. - */ - if (entity_before(se, entry)) { - link = &parent->rb_left; - } else { - link = &parent->rb_right; - leftmost = false; - } + WARN_ON_ONCE(!se->on_rq); + + 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); +} + +/* + * Entity is eligible once it received less service than it ought to have, + * eg. lag >= 0. + * + * lag_i = S - s_i = w_i*(V - v_i) + * + * lag_i >= 0 -> V >= v_i + * + * \Sum (v_i - v)*w_i + * V = ------------------ + v + * \Sum w_i + * + * lag_i >= 0 -> \Sum (v_i - v)*w_i >= (v_i - v)*(\Sum w_i) + * + * Note: using 'avg_vruntime() > se->vruntime' is inaccurate due + * to the loss in precision caused by the division. + */ +static int vruntime_eligible(struct cfs_rq *cfs_rq, u64 vruntime) +{ + struct sched_entity *curr = cfs_rq->curr; + s64 avg = cfs_rq->avg_vruntime; + long load = cfs_rq->avg_load; + + if (curr && curr->on_rq) { + unsigned long weight = scale_load_down(curr->load.weight); + + avg += entity_key(cfs_rq, curr) * weight; + load += weight; } - rb_link_node(&se->run_node, parent, link); - rb_insert_color_cached(&se->run_node, - &cfs_rq->tasks_timeline, leftmost); + return avg >= (s64)(vruntime - cfs_rq->zero_vruntime) * load; } -static void __dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) +int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se) { - rb_erase_cached(&se->run_node, &cfs_rq->tasks_timeline); + return vruntime_eligible(cfs_rq, se->vruntime); } -struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) +static void update_zero_vruntime(struct cfs_rq *cfs_rq) { - struct rb_node *left = rb_first_cached(&cfs_rq->tasks_timeline); + u64 vruntime = avg_vruntime(cfs_rq); + s64 delta = (s64)(vruntime - cfs_rq->zero_vruntime); - if (!left) - return NULL; + avg_vruntime_update(cfs_rq, delta); - return rb_entry(left, struct sched_entity, run_node); + cfs_rq->zero_vruntime = vruntime; } -static struct sched_entity *__pick_next_entity(struct sched_entity *se) +static inline u64 cfs_rq_min_slice(struct cfs_rq *cfs_rq) { - struct rb_node *next = rb_next(&se->run_node); + struct sched_entity *root = __pick_root_entity(cfs_rq); + struct sched_entity *curr = cfs_rq->curr; + u64 min_slice = ~0ULL; - if (!next) - return NULL; + if (curr && curr->on_rq) + min_slice = curr->slice; + + if (root) + min_slice = min(min_slice, root->min_slice); - return rb_entry(next, struct sched_entity, run_node); + return min_slice; } -#ifdef CONFIG_SCHED_DEBUG -struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) +static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) { - struct rb_node *last = rb_last(&cfs_rq->tasks_timeline.rb_root); + return entity_before(__node_2_se(a), __node_2_se(b)); +} - if (!last) - return NULL; +#define vruntime_gt(field, lse, rse) ({ (s64)((lse)->field - (rse)->field) > 0; }) - return rb_entry(last, struct sched_entity, run_node); +static inline void __min_vruntime_update(struct sched_entity *se, struct rb_node *node) +{ + if (node) { + struct sched_entity *rse = __node_2_se(node); + if (vruntime_gt(min_vruntime, se, rse)) + se->min_vruntime = rse->min_vruntime; + } } -/************************************************************** - * Scheduling class statistics methods: - */ - -int sched_proc_update_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) +static inline void __min_slice_update(struct sched_entity *se, struct rb_node *node) { - int ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); - unsigned int factor = get_update_sysctl_factor(); + if (node) { + struct sched_entity *rse = __node_2_se(node); + if (rse->min_slice < se->min_slice) + se->min_slice = rse->min_slice; + } +} - if (ret || !write) - return ret; +/* + * se->min_vruntime = min(se->vruntime, {left,right}->min_vruntime) + */ +static inline bool min_vruntime_update(struct sched_entity *se, bool exit) +{ + u64 old_min_vruntime = se->min_vruntime; + u64 old_min_slice = se->min_slice; + struct rb_node *node = &se->run_node; - sched_nr_latency = DIV_ROUND_UP(sysctl_sched_latency, - sysctl_sched_min_granularity); + se->min_vruntime = se->vruntime; + __min_vruntime_update(se, node->rb_right); + __min_vruntime_update(se, node->rb_left); -#define WRT_SYSCTL(name) \ - (normalized_sysctl_##name = sysctl_##name / (factor)) - WRT_SYSCTL(sched_min_granularity); - WRT_SYSCTL(sched_latency); - WRT_SYSCTL(sched_wakeup_granularity); -#undef WRT_SYSCTL + se->min_slice = se->slice; + __min_slice_update(se, node->rb_right); + __min_slice_update(se, node->rb_left); - return 0; + return se->min_vruntime == old_min_vruntime && + se->min_slice == old_min_slice; } -#endif + +RB_DECLARE_CALLBACKS(static, min_vruntime_cb, struct sched_entity, + run_node, min_vruntime, min_vruntime_update); /* - * delta /= w + * Enqueue an entity into the rb-tree: */ -static inline u64 calc_delta_fair(u64 delta, struct sched_entity *se) +static void __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (unlikely(se->load.weight != NICE_0_LOAD)) - delta = __calc_delta(delta, NICE_0_LOAD, &se->load); + avg_vruntime_add(cfs_rq, se); + update_zero_vruntime(cfs_rq); + se->min_vruntime = se->vruntime; + se->min_slice = se->slice; + rb_add_augmented_cached(&se->run_node, &cfs_rq->tasks_timeline, + __entity_less, &min_vruntime_cb); +} - return delta; +static void __dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + rb_erase_augmented_cached(&se->run_node, &cfs_rq->tasks_timeline, + &min_vruntime_cb); + avg_vruntime_sub(cfs_rq, se); + update_zero_vruntime(cfs_rq); +} + +struct sched_entity *__pick_root_entity(struct cfs_rq *cfs_rq) +{ + struct rb_node *root = cfs_rq->tasks_timeline.rb_root.rb_node; + + if (!root) + return NULL; + + return __node_2_se(root); +} + +struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) +{ + struct rb_node *left = rb_first_cached(&cfs_rq->tasks_timeline); + + if (!left) + return NULL; + + return __node_2_se(left); } /* - * The idea is to set a period in which each task runs once. - * - * When there are too many tasks (sched_nr_latency) we have to stretch - * this period because otherwise the slices get too small. - * - * p = (nr <= nl) ? l : l*nr/nl + * Set the vruntime up to which an entity can run before looking + * for another entity to pick. + * In case of run to parity, we use the shortest slice of the enqueued + * entities to set the protected period. + * When run to parity is disabled, we give a minimum quantum to the running + * entity to ensure progress. */ -static u64 __sched_period(unsigned long nr_running) +static inline void set_protect_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if (unlikely(nr_running > sched_nr_latency)) - return nr_running * sysctl_sched_min_granularity; - else - return sysctl_sched_latency; + u64 slice = normalized_sysctl_sched_base_slice; + u64 vprot = se->deadline; + + if (sched_feat(RUN_TO_PARITY)) + slice = cfs_rq_min_slice(cfs_rq); + + slice = min(slice, se->slice); + if (slice != se->slice) + vprot = min_vruntime(vprot, se->vruntime + calc_delta_fair(slice, se)); + + se->vprot = vprot; +} + +static inline void update_protect_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + u64 slice = cfs_rq_min_slice(cfs_rq); + + se->vprot = min_vruntime(se->vprot, se->vruntime + calc_delta_fair(slice, se)); +} + +static inline bool protect_slice(struct sched_entity *se) +{ + return ((s64)(se->vprot - se->vruntime) > 0); +} + +static inline void cancel_protect_slice(struct sched_entity *se) +{ + if (protect_slice(se)) + se->vprot = se->vruntime; } /* - * We calculate the wall-time slice from the period by taking a part - * proportional to the weight. + * Earliest Eligible Virtual Deadline First + * + * In order to provide latency guarantees for different request sizes + * EEVDF selects the best runnable task from two criteria: + * + * 1) the task must be eligible (must be owed service) + * + * 2) from those tasks that meet 1), we select the one + * with the earliest virtual deadline. + * + * We can do this in O(log n) time due to an augmented RB-tree. The + * tree keeps the entries sorted on deadline, but also functions as a + * heap based on the vruntime by keeping: * - * s = p*P[w/rw] + * se->min_vruntime = min(se->vruntime, se->{left,right}->min_vruntime) + * + * Which allows tree pruning through eligibility. */ -static u64 sched_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) +static struct sched_entity *__pick_eevdf(struct cfs_rq *cfs_rq, bool protect) { - u64 slice = __sched_period(cfs_rq->nr_running + !se->on_rq); + struct rb_node *node = cfs_rq->tasks_timeline.rb_root.rb_node; + struct sched_entity *se = __pick_first_entity(cfs_rq); + struct sched_entity *curr = cfs_rq->curr; + struct sched_entity *best = NULL; - for_each_sched_entity(se) { - struct load_weight *load; - struct load_weight lw; + /* + * We can safely skip eligibility check if there is only one entity + * in this cfs_rq, saving some cycles. + */ + if (cfs_rq->nr_queued == 1) + return curr && curr->on_rq ? curr : se; - cfs_rq = cfs_rq_of(se); - load = &cfs_rq->load; + /* + * Picking the ->next buddy will affect latency but not fairness. + */ + if (sched_feat(PICK_BUDDY) && + cfs_rq->next && entity_eligible(cfs_rq, cfs_rq->next)) { + /* ->next will never be delayed */ + WARN_ON_ONCE(cfs_rq->next->sched_delayed); + return cfs_rq->next; + } + + if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr))) + curr = NULL; + + if (curr && protect && protect_slice(curr)) + return curr; - if (unlikely(!se->on_rq)) { - lw = cfs_rq->load; + /* Pick the leftmost entity if it's eligible */ + if (se && entity_eligible(cfs_rq, se)) { + best = se; + goto found; + } + + /* Heap search for the EEVD entity */ + while (node) { + struct rb_node *left = node->rb_left; - update_load_add(&lw, se->load.weight); - load = &lw; + /* + * Eligible entities in left subtree are always better + * choices, since they have earlier deadlines. + */ + if (left && vruntime_eligible(cfs_rq, + __node_2_se(left)->min_vruntime)) { + node = left; + continue; + } + + se = __node_2_se(node); + + /* + * The left subtree either is empty or has no eligible + * entity, so check the current node since it is the one + * with earliest deadline that might be eligible. + */ + if (entity_eligible(cfs_rq, se)) { + best = se; + break; } - slice = __calc_delta(slice, se->load.weight, load); + + node = node->rb_right; } - return slice; +found: + if (!best || (curr && entity_before(curr, best))) + best = curr; + + return best; +} + +static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) +{ + return __pick_eevdf(cfs_rq, true); +} + +struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) +{ + struct rb_node *last = rb_last(&cfs_rq->tasks_timeline.rb_root); + + if (!last) + return NULL; + + return __node_2_se(last); } +/************************************************************** + * Scheduling class statistics methods: + */ +int sched_update_scaling(void) +{ + unsigned int factor = get_update_sysctl_factor(); + +#define WRT_SYSCTL(name) \ + (normalized_sysctl_##name = sysctl_##name / (factor)) + WRT_SYSCTL(sched_base_slice); +#undef WRT_SYSCTL + + return 0; +} + +static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se); + /* - * We calculate the vruntime slice of a to-be-inserted task. - * - * vs = s/w + * XXX: strictly: vd_i += N*r_i/w_i such that: vd_i > ve_i + * this is probably good enough. */ -static u64 sched_vslice(struct cfs_rq *cfs_rq, struct sched_entity *se) +static bool update_deadline(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return calc_delta_fair(sched_slice(cfs_rq, se), se); + if ((s64)(se->vruntime - se->deadline) < 0) + return false; + + /* + * For EEVDF the virtual time slope is determined by w_i (iow. + * nice) while the request time r_i is determined by + * sysctl_sched_base_slice. + */ + if (!se->custom_slice) + se->slice = sysctl_sched_base_slice; + + /* + * EEVDF: vd_i = ve_i + r_i / w_i + */ + se->deadline = se->vruntime + calc_delta_fair(se->slice, se); + + /* + * The task has consumed its request, reschedule. + */ + return true; } -#ifdef CONFIG_SMP #include "pelt.h" -#include "sched-pelt.h" static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu); static unsigned long task_h_load(struct task_struct *p); @@ -708,21 +1066,17 @@ void init_entity_runnable_average(struct sched_entity *se) * nothing has been attached to the task group yet. */ if (entity_is_task(se)) - sa->runnable_load_avg = sa->load_avg = scale_load_down(se->load.weight); - - se->runnable_weight = se->load.weight; + sa->load_avg = scale_load_down(se->load.weight); - /* when this task enqueue'ed, it will contribute to its cfs_rq's load_avg */ + /* when this task is enqueued, it will contribute to its cfs_rq's load_avg */ } -static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq); -static void attach_entity_cfs_rq(struct sched_entity *se); - /* * With new tasks being created, their initial util_avgs are extrapolated * based on the cfs_rq's current util_avg: * - * util_avg = cfs_rq->util_avg / (cfs_rq->load_avg + 1) * se.load.weight + * util_avg = cfs_rq->avg.util_avg / (cfs_rq->avg.load_avg + 1) + * * se_weight(se) * * However, in many cases, the above util_avg does not give a desired * value. Moreover, the sum of the util_avgs may be divergent, such @@ -744,16 +1098,32 @@ static void attach_entity_cfs_rq(struct sched_entity *se); * Finally, that extrapolated util_avg is clamped to the cap (util_avg_cap) * if util_avg > util_avg_cap. */ -void post_init_entity_util_avg(struct sched_entity *se) +void post_init_entity_util_avg(struct task_struct *p) { + struct sched_entity *se = &p->se; struct cfs_rq *cfs_rq = cfs_rq_of(se); struct sched_avg *sa = &se->avg; - long cpu_scale = arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq))); + long cpu_scale = arch_scale_cpu_capacity(cpu_of(rq_of(cfs_rq))); long cap = (long)(cpu_scale - cfs_rq->avg.util_avg) / 2; + if (p->sched_class != &fair_sched_class) { + /* + * For !fair tasks do: + * + update_cfs_rq_load_avg(now, cfs_rq); + attach_entity_load_avg(cfs_rq, se); + switched_from_fair(rq, p); + * + * such that the next switched_to_fair() has the + * expected state. + */ + se->avg.last_update_time = cfs_rq_clock_pelt(cfs_rq); + return; + } + if (cap > 0) { if (cfs_rq->avg.util_avg != 0) { - sa->util_avg = cfs_rq->avg.util_avg * se->load.weight; + sa->util_avg = cfs_rq->avg.util_avg * se_weight(se); sa->util_avg /= (cfs_rq->avg.load_avg + 1); if (sa->util_avg > cap) @@ -763,205 +1133,181 @@ void post_init_entity_util_avg(struct sched_entity *se) } } + sa->runnable_avg = sa->util_avg; +} + +static s64 update_se(struct rq *rq, struct sched_entity *se) +{ + u64 now = rq_clock_task(rq); + s64 delta_exec; + + delta_exec = now - se->exec_start; + if (unlikely(delta_exec <= 0)) + return delta_exec; + + se->exec_start = now; if (entity_is_task(se)) { - struct task_struct *p = task_of(se); - if (p->sched_class != &fair_sched_class) { - /* - * For !fair tasks do: - * - update_cfs_rq_load_avg(now, cfs_rq); - attach_entity_load_avg(cfs_rq, se, 0); - switched_from_fair(rq, p); - * - * such that the next switched_to_fair() has the - * expected state. - */ - se->avg.last_update_time = cfs_rq_clock_task(cfs_rq); - return; - } + struct task_struct *donor = task_of(se); + struct task_struct *running = rq->curr; + /* + * If se is a task, we account the time against the running + * task, as w/ proxy-exec they may not be the same. + */ + running->se.exec_start = now; + running->se.sum_exec_runtime += delta_exec; + + trace_sched_stat_runtime(running, delta_exec); + account_group_exec_runtime(running, delta_exec); + + /* cgroup time is always accounted against the donor */ + cgroup_account_cputime(donor, delta_exec); + } else { + /* If not task, account the time against donor se */ + se->sum_exec_runtime += delta_exec; } - attach_entity_cfs_rq(se); -} + if (schedstat_enabled()) { + struct sched_statistics *stats; -#else /* !CONFIG_SMP */ -void init_entity_runnable_average(struct sched_entity *se) -{ -} -void post_init_entity_util_avg(struct sched_entity *se) -{ + stats = __schedstats_from_se(se); + __schedstat_set(stats->exec_max, + max(delta_exec, stats->exec_max)); + } + + return delta_exec; } -static void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) + +static void set_next_buddy(struct sched_entity *se); + +/* + * Used by other classes to account runtime. + */ +s64 update_curr_common(struct rq *rq) { + return update_se(rq, &rq->donor->se); } -#endif /* CONFIG_SMP */ /* * Update the current task's runtime statistics. */ static void update_curr(struct cfs_rq *cfs_rq) { + /* + * Note: cfs_rq->curr corresponds to the task picked to + * run (ie: rq->donor.se) which due to proxy-exec may + * not necessarily be the actual task running + * (rq->curr.se). This is easy to confuse! + */ struct sched_entity *curr = cfs_rq->curr; - u64 now = rq_clock_task(rq_of(cfs_rq)); - u64 delta_exec; + struct rq *rq = rq_of(cfs_rq); + s64 delta_exec; + bool resched; if (unlikely(!curr)) return; - delta_exec = now - curr->exec_start; - if (unlikely((s64)delta_exec <= 0)) + delta_exec = update_se(rq, curr); + if (unlikely(delta_exec <= 0)) return; - curr->exec_start = now; - - schedstat_set(curr->statistics.exec_max, - max(delta_exec, curr->statistics.exec_max)); - - curr->sum_exec_runtime += delta_exec; - schedstat_add(cfs_rq->exec_clock, delta_exec); - curr->vruntime += calc_delta_fair(delta_exec, curr); - update_min_vruntime(cfs_rq); + resched = update_deadline(cfs_rq, curr); if (entity_is_task(curr)) { - struct task_struct *curtask = task_of(curr); - - trace_sched_stat_runtime(curtask, delta_exec, curr->vruntime); - cgroup_account_cputime(curtask, delta_exec); - account_group_exec_runtime(curtask, delta_exec); + /* + * 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. + */ + dl_server_update(&rq->fair_server, delta_exec); } account_cfs_rq_runtime(cfs_rq, delta_exec); + + if (cfs_rq->nr_queued == 1) + return; + + if (resched || !protect_slice(curr)) { + resched_curr_lazy(rq); + clear_buddies(cfs_rq, curr); + } } static void update_curr_fair(struct rq *rq) { - update_curr(cfs_rq_of(&rq->curr->se)); + update_curr(cfs_rq_of(&rq->donor->se)); } static inline void -update_stats_wait_start(struct cfs_rq *cfs_rq, struct sched_entity *se) +update_stats_wait_start_fair(struct cfs_rq *cfs_rq, struct sched_entity *se) { - u64 wait_start, prev_wait_start; + struct sched_statistics *stats; + struct task_struct *p = NULL; if (!schedstat_enabled()) return; - wait_start = rq_clock(rq_of(cfs_rq)); - prev_wait_start = schedstat_val(se->statistics.wait_start); + stats = __schedstats_from_se(se); - if (entity_is_task(se) && task_on_rq_migrating(task_of(se)) && - likely(wait_start > prev_wait_start)) - wait_start -= prev_wait_start; + if (entity_is_task(se)) + p = task_of(se); - __schedstat_set(se->statistics.wait_start, wait_start); + __update_stats_wait_start(rq_of(cfs_rq), p, stats); } static inline void -update_stats_wait_end(struct cfs_rq *cfs_rq, struct sched_entity *se) +update_stats_wait_end_fair(struct cfs_rq *cfs_rq, struct sched_entity *se) { - struct task_struct *p; - u64 delta; + struct sched_statistics *stats; + struct task_struct *p = NULL; if (!schedstat_enabled()) return; - delta = rq_clock(rq_of(cfs_rq)) - schedstat_val(se->statistics.wait_start); + stats = __schedstats_from_se(se); - if (entity_is_task(se)) { + /* + * When the sched_schedstat changes from 0 to 1, some sched se + * maybe already in the runqueue, the se->statistics.wait_start + * will be 0.So it will let the delta wrong. We need to avoid this + * scenario. + */ + if (unlikely(!schedstat_val(stats->wait_start))) + return; + + if (entity_is_task(se)) p = task_of(se); - if (task_on_rq_migrating(p)) { - /* - * Preserve migrating task's wait time so wait_start - * time stamp can be adjusted to accumulate wait time - * prior to migration. - */ - __schedstat_set(se->statistics.wait_start, delta); - return; - } - trace_sched_stat_wait(p, delta); - } - __schedstat_set(se->statistics.wait_max, - max(schedstat_val(se->statistics.wait_max), delta)); - __schedstat_inc(se->statistics.wait_count); - __schedstat_add(se->statistics.wait_sum, delta); - __schedstat_set(se->statistics.wait_start, 0); + __update_stats_wait_end(rq_of(cfs_rq), p, stats); } static inline void -update_stats_enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se) +update_stats_enqueue_sleeper_fair(struct cfs_rq *cfs_rq, struct sched_entity *se) { + struct sched_statistics *stats; struct task_struct *tsk = NULL; - u64 sleep_start, block_start; if (!schedstat_enabled()) return; - sleep_start = schedstat_val(se->statistics.sleep_start); - block_start = schedstat_val(se->statistics.block_start); + stats = __schedstats_from_se(se); if (entity_is_task(se)) tsk = task_of(se); - if (sleep_start) { - u64 delta = rq_clock(rq_of(cfs_rq)) - sleep_start; - - if ((s64)delta < 0) - delta = 0; - - if (unlikely(delta > schedstat_val(se->statistics.sleep_max))) - __schedstat_set(se->statistics.sleep_max, delta); - - __schedstat_set(se->statistics.sleep_start, 0); - __schedstat_add(se->statistics.sum_sleep_runtime, delta); - - if (tsk) { - account_scheduler_latency(tsk, delta >> 10, 1); - trace_sched_stat_sleep(tsk, delta); - } - } - if (block_start) { - u64 delta = rq_clock(rq_of(cfs_rq)) - block_start; - - if ((s64)delta < 0) - delta = 0; - - if (unlikely(delta > schedstat_val(se->statistics.block_max))) - __schedstat_set(se->statistics.block_max, delta); - - __schedstat_set(se->statistics.block_start, 0); - __schedstat_add(se->statistics.sum_sleep_runtime, delta); - - if (tsk) { - if (tsk->in_iowait) { - __schedstat_add(se->statistics.iowait_sum, delta); - __schedstat_inc(se->statistics.iowait_count); - trace_sched_stat_iowait(tsk, delta); - } - - trace_sched_stat_blocked(tsk, delta); - - /* - * Blocking time is in units of nanosecs, so shift by - * 20 to get a milliseconds-range estimation of the - * amount of time that the task spent sleeping: - */ - if (unlikely(prof_on == SLEEP_PROFILING)) { - profile_hits(SLEEP_PROFILING, - (void *)get_wchan(tsk), - delta >> 20); - } - account_scheduler_latency(tsk, delta >> 10, 0); - } - } + __update_stats_enqueue_sleeper(rq_of(cfs_rq), tsk, stats); } /* * Task is being enqueued - update stats: */ static inline void -update_stats_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) +update_stats_enqueue_fair(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { if (!schedstat_enabled()) return; @@ -971,14 +1317,14 @@ update_stats_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * a dequeue/enqueue event is a NOP) */ if (se != cfs_rq->curr) - update_stats_wait_start(cfs_rq, se); + update_stats_wait_start_fair(cfs_rq, se); if (flags & ENQUEUE_WAKEUP) - update_stats_enqueue_sleeper(cfs_rq, se); + update_stats_enqueue_sleeper_fair(cfs_rq, se); } static inline void -update_stats_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) +update_stats_dequeue_fair(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { if (!schedstat_enabled()) @@ -989,16 +1335,19 @@ update_stats_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * waiting task: */ if (se != cfs_rq->curr) - update_stats_wait_end(cfs_rq, se); + update_stats_wait_end_fair(cfs_rq, se); if ((flags & DEQUEUE_SLEEP) && entity_is_task(se)) { struct task_struct *tsk = task_of(se); + unsigned int state; - if (tsk->state & TASK_INTERRUPTIBLE) - __schedstat_set(se->statistics.sleep_start, + /* XXX racy against TTWU */ + state = READ_ONCE(tsk->__state); + if (state & TASK_INTERRUPTIBLE) + __schedstat_set(tsk->stats.sleep_start, rq_clock(rq_of(cfs_rq))); - if (tsk->state & TASK_UNINTERRUPTIBLE) - __schedstat_set(se->statistics.block_start, + if (state & TASK_UNINTERRUPTIBLE) + __schedstat_set(tsk->stats.block_start, rq_clock(rq_of(cfs_rq))); } } @@ -1019,6 +1368,50 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se) * Scheduling class queueing methods: */ +static inline bool is_core_idle(int cpu) +{ +#ifdef CONFIG_SCHED_SMT + int sibling; + + for_each_cpu(sibling, cpu_smt_mask(cpu)) { + if (cpu == sibling) + continue; + + if (!idle_cpu(sibling)) + return false; + } +#endif + + return true; +} + +#ifdef CONFIG_NUMA +#define NUMA_IMBALANCE_MIN 2 + +static inline long +adjust_numa_imbalance(int imbalance, int dst_running, int imb_numa_nr) +{ + /* + * Allow a NUMA imbalance if busy CPUs is less than the maximum + * threshold. Above this threshold, individual tasks may be contending + * for both memory bandwidth and any shared HT resources. This is an + * approximation as the number of running tasks may not be related to + * the number of busy CPUs due to sched_setaffinity. + */ + if (dst_running > imb_numa_nr) + return imbalance; + + /* + * Allow a small imbalance based on a simple pair of communicating + * tasks that remain local when the destination is lightly loaded. + */ + if (imbalance <= NUMA_IMBALANCE_MIN) + return 0; + + return imbalance; +} +#endif /* CONFIG_NUMA */ + #ifdef CONFIG_NUMA_BALANCING /* * Approximate time to scan a full NUMA task in ms. The task scan period is @@ -1034,8 +1427,11 @@ unsigned int sysctl_numa_balancing_scan_size = 256; /* Scan @scan_size MB every @scan_period after an initial @scan_delay in ms */ unsigned int sysctl_numa_balancing_scan_delay = 1000; +/* The page with hint page fault latency < threshold in ms is considered hot */ +unsigned int sysctl_numa_balancing_hot_threshold = MSEC_PER_SEC; + struct numa_group { - atomic_t refcount; + refcount_t refcount; spinlock_t lock; /* nr_tasks, tasks */ int nr_tasks; @@ -1046,14 +1442,30 @@ struct numa_group { unsigned long total_faults; unsigned long max_faults_cpu; /* + * faults[] array is split into two regions: faults_mem and faults_cpu. + * * Faults_cpu is used to decide whether memory should move * towards the CPU. As a consequence, these stats are weighted * more by CPU use than by memory faults. */ - unsigned long *faults_cpu; - unsigned long faults[0]; + unsigned long faults[]; }; +/* + * For functions that can be called in multiple contexts that permit reading + * ->numa_group (see struct task_struct for locking rules). + */ +static struct numa_group *deref_task_numa_group(struct task_struct *p) +{ + return rcu_dereference_check(p->numa_group, p == current || + (lockdep_is_held(__rq_lockp(task_rq(p))) && !READ_ONCE(p->on_cpu))); +} + +static struct numa_group *deref_curr_numa_group(struct task_struct *p) +{ + return rcu_dereference_protected(p->numa_group, p == current); +} + static inline unsigned long group_faults_priv(struct numa_group *ng); static inline unsigned long group_faults_shared(struct numa_group *ng); @@ -1067,7 +1479,7 @@ static unsigned int task_nr_scan_windows(struct task_struct *p) * by the PTE scanner and NUMA hinting faults should be trapped based * on resident pages */ - nr_scan_pages = sysctl_numa_balancing_scan_size << (20 - PAGE_SHIFT); + nr_scan_pages = MB_TO_PAGES(sysctl_numa_balancing_scan_size); rss = get_mm_rss(p->mm); if (!rss) rss = nr_scan_pages; @@ -1076,7 +1488,7 @@ static unsigned int task_nr_scan_windows(struct task_struct *p) return rss / nr_scan_pages; } -/* For sanitys sake, never scan more PTEs than MAX_SCAN_WINDOW MB/sec. */ +/* For sanity's sake, never scan more PTEs than MAX_SCAN_WINDOW MB/sec. */ #define MAX_SCAN_WINDOW 2560 static unsigned int task_scan_min(struct task_struct *p) @@ -1097,17 +1509,20 @@ static unsigned int task_scan_start(struct task_struct *p) { unsigned long smin = task_scan_min(p); unsigned long period = smin; + struct numa_group *ng; /* Scale the maximum scan period with the amount of shared memory. */ - if (p->numa_group) { - struct numa_group *ng = p->numa_group; + rcu_read_lock(); + ng = rcu_dereference(p->numa_group); + if (ng) { unsigned long shared = group_faults_shared(ng); unsigned long private = group_faults_priv(ng); - period *= atomic_read(&ng->refcount); + period *= refcount_read(&ng->refcount); period *= shared + 1; period /= private + shared + 1; } + rcu_read_unlock(); return max(smin, period); } @@ -1116,18 +1531,19 @@ static unsigned int task_scan_max(struct task_struct *p) { unsigned long smin = task_scan_min(p); unsigned long smax; + struct numa_group *ng; /* Watch for min being lower than max due to floor calculations */ smax = sysctl_numa_balancing_scan_period_max / task_nr_scan_windows(p); /* Scale the maximum scan period with the amount of shared memory. */ - if (p->numa_group) { - struct numa_group *ng = p->numa_group; + ng = deref_curr_numa_group(p); + if (ng) { unsigned long shared = group_faults_shared(ng); unsigned long private = group_faults_priv(ng); unsigned long period = smax; - period *= atomic_read(&ng->refcount); + period *= refcount_read(&ng->refcount); period *= shared + 1; period /= private + shared + 1; @@ -1137,56 +1553,15 @@ static unsigned int task_scan_max(struct task_struct *p) return max(smin, smax); } -void init_numa_balancing(unsigned long clone_flags, struct task_struct *p) -{ - int mm_users = 0; - struct mm_struct *mm = p->mm; - - if (mm) { - mm_users = atomic_read(&mm->mm_users); - if (mm_users == 1) { - mm->numa_next_scan = jiffies + msecs_to_jiffies(sysctl_numa_balancing_scan_delay); - mm->numa_scan_seq = 0; - } - } - p->node_stamp = 0; - p->numa_scan_seq = mm ? mm->numa_scan_seq : 0; - p->numa_scan_period = sysctl_numa_balancing_scan_delay; - p->numa_work.next = &p->numa_work; - p->numa_faults = NULL; - p->numa_group = NULL; - p->last_task_numa_placement = 0; - p->last_sum_exec_runtime = 0; - - /* New address space, reset the preferred nid */ - if (!(clone_flags & CLONE_VM)) { - p->numa_preferred_nid = -1; - return; - } - - /* - * New thread, keep existing numa_preferred_nid which should be copied - * already by arch_dup_task_struct but stagger when scans start. - */ - if (mm) { - unsigned int delay; - - delay = min_t(unsigned int, task_scan_max(current), - current->numa_scan_period * mm_users * NSEC_PER_MSEC); - delay += 2 * TICK_NSEC; - p->node_stamp = delay; - } -} - static void account_numa_enqueue(struct rq *rq, struct task_struct *p) { - rq->nr_numa_running += (p->numa_preferred_nid != -1); + rq->nr_numa_running += (p->numa_preferred_nid != NUMA_NO_NODE); rq->nr_preferred_running += (p->numa_preferred_nid == task_node(p)); } static void account_numa_dequeue(struct rq *rq, struct task_struct *p) { - rq->nr_numa_running -= (p->numa_preferred_nid != -1); + rq->nr_numa_running -= (p->numa_preferred_nid != NUMA_NO_NODE); rq->nr_preferred_running -= (p->numa_preferred_nid == task_node(p)); } @@ -1201,7 +1576,16 @@ static void account_numa_dequeue(struct rq *rq, struct task_struct *p) pid_t task_numa_group_id(struct task_struct *p) { - return p->numa_group ? p->numa_group->gid : 0; + struct numa_group *ng; + pid_t gid = 0; + + rcu_read_lock(); + ng = rcu_dereference(p->numa_group); + if (ng) + gid = ng->gid; + rcu_read_unlock(); + + return gid; } /* @@ -1226,17 +1610,19 @@ static inline unsigned long task_faults(struct task_struct *p, int nid) static inline unsigned long group_faults(struct task_struct *p, int nid) { - if (!p->numa_group) + struct numa_group *ng = deref_task_numa_group(p); + + if (!ng) return 0; - return p->numa_group->faults[task_faults_idx(NUMA_MEM, nid, 0)] + - p->numa_group->faults[task_faults_idx(NUMA_MEM, nid, 1)]; + return ng->faults[task_faults_idx(NUMA_MEM, nid, 0)] + + ng->faults[task_faults_idx(NUMA_MEM, nid, 1)]; } static inline unsigned long group_faults_cpu(struct numa_group *group, int nid) { - return group->faults_cpu[task_faults_idx(NUMA_MEM, nid, 0)] + - group->faults_cpu[task_faults_idx(NUMA_MEM, nid, 1)]; + return group->faults[task_faults_idx(NUMA_CPU, nid, 0)] + + group->faults[task_faults_idx(NUMA_CPU, nid, 1)]; } static inline unsigned long group_faults_priv(struct numa_group *ng) @@ -1277,10 +1663,10 @@ static bool numa_is_active_node(int nid, struct numa_group *ng) /* Handle placement on systems where not all nodes are directly connected. */ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, - int maxdist, bool task) + int lim_dist, bool task) { unsigned long score = 0; - int node; + int node, max_dist; /* * All nodes are directly connected, and the same distance @@ -1289,9 +1675,11 @@ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, if (sched_numa_topology_type == NUMA_DIRECT) return 0; + /* sched_max_numa_distance may be changed in parallel. */ + max_dist = READ_ONCE(sched_max_numa_distance); /* * This code is called for each node, introducing N^2 complexity, - * which should be ok given the number of nodes rarely exceeds 8. + * which should be OK given the number of nodes rarely exceeds 8. */ for_each_online_node(node) { unsigned long faults; @@ -1301,7 +1689,7 @@ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, * The furthest away nodes in the system are not interesting * for placement; nid was already counted. */ - if (dist == sched_max_numa_distance || node == nid) + if (dist >= max_dist || node == nid) continue; /* @@ -1311,8 +1699,7 @@ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, * "hoplimit", only nodes closer by than "hoplimit" are part * of each group. Skip other nodes. */ - if (sched_numa_topology_type == NUMA_BACKPLANE && - dist >= maxdist) + if (sched_numa_topology_type == NUMA_BACKPLANE && dist >= lim_dist) continue; /* Add up the faults from nearby nodes. */ @@ -1330,8 +1717,8 @@ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, * This seems to result in good task placement. */ if (sched_numa_topology_type == NUMA_GLUELESS_MESH) { - faults *= (sched_max_numa_distance - dist); - faults /= (sched_max_numa_distance - LOCAL_DISTANCE); + faults *= (max_dist - dist); + faults /= (max_dist - LOCAL_DISTANCE); } score += faults; @@ -1368,12 +1755,13 @@ static inline unsigned long task_weight(struct task_struct *p, int nid, static inline unsigned long group_weight(struct task_struct *p, int nid, int dist) { + struct numa_group *ng = deref_task_numa_group(p); unsigned long faults, total_faults; - if (!p->numa_group) + if (!ng) return 0; - total_faults = p->numa_group->total_faults; + total_faults = ng->total_faults; if (!total_faults) return 0; @@ -1384,15 +1772,169 @@ static inline unsigned long group_weight(struct task_struct *p, int nid, return 1000 * faults / total_faults; } -bool should_numa_migrate_memory(struct task_struct *p, struct page * page, +/* + * If memory tiering mode is enabled, cpupid of slow memory page is + * used to record scan time instead of CPU and PID. When tiering mode + * is disabled at run time, the scan time (in cpupid) will be + * interpreted as CPU and PID. So CPU needs to be checked to avoid to + * access out of array bound. + */ +static inline bool cpupid_valid(int cpupid) +{ + return cpupid_to_cpu(cpupid) < nr_cpu_ids; +} + +/* + * For memory tiering mode, if there are enough free pages (more than + * enough watermark defined here) in fast memory node, to take full + * advantage of fast memory capacity, all recently accessed slow + * memory pages will be migrated to fast memory node without + * considering hot threshold. + */ +static bool pgdat_free_space_enough(struct pglist_data *pgdat) +{ + int z; + unsigned long enough_wmark; + + enough_wmark = max(1UL * 1024 * 1024 * 1024 >> PAGE_SHIFT, + pgdat->node_present_pages >> 4); + for (z = pgdat->nr_zones - 1; z >= 0; z--) { + struct zone *zone = pgdat->node_zones + z; + + if (!populated_zone(zone)) + continue; + + if (zone_watermark_ok(zone, 0, + promo_wmark_pages(zone) + enough_wmark, + ZONE_MOVABLE, 0)) + return true; + } + return false; +} + +/* + * For memory tiering mode, when page tables are scanned, the scan + * time will be recorded in struct page in addition to make page + * PROT_NONE for slow memory page. So when the page is accessed, in + * hint page fault handler, the hint page fault latency is calculated + * via, + * + * hint page fault latency = hint page fault time - scan time + * + * The smaller the hint page fault latency, the higher the possibility + * for the page to be hot. + */ +static int numa_hint_fault_latency(struct folio *folio) +{ + int last_time, time; + + time = jiffies_to_msecs(jiffies); + last_time = folio_xchg_access_time(folio, time); + + return (time - last_time) & PAGE_ACCESS_TIME_MASK; +} + +/* + * For memory tiering mode, too high promotion/demotion throughput may + * hurt application latency. So we provide a mechanism to rate limit + * the number of pages that are tried to be promoted. + */ +static bool numa_promotion_rate_limit(struct pglist_data *pgdat, + unsigned long rate_limit, int nr) +{ + unsigned long nr_cand; + unsigned int now, start; + + now = jiffies_to_msecs(jiffies); + mod_node_page_state(pgdat, PGPROMOTE_CANDIDATE, nr); + nr_cand = node_page_state(pgdat, PGPROMOTE_CANDIDATE); + start = pgdat->nbp_rl_start; + if (now - start > MSEC_PER_SEC && + cmpxchg(&pgdat->nbp_rl_start, start, now) == start) + pgdat->nbp_rl_nr_cand = nr_cand; + if (nr_cand - pgdat->nbp_rl_nr_cand >= rate_limit) + return true; + return false; +} + +#define NUMA_MIGRATION_ADJUST_STEPS 16 + +static void numa_promotion_adjust_threshold(struct pglist_data *pgdat, + unsigned long rate_limit, + unsigned int ref_th) +{ + unsigned int now, start, th_period, unit_th, th; + unsigned long nr_cand, ref_cand, diff_cand; + + now = jiffies_to_msecs(jiffies); + th_period = sysctl_numa_balancing_scan_period_max; + start = pgdat->nbp_th_start; + if (now - start > th_period && + cmpxchg(&pgdat->nbp_th_start, start, now) == start) { + ref_cand = rate_limit * + sysctl_numa_balancing_scan_period_max / MSEC_PER_SEC; + nr_cand = node_page_state(pgdat, PGPROMOTE_CANDIDATE); + diff_cand = nr_cand - pgdat->nbp_th_nr_cand; + unit_th = ref_th * 2 / NUMA_MIGRATION_ADJUST_STEPS; + th = pgdat->nbp_threshold ? : ref_th; + if (diff_cand > ref_cand * 11 / 10) + th = max(th - unit_th, unit_th); + else if (diff_cand < ref_cand * 9 / 10) + th = min(th + unit_th, ref_th * 2); + pgdat->nbp_th_nr_cand = nr_cand; + pgdat->nbp_threshold = th; + } +} + +bool should_numa_migrate_memory(struct task_struct *p, struct folio *folio, int src_nid, int dst_cpu) { - struct numa_group *ng = p->numa_group; + struct numa_group *ng = deref_curr_numa_group(p); int dst_nid = cpu_to_node(dst_cpu); int last_cpupid, this_cpupid; + /* + * Cannot migrate to memoryless nodes. + */ + if (!node_state(dst_nid, N_MEMORY)) + return false; + + /* + * The pages in slow memory node should be migrated according + * to hot/cold instead of private/shared. + */ + if (folio_use_access_time(folio)) { + struct pglist_data *pgdat; + unsigned long rate_limit; + unsigned int latency, th, def_th; + long nr = folio_nr_pages(folio); + + pgdat = NODE_DATA(dst_nid); + if (pgdat_free_space_enough(pgdat)) { + /* workload changed, reset hot threshold */ + pgdat->nbp_threshold = 0; + mod_node_page_state(pgdat, PGPROMOTE_CANDIDATE_NRL, nr); + return true; + } + + def_th = sysctl_numa_balancing_hot_threshold; + rate_limit = MB_TO_PAGES(sysctl_numa_balancing_promote_rate_limit); + numa_promotion_adjust_threshold(pgdat, rate_limit, def_th); + + th = pgdat->nbp_threshold ? : def_th; + latency = numa_hint_fault_latency(folio); + if (latency >= th) + return false; + + return !numa_promotion_rate_limit(pgdat, rate_limit, nr); + } + this_cpupid = cpu_pid_to_cpupid(dst_cpu, current->pid); - last_cpupid = page_cpupid_xchg_last(page, this_cpupid); + last_cpupid = folio_xchg_last_cpupid(folio, this_cpupid); + + if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) && + !node_is_toptier(src_nid) && !cpupid_valid(last_cpupid)) + return false; /* * Allow first faults or private faults to migrate immediately early in @@ -1400,7 +1942,7 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page, * two full passes of the "multi-stage node selection" test that is * executed below. */ - if ((p->numa_preferred_nid == -1 || p->numa_scan_seq <= 4) && + if ((p->numa_preferred_nid == NUMA_NO_NODE || p->numa_scan_seq <= 4) && (cpupid_pid_unset(last_cpupid) || cpupid_match_pid(p, last_cpupid))) return true; @@ -1453,40 +1995,43 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page, group_faults_cpu(ng, src_nid) * group_faults(p, dst_nid) * 4; } -static unsigned long weighted_cpuload(struct rq *rq); -static unsigned long source_load(int cpu, int type); -static unsigned long target_load(int cpu, int type); +/* + * 'numa_type' describes the node at the moment of load balancing. + */ +enum numa_type { + /* The node has spare capacity that can be used to run more tasks. */ + node_has_spare = 0, + /* + * The node is fully used and the tasks don't compete for more CPU + * cycles. Nevertheless, some tasks might wait before running. + */ + node_fully_busy, + /* + * The node is overloaded and can't provide expected CPU cycles to all + * tasks. + */ + node_overloaded +}; /* Cached statistics for all CPUs within a node */ struct numa_stats { unsigned long load; - + unsigned long runnable; + unsigned long util; /* Total compute capacity of CPUs on a node */ unsigned long compute_capacity; + unsigned int nr_running; + unsigned int weight; + enum numa_type node_type; + int idle_cpu; }; -/* - * XXX borrowed from update_sg_lb_stats - */ -static void update_numa_stats(struct numa_stats *ns, int nid) -{ - int cpu; - - memset(ns, 0, sizeof(*ns)); - for_each_cpu(cpu, cpumask_of_node(nid)) { - struct rq *rq = cpu_rq(cpu); - - ns->load += weighted_cpuload(rq); - ns->compute_capacity += capacity_of(cpu); - } - -} - struct task_numa_env { struct task_struct *p; int src_cpu, src_nid; int dst_cpu, dst_nid; + int imb_numa_nr; struct numa_stats src_stats, dst_stats; @@ -1498,20 +2043,130 @@ struct task_numa_env { int best_cpu; }; +static unsigned long cpu_load(struct rq *rq); +static unsigned long cpu_runnable(struct rq *rq); + +static inline enum +numa_type numa_classify(unsigned int imbalance_pct, + struct numa_stats *ns) +{ + if ((ns->nr_running > ns->weight) && + (((ns->compute_capacity * 100) < (ns->util * imbalance_pct)) || + ((ns->compute_capacity * imbalance_pct) < (ns->runnable * 100)))) + return node_overloaded; + + if ((ns->nr_running < ns->weight) || + (((ns->compute_capacity * 100) > (ns->util * imbalance_pct)) && + ((ns->compute_capacity * imbalance_pct) > (ns->runnable * 100)))) + return node_has_spare; + + return node_fully_busy; +} + +#ifdef CONFIG_SCHED_SMT +/* Forward declarations of select_idle_sibling helpers */ +static inline bool test_idle_cores(int cpu); +static inline int numa_idle_core(int idle_core, int cpu) +{ + if (!static_branch_likely(&sched_smt_present) || + idle_core >= 0 || !test_idle_cores(cpu)) + return idle_core; + + /* + * Prefer cores instead of packing HT siblings + * and triggering future load balancing. + */ + if (is_core_idle(cpu)) + idle_core = cpu; + + return idle_core; +} +#else /* !CONFIG_SCHED_SMT: */ +static inline int numa_idle_core(int idle_core, int cpu) +{ + return idle_core; +} +#endif /* !CONFIG_SCHED_SMT */ + +/* + * Gather all necessary information to make NUMA balancing placement + * decisions that are compatible with standard load balancer. This + * borrows code and logic from update_sg_lb_stats but sharing a + * common implementation is impractical. + */ +static void update_numa_stats(struct task_numa_env *env, + struct numa_stats *ns, int nid, + bool find_idle) +{ + int cpu, idle_core = -1; + + memset(ns, 0, sizeof(*ns)); + ns->idle_cpu = -1; + + rcu_read_lock(); + for_each_cpu(cpu, cpumask_of_node(nid)) { + struct rq *rq = cpu_rq(cpu); + + ns->load += cpu_load(rq); + ns->runnable += cpu_runnable(rq); + ns->util += cpu_util_cfs(cpu); + 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)) { + if (READ_ONCE(rq->numa_migrate_on) || + !cpumask_test_cpu(cpu, env->p->cpus_ptr)) + continue; + + if (ns->idle_cpu == -1) + ns->idle_cpu = cpu; + + idle_core = numa_idle_core(idle_core, cpu); + } + } + rcu_read_unlock(); + + ns->weight = cpumask_weight(cpumask_of_node(nid)); + + ns->node_type = numa_classify(env->imbalance_pct, ns); + + if (idle_core >= 0) + ns->idle_cpu = idle_core; +} + static void task_numa_assign(struct task_numa_env *env, struct task_struct *p, long imp) { struct rq *rq = cpu_rq(env->dst_cpu); - /* Bail out if run-queue part of active NUMA balance. */ - if (xchg(&rq->numa_migrate_on, 1)) + /* Check if run-queue part of active NUMA balance. */ + if (env->best_cpu != env->dst_cpu && xchg(&rq->numa_migrate_on, 1)) { + int cpu; + int start = env->dst_cpu; + + /* Find alternative idle CPU. */ + for_each_cpu_wrap(cpu, cpumask_of_node(env->dst_nid), start + 1) { + if (cpu == env->best_cpu || !idle_cpu(cpu) || + !cpumask_test_cpu(cpu, env->p->cpus_ptr)) { + continue; + } + + env->dst_cpu = cpu; + rq = cpu_rq(env->dst_cpu); + if (!xchg(&rq->numa_migrate_on, 1)) + goto assign; + } + + /* Failed to find an alternative idle CPU */ return; + } +assign: /* * Clear previous best_cpu/rq numa-migrate flag, since task now * found a better CPU to move/swap. */ - if (env->best_cpu != -1) { + if (env->best_cpu != -1 && env->best_cpu != env->dst_cpu) { rq = cpu_rq(env->best_cpu); WRITE_ONCE(rq->numa_migrate_on, 0); } @@ -1567,31 +2222,36 @@ static bool load_too_imbalanced(long src_load, long dst_load, * into account that it might be best if task running on the dst_cpu should * be exchanged with the source task */ -static void task_numa_compare(struct task_numa_env *env, +static bool task_numa_compare(struct task_numa_env *env, long taskimp, long groupimp, bool maymove) { + struct numa_group *cur_ng, *p_ng = deref_curr_numa_group(env->p); struct rq *dst_rq = cpu_rq(env->dst_cpu); + long imp = p_ng ? groupimp : taskimp; struct task_struct *cur; long src_load, dst_load; - long load; - long imp = env->p->numa_group ? groupimp : taskimp; - long moveimp = imp; int dist = env->dist; + long moveimp = imp; + long load; + bool stopsearch = false; if (READ_ONCE(dst_rq->numa_migrate_on)) - return; + return false; rcu_read_lock(); - cur = task_rcu_dereference(&dst_rq->curr); - if (cur && ((cur->flags & PF_EXITING) || is_idle_task(cur))) + cur = rcu_dereference(dst_rq->curr); + if (cur && ((cur->flags & (PF_EXITING | PF_KTHREAD)) || + !cur->mm)) cur = NULL; /* * Because we have preemption enabled we can get migrated around and * end try selecting ourselves (current == env->p) as a swap candidate. */ - if (cur == env->p) + if (cur == env->p) { + stopsearch = true; goto unlock; + } if (!cur) { if (maymove && moveimp >= env->best_imp) @@ -1600,36 +2260,55 @@ static void task_numa_compare(struct task_numa_env *env, goto unlock; } + /* Skip this swap candidate if cannot move to the source cpu. */ + if (!cpumask_test_cpu(env->src_cpu, cur->cpus_ptr)) + goto unlock; + + /* + * Skip this swap candidate if it is not moving to its preferred + * node and the best task is. + */ + if (env->best_task && + env->best_task->numa_preferred_nid == env->src_nid && + cur->numa_preferred_nid != env->src_nid) { + goto unlock; + } + /* * "imp" is the fault differential for the source task between the * source and destination node. Calculate the total differential for * the source task and potential destination task. The more negative * the value is, the more remote accesses that would be expected to * be incurred if the tasks were swapped. - */ - /* Skip this swap candidate if cannot move to the source cpu */ - if (!cpumask_test_cpu(env->src_cpu, &cur->cpus_allowed)) - goto unlock; - - /* + * * If dst and source tasks are in the same NUMA group, or not * in any group then look only at task weights. */ - if (cur->numa_group == env->p->numa_group) { + cur_ng = rcu_dereference(cur->numa_group); + if (cur_ng == p_ng) { + /* + * Do not swap within a group or between tasks that have + * no group if there is spare capacity. Swapping does + * not address the load imbalance and helps one task at + * the cost of punishing another. + */ + if (env->dst_stats.node_type == node_has_spare) + goto unlock; + imp = taskimp + task_weight(cur, env->src_nid, dist) - task_weight(cur, env->dst_nid, dist); /* * Add some hysteresis to prevent swapping the * tasks within a group over tiny differences. */ - if (cur->numa_group) + if (cur_ng) imp -= imp / 16; } else { /* * Compare the group weights. If a task is all by itself * (not part of a group), use the task weight instead. */ - if (cur->numa_group && env->p->numa_group) + if (cur_ng && p_ng) imp += group_weight(cur, env->src_nid, dist) - group_weight(cur, env->dst_nid, dist); else @@ -1637,6 +2316,19 @@ static void task_numa_compare(struct task_numa_env *env, task_weight(cur, env->dst_nid, dist); } + /* Discourage picking a task already on its preferred node */ + if (cur->numa_preferred_nid == env->dst_nid) + imp -= imp / 16; + + /* + * Encourage picking a task that moves to its preferred node. + * This potentially makes imp larger than it's maximum of + * 1998 (see SMALLIMP and task_weight for why) but in this + * case, it does not matter. + */ + if (cur->numa_preferred_nid == env->src_nid) + imp += imp / 8; + if (maymove && moveimp > imp && moveimp > env->best_imp) { imp = moveimp; cur = NULL; @@ -1644,6 +2336,15 @@ static void task_numa_compare(struct task_numa_env *env, } /* + * Prefer swapping with a task moving to its preferred node over a + * task that is not. + */ + if (env->best_task && cur->numa_preferred_nid == env->src_nid && + env->best_task->numa_preferred_nid != env->src_nid) { + goto assign; + } + + /* * If the NUMA importance is less than SMALLIMP, * task migration might only result in ping pong * of tasks and also hurt performance due to cache @@ -1666,50 +2367,105 @@ static void task_numa_compare(struct task_numa_env *env, goto unlock; assign: - /* - * One idle CPU per node is evaluated for a task numa move. - * Call select_idle_sibling to maybe find a better one. - */ + /* Evaluate an idle CPU for a task numa move. */ if (!cur) { + int cpu = env->dst_stats.idle_cpu; + + /* Nothing cached so current CPU went idle since the search. */ + if (cpu < 0) + cpu = env->dst_cpu; + /* - * select_idle_siblings() uses an per-CPU cpumask that - * can be used from IRQ context. + * If the CPU is no longer truly idle and the previous best CPU + * is, keep using it. */ - local_irq_disable(); - env->dst_cpu = select_idle_sibling(env->p, env->src_cpu, - env->dst_cpu); - local_irq_enable(); + if (!idle_cpu(cpu) && env->best_cpu >= 0 && + idle_cpu(env->best_cpu)) { + cpu = env->best_cpu; + } + + env->dst_cpu = cpu; } task_numa_assign(env, cur, imp); + + /* + * If a move to idle is allowed because there is capacity or load + * balance improves then stop the search. While a better swap + * candidate may exist, a search is not free. + */ + if (maymove && !cur && env->best_cpu >= 0 && idle_cpu(env->best_cpu)) + stopsearch = true; + + /* + * If a swap candidate must be identified and the current best task + * moves its preferred node then stop the search. + */ + if (!maymove && env->best_task && + env->best_task->numa_preferred_nid == env->src_nid) { + stopsearch = true; + } unlock: rcu_read_unlock(); + + return stopsearch; } static void task_numa_find_cpu(struct task_numa_env *env, long taskimp, long groupimp) { - long src_load, dst_load, load; bool maymove = false; int cpu; - load = task_h_load(env->p); - dst_load = env->dst_stats.load + load; - src_load = env->src_stats.load - load; - /* - * If the improvement from just moving env->p direction is better - * than swapping tasks around, check if a move is possible. + * If dst node has spare capacity, then check if there is an + * imbalance that would be overruled by the load balancer. */ - maymove = !load_too_imbalanced(src_load, dst_load, env); + if (env->dst_stats.node_type == node_has_spare) { + unsigned int imbalance; + int src_running, dst_running; + + /* + * Would movement cause an imbalance? Note that if src has + * more running tasks that the imbalance is ignored as the + * move improves the imbalance from the perspective of the + * CPU load balancer. + * */ + src_running = env->src_stats.nr_running - 1; + dst_running = env->dst_stats.nr_running + 1; + imbalance = max(0, dst_running - src_running); + imbalance = adjust_numa_imbalance(imbalance, dst_running, + env->imb_numa_nr); + + /* Use idle CPU if there is no imbalance */ + if (!imbalance) { + maymove = true; + if (env->dst_stats.idle_cpu >= 0) { + env->dst_cpu = env->dst_stats.idle_cpu; + task_numa_assign(env, NULL, 0); + return; + } + } + } else { + long src_load, dst_load, load; + /* + * If the improvement from just moving env->p direction is better + * than swapping tasks around, check if a move is possible. + */ + load = task_h_load(env->p); + dst_load = env->dst_stats.load + load; + src_load = env->src_stats.load - load; + maymove = !load_too_imbalanced(src_load, dst_load, env); + } for_each_cpu(cpu, cpumask_of_node(env->dst_nid)) { /* Skip this CPU if the source task cannot migrate */ - if (!cpumask_test_cpu(cpu, &env->p->cpus_allowed)) + if (!cpumask_test_cpu(cpu, env->p->cpus_ptr)) continue; env->dst_cpu = cpu; - task_numa_compare(env, taskimp, groupimp, maymove); + if (task_numa_compare(env, taskimp, groupimp, maymove)) + break; } } @@ -1727,11 +2483,12 @@ static int task_numa_migrate(struct task_struct *p) .best_imp = 0, .best_cpu = -1, }; + unsigned long taskweight, groupweight; struct sched_domain *sd; + long taskimp, groupimp; + struct numa_group *ng; struct rq *best_rq; - unsigned long taskweight, groupweight; int nid, ret, dist; - long taskimp, groupimp; /* * Pick the lowest SD_NUMA domain, as that would have the smallest @@ -1743,8 +2500,10 @@ static int task_numa_migrate(struct task_struct *p) */ rcu_read_lock(); sd = rcu_dereference(per_cpu(sd_numa, env.src_cpu)); - if (sd) + if (sd) { env.imbalance_pct = 100 + (sd->imbalance_pct - 100) / 2; + env.imb_numa_nr = sd->imb_numa_nr; + } rcu_read_unlock(); /* @@ -1762,10 +2521,10 @@ static int task_numa_migrate(struct task_struct *p) dist = env.dist = node_distance(env.src_nid, env.dst_nid); taskweight = task_weight(p, env.src_nid, dist); groupweight = group_weight(p, env.src_nid, dist); - update_numa_stats(&env.src_stats, env.src_nid); + update_numa_stats(&env, &env.src_stats, env.src_nid, false); taskimp = task_weight(p, env.dst_nid, dist) - taskweight; groupimp = group_weight(p, env.dst_nid, dist) - groupweight; - update_numa_stats(&env.dst_stats, env.dst_nid); + update_numa_stats(&env, &env.dst_stats, env.dst_nid, true); /* Try to find a spot on the preferred nid. */ task_numa_find_cpu(&env, taskimp, groupimp); @@ -1777,8 +2536,9 @@ static int task_numa_migrate(struct task_struct *p) * multiple NUMA nodes; in order to better consolidate the group, * we need to check other locations. */ - if (env.best_cpu == -1 || (p->numa_group && p->numa_group->active_nodes > 1)) { - for_each_online_node(nid) { + ng = deref_curr_numa_group(p); + if (env.best_cpu == -1 || (ng && ng->active_nodes > 1)) { + for_each_node_state(nid, N_CPU) { if (nid == env.src_nid || nid == p->numa_preferred_nid) continue; @@ -1797,7 +2557,7 @@ static int task_numa_migrate(struct task_struct *p) env.dist = dist; env.dst_nid = nid; - update_numa_stats(&env.dst_stats, env.dst_nid); + update_numa_stats(&env, &env.dst_stats, env.dst_nid, true); task_numa_find_cpu(&env, taskimp, groupimp); } } @@ -1810,7 +2570,7 @@ static int task_numa_migrate(struct task_struct *p) * A task that migrated to a second choice node will be better off * trying for a better one later. Do not set the preferred node here. */ - if (p->numa_group) { + if (ng) { if (env.best_cpu == -1) nid = env.src_nid; else @@ -1821,15 +2581,17 @@ static int task_numa_migrate(struct task_struct *p) } /* No better CPU than the current one was found. */ - if (env.best_cpu == -1) + if (env.best_cpu == -1) { + trace_sched_stick_numa(p, env.src_cpu, NULL, -1); return -EAGAIN; + } best_rq = cpu_rq(env.best_cpu); if (env.best_task == NULL) { ret = migrate_task_to(p, env.best_cpu); WRITE_ONCE(best_rq->numa_migrate_on, 0); if (ret != 0) - trace_sched_stick_numa(p, env.src_cpu, env.best_cpu); + trace_sched_stick_numa(p, env.src_cpu, NULL, env.best_cpu); return ret; } @@ -1837,7 +2599,7 @@ static int task_numa_migrate(struct task_struct *p) WRITE_ONCE(best_rq->numa_migrate_on, 0); if (ret != 0) - trace_sched_stick_numa(p, env.src_cpu, task_cpu(env.best_task)); + trace_sched_stick_numa(p, env.src_cpu, env.best_task, env.best_cpu); put_task_struct(env.best_task); return ret; } @@ -1848,7 +2610,7 @@ static void numa_migrate_preferred(struct task_struct *p) unsigned long interval = HZ; /* This task has no NUMA fault statistics yet */ - if (unlikely(p->numa_preferred_nid == -1 || !p->numa_faults)) + if (unlikely(p->numa_preferred_nid == NUMA_NO_NODE || !p->numa_faults)) return; /* Periodically retry migrating the task to the preferred node */ @@ -1864,7 +2626,7 @@ static void numa_migrate_preferred(struct task_struct *p) } /* - * Find out how many nodes on the workload is actively running on. Do this by + * Find out how many nodes the workload is actively running on. Do this by * tracking the nodes from which NUMA hinting faults are triggered. This can * be different from the set of nodes where the workload's memory is currently * located. @@ -1874,13 +2636,13 @@ static void numa_group_count_active_nodes(struct numa_group *numa_group) unsigned long faults, max_faults = 0; int nid, active_nodes = 0; - for_each_online_node(nid) { + for_each_node_state(nid, N_CPU) { faults = group_faults_cpu(numa_group, nid); if (faults > max_faults) max_faults = faults; } - for_each_online_node(nid) { + for_each_node_state(nid, N_CPU) { faults = group_faults_cpu(numa_group, nid); if (faults * ACTIVE_NODE_FRACTION > max_faults) active_nodes++; @@ -1918,7 +2680,7 @@ static void update_task_scan_period(struct task_struct *p, /* * If there were no record hinting faults then either the task is - * completely idle or all activity is areas that are not of interest + * completely idle or all activity is in areas that are not of interest * to automatic numa balancing. Related to that, if there were failed * migration then it implies we are migrating too quickly or the local * node is overloaded. In either case, scan slower @@ -1994,6 +2756,10 @@ static u64 numa_get_avg_runtime(struct task_struct *p, u64 *period) if (p->last_task_numa_placement) { delta = runtime - p->last_sum_exec_runtime; *period = now - p->last_task_numa_placement; + + /* Avoid time going backwards, prevent potential divide error: */ + if (unlikely((s64)*period < 0)) + *period = 0; } else { delta = p->se.avg.load_sum; *period = LOAD_AVG_MAX; @@ -2030,7 +2796,7 @@ static int preferred_group_nid(struct task_struct *p, int nid) dist = sched_max_numa_distance; - for_each_online_node(node) { + for_each_node_state(node, N_CPU) { score = group_weight(p, node, dist); if (score > max_score) { max_score = score; @@ -2049,7 +2815,7 @@ static int preferred_group_nid(struct task_struct *p, int nid) * inside the highest scoring group of nodes. The nodemask tricks * keep the complexity of the search down. */ - nodes = node_online_map; + nodes = node_states[N_CPU]; for (dist = sched_max_numa_distance; dist > LOCAL_DISTANCE; dist--) { unsigned long max_faults = 0; nodemask_t max_group = NODE_MASK_NONE; @@ -2095,12 +2861,13 @@ static int preferred_group_nid(struct task_struct *p, int nid) static void task_numa_placement(struct task_struct *p) { - int seq, nid, max_nid = -1; + int seq, nid, max_nid = NUMA_NO_NODE; unsigned long max_faults = 0; unsigned long fault_types[2] = { 0, 0 }; unsigned long total_faults; u64 runtime, period; spinlock_t *group_lock = NULL; + struct numa_group *ng; /* * The p->mm->numa_scan_seq field gets updated without @@ -2118,8 +2885,9 @@ static void task_numa_placement(struct task_struct *p) runtime = numa_get_avg_runtime(p, &period); /* If the task is part of a group prevent parallel updates to group stats */ - if (p->numa_group) { - group_lock = &p->numa_group->lock; + ng = deref_curr_numa_group(p); + if (ng) { + group_lock = &ng->lock; spin_lock_irq(group_lock); } @@ -2160,7 +2928,7 @@ static void task_numa_placement(struct task_struct *p) p->numa_faults[cpu_idx] += f_diff; faults += p->numa_faults[mem_idx]; p->total_numa_faults += diff; - if (p->numa_group) { + if (ng) { /* * safe because we can only change our own group * @@ -2168,14 +2936,14 @@ static void task_numa_placement(struct task_struct *p) * nid and priv in a specific region because it * is at the beginning of the numa_faults array. */ - p->numa_group->faults[mem_idx] += diff; - p->numa_group->faults_cpu[mem_idx] += f_diff; - p->numa_group->total_faults += diff; - group_faults += p->numa_group->faults[mem_idx]; + ng->faults[mem_idx] += diff; + ng->faults[cpu_idx] += f_diff; + ng->total_faults += diff; + group_faults += ng->faults[mem_idx]; } } - if (!p->numa_group) { + if (!ng) { if (faults > max_faults) { max_faults = faults; max_nid = nid; @@ -2186,8 +2954,11 @@ static void task_numa_placement(struct task_struct *p) } } - if (p->numa_group) { - numa_group_count_active_nodes(p->numa_group); + /* Cannot migrate task to CPU-less node */ + max_nid = numa_nearest_node(max_nid, N_CPU); + + if (ng) { + numa_group_count_active_nodes(ng); spin_unlock_irq(group_lock); max_nid = preferred_group_nid(p, max_nid); } @@ -2203,12 +2974,12 @@ static void task_numa_placement(struct task_struct *p) static inline int get_numa_group(struct numa_group *grp) { - return atomic_inc_not_zero(&grp->refcount); + return refcount_inc_not_zero(&grp->refcount); } static inline void put_numa_group(struct numa_group *grp) { - if (atomic_dec_and_test(&grp->refcount)) + if (refcount_dec_and_test(&grp->refcount)) kfree_rcu(grp, rcu); } @@ -2221,22 +2992,20 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, int cpu = cpupid_to_cpu(cpupid); int i; - if (unlikely(!p->numa_group)) { + if (unlikely(!deref_curr_numa_group(p))) { unsigned int size = sizeof(struct numa_group) + - 4*nr_node_ids*sizeof(unsigned long); + NR_NUMA_HINT_FAULT_STATS * + nr_node_ids * sizeof(unsigned long); grp = kzalloc(size, GFP_KERNEL | __GFP_NOWARN); if (!grp) return; - atomic_set(&grp->refcount, 1); + refcount_set(&grp->refcount, 1); grp->active_nodes = 1; grp->max_faults_cpu = 0; spin_lock_init(&grp->lock); grp->gid = p->pid; - /* Second half of the array tracks nids where faults happen */ - grp->faults_cpu = grp->faults + NR_NUMA_HINT_FAULT_TYPES * - nr_node_ids; for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++) grp->faults[i] = p->numa_faults[i]; @@ -2257,7 +3026,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, if (!grp) goto no_join; - my_grp = p->numa_group; + my_grp = deref_curr_numa_group(p); if (grp == my_grp) goto no_join; @@ -2293,7 +3062,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, if (!join) return; - BUG_ON(irqs_disabled()); + WARN_ON_ONCE(irqs_disabled()); double_lock_irq(&my_grp->lock, &grp->lock); for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++) { @@ -2319,13 +3088,24 @@ no_join: return; } -void task_numa_free(struct task_struct *p) +/* + * Get rid of NUMA statistics associated with a task (either current or dead). + * If @final is set, the task is dead and has reached refcount zero, so we can + * safely free all relevant data structures. Otherwise, there might be + * concurrent reads from places like load balancing and procfs, and we should + * reset the data back to default state without freeing ->numa_faults. + */ +void task_numa_free(struct task_struct *p, bool final) { - struct numa_group *grp = p->numa_group; - void *numa_faults = p->numa_faults; + /* safe: p either is current or is being freed by current */ + struct numa_group *grp = rcu_dereference_raw(p->numa_group); + unsigned long *numa_faults = p->numa_faults; unsigned long flags; int i; + if (!numa_faults) + return; + if (grp) { spin_lock_irqsave(&grp->lock, flags); for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++) @@ -2338,8 +3118,14 @@ void task_numa_free(struct task_struct *p) put_numa_group(grp); } - p->numa_faults = NULL; - kfree(numa_faults); + if (final) { + p->numa_faults = NULL; + kfree(numa_faults); + } else { + p->total_numa_faults = 0; + for (i = 0; i < NR_NUMA_HINT_FAULT_STATS * nr_node_ids; i++) + numa_faults[i] = 0; + } } /* @@ -2361,6 +3147,15 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags) if (!p->mm) return; + /* + * NUMA faults statistics are unnecessary for the slow memory + * node for memory tiering mode. + */ + if (!node_is_toptier(mem_node) && + (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING || + !cpupid_valid(last_cpupid))) + return; + /* Allocate buffer to track faults on a per-node basis */ if (unlikely(!p->numa_faults)) { int size = sizeof(*p->numa_faults) * @@ -2392,7 +3187,7 @@ void task_numa_fault(int last_cpupid, int mem_node, int pages, int flags) * actively using should be counted as local. This allows the * scan rate to slow down when a workload has settled down. */ - ng = p->numa_group; + ng = deref_curr_numa_group(p); if (!priv && !local && ng && ng->active_nodes > 1 && numa_is_active_node(cpu_node, ng) && numa_is_active_node(mem_node, ng)) @@ -2431,11 +3226,50 @@ static void reset_ptenuma_scan(struct task_struct *p) p->mm->numa_scan_offset = 0; } +static bool vma_is_accessed(struct mm_struct *mm, struct vm_area_struct *vma) +{ + unsigned long pids; + /* + * Allow unconditional access first two times, so that all the (pages) + * of VMAs get prot_none fault introduced irrespective of accesses. + * This is also done to avoid any side effect of task scanning + * amplifying the unfairness of disjoint set of VMAs' access. + */ + if ((READ_ONCE(current->mm->numa_scan_seq) - vma->numab_state->start_scan_seq) < 2) + return true; + + pids = vma->numab_state->pids_active[0] | vma->numab_state->pids_active[1]; + if (test_bit(hash_32(current->pid, ilog2(BITS_PER_LONG)), &pids)) + return true; + + /* + * Complete a scan that has already started regardless of PID access, or + * some VMAs may never be scanned in multi-threaded applications: + */ + if (mm->numa_scan_offset > vma->vm_start) { + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_IGNORE_PID); + return true; + } + + /* + * This vma has not been accessed for a while, and if the number + * the threads in the same process is low, which means no other + * threads can help scan this vma, force a vma scan. + */ + if (READ_ONCE(mm->numa_scan_seq) > + (vma->numab_state->prev_scan_seq + get_nr_threads(current))) + return true; + + return false; +} + +#define VMA_PID_RESET_PERIOD (4 * sysctl_numa_balancing_scan_delay) + /* * The expensive part of numa migration is done from task_work context. * Triggered from task_tick_numa(). */ -void task_numa_work(struct callback_head *work) +static void task_numa_work(struct callback_head *work) { unsigned long migrate, next_scan, now = jiffies; struct task_struct *p = current; @@ -2445,10 +3279,13 @@ void task_numa_work(struct callback_head *work) unsigned long start, end; unsigned long nr_pte_updates = 0; long pages, virtpages; + struct vma_iterator vmi; + bool vma_pids_skipped; + bool vma_pids_forced = false; - SCHED_WARN_ON(p != container_of(work, struct task_struct, numa_work)); + WARN_ON_ONCE(p != container_of(work, struct task_struct, numa_work)); - work->next = work; /* protect against double add */ + work->next = work; /* * Who cares about NUMA placement when they're dying. * @@ -2460,6 +3297,15 @@ void task_numa_work(struct callback_head *work) if (p->flags & PF_EXITING) return; + /* + * Memory is pinned to only one NUMA node via cpuset.mems, naturally + * no page can be migrated. + */ + if (cpusets_enabled() && nodes_weight(cpuset_current_mems_allowed) == 1) { + trace_sched_skip_cpuset_numa(current, &cpuset_current_mems_allowed); + return; + } + if (!mm->numa_next_scan) { mm->numa_next_scan = now + msecs_to_jiffies(sysctl_numa_balancing_scan_delay); @@ -2478,7 +3324,7 @@ void task_numa_work(struct callback_head *work) } next_scan = now + msecs_to_jiffies(p->numa_scan_period); - if (cmpxchg(&mm->numa_next_scan, migrate, next_scan) != migrate) + if (!try_cmpxchg(&mm->numa_next_scan, &migrate, next_scan)) return; /* @@ -2487,7 +3333,6 @@ void task_numa_work(struct callback_head *work) */ p->node_stamp += 2 * TICK_NSEC; - start = mm->numa_scan_offset; pages = sysctl_numa_balancing_scan_size; pages <<= 20 - PAGE_SHIFT; /* MB in pages */ virtpages = pages * 8; /* Scan up to this much virtual space */ @@ -2495,36 +3340,120 @@ void task_numa_work(struct callback_head *work) return; - if (!down_read_trylock(&mm->mmap_sem)) + if (!mmap_read_trylock(mm)) return; - vma = find_vma(mm, start); + + /* + * VMAs are skipped if the current PID has not trapped a fault within + * the VMA recently. Allow scanning to be forced if there is no + * suitable VMA remaining. + */ + vma_pids_skipped = false; + +retry_pids: + start = mm->numa_scan_offset; + vma_iter_init(&vmi, mm, start); + vma = vma_next(&vmi); if (!vma) { reset_ptenuma_scan(p); start = 0; - vma = mm->mmap; + vma_iter_set(&vmi, start); + vma = vma_next(&vmi); } - for (; vma; vma = vma->vm_next) { + + for (; vma; vma = vma_next(&vmi)) { if (!vma_migratable(vma) || !vma_policy_mof(vma) || is_vm_hugetlb_page(vma) || (vma->vm_flags & VM_MIXEDMAP)) { + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_UNSUITABLE); continue; } /* * Shared library pages mapped by multiple processes are not * migrated as it is expected they are cache replicated. Avoid - * hinting faults in read-only file-backed mappings or the vdso + * hinting faults in read-only file-backed mappings or the vDSO * as migrating the pages will be of marginal benefit. */ if (!vma->vm_mm || - (vma->vm_file && (vma->vm_flags & (VM_READ|VM_WRITE)) == (VM_READ))) + (vma->vm_file && (vma->vm_flags & (VM_READ|VM_WRITE)) == (VM_READ))) { + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_SHARED_RO); continue; + } /* * Skip inaccessible VMAs to avoid any confusion between - * PROT_NONE and NUMA hinting ptes + * PROT_NONE and NUMA hinting PTEs */ - if (!(vma->vm_flags & (VM_READ | VM_EXEC | VM_WRITE))) + if (!vma_is_accessible(vma)) { + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_INACCESSIBLE); continue; + } + + /* Initialise new per-VMA NUMAB state. */ + if (!vma->numab_state) { + struct vma_numab_state *ptr; + + ptr = kzalloc(sizeof(*ptr), GFP_KERNEL); + if (!ptr) + continue; + + if (cmpxchg(&vma->numab_state, NULL, ptr)) { + kfree(ptr); + continue; + } + + vma->numab_state->start_scan_seq = mm->numa_scan_seq; + + vma->numab_state->next_scan = now + + msecs_to_jiffies(sysctl_numa_balancing_scan_delay); + + /* Reset happens after 4 times scan delay of scan start */ + vma->numab_state->pids_active_reset = vma->numab_state->next_scan + + msecs_to_jiffies(VMA_PID_RESET_PERIOD); + + /* + * Ensure prev_scan_seq does not match numa_scan_seq, + * to prevent VMAs being skipped prematurely on the + * first scan: + */ + vma->numab_state->prev_scan_seq = mm->numa_scan_seq - 1; + } + + /* + * Scanning the VMAs of short lived tasks add more overhead. So + * delay the scan for new VMAs. + */ + if (mm->numa_scan_seq && time_before(jiffies, + vma->numab_state->next_scan)) { + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_SCAN_DELAY); + continue; + } + + /* RESET access PIDs regularly for old VMAs. */ + if (mm->numa_scan_seq && + time_after(jiffies, vma->numab_state->pids_active_reset)) { + vma->numab_state->pids_active_reset = vma->numab_state->pids_active_reset + + msecs_to_jiffies(VMA_PID_RESET_PERIOD); + vma->numab_state->pids_active[0] = READ_ONCE(vma->numab_state->pids_active[1]); + vma->numab_state->pids_active[1] = 0; + } + + /* Do not rescan VMAs twice within the same sequence. */ + if (vma->numab_state->prev_scan_seq == mm->numa_scan_seq) { + mm->numa_scan_offset = vma->vm_end; + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_SEQ_COMPLETED); + continue; + } + + /* + * Do not scan the VMA if task has not accessed it, unless no other + * VMA candidate exists. + */ + if (!vma_pids_forced && !vma_is_accessed(mm, vma)) { + vma_pids_skipped = true; + trace_sched_skip_vma_numa(mm, vma, NUMAB_SKIP_PID_INACTIVE); + continue; + } do { start = max(start, vma->vm_start); @@ -2535,7 +3464,7 @@ void task_numa_work(struct callback_head *work) /* * Try to scan sysctl_numa_balancing_size worth of * hpages that have at least one present PTE that - * is not already pte-numa. If the VMA contains + * is not already PTE-numa. If the VMA contains * areas that are unused or already full of prot_numa * PTEs, scan up to virtpages, to skip through those * areas faster. @@ -2550,6 +3479,26 @@ void task_numa_work(struct callback_head *work) cond_resched(); } while (end != vma->vm_end); + + /* VMA scan is complete, do not scan until next sequence. */ + vma->numab_state->prev_scan_seq = mm->numa_scan_seq; + + /* + * Only force scan within one VMA at a time, to limit the + * cost of scanning a potentially uninteresting VMA. + */ + if (vma_pids_forced) + break; + } + + /* + * If no VMAs are remaining and VMAs were skipped due to the PID + * not accessing the VMA previously, then force a scan to ensure + * forward progress: + */ + if (!vma && !vma_pids_forced && vma_pids_skipped) { + vma_pids_forced = true; + goto retry_pids; } out: @@ -2563,7 +3512,7 @@ out: mm->numa_scan_offset = start; else reset_ptenuma_scan(p); - up_read(&mm->mmap_sem); + mmap_read_unlock(mm); /* * Make sure tasks use at least 32x as much time to run other code @@ -2577,10 +3526,57 @@ out: } } +void init_numa_balancing(u64 clone_flags, struct task_struct *p) +{ + int mm_users = 0; + struct mm_struct *mm = p->mm; + + if (mm) { + mm_users = atomic_read(&mm->mm_users); + if (mm_users == 1) { + mm->numa_next_scan = jiffies + msecs_to_jiffies(sysctl_numa_balancing_scan_delay); + mm->numa_scan_seq = 0; + } + } + p->node_stamp = 0; + p->numa_scan_seq = mm ? mm->numa_scan_seq : 0; + p->numa_scan_period = sysctl_numa_balancing_scan_delay; + p->numa_migrate_retry = 0; + /* Protect against double add, see task_tick_numa and task_numa_work */ + p->numa_work.next = &p->numa_work; + p->numa_faults = NULL; + p->numa_pages_migrated = 0; + p->total_numa_faults = 0; + RCU_INIT_POINTER(p->numa_group, NULL); + p->last_task_numa_placement = 0; + p->last_sum_exec_runtime = 0; + + init_task_work(&p->numa_work, task_numa_work); + + /* New address space, reset the preferred nid */ + if (!(clone_flags & CLONE_VM)) { + p->numa_preferred_nid = NUMA_NO_NODE; + return; + } + + /* + * New thread, keep existing numa_preferred_nid which should be copied + * already by arch_dup_task_struct but stagger when scans start. + */ + if (mm) { + unsigned int delay; + + delay = min_t(unsigned int, task_scan_max(current), + current->numa_scan_period * mm_users * NSEC_PER_MSEC); + delay += 2 * TICK_NSEC; + p->node_stamp = delay; + } +} + /* * Drive the periodic memory faults.. */ -void task_tick_numa(struct rq *rq, struct task_struct *curr) +static void task_tick_numa(struct rq *rq, struct task_struct *curr) { struct callback_head *work = &curr->numa_work; u64 period, now; @@ -2588,7 +3584,7 @@ void task_tick_numa(struct rq *rq, struct task_struct *curr) /* * We don't care about NUMA placement if we don't have memory. */ - if (!curr->mm || (curr->flags & PF_EXITING) || work->next != work) + if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) || work->next != work) return; /* @@ -2605,10 +3601,8 @@ void task_tick_numa(struct rq *rq, struct task_struct *curr) curr->numa_scan_period = task_scan_start(curr); curr->node_stamp += period; - if (!time_before(jiffies, curr->mm->numa_next_scan)) { - init_task_work(work, task_numa_work); /* TODO: move this into sched_fork() */ - task_work_add(curr, work, true); - } + if (!time_before(jiffies, curr->mm->numa_next_scan)) + task_work_add(curr, work, TWA_RESUME); } } @@ -2638,14 +3632,16 @@ static void update_scan_period(struct task_struct *p, int new_cpu) * the preferred node. */ if (dst_nid == p->numa_preferred_nid || - (p->numa_preferred_nid != -1 && src_nid != p->numa_preferred_nid)) + (p->numa_preferred_nid != NUMA_NO_NODE && + src_nid != p->numa_preferred_nid)) return; } p->numa_scan_period = task_scan_start(p); } -#else +#else /* !CONFIG_NUMA_BALANCING: */ + static void task_tick_numa(struct rq *rq, struct task_struct *curr) { } @@ -2662,38 +3658,30 @@ static inline void update_scan_period(struct task_struct *p, int new_cpu) { } -#endif /* CONFIG_NUMA_BALANCING */ +#endif /* !CONFIG_NUMA_BALANCING */ static void account_entity_enqueue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_add(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - update_load_add(&rq_of(cfs_rq)->load, se->load.weight); -#ifdef CONFIG_SMP if (entity_is_task(se)) { struct rq *rq = rq_of(cfs_rq); account_numa_enqueue(rq, task_of(se)); list_add(&se->group_node, &rq->cfs_tasks); } -#endif - cfs_rq->nr_running++; + cfs_rq->nr_queued++; } static void account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) { update_load_sub(&cfs_rq->load, se->load.weight); - if (!parent_entity(se)) - update_load_sub(&rq_of(cfs_rq)->load, se->load.weight); -#ifdef CONFIG_SMP if (entity_is_task(se)) { account_numa_dequeue(rq_of(cfs_rq), task_of(se)); list_del_init(&se->group_node); } -#endif - cfs_rq->nr_running--; + cfs_rq->nr_queued--; } /* @@ -2744,26 +3732,6 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) *ptr -= min_t(typeof(*ptr), *ptr, _val); \ } while (0) -#ifdef CONFIG_SMP -static inline void -enqueue_runnable_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ - cfs_rq->runnable_weight += se->runnable_weight; - - cfs_rq->avg.runnable_load_avg += se->avg.runnable_load_avg; - cfs_rq->avg.runnable_load_sum += se_runnable(se) * se->avg.runnable_load_sum; -} - -static inline void -dequeue_runnable_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ - cfs_rq->runnable_weight -= se->runnable_weight; - - sub_positive(&cfs_rq->avg.runnable_load_avg, se->avg.runnable_load_avg); - sub_positive(&cfs_rq->avg.runnable_load_sum, - se_runnable(se) * se->avg.runnable_load_sum); -} - static inline void enqueue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { @@ -2776,63 +3744,71 @@ dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { sub_positive(&cfs_rq->avg.load_avg, se->avg.load_avg); sub_positive(&cfs_rq->avg.load_sum, se_weight(se) * se->avg.load_sum); + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum, + cfs_rq->avg.load_avg * PELT_MIN_DIVIDER); } -#else -static inline void -enqueue_runnable_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } -static inline void -dequeue_runnable_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } -static inline void -enqueue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } -static inline void -dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { } -#endif + +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, unsigned long runnable) + unsigned long weight) { + bool curr = cfs_rq->curr == se; + if (se->on_rq) { /* commit outstanding execution time */ - if (cfs_rq->curr == se) - update_curr(cfs_rq); - account_entity_dequeue(cfs_rq, se); - dequeue_runnable_load_avg(cfs_rq, se); + update_curr(cfs_rq); + update_entity_lag(cfs_rq, se); + se->deadline -= se->vruntime; + se->rel_deadline = 1; + cfs_rq->nr_queued--; + if (!curr) + __dequeue_entity(cfs_rq, se); + update_load_sub(&cfs_rq->load, se->load.weight); } dequeue_load_avg(cfs_rq, se); - se->runnable_weight = runnable; + /* + * 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); -#ifdef CONFIG_SMP do { - u32 divider = LOAD_AVG_MAX - 1024 + se->avg.period_contrib; + u32 divider = get_pelt_divider(&se->avg); se->avg.load_avg = div_u64(se_weight(se) * se->avg.load_sum, divider); - se->avg.runnable_load_avg = - div_u64(se_runnable(se) * se->avg.runnable_load_sum, divider); } while (0); -#endif enqueue_load_avg(cfs_rq, se); if (se->on_rq) { - account_entity_enqueue(cfs_rq, se); - enqueue_runnable_load_avg(cfs_rq, se); + place_entity(cfs_rq, se, 0); + update_load_add(&cfs_rq->load, se->load.weight); + if (!curr) + __enqueue_entity(cfs_rq, se); + cfs_rq->nr_queued++; } } -void reweight_task(struct task_struct *p, int prio) +static void reweight_task_fair(struct rq *rq, struct task_struct *p, + const struct load_weight *lw) { struct sched_entity *se = &p->se; struct cfs_rq *cfs_rq = cfs_rq_of(se); struct load_weight *load = &se->load; - unsigned long weight = scale_load(sched_prio_to_weight[prio]); - reweight_entity(cfs_rq, se, weight, weight); - load->inv_weight = sched_prio_to_wmult[prio]; + reweight_entity(cfs_rq, se, lw->weight); + load->inv_weight = lw->inv_weight; } +static inline int throttled_hierarchy(struct cfs_rq *cfs_rq); + #ifdef CONFIG_FAIR_GROUP_SCHED -#ifdef CONFIG_SMP /* * All this does is approximate the hierarchical proportion which includes that * global sum we all love to hate. @@ -2842,7 +3818,7 @@ void reweight_task(struct task_struct *p, int prio) * * tg->weight * grq->load.weight * ge->load.weight = ----------------------------- (1) - * \Sum grq->load.weight + * \Sum grq->load.weight * * Now, because computing that sum is prohibitively expensive to compute (been * there, done that) we approximate it with this average stuff. The average @@ -2856,7 +3832,7 @@ void reweight_task(struct task_struct *p, int prio) * * tg->weight * grq->avg.load_avg * ge->load.weight = ------------------------------ (3) - * tg->load_avg + * tg->load_avg * * Where: tg->load_avg ~= \Sum grq->avg.load_avg * @@ -2872,7 +3848,7 @@ void reweight_task(struct task_struct *p, int prio) * * tg->weight * grq->load.weight * ge->load.weight = ----------------------------- = tg->weight (4) - * grp->load.weight + * grp->load.weight * * That is, the sum collapses because all other CPUs are idle; the UP scenario. * @@ -2891,7 +3867,7 @@ void reweight_task(struct task_struct *p, int prio) * * tg->weight * grq->load.weight * ge->load.weight = ----------------------------- (6) - * tg_load_avg' + * tg_load_avg' * * Where: * @@ -2941,91 +3917,37 @@ static long calc_group_shares(struct cfs_rq *cfs_rq) } /* - * This calculates the effective runnable weight for a group entity based on - * the group entity weight calculated above. - * - * Because of the above approximation (2), our group entity weight is - * an load_avg based ratio (3). This means that it includes blocked load and - * does not represent the runnable weight. - * - * Approximate the group entity's runnable weight per ratio from the group - * runqueue: - * - * grq->avg.runnable_load_avg - * ge->runnable_weight = ge->load.weight * -------------------------- (7) - * grq->avg.load_avg - * - * However, analogous to above, since the avg numbers are slow, this leads to - * transients in the from-idle case. Instead we use: - * - * ge->runnable_weight = ge->load.weight * - * - * max(grq->avg.runnable_load_avg, grq->runnable_weight) - * ----------------------------------------------------- (8) - * max(grq->avg.load_avg, grq->load.weight) - * - * Where these max() serve both to use the 'instant' values to fix the slow - * from-idle and avoid the /0 on to-idle, similar to (6). - */ -static long calc_group_runnable(struct cfs_rq *cfs_rq, long shares) -{ - long runnable, load_avg; - - load_avg = max(cfs_rq->avg.load_avg, - scale_load_down(cfs_rq->load.weight)); - - runnable = max(cfs_rq->avg.runnable_load_avg, - scale_load_down(cfs_rq->runnable_weight)); - - runnable *= shares; - if (load_avg) - runnable /= load_avg; - - return clamp_t(long, runnable, MIN_SHARES, shares); -} -#endif /* CONFIG_SMP */ - -static inline int throttled_hierarchy(struct cfs_rq *cfs_rq); - -/* * Recomputes the group entity based on the current state of its group * runqueue. */ static void update_cfs_group(struct sched_entity *se) { struct cfs_rq *gcfs_rq = group_cfs_rq(se); - long shares, runnable; - - if (!gcfs_rq) - return; - - if (throttled_hierarchy(gcfs_rq)) - return; - -#ifndef CONFIG_SMP - runnable = shares = READ_ONCE(gcfs_rq->tg->shares); + long shares; - if (likely(se->load.weight == shares)) + /* + * When a group becomes empty, preserve its weight. This matters for + * DELAY_DEQUEUE. + */ + if (!gcfs_rq || !gcfs_rq->load.weight) return; -#else - shares = calc_group_shares(gcfs_rq); - runnable = calc_group_runnable(gcfs_rq, shares); -#endif - reweight_entity(cfs_rq_of(se), se, shares, runnable); + shares = calc_group_shares(gcfs_rq); + if (unlikely(se->load.weight != shares)) + reweight_entity(cfs_rq_of(se), se, shares); } -#else /* CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static inline void update_cfs_group(struct sched_entity *se) { } -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* !CONFIG_FAIR_GROUP_SCHED */ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq, int flags) { struct rq *rq = rq_of(cfs_rq); - if (&rq->cfs == cfs_rq || (flags & SCHED_CPUFREQ_MIGRATION)) { + if (&rq->cfs == cfs_rq) { /* * There are a few boundary cases this might miss but it should * get called often enough that that should (hopefully) not be @@ -3038,18 +3960,89 @@ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq, int flags) * As is, the util number is not freq-invariant (we'd have to * implement arch_scale_freq_capacity() for that). * - * See cpu_util(). + * See cpu_util_cfs(). */ cpufreq_update_util(rq, flags); } } -#ifdef CONFIG_SMP +static inline bool load_avg_is_decayed(struct sched_avg *sa) +{ + if (sa->load_sum) + return false; + + if (sa->util_sum) + return false; + + if (sa->runnable_sum) + return false; + + /* + * _avg must be null when _sum are null because _avg = _sum / divider + * Make sure that rounding and/or propagation of PELT values never + * break this. + */ + WARN_ON_ONCE(sa->load_avg || + sa->util_avg || + sa->runnable_avg); + + return true; +} + +static inline u64 cfs_rq_last_update_time(struct cfs_rq *cfs_rq) +{ + return u64_u32_load_copy(cfs_rq->avg.last_update_time, + cfs_rq->last_update_time_copy); +} #ifdef CONFIG_FAIR_GROUP_SCHED +/* + * Because list_add_leaf_cfs_rq always places a child cfs_rq on the list + * immediately before a parent cfs_rq, and cfs_rqs are removed from the list + * bottom-up, we only have to test whether the cfs_rq before us on the list + * is our child. + * If cfs_rq is not on the list, test whether a child needs its to be added to + * connect a branch to the tree * (see list_add_leaf_cfs_rq() for details). + */ +static inline bool child_cfs_rq_on_list(struct cfs_rq *cfs_rq) +{ + struct cfs_rq *prev_cfs_rq; + struct list_head *prev; + struct rq *rq = rq_of(cfs_rq); + + if (cfs_rq->on_list) { + prev = cfs_rq->leaf_cfs_rq_list.prev; + } else { + prev = rq->tmp_alone_branch; + } + + if (prev == &rq->leaf_cfs_rq_list) + return false; + + prev_cfs_rq = container_of(prev, struct cfs_rq, leaf_cfs_rq_list); + + return (prev_cfs_rq->tg->parent == cfs_rq->tg); +} + +static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) +{ + if (cfs_rq->load.weight) + return false; + + if (!load_avg_is_decayed(&cfs_rq->avg)) + return false; + + if (child_cfs_rq_on_list(cfs_rq)) + return false; + + if (cfs_rq->tg_load_avg_contrib) + return false; + + return true; +} + /** * update_tg_load_avg - update the tg's load avg * @cfs_rq: the cfs_rq whose avg changed - * @force: update regardless of how small the difference * * This function 'ensures': tg->load_avg := \Sum tg->cfs_rq[]->avg.load. * However, because tg->load_avg is a global value there are performance @@ -3061,9 +4054,10 @@ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq, int flags) * * Updating tg's load_avg is necessary before update_cfs_share(). */ -static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) +static inline void update_tg_load_avg(struct cfs_rq *cfs_rq) { - long delta = cfs_rq->avg.load_avg - cfs_rq->tg_load_avg_contrib; + long delta; + u64 now; /* * No need to update load_avg for root_task_group as it is not used. @@ -3071,10 +4065,67 @@ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) if (cfs_rq->tg == &root_task_group) return; - if (force || abs(delta) > cfs_rq->tg_load_avg_contrib / 64) { + /* rq has been offline and doesn't contribute to the share anymore: */ + if (!cpu_active(cpu_of(rq_of(cfs_rq)))) + return; + + /* + * For migration heavy workloads, access to tg->load_avg can be + * unbound. Limit the update rate to at most once per ms. + */ + now = sched_clock_cpu(cpu_of(rq_of(cfs_rq))); + if (now - cfs_rq->last_update_tg_load_avg < NSEC_PER_MSEC) + return; + + delta = cfs_rq->avg.load_avg - cfs_rq->tg_load_avg_contrib; + if (abs(delta) > cfs_rq->tg_load_avg_contrib / 64) { atomic_long_add(delta, &cfs_rq->tg->load_avg); cfs_rq->tg_load_avg_contrib = cfs_rq->avg.load_avg; + cfs_rq->last_update_tg_load_avg = now; + } +} + +static inline void clear_tg_load_avg(struct cfs_rq *cfs_rq) +{ + long delta; + u64 now; + + /* + * No need to update load_avg for root_task_group, as it is not used. + */ + if (cfs_rq->tg == &root_task_group) + return; + + now = sched_clock_cpu(cpu_of(rq_of(cfs_rq))); + delta = 0 - cfs_rq->tg_load_avg_contrib; + atomic_long_add(delta, &cfs_rq->tg->load_avg); + cfs_rq->tg_load_avg_contrib = 0; + cfs_rq->last_update_tg_load_avg = now; +} + +/* CPU offline callback: */ +static void __maybe_unused clear_tg_offline_cfs_rqs(struct rq *rq) +{ + struct task_group *tg; + + lockdep_assert_rq_held(rq); + + /* + * The rq clock has already been updated in + * set_rq_offline(), so we should skip updating + * the rq clock again in unthrottle_cfs_rq(). + */ + rq_clock_start_loop_update(rq); + + rcu_read_lock(); + list_for_each_entry_rcu(tg, &task_groups, list) { + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; + + clear_tg_load_avg(cfs_rq); } + rcu_read_unlock(); + + rq_clock_stop_loop_update(rq); } /* @@ -3101,32 +4152,13 @@ void set_task_rq_fair(struct sched_entity *se, if (!(se->avg.last_update_time && prev)) return; -#ifndef CONFIG_64BIT - { - u64 p_last_update_time_copy; - u64 n_last_update_time_copy; - - do { - p_last_update_time_copy = prev->load_last_update_time_copy; - n_last_update_time_copy = next->load_last_update_time_copy; - - smp_rmb(); - - p_last_update_time = prev->avg.last_update_time; - n_last_update_time = next->avg.last_update_time; + p_last_update_time = cfs_rq_last_update_time(prev); + n_last_update_time = cfs_rq_last_update_time(next); - } while (p_last_update_time != p_last_update_time_copy || - n_last_update_time != n_last_update_time_copy); - } -#else - p_last_update_time = prev->avg.last_update_time; - n_last_update_time = next->avg.last_update_time; -#endif - __update_load_avg_blocked_se(p_last_update_time, cpu_of(rq_of(prev)), se); + __update_load_avg_blocked_se(p_last_update_time, se); se->avg.last_update_time = n_last_update_time; } - /* * When on migration a sched_entity joins/leaves the PELT hierarchy, we need to * propagate its contribution. The key to this propagation is the invariant @@ -3137,11 +4169,11 @@ void set_task_rq_fair(struct sched_entity *se, * _IFF_ we look at the pure running and runnable sums. Because they * represent the very same entity, just at different points in the hierarchy. * - * Per the above update_tg_cfs_util() is trivial and simply copies the running - * sum over (but still wrong, because the group entity and group rq do not have - * their PELT windows aligned). + * Per the above update_tg_cfs_util() and update_tg_cfs_runnable() are trivial + * and simply copies the running/runnable sum over (but still wrong, because + * the group entity and group rq do not have their PELT windows aligned). * - * However, update_tg_cfs_runnable() is more complex. So we have: + * However, update_tg_cfs_load() is more complex. So we have: * * ge->avg.load_avg = ge->load.weight * ge->avg.runnable_avg (2) * @@ -3194,60 +4226,102 @@ void set_task_rq_fair(struct sched_entity *se, * XXX: only do this for the part of runnable > running ? * */ - static inline void update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq) { - long delta = gcfs_rq->avg.util_avg - se->avg.util_avg; + long delta_sum, delta_avg = gcfs_rq->avg.util_avg - se->avg.util_avg; + u32 new_sum, divider; /* Nothing to update */ - if (!delta) + if (!delta_avg) return; /* - * The relation between sum and avg is: - * - * LOAD_AVG_MAX - 1024 + sa->period_contrib - * - * however, the PELT windows are not aligned between grq and gse. + * cfs_rq->avg.period_contrib can be used for both cfs_rq and se. + * See ___update_load_avg() for details. */ + divider = get_pelt_divider(&cfs_rq->avg); + /* Set new sched_entity's utilization */ se->avg.util_avg = gcfs_rq->avg.util_avg; - se->avg.util_sum = se->avg.util_avg * LOAD_AVG_MAX; + new_sum = se->avg.util_avg * divider; + delta_sum = (long)new_sum - (long)se->avg.util_sum; + se->avg.util_sum = new_sum; /* Update parent cfs_rq utilization */ - add_positive(&cfs_rq->avg.util_avg, delta); - cfs_rq->avg.util_sum = cfs_rq->avg.util_avg * LOAD_AVG_MAX; + add_positive(&cfs_rq->avg.util_avg, delta_avg); + add_positive(&cfs_rq->avg.util_sum, delta_sum); + + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum, + cfs_rq->avg.util_avg * PELT_MIN_DIVIDER); } static inline void update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq) { + long delta_sum, delta_avg = gcfs_rq->avg.runnable_avg - se->avg.runnable_avg; + u32 new_sum, divider; + + /* Nothing to update */ + if (!delta_avg) + return; + + /* + * cfs_rq->avg.period_contrib can be used for both cfs_rq and se. + * See ___update_load_avg() for details. + */ + divider = get_pelt_divider(&cfs_rq->avg); + + /* Set new sched_entity's runnable */ + se->avg.runnable_avg = gcfs_rq->avg.runnable_avg; + new_sum = se->avg.runnable_avg * divider; + delta_sum = (long)new_sum - (long)se->avg.runnable_sum; + se->avg.runnable_sum = new_sum; + + /* Update parent cfs_rq runnable */ + add_positive(&cfs_rq->avg.runnable_avg, delta_avg); + add_positive(&cfs_rq->avg.runnable_sum, delta_sum); + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum, + cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER); +} + +static inline void +update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq *gcfs_rq) +{ long delta_avg, running_sum, runnable_sum = gcfs_rq->prop_runnable_sum; - unsigned long runnable_load_avg, load_avg; - u64 runnable_load_sum, load_sum = 0; + unsigned long load_avg; + u64 load_sum = 0; s64 delta_sum; + u32 divider; if (!runnable_sum) return; gcfs_rq->prop_runnable_sum = 0; + /* + * cfs_rq->avg.period_contrib can be used for both cfs_rq and se. + * See ___update_load_avg() for details. + */ + divider = get_pelt_divider(&cfs_rq->avg); + if (runnable_sum >= 0) { /* * Add runnable; clip at LOAD_AVG_MAX. Reflects that until * the CPU is saturated running == runnable. */ runnable_sum += se->avg.load_sum; - runnable_sum = min(runnable_sum, (long)LOAD_AVG_MAX); + runnable_sum = min_t(long, runnable_sum, divider); } else { /* * Estimate the new unweighted runnable_sum of the gcfs_rq by * assuming all tasks are equally runnable. */ if (scale_load_down(gcfs_rq->load.weight)) { - load_sum = div_s64(gcfs_rq->avg.load_sum, + load_sum = div_u64(gcfs_rq->avg.load_sum, scale_load_down(gcfs_rq->load.weight)); } @@ -3257,36 +4331,29 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf /* * runnable_sum can't be lower than running_sum - * As running sum is scale with CPU capacity wehreas the runnable sum - * is not we rescale running_sum 1st + * Rescale running sum to be in the same range as runnable sum + * running_sum is in [0 : LOAD_AVG_MAX << SCHED_CAPACITY_SHIFT] + * runnable_sum is in [0 : LOAD_AVG_MAX] */ - running_sum = se->avg.util_sum / - arch_scale_cpu_capacity(NULL, cpu_of(rq_of(cfs_rq))); + running_sum = se->avg.util_sum >> SCHED_CAPACITY_SHIFT; runnable_sum = max(runnable_sum, running_sum); - load_sum = (s64)se_weight(se) * runnable_sum; - load_avg = div_s64(load_sum, LOAD_AVG_MAX); + load_sum = se_weight(se) * runnable_sum; + load_avg = div_u64(load_sum, divider); - delta_sum = load_sum - (s64)se_weight(se) * se->avg.load_sum; delta_avg = load_avg - se->avg.load_avg; + if (!delta_avg) + return; + + delta_sum = load_sum - (s64)se_weight(se) * se->avg.load_sum; se->avg.load_sum = runnable_sum; se->avg.load_avg = load_avg; add_positive(&cfs_rq->avg.load_avg, delta_avg); add_positive(&cfs_rq->avg.load_sum, delta_sum); - - runnable_load_sum = (s64)se_runnable(se) * runnable_sum; - runnable_load_avg = div_s64(runnable_load_sum, LOAD_AVG_MAX); - delta_sum = runnable_load_sum - se_weight(se) * se->avg.runnable_load_sum; - delta_avg = runnable_load_avg - se->avg.runnable_load_avg; - - se->avg.runnable_load_sum = runnable_sum; - se->avg.runnable_load_avg = runnable_load_avg; - - if (se->on_rq) { - add_positive(&cfs_rq->avg.runnable_load_avg, delta_avg); - add_positive(&cfs_rq->avg.runnable_load_sum, delta_sum); - } + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum, + cfs_rq->avg.load_avg * PELT_MIN_DIVIDER); } static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum) @@ -3315,6 +4382,10 @@ static inline int propagate_entity_load_avg(struct sched_entity *se) update_tg_cfs_util(cfs_rq, se, gcfs_rq); update_tg_cfs_runnable(cfs_rq, se, gcfs_rq); + update_tg_cfs_load(cfs_rq, se, gcfs_rq); + + trace_pelt_cfs_tp(cfs_rq); + trace_pelt_se_tp(se); return 1; } @@ -3349,9 +4420,11 @@ static inline bool skip_blocked_update(struct sched_entity *se) return true; } -#else /* CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED: */ + +static inline void update_tg_load_avg(struct cfs_rq *cfs_rq) {} -static inline void update_tg_load_avg(struct cfs_rq *cfs_rq, int force) {} +static inline void clear_tg_offline_cfs_rqs(struct rq *rq) {} static inline int propagate_entity_load_avg(struct sched_entity *se) { @@ -3360,20 +4433,102 @@ static inline int propagate_entity_load_avg(struct sched_entity *se) static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum) {} -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* !CONFIG_FAIR_GROUP_SCHED */ + +#ifdef CONFIG_NO_HZ_COMMON +static inline void migrate_se_pelt_lag(struct sched_entity *se) +{ + u64 throttled = 0, now, lut; + struct cfs_rq *cfs_rq; + struct rq *rq; + bool is_idle; + + if (load_avg_is_decayed(&se->avg)) + return; + + cfs_rq = cfs_rq_of(se); + rq = rq_of(cfs_rq); + + rcu_read_lock(); + is_idle = is_idle_task(rcu_dereference(rq->curr)); + rcu_read_unlock(); + + /* + * The lag estimation comes with a cost we don't want to pay all the + * time. Hence, limiting to the case where the source CPU is idle and + * we know we are at the greatest risk to have an outdated clock. + */ + if (!is_idle) + return; + + /* + * Estimated "now" is: last_update_time + cfs_idle_lag + rq_idle_lag, where: + * + * last_update_time (the cfs_rq's last_update_time) + * = cfs_rq_clock_pelt()@cfs_rq_idle + * = rq_clock_pelt()@cfs_rq_idle + * - cfs->throttled_clock_pelt_time@cfs_rq_idle + * + * cfs_idle_lag (delta between rq's update and cfs_rq's update) + * = rq_clock_pelt()@rq_idle - rq_clock_pelt()@cfs_rq_idle + * + * rq_idle_lag (delta between now and rq's update) + * = sched_clock_cpu() - rq_clock()@rq_idle + * + * We can then write: + * + * now = rq_clock_pelt()@rq_idle - cfs->throttled_clock_pelt_time + + * sched_clock_cpu() - rq_clock()@rq_idle + * Where: + * rq_clock_pelt()@rq_idle is rq->clock_pelt_idle + * rq_clock()@rq_idle is rq->clock_idle + * cfs->throttled_clock_pelt_time@cfs_rq_idle + * is cfs_rq->throttled_pelt_idle + */ + +#ifdef CONFIG_CFS_BANDWIDTH + throttled = u64_u32_load(cfs_rq->throttled_pelt_idle); + /* The clock has been stopped for throttling */ + if (throttled == U64_MAX) + return; +#endif + now = u64_u32_load(rq->clock_pelt_idle); + /* + * Paired with _update_idle_rq_clock_pelt(). It ensures at the worst case + * is observed the old clock_pelt_idle value and the new clock_idle, + * which lead to an underestimation. The opposite would lead to an + * overestimation. + */ + smp_rmb(); + lut = cfs_rq_last_update_time(cfs_rq); + + now -= throttled; + if (now < lut) + /* + * cfs_rq->avg.last_update_time is more recent than our + * estimation, let's use it. + */ + now = lut; + else + now += sched_clock_cpu(cpu_of(rq)) - u64_u32_load(rq->clock_idle); + + __update_load_avg_blocked_se(now, se); +} +#else /* !CONFIG_NO_HZ_COMMON: */ +static void migrate_se_pelt_lag(struct sched_entity *se) {} +#endif /* !CONFIG_NO_HZ_COMMON */ /** * update_cfs_rq_load_avg - update the cfs_rq's load/util averages - * @now: current time, as per cfs_rq_clock_task() + * @now: current time, as per cfs_rq_clock_pelt() * @cfs_rq: cfs_rq to update * * The cfs_rq avg is the direct sum of all its entities (blocked and runnable) - * avg. The immediate corollary is that all (fair) tasks must be attached, see - * post_init_entity_util_avg(). + * avg. The immediate corollary is that all (fair) tasks must be attached. * * cfs_rq->avg is used for task_h_load() and update_cfs_share() for example. * - * Returns true if the load decayed or we removed load. + * Return: true if the load decayed or we removed load. * * Since both these conditions indicate a changed cfs_rq->avg.load we should * call update_tg_load_avg() when this function returns true. @@ -3381,44 +4536,64 @@ static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum static inline int update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq) { - unsigned long removed_load = 0, removed_util = 0, removed_runnable_sum = 0; + unsigned long removed_load = 0, removed_util = 0, removed_runnable = 0; struct sched_avg *sa = &cfs_rq->avg; int decayed = 0; if (cfs_rq->removed.nr) { unsigned long r; - u32 divider = LOAD_AVG_MAX - 1024 + sa->period_contrib; + u32 divider = get_pelt_divider(&cfs_rq->avg); raw_spin_lock(&cfs_rq->removed.lock); swap(cfs_rq->removed.util_avg, removed_util); swap(cfs_rq->removed.load_avg, removed_load); - swap(cfs_rq->removed.runnable_sum, removed_runnable_sum); + swap(cfs_rq->removed.runnable_avg, removed_runnable); cfs_rq->removed.nr = 0; raw_spin_unlock(&cfs_rq->removed.lock); r = removed_load; sub_positive(&sa->load_avg, r); sub_positive(&sa->load_sum, r * divider); + /* See sa->util_sum below */ + sa->load_sum = max_t(u32, sa->load_sum, sa->load_avg * PELT_MIN_DIVIDER); r = removed_util; sub_positive(&sa->util_avg, r); sub_positive(&sa->util_sum, r * divider); + /* + * Because of rounding, se->util_sum might ends up being +1 more than + * cfs->util_sum. Although this is not a problem by itself, detaching + * a lot of tasks with the rounding problem between 2 updates of + * util_avg (~1ms) can make cfs->util_sum becoming null whereas + * cfs_util_avg is not. + * Check that util_sum is still above its lower bound for the new + * util_avg. Given that period_contrib might have moved since the last + * sync, we are only sure that util_sum must be above or equal to + * util_avg * minimum possible divider + */ + sa->util_sum = max_t(u32, sa->util_sum, sa->util_avg * PELT_MIN_DIVIDER); - add_tg_cfs_propagate(cfs_rq, -(long)removed_runnable_sum); + r = removed_runnable; + sub_positive(&sa->runnable_avg, r); + sub_positive(&sa->runnable_sum, r * divider); + /* See sa->util_sum above */ + sa->runnable_sum = max_t(u32, sa->runnable_sum, + sa->runnable_avg * PELT_MIN_DIVIDER); + + /* + * removed_runnable is the unweighted version of removed_load so we + * can use it to estimate removed_load_sum. + */ + add_tg_cfs_propagate(cfs_rq, + -(long)(removed_runnable * divider) >> SCHED_CAPACITY_SHIFT); decayed = 1; } - decayed |= __update_load_avg_cfs_rq(now, cpu_of(rq_of(cfs_rq)), cfs_rq); - -#ifndef CONFIG_64BIT - smp_wmb(); - cfs_rq->load_last_update_time_copy = sa->last_update_time; -#endif - - if (decayed) - cfs_rq_util_change(cfs_rq, 0); - + decayed |= __update_load_avg_cfs_rq(now, cfs_rq); + u64_u32_store_copy(sa->last_update_time, + cfs_rq->last_update_time_copy, + sa->last_update_time); return decayed; } @@ -3426,14 +4601,17 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq) * attach_entity_load_avg - attach this entity to its cfs_rq load avg * @cfs_rq: cfs_rq to attach to * @se: sched_entity to attach - * @flags: migration hints * * Must call update_cfs_rq_load_avg() before this, since we rely on * cfs_rq->avg.last_update_time being current. */ -static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) +static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { - u32 divider = LOAD_AVG_MAX - 1024 + cfs_rq->avg.period_contrib; + /* + * cfs_rq->avg.period_contrib can be used for both cfs_rq and se. + * See ___update_load_avg() for details. + */ + u32 divider = get_pelt_divider(&cfs_rq->avg); /* * When we attach the @se to the @cfs_rq, we must align the decay @@ -3453,21 +4631,25 @@ static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s */ se->avg.util_sum = se->avg.util_avg * divider; - se->avg.load_sum = divider; - if (se_weight(se)) { - se->avg.load_sum = - div_u64(se->avg.load_avg * se->avg.load_sum, se_weight(se)); - } + se->avg.runnable_sum = se->avg.runnable_avg * divider; - se->avg.runnable_load_sum = se->avg.load_sum; + se->avg.load_sum = se->avg.load_avg * divider; + if (se_weight(se) < se->avg.load_sum) + se->avg.load_sum = div_u64(se->avg.load_sum, se_weight(se)); + else + se->avg.load_sum = 1; enqueue_load_avg(cfs_rq, se); cfs_rq->avg.util_avg += se->avg.util_avg; cfs_rq->avg.util_sum += se->avg.util_sum; + cfs_rq->avg.runnable_avg += se->avg.runnable_avg; + cfs_rq->avg.runnable_sum += se->avg.runnable_sum; add_tg_cfs_propagate(cfs_rq, se->avg.load_sum); - cfs_rq_util_change(cfs_rq, flags); + cfs_rq_util_change(cfs_rq, 0); + + trace_pelt_cfs_tp(cfs_rq); } /** @@ -3483,10 +4665,21 @@ static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s dequeue_load_avg(cfs_rq, se); sub_positive(&cfs_rq->avg.util_avg, se->avg.util_avg); sub_positive(&cfs_rq->avg.util_sum, se->avg.util_sum); + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum, + cfs_rq->avg.util_avg * PELT_MIN_DIVIDER); + + sub_positive(&cfs_rq->avg.runnable_avg, se->avg.runnable_avg); + sub_positive(&cfs_rq->avg.runnable_sum, se->avg.runnable_sum); + /* See update_cfs_rq_load_avg() */ + cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum, + cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER); add_tg_cfs_propagate(cfs_rq, -se->avg.load_sum); cfs_rq_util_change(cfs_rq, 0); + + trace_pelt_cfs_tp(cfs_rq); } /* @@ -3495,21 +4688,20 @@ static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s #define UPDATE_TG 0x1 #define SKIP_AGE_LOAD 0x2 #define DO_ATTACH 0x4 +#define DO_DETACH 0x8 /* Update task and its cfs_rq load average */ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - u64 now = cfs_rq_clock_task(cfs_rq); - struct rq *rq = rq_of(cfs_rq); - int cpu = cpu_of(rq); + u64 now = cfs_rq_clock_pelt(cfs_rq); int decayed; /* * Track task load average for carrying it to new CPU after migrated, and - * track group sched_entity load average for task_h_load calc in migration + * track group sched_entity load average for task_h_load calculation in migration */ if (se->avg.last_update_time && !(flags & SKIP_AGE_LOAD)) - __update_load_avg_se(now, cpu, cfs_rq, se); + __update_load_avg_se(now, cfs_rq, se); decayed = update_cfs_rq_load_avg(now, cfs_rq); decayed |= propagate_entity_load_avg(se); @@ -3523,64 +4715,50 @@ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s * * IOW we're enqueueing a task on a new CPU. */ - attach_entity_load_avg(cfs_rq, se, SCHED_CPUFREQ_MIGRATION); - update_tg_load_avg(cfs_rq, 0); - - } else if (decayed && (flags & UPDATE_TG)) - update_tg_load_avg(cfs_rq, 0); -} - -#ifndef CONFIG_64BIT -static inline u64 cfs_rq_last_update_time(struct cfs_rq *cfs_rq) -{ - u64 last_update_time_copy; - u64 last_update_time; + attach_entity_load_avg(cfs_rq, se); + update_tg_load_avg(cfs_rq); - do { - last_update_time_copy = cfs_rq->load_last_update_time_copy; - smp_rmb(); - last_update_time = cfs_rq->avg.last_update_time; - } while (last_update_time != last_update_time_copy); + } else if (flags & DO_DETACH) { + /* + * DO_DETACH means we're here from dequeue_entity() + * and we are migrating task out of the CPU. + */ + detach_entity_load_avg(cfs_rq, se); + update_tg_load_avg(cfs_rq); + } else if (decayed) { + cfs_rq_util_change(cfs_rq, 0); - return last_update_time; -} -#else -static inline u64 cfs_rq_last_update_time(struct cfs_rq *cfs_rq) -{ - return cfs_rq->avg.last_update_time; + if (flags & UPDATE_TG) + update_tg_load_avg(cfs_rq); + } } -#endif /* * Synchronize entity load avg of dequeued entity without locking * the previous rq. */ -void sync_entity_load_avg(struct sched_entity *se) +static void sync_entity_load_avg(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); u64 last_update_time; last_update_time = cfs_rq_last_update_time(cfs_rq); - __update_load_avg_blocked_se(last_update_time, cpu_of(rq_of(cfs_rq)), se); + __update_load_avg_blocked_se(last_update_time, se); } /* * Task first catches up with cfs_rq, and then subtract * itself from the cfs_rq (task must be off the queue now). */ -void remove_entity_load_avg(struct sched_entity *se) +static void remove_entity_load_avg(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); unsigned long flags; /* * tasks cannot exit without having gone through wake_up_new_task() -> - * post_init_entity_util_avg() which will have added things to the - * cfs_rq, so we can remove unconditionally. - * - * Similarly for groups, they will have passed through - * post_init_entity_util_avg() before unregister_sched_fair_group() - * calls this. + * enqueue_task_fair() which will have added things to the cfs_rq, + * so we can remove unconditionally. */ sync_entity_load_avg(se); @@ -3589,13 +4767,13 @@ void remove_entity_load_avg(struct sched_entity *se) ++cfs_rq->removed.nr; cfs_rq->removed.util_avg += se->avg.util_avg; cfs_rq->removed.load_avg += se->avg.load_avg; - cfs_rq->removed.runnable_sum += se->avg.load_sum; /* == runnable_sum */ + cfs_rq->removed.runnable_avg += se->avg.runnable_avg; raw_spin_unlock_irqrestore(&cfs_rq->removed.lock, flags); } -static inline unsigned long cfs_rq_runnable_load_avg(struct cfs_rq *cfs_rq) +static inline unsigned long cfs_rq_runnable_avg(struct cfs_rq *cfs_rq) { - return cfs_rq->avg.runnable_load_avg; + return cfs_rq->avg.runnable_avg; } static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) @@ -3603,18 +4781,21 @@ static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) return cfs_rq->avg.load_avg; } -static int idle_balance(struct rq *this_rq, struct rq_flags *rf); +static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf); static inline unsigned long task_util(struct task_struct *p) { return READ_ONCE(p->se.avg.util_avg); } -static inline unsigned long _task_util_est(struct task_struct *p) +static inline unsigned long task_runnable(struct task_struct *p) { - struct util_est ue = READ_ONCE(p->se.avg.util_est); + return READ_ONCE(p->se.avg.runnable_avg); +} - return (max(ue.ewma, ue.enqueued) | UTIL_AVG_UNCHANGED); +static inline unsigned long _task_util_est(struct task_struct *p) +{ + return READ_ONCE(p->se.avg.util_est) & ~UTIL_AVG_UNCHANGED; } static inline unsigned long task_util_est(struct task_struct *p) @@ -3631,37 +4812,39 @@ static inline void util_est_enqueue(struct cfs_rq *cfs_rq, return; /* Update root cfs_rq's estimated utilization */ - enqueued = cfs_rq->avg.util_est.enqueued; + enqueued = cfs_rq->avg.util_est; enqueued += _task_util_est(p); - WRITE_ONCE(cfs_rq->avg.util_est.enqueued, enqueued); -} + WRITE_ONCE(cfs_rq->avg.util_est, enqueued); -/* - * Check if a (signed) value is within a specified (unsigned) margin, - * based on the observation that: - * - * abs(x) < y := (unsigned)(x + y - 1) < (2 * y - 1) - * - * NOTE: this only works when value + maring < INT_MAX. - */ -static inline bool within_margin(int value, int margin) -{ - return ((unsigned int)(value + margin - 1) < (2 * margin - 1)); + trace_sched_util_est_cfs_tp(cfs_rq); } -static void -util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) +static inline void util_est_dequeue(struct cfs_rq *cfs_rq, + struct task_struct *p) { - long last_ewma_diff; - struct util_est ue; + unsigned int enqueued; if (!sched_feat(UTIL_EST)) return; /* Update root cfs_rq's estimated utilization */ - ue.enqueued = cfs_rq->avg.util_est.enqueued; - ue.enqueued -= min_t(unsigned int, ue.enqueued, _task_util_est(p)); - WRITE_ONCE(cfs_rq->avg.util_est.enqueued, ue.enqueued); + enqueued = cfs_rq->avg.util_est; + enqueued -= min_t(unsigned int, enqueued, _task_util_est(p)); + WRITE_ONCE(cfs_rq->avg.util_est, enqueued); + + trace_sched_util_est_cfs_tp(cfs_rq); +} + +#define UTIL_EST_MARGIN (SCHED_CAPACITY_SCALE / 100) + +static inline void util_est_update(struct cfs_rq *cfs_rq, + struct task_struct *p, + bool task_sleep) +{ + unsigned int ewma, dequeued, last_ewma_diff; + + if (!sched_feat(UTIL_EST)) + return; /* * Skip update of task's estimated utilization when the task has not @@ -3670,261 +4853,429 @@ util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) if (!task_sleep) return; + /* Get current estimate of utilization */ + ewma = READ_ONCE(p->se.avg.util_est); + /* * If the PELT values haven't changed since enqueue time, * skip the util_est update. */ - ue = p->se.avg.util_est; - if (ue.enqueued & UTIL_AVG_UNCHANGED) + if (ewma & UTIL_AVG_UNCHANGED) return; + /* Get utilization at dequeue */ + dequeued = task_util(p); + /* - * Skip update of task's estimated utilization when its EWMA is + * Reset EWMA on utilization increases, the moving average is used only + * to smooth utilization decreases. + */ + if (ewma <= dequeued) { + ewma = dequeued; + goto done; + } + + /* + * Skip update of task's estimated utilization when its members are * already ~1% close to its last activation value. */ - ue.enqueued = (task_util(p) | UTIL_AVG_UNCHANGED); - last_ewma_diff = ue.enqueued - ue.ewma; - if (within_margin(last_ewma_diff, (SCHED_CAPACITY_SCALE / 100))) - return; + last_ewma_diff = ewma - dequeued; + if (last_ewma_diff < UTIL_EST_MARGIN) + goto done; + + /* + * To avoid underestimate of task utilization, skip updates of EWMA if + * we cannot grant that thread got all CPU time it wanted. + */ + if ((dequeued + UTIL_EST_MARGIN) < task_runnable(p)) + goto done; + /* * Update Task's estimated utilization * * When *p completes an activation we can consolidate another sample - * of the task size. This is done by storing the current PELT value - * as ue.enqueued and by using this value to update the Exponential - * Weighted Moving Average (EWMA): + * of the task size. This is done by using this value to update the + * Exponential Weighted Moving Average (EWMA): * * ewma(t) = w * task_util(p) + (1-w) * ewma(t-1) * = w * task_util(p) + ewma(t-1) - w * ewma(t-1) * = w * (task_util(p) - ewma(t-1)) + ewma(t-1) - * = w * ( last_ewma_diff ) + ewma(t-1) - * = w * (last_ewma_diff + ewma(t-1) / w) + * = w * ( -last_ewma_diff ) + ewma(t-1) + * = w * (-last_ewma_diff + ewma(t-1) / w) * * Where 'w' is the weight of new samples, which is configured to be * 0.25, thus making w=1/4 ( >>= UTIL_EST_WEIGHT_SHIFT) */ - ue.ewma <<= UTIL_EST_WEIGHT_SHIFT; - ue.ewma += last_ewma_diff; - ue.ewma >>= UTIL_EST_WEIGHT_SHIFT; - WRITE_ONCE(p->se.avg.util_est, ue); + ewma <<= UTIL_EST_WEIGHT_SHIFT; + ewma -= last_ewma_diff; + ewma >>= UTIL_EST_WEIGHT_SHIFT; +done: + ewma |= UTIL_AVG_UNCHANGED; + WRITE_ONCE(p->se.avg.util_est, ewma); + + trace_sched_util_est_se_tp(&p->se); } -static inline int task_fits_capacity(struct task_struct *p, long capacity) +static inline unsigned long get_actual_cpu_capacity(int cpu) { - return capacity * 1024 > task_util_est(p) * capacity_margin; + unsigned long capacity = arch_scale_cpu_capacity(cpu); + + capacity -= max(hw_load_avg(cpu_rq(cpu)), cpufreq_get_pressure(cpu)); + + return capacity; } -static inline void update_misfit_status(struct task_struct *p, struct rq *rq) +static inline int util_fits_cpu(unsigned long util, + unsigned long uclamp_min, + unsigned long uclamp_max, + int cpu) { - if (!static_branch_unlikely(&sched_asym_cpucapacity)) - return; + unsigned long capacity = capacity_of(cpu); + unsigned long capacity_orig; + bool fits, uclamp_max_fits; - if (!p) { - rq->misfit_task_load = 0; - return; - } + /* + * Check if the real util fits without any uclamp boost/cap applied. + */ + fits = fits_capacity(util, capacity); - if (task_fits_capacity(p, capacity_of(cpu_of(rq)))) { - rq->misfit_task_load = 0; - return; - } + if (!uclamp_is_used()) + return fits; - rq->misfit_task_load = task_h_load(p); -} + /* + * We must use arch_scale_cpu_capacity() for comparing against uclamp_min and + * uclamp_max. We only care about capacity pressure (by using + * capacity_of()) for comparing against the real util. + * + * If a task is boosted to 1024 for example, we don't want a tiny + * pressure to skew the check whether it fits a CPU or not. + * + * Similarly if a task is capped to arch_scale_cpu_capacity(little_cpu), it + * should fit a little cpu even if there's some pressure. + * + * Only exception is for HW or cpufreq pressure since it has a direct impact + * on available OPP of the system. + * + * We honour it for uclamp_min only as a drop in performance level + * could result in not getting the requested minimum performance level. + * + * For uclamp_max, we can tolerate a drop in performance level as the + * goal is to cap the task. So it's okay if it's getting less. + */ + capacity_orig = arch_scale_cpu_capacity(cpu); + + /* + * We want to force a task to fit a cpu as implied by uclamp_max. + * But we do have some corner cases to cater for.. + * + * + * C=z + * | ___ + * | C=y | | + * |_ _ _ _ _ _ _ _ _ ___ _ _ _ | _ | _ _ _ _ _ uclamp_max + * | C=x | | | | + * | ___ | | | | + * | | | | | | | (util somewhere in this region) + * | | | | | | | + * | | | | | | | + * +---------------------------------------- + * CPU0 CPU1 CPU2 + * + * In the above example if a task is capped to a specific performance + * point, y, then when: + * + * * util = 80% of x then it does not fit on CPU0 and should migrate + * to CPU1 + * * util = 80% of y then it is forced to fit on CPU1 to honour + * uclamp_max request. + * + * which is what we're enforcing here. A task always fits if + * uclamp_max <= capacity_orig. But when uclamp_max > capacity_orig, + * the normal upmigration rules should withhold still. + * + * Only exception is when we are on max capacity, then we need to be + * careful not to block overutilized state. This is so because: + * + * 1. There's no concept of capping at max_capacity! We can't go + * beyond this performance level anyway. + * 2. The system is being saturated when we're operating near + * max capacity, it doesn't make sense to block overutilized. + */ + uclamp_max_fits = (capacity_orig == SCHED_CAPACITY_SCALE) && (uclamp_max == SCHED_CAPACITY_SCALE); + uclamp_max_fits = !uclamp_max_fits && (uclamp_max <= capacity_orig); + fits = fits || uclamp_max_fits; -#else /* CONFIG_SMP */ + /* + * + * C=z + * | ___ (region a, capped, util >= uclamp_max) + * | C=y | | + * |_ _ _ _ _ _ _ _ _ ___ _ _ _ | _ | _ _ _ _ _ uclamp_max + * | C=x | | | | + * | ___ | | | | (region b, uclamp_min <= util <= uclamp_max) + * |_ _ _|_ _|_ _ _ _| _ | _ _ _| _ | _ _ _ _ _ uclamp_min + * | | | | | | | + * | | | | | | | (region c, boosted, util < uclamp_min) + * +---------------------------------------- + * CPU0 CPU1 CPU2 + * + * a) If util > uclamp_max, then we're capped, we don't care about + * actual fitness value here. We only care if uclamp_max fits + * capacity without taking margin/pressure into account. + * See comment above. + * + * b) If uclamp_min <= util <= uclamp_max, then the normal + * fits_capacity() rules apply. Except we need to ensure that we + * enforce we remain within uclamp_max, see comment above. + * + * c) If util < uclamp_min, then we are boosted. Same as (b) but we + * need to take into account the boosted value fits the CPU without + * taking margin/pressure into account. + * + * Cases (a) and (b) are handled in the 'fits' variable already. We + * just need to consider an extra check for case (c) after ensuring we + * handle the case uclamp_min > uclamp_max. + */ + uclamp_min = min(uclamp_min, uclamp_max); + if (fits && (util < uclamp_min) && + (uclamp_min > get_actual_cpu_capacity(cpu))) + return -1; -#define UPDATE_TG 0x0 -#define SKIP_AGE_LOAD 0x0 -#define DO_ATTACH 0x0 + return fits; +} -static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int not_used1) +static inline int task_fits_cpu(struct task_struct *p, int cpu) { - cfs_rq_util_change(cfs_rq, 0); + unsigned long uclamp_min = uclamp_eff_value(p, UCLAMP_MIN); + unsigned long uclamp_max = uclamp_eff_value(p, UCLAMP_MAX); + unsigned long util = task_util_est(p); + /* + * Return true only if the cpu fully fits the task requirements, which + * include the utilization but also the performance hints. + */ + return (util_fits_cpu(util, uclamp_min, uclamp_max, cpu) > 0); } -static inline void remove_entity_load_avg(struct sched_entity *se) {} - -static inline void -attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) {} -static inline void -detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} - -static inline int idle_balance(struct rq *rq, struct rq_flags *rf) +static inline void update_misfit_status(struct task_struct *p, struct rq *rq) { - return 0; -} + int cpu = cpu_of(rq); -static inline void -util_est_enqueue(struct cfs_rq *cfs_rq, struct task_struct *p) {} + if (!sched_asym_cpucap_active()) + return; -static inline void -util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p, - bool task_sleep) {} -static inline void update_misfit_status(struct task_struct *p, struct rq *rq) {} + /* + * Affinity allows us to go somewhere higher? Or are we on biggest + * available CPU already? Or do we fit into this CPU ? + */ + if (!p || (p->nr_cpus_allowed == 1) || + (arch_scale_cpu_capacity(cpu) == p->max_allowed_capacity) || + task_fits_cpu(p, cpu)) { -#endif /* CONFIG_SMP */ + rq->misfit_task_load = 0; + return; + } -static void check_spread(struct cfs_rq *cfs_rq, struct sched_entity *se) -{ -#ifdef CONFIG_SCHED_DEBUG - s64 d = se->vruntime - cfs_rq->min_vruntime; + /* + * Make sure that misfit_task_load will not be null even if + * task_h_load() returns 0. + */ + rq->misfit_task_load = max_t(unsigned long, task_h_load(p), 1); +} - if (d < 0) - d = -d; +void __setparam_fair(struct task_struct *p, const struct sched_attr *attr) +{ + struct sched_entity *se = &p->se; - if (d > 3*sysctl_sched_latency) - schedstat_inc(cfs_rq->nr_spread_over); -#endif + 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 initial) +place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - u64 vruntime = cfs_rq->min_vruntime; + u64 vslice, vruntime = avg_vruntime(cfs_rq); + s64 lag = 0; + + if (!se->custom_slice) + se->slice = sysctl_sched_base_slice; + vslice = calc_delta_fair(se->slice, se); /* - * The 'current' period is already promised to the current tasks, - * however the extra weight of the new task will slow them down a - * little, place the new task so that it fits in the slot that - * stays open at the end. + * Due to how V is constructed as the weighted average of entities, + * adding tasks with positive lag, or removing tasks with negative lag + * will move 'time' backwards, this can screw around with the lag of + * other tasks. + * + * EEVDF: placement strategy #1 / #2 */ - if (initial && sched_feat(START_DEBIT)) - vruntime += sched_vslice(cfs_rq, se); + if (sched_feat(PLACE_LAG) && cfs_rq->nr_queued && se->vlag) { + struct sched_entity *curr = cfs_rq->curr; + unsigned long load; - /* sleeps up to a single latency don't count. */ - if (!initial) { - unsigned long thresh = sysctl_sched_latency; + lag = se->vlag; /* - * Halve their sleep time's effect, to allow - * for a gentler effect of sleepers: + * If we want to place a task and preserve lag, we have to + * consider the effect of the new entity on the weighted + * average and compensate for this, otherwise lag can quickly + * evaporate. + * + * Lag is defined as: + * + * lag_i = S - s_i = w_i * (V - v_i) + * + * To avoid the 'w_i' term all over the place, we only track + * the virtual lag: + * + * vl_i = V - v_i <=> v_i = V - vl_i + * + * And we take V to be the weighted average of all v: + * + * V = (\Sum w_j*v_j) / W + * + * Where W is: \Sum w_j + * + * Then, the weighted average after adding an entity with lag + * vl_i is given by: + * + * V' = (\Sum w_j*v_j + w_i*v_i) / (W + w_i) + * = (W*V + w_i*(V - vl_i)) / (W + w_i) + * = (W*V + w_i*V - w_i*vl_i) / (W + w_i) + * = (V*(W + w_i) - w_i*vl_i) / (W + w_i) + * = V - w_i*vl_i / (W + w_i) + * + * And the actual lag after adding an entity with vl_i is: + * + * vl'_i = V' - v_i + * = V - w_i*vl_i / (W + w_i) - (V - vl_i) + * = vl_i - w_i*vl_i / (W + w_i) + * + * Which is strictly less than vl_i. So in order to preserve lag + * we should inflate the lag before placement such that the + * effective lag after placement comes out right. + * + * As such, invert the above relation for vl'_i to get the vl_i + * we need to use such that the lag after placement is the lag + * we computed before dequeue. + * + * vl'_i = vl_i - w_i*vl_i / (W + w_i) + * = ((W + w_i)*vl_i - w_i*vl_i) / (W + w_i) + * + * (W + w_i)*vl'_i = (W + w_i)*vl_i - w_i*vl_i + * = W*vl_i + * + * vl_i = (W + w_i)*vl'_i / W */ - if (sched_feat(GENTLE_FAIR_SLEEPERS)) - thresh >>= 1; + load = cfs_rq->avg_load; + if (curr && curr->on_rq) + load += scale_load_down(curr->load.weight); - vruntime -= thresh; + lag *= load + scale_load_down(se->load.weight); + if (WARN_ON_ONCE(!load)) + load = 1; + lag = div_s64(lag, load); } - /* ensure we never gain time by being placed backwards. */ - se->vruntime = max_vruntime(se->vruntime, vruntime); -} + se->vruntime = vruntime - lag; -static void check_enqueue_throttle(struct cfs_rq *cfs_rq); - -static inline void check_schedstat_required(void) -{ -#ifdef CONFIG_SCHEDSTATS - if (schedstat_enabled()) + if (se->rel_deadline) { + se->deadline += se->vruntime; + se->rel_deadline = 0; return; - - /* Force schedstat enabled if a dependent tracepoint is active */ - if (trace_sched_stat_wait_enabled() || - trace_sched_stat_sleep_enabled() || - trace_sched_stat_iowait_enabled() || - trace_sched_stat_blocked_enabled() || - trace_sched_stat_runtime_enabled()) { - printk_deferred_once("Scheduler tracepoints stat_sleep, stat_iowait, " - "stat_blocked and stat_runtime require the " - "kernel parameter schedstats=enable or " - "kernel.sched_schedstats=1\n"); } -#endif + + /* + * When joining the competition; the existing tasks will be, + * on average, halfway through their slice, as such start tasks + * off with half a slice to ease into the competition. + */ + if (sched_feat(PLACE_DEADLINE_INITIAL) && (flags & ENQUEUE_INITIAL)) + vslice /= 2; + + /* + * EEVDF: vd_i = ve_i + r_i/w_i + */ + se->deadline = se->vruntime + vslice; } +static void check_enqueue_throttle(struct cfs_rq *cfs_rq); +static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq); -/* - * MIGRATION - * - * dequeue - * update_curr() - * update_min_vruntime() - * vruntime -= min_vruntime - * - * enqueue - * update_curr() - * update_min_vruntime() - * vruntime += min_vruntime - * - * this way the vruntime transition between RQs is done when both - * min_vruntime are up-to-date. - * - * WAKEUP (remote) - * - * ->migrate_task_rq_fair() (p->state == TASK_WAKING) - * vruntime -= min_vruntime - * - * enqueue - * update_curr() - * update_min_vruntime() - * vruntime += min_vruntime - * - * this way we don't have the most up-to-date min_vruntime on the originating - * CPU and an up-to-date min_vruntime on the destination CPU. - */ +static void +requeue_delayed_entity(struct sched_entity *se); static void enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - bool renorm = !(flags & ENQUEUE_WAKEUP) || (flags & ENQUEUE_MIGRATED); bool curr = cfs_rq->curr == se; /* * If we're the current task, we must renormalise before calling * update_curr(). */ - if (renorm && curr) - se->vruntime += cfs_rq->min_vruntime; + if (curr) + place_entity(cfs_rq, se, flags); update_curr(cfs_rq); /* - * Otherwise, renormalise after, such that we're placed at the current - * moment in time, instead of some random moment in the past. Being - * placed in the past could significantly boost this task to the - * fairness detriment of existing tasks. - */ - if (renorm && !curr) - se->vruntime += cfs_rq->min_vruntime; - - /* * When enqueuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. - * - Add its load to cfs_rq->runnable_avg + * - For group_entity, update its runnable_weight to reflect the new + * 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 */ update_load_avg(cfs_rq, se, UPDATE_TG | DO_ATTACH); + se_update_runnable(se); + /* + * XXX update_load_avg() above will have attached us to the pelt sum; + * but update_cfs_group() here will re-adjust the weight and have to + * undo/redo all that. Seems wasteful. + */ update_cfs_group(se); - enqueue_runnable_load_avg(cfs_rq, se); + + /* + * XXX now that the entity has been re-weighted, and it's lag adjusted, + * we can place the entity. + */ + if (!curr) + place_entity(cfs_rq, se, flags); + account_entity_enqueue(cfs_rq, se); - if (flags & ENQUEUE_WAKEUP) - place_entity(cfs_rq, se, 0); + /* Entity has migrated, no longer consider this task hot */ + if (flags & ENQUEUE_MIGRATED) + se->exec_start = 0; check_schedstat_required(); - update_stats_enqueue(cfs_rq, se, flags); - check_spread(cfs_rq, se); + update_stats_enqueue_fair(cfs_rq, se, flags); if (!curr) __enqueue_entity(cfs_rq, se); se->on_rq = 1; - if (cfs_rq->nr_running == 1) { - list_add_leaf_cfs_rq(cfs_rq); + if (cfs_rq->nr_queued == 1) { check_enqueue_throttle(cfs_rq); - } -} - -static void __clear_buddies_last(struct sched_entity *se) -{ - for_each_sched_entity(se) { - struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq->last != se) - break; + list_add_leaf_cfs_rq(cfs_rq); +#ifdef CONFIG_CFS_BANDWIDTH + if (cfs_rq->pelt_clock_throttled) { + struct rq *rq = rq_of(cfs_rq); - cfs_rq->last = NULL; + cfs_rq->throttled_clock_pelt_time += rq_clock_pelt(rq) - + cfs_rq->throttled_clock_pelt; + cfs_rq->pelt_clock_throttled = 0; + } +#endif } } @@ -3939,126 +5290,147 @@ static void __clear_buddies_next(struct sched_entity *se) } } -static void __clear_buddies_skip(struct sched_entity *se) +static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + if (cfs_rq->next == se) + __clear_buddies_next(se); +} + +static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); + +static void set_delayed(struct sched_entity *se) { + se->sched_delayed = 1; + + /* + * Delayed se of cfs_rq have no tasks queued on them. + * Do not adjust h_nr_runnable since dequeue_entities() + * will account it for blocked tasks. + */ + if (!entity_is_task(se)) + return; + for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq->skip != se) - break; - cfs_rq->skip = NULL; + cfs_rq->h_nr_runnable--; } } -static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se) +static void clear_delayed(struct sched_entity *se) { - if (cfs_rq->last == se) - __clear_buddies_last(se); + se->sched_delayed = 0; - if (cfs_rq->next == se) - __clear_buddies_next(se); + /* + * Delayed se of cfs_rq have no tasks queued on them. + * Do not adjust h_nr_runnable since a dequeue has + * already accounted for it or an enqueue of a task + * below it will account for it in enqueue_task_fair(). + */ + if (!entity_is_task(se)) + return; + + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq->skip == se) - __clear_buddies_skip(se); + cfs_rq->h_nr_runnable++; + } } -static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq); +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; +} -static void +static bool dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { - /* - * Update run-time statistics of the 'current'. - */ + bool sleep = flags & DEQUEUE_SLEEP; + int action = UPDATE_TG; + update_curr(cfs_rq); + clear_buddies(cfs_rq, se); + + if (flags & DEQUEUE_DELAYED) { + WARN_ON_ONCE(!se->sched_delayed); + } else { + bool delay = sleep; + /* + * DELAY_DEQUEUE relies on spurious wakeups, special task + * states must not suffer spurious wakeups, excempt them. + */ + if (flags & (DEQUEUE_SPECIAL | DEQUEUE_THROTTLE)) + delay = false; + + WARN_ON_ONCE(delay && se->sched_delayed); + + if (sched_feat(DELAY_DEQUEUE) && delay && + !entity_eligible(cfs_rq, se)) { + update_load_avg(cfs_rq, se, 0); + set_delayed(se); + return false; + } + } + + if (entity_is_task(se) && task_on_rq_migrating(task_of(se))) + action |= DO_DETACH; /* * When dequeuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. - * - Subtract its load from the cfs_rq->runnable_avg. + * - For group_entity, update its runnable_weight to reflect the new + * 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. */ - update_load_avg(cfs_rq, se, UPDATE_TG); - dequeue_runnable_load_avg(cfs_rq, se); + update_load_avg(cfs_rq, se, action); + se_update_runnable(se); - update_stats_dequeue(cfs_rq, se, 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; + se->rel_deadline = 1; + } if (se != cfs_rq->curr) __dequeue_entity(cfs_rq, se); se->on_rq = 0; account_entity_dequeue(cfs_rq, se); - /* - * Normalize after update_curr(); which will also have moved - * min_vruntime if @se is the one holding it back. But before doing - * update_min_vruntime() again, which will discount @se's position and - * can move min_vruntime forward still more. - */ - if (!(flags & DEQUEUE_SLEEP)) - se->vruntime -= cfs_rq->min_vruntime; - /* return excess runtime on last dequeue */ return_cfs_rq_runtime(cfs_rq); update_cfs_group(se); - /* - * Now advance min_vruntime if @se was the entity holding it back, - * except when: DEQUEUE_SAVE && !DEQUEUE_MOVE, in this case we'll be - * put back on, and if we advance min_vruntime, we'll be placed back - * further than we started -- ie. we'll be penalized. - */ - if ((flags & (DEQUEUE_SAVE | DEQUEUE_MOVE)) != DEQUEUE_SAVE) - update_min_vruntime(cfs_rq); -} + if (flags & DEQUEUE_DELAYED) + finish_delayed_dequeue_entity(se); -/* - * Preempt the current task with a newly woken task if needed: - */ -static void -check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr) -{ - unsigned long ideal_runtime, delta_exec; - struct sched_entity *se; - s64 delta; + if (cfs_rq->nr_queued == 0) { + update_idle_cfs_rq_clock_pelt(cfs_rq); +#ifdef CONFIG_CFS_BANDWIDTH + if (throttled_hierarchy(cfs_rq)) { + struct rq *rq = rq_of(cfs_rq); - ideal_runtime = sched_slice(cfs_rq, curr); - delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime; - if (delta_exec > ideal_runtime) { - resched_curr(rq_of(cfs_rq)); - /* - * The current task ran long enough, ensure it doesn't get - * re-elected due to buddy favours. - */ - clear_buddies(cfs_rq, curr); - return; + list_del_leaf_cfs_rq(cfs_rq); + cfs_rq->throttled_clock_pelt = rq_clock_pelt(rq); + cfs_rq->pelt_clock_throttled = 1; + } +#endif } - /* - * Ensure that a task that missed wakeup preemption by a - * narrow margin doesn't have to wait for a full slice. - * This also mitigates buddy induced latencies under load. - */ - if (delta_exec < sysctl_sched_min_granularity) - return; - - se = __pick_first_entity(cfs_rq); - delta = curr->vruntime - se->vruntime; - - if (delta < 0) - return; - - if (delta > ideal_runtime) - resched_curr(rq_of(cfs_rq)); + return true; } static void set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { + clear_buddies(cfs_rq, se); + /* 'current' is not kept within the tree. */ if (se->on_rq) { /* @@ -4066,30 +5438,36 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) * a CPU. So account for the time it spent waiting on the * runqueue. */ - update_stats_wait_end(cfs_rq, se); + update_stats_wait_end_fair(cfs_rq, se); __dequeue_entity(cfs_rq, se); update_load_avg(cfs_rq, se, UPDATE_TG); + + set_protect_slice(cfs_rq, se); } update_stats_curr_start(cfs_rq, se); + WARN_ON_ONCE(cfs_rq->curr); cfs_rq->curr = se; /* * Track our maximum slice length, if the CPU's load is at - * least twice that of our own weight (i.e. dont track it + * least twice that of our own weight (i.e. don't track it * when there are only lesser-weight tasks around): */ - if (schedstat_enabled() && rq_of(cfs_rq)->load.weight >= 2*se->load.weight) { - schedstat_set(se->statistics.slice_max, - max((u64)schedstat_val(se->statistics.slice_max), - se->sum_exec_runtime - se->prev_sum_exec_runtime)); + if (schedstat_enabled() && + rq_of(cfs_rq)->cfs.load.weight >= 2*se->load.weight) { + struct sched_statistics *stats; + + stats = __schedstats_from_se(se); + __schedstat_set(stats->slice_max, + max((u64)stats->slice_max, + se->sum_exec_runtime - se->prev_sum_exec_runtime)); } se->prev_sum_exec_runtime = se->sum_exec_runtime; } -static int -wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); +static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags); /* * Pick the next process, keeping these things in mind, in this order: @@ -4099,53 +5477,18 @@ wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se); * 4) do not run the "skip" process, if something else is available */ static struct sched_entity * -pick_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *curr) +pick_next_entity(struct rq *rq, struct cfs_rq *cfs_rq) { - struct sched_entity *left = __pick_first_entity(cfs_rq); struct sched_entity *se; - /* - * If curr is set we have to see if its left of the leftmost entity - * still in the tree, provided there was anything in the tree at all. - */ - if (!left || (curr && entity_before(curr, left))) - left = curr; - - se = left; /* ideally we run the leftmost entity */ - - /* - * Avoid running the skip buddy, if running something else can - * be done without getting too unfair. - */ - if (cfs_rq->skip == se) { - struct sched_entity *second; - - if (se == curr) { - second = __pick_first_entity(cfs_rq); - } else { - second = __pick_next_entity(se); - if (!second || (curr && entity_before(curr, second))) - second = curr; - } - - if (second && wakeup_preempt_entity(second, left) < 1) - se = second; + se = pick_eevdf(cfs_rq); + if (se->sched_delayed) { + dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); + /* + * Must not reference @se again, see __block_task(). + */ + return NULL; } - - /* - * Prefer last buddy, try to return the CPU to a preempted task. - */ - if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1) - se = cfs_rq->last; - - /* - * Someone really wants this to run. If it's not unfair, run it. - */ - if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) - se = cfs_rq->next; - - clear_buddies(cfs_rq, se); - return se; } @@ -4163,15 +5506,14 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *prev) /* throttle cfs_rqs exceeding runtime */ check_cfs_rq_runtime(cfs_rq); - check_spread(cfs_rq, prev); - if (prev->on_rq) { - update_stats_wait_start(cfs_rq, prev); + update_stats_wait_start_fair(cfs_rq, prev); /* Put 'current' back into the tree. */ __enqueue_entity(cfs_rq, prev); /* in !on_rq case, update occurred at dequeue */ update_load_avg(cfs_rq, prev, 0); } + WARN_ON_ONCE(cfs_rq->curr != prev); cfs_rq->curr = NULL; } @@ -4195,19 +5537,10 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * validating it and just reschedule. */ if (queued) { - resched_curr(rq_of(cfs_rq)); + resched_curr_lazy(rq_of(cfs_rq)); return; } - /* - * don't let the period tick interfere with the hrtick preemption - */ - if (!sched_feat(DOUBLE_TICK) && - hrtimer_active(&rq_of(cfs_rq)->hrtick_timer)) - return; #endif - - if (cfs_rq->nr_running > 1) - check_preempt_tick(cfs_rq, curr); } @@ -4234,7 +5567,7 @@ void cfs_bandwidth_usage_dec(void) { static_key_slow_dec_cpuslocked(&__cfs_bandwidth_used); } -#else /* CONFIG_JUMP_LABEL */ +#else /* !CONFIG_JUMP_LABEL: */ static bool cfs_bandwidth_used(void) { return true; @@ -4242,16 +5575,7 @@ static bool cfs_bandwidth_used(void) void cfs_bandwidth_usage_inc(void) {} void cfs_bandwidth_usage_dec(void) {} -#endif /* CONFIG_JUMP_LABEL */ - -/* - * default period for cfs group bandwidth. - * default: 0.1s, units: nanoseconds - */ -static inline u64 default_cfs_period(void) -{ - return 100000000ULL; -} +#endif /* !CONFIG_JUMP_LABEL */ static inline u64 sched_cfs_bandwidth_slice(void) { @@ -4259,23 +5583,28 @@ static inline u64 sched_cfs_bandwidth_slice(void) } /* - * Replenish runtime according to assigned quota and update expiration time. - * We use sched_clock_cpu directly instead of rq->clock to avoid adding - * additional synchronization around rq->lock. + * Replenish runtime according to assigned quota. We use sched_clock_cpu + * directly instead of rq->clock to avoid adding additional synchronization + * around rq->lock. * * requires cfs_b->lock */ void __refill_cfs_bandwidth_runtime(struct cfs_bandwidth *cfs_b) { - u64 now; + s64 runtime; - if (cfs_b->quota == RUNTIME_INF) + if (unlikely(cfs_b->quota == RUNTIME_INF)) return; - now = sched_clock_cpu(smp_processor_id()); - cfs_b->runtime = cfs_b->quota; - cfs_b->runtime_expires = now + ktime_to_ns(cfs_b->period); - cfs_b->expires_seq++; + cfs_b->runtime += cfs_b->quota; + runtime = cfs_b->runtime_snap - cfs_b->runtime; + if (runtime > 0) { + cfs_b->burst_time += runtime; + cfs_b->nr_burst++; + } + + cfs_b->runtime = min(cfs_b->runtime, cfs_b->quota + cfs_b->burst); + cfs_b->runtime_snap = cfs_b->runtime; } static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) @@ -4283,27 +5612,17 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) return &tg->cfs_bandwidth; } -/* rq->task_clock normalized against any time this cfs_rq has spent throttled */ -static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq) -{ - if (unlikely(cfs_rq->throttle_count)) - return cfs_rq->throttled_clock_task - cfs_rq->throttled_clock_task_time; - - return rq_clock_task(rq_of(cfs_rq)) - cfs_rq->throttled_clock_task_time; -} - /* returns 0 on failure to allocate runtime */ -static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) +static int __assign_cfs_rq_runtime(struct cfs_bandwidth *cfs_b, + struct cfs_rq *cfs_rq, u64 target_runtime) { - struct task_group *tg = cfs_rq->tg; - struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(tg); - u64 amount = 0, min_amount, expires; - int expires_seq; + u64 min_amount, amount = 0; + + lockdep_assert_held(&cfs_b->lock); /* note: this is a positive sum as runtime_remaining <= 0 */ - min_amount = sched_cfs_bandwidth_slice() - cfs_rq->runtime_remaining; + min_amount = target_runtime - cfs_rq->runtime_remaining; - raw_spin_lock(&cfs_b->lock); if (cfs_b->quota == RUNTIME_INF) amount = min_amount; else { @@ -4315,65 +5634,35 @@ static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) cfs_b->idle = 0; } } - expires_seq = cfs_b->expires_seq; - expires = cfs_b->runtime_expires; - raw_spin_unlock(&cfs_b->lock); cfs_rq->runtime_remaining += amount; - /* - * we may have advanced our local expiration to account for allowed - * spread between our sched_clock and the one on which runtime was - * issued. - */ - if (cfs_rq->expires_seq != expires_seq) { - cfs_rq->expires_seq = expires_seq; - cfs_rq->runtime_expires = expires; - } return cfs_rq->runtime_remaining > 0; } -/* - * Note: This depends on the synchronization provided by sched_clock and the - * fact that rq->clock snapshots this value. - */ -static void expire_cfs_rq_runtime(struct cfs_rq *cfs_rq) +/* returns 0 on failure to allocate runtime */ +static int assign_cfs_rq_runtime(struct cfs_rq *cfs_rq) { struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); + int ret; - /* if the deadline is ahead of our clock, nothing to do */ - if (likely((s64)(rq_clock(rq_of(cfs_rq)) - cfs_rq->runtime_expires) < 0)) - return; - - if (cfs_rq->runtime_remaining < 0) - return; + raw_spin_lock(&cfs_b->lock); + ret = __assign_cfs_rq_runtime(cfs_b, cfs_rq, sched_cfs_bandwidth_slice()); + raw_spin_unlock(&cfs_b->lock); - /* - * If the local deadline has passed we have to consider the - * possibility that our sched_clock is 'fast' and the global deadline - * has not truly expired. - * - * Fortunately we can check determine whether this the case by checking - * whether the global deadline(cfs_b->expires_seq) has advanced. - */ - if (cfs_rq->expires_seq == cfs_b->expires_seq) { - /* extend local deadline, drift is bounded above by 2 ticks */ - cfs_rq->runtime_expires += TICK_NSEC; - } else { - /* global deadline is ahead, expiration has passed */ - cfs_rq->runtime_remaining = 0; - } + return ret; } static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec) { /* dock delta_exec before expiring quota (as it could span periods) */ cfs_rq->runtime_remaining -= delta_exec; - expire_cfs_rq_runtime(cfs_rq); if (likely(cfs_rq->runtime_remaining > 0)) return; + if (cfs_rq->throttled) + return; /* * if we're unable to extend our runtime we resched so that the active * hierarchy can be throttled @@ -4396,200 +5685,500 @@ static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) return cfs_bandwidth_used() && cfs_rq->throttled; } +static inline bool cfs_rq_pelt_clock_throttled(struct cfs_rq *cfs_rq) +{ + return cfs_bandwidth_used() && cfs_rq->pelt_clock_throttled; +} + /* check whether cfs_rq, or any parent, is throttled */ static inline int throttled_hierarchy(struct cfs_rq *cfs_rq) { return cfs_bandwidth_used() && cfs_rq->throttle_count; } +static inline int lb_throttled_hierarchy(struct task_struct *p, int dst_cpu) +{ + return throttled_hierarchy(task_group(p)->cfs_rq[dst_cpu]); +} + +static inline bool task_is_throttled(struct task_struct *p) +{ + return cfs_bandwidth_used() && p->throttled; +} + +static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags); +static void throttle_cfs_rq_work(struct callback_head *work) +{ + struct task_struct *p = container_of(work, struct task_struct, sched_throttle_work); + struct sched_entity *se; + struct cfs_rq *cfs_rq; + struct rq *rq; + + WARN_ON_ONCE(p != current); + p->sched_throttle_work.next = &p->sched_throttle_work; + + /* + * If task is exiting, then there won't be a return to userspace, so we + * don't have to bother with any of this. + */ + if ((p->flags & PF_EXITING)) + return; + + scoped_guard(task_rq_lock, p) { + se = &p->se; + cfs_rq = cfs_rq_of(se); + + /* Raced, forget */ + if (p->sched_class != &fair_sched_class) + return; + + /* + * If not in limbo, then either replenish has happened or this + * task got migrated out of the throttled cfs_rq, move along. + */ + if (!cfs_rq->throttle_count) + return; + rq = scope.rq; + update_rq_clock(rq); + WARN_ON_ONCE(p->throttled || !list_empty(&p->throttle_node)); + dequeue_task_fair(rq, p, DEQUEUE_SLEEP | DEQUEUE_THROTTLE); + list_add(&p->throttle_node, &cfs_rq->throttled_limbo_list); + /* + * Must not set throttled before dequeue or dequeue will + * mistakenly regard this task as an already throttled one. + */ + p->throttled = true; + resched_curr(rq); + } +} + +void init_cfs_throttle_work(struct task_struct *p) +{ + init_task_work(&p->sched_throttle_work, throttle_cfs_rq_work); + /* Protect against double add, see throttle_cfs_rq() and throttle_cfs_rq_work() */ + p->sched_throttle_work.next = &p->sched_throttle_work; + INIT_LIST_HEAD(&p->throttle_node); +} + /* - * Ensure that neither of the group entities corresponding to src_cpu or - * dest_cpu are members of a throttled hierarchy when performing group - * load-balance operations. + * Task is throttled and someone wants to dequeue it again: + * it could be sched/core when core needs to do things like + * task affinity change, task group change, task sched class + * change etc. and in these cases, DEQUEUE_SLEEP is not set; + * or the task is blocked after throttled due to freezer etc. + * and in these cases, DEQUEUE_SLEEP is set. */ -static inline int throttled_lb_pair(struct task_group *tg, - int src_cpu, int dest_cpu) +static void detach_task_cfs_rq(struct task_struct *p); +static void dequeue_throttled_task(struct task_struct *p, int flags) { - struct cfs_rq *src_cfs_rq, *dest_cfs_rq; + WARN_ON_ONCE(p->se.on_rq); + list_del_init(&p->throttle_node); - src_cfs_rq = tg->cfs_rq[src_cpu]; - dest_cfs_rq = tg->cfs_rq[dest_cpu]; + /* task blocked after throttled */ + if (flags & DEQUEUE_SLEEP) { + p->throttled = false; + return; + } - return throttled_hierarchy(src_cfs_rq) || - throttled_hierarchy(dest_cfs_rq); + /* + * task is migrating off its old cfs_rq, detach + * the task's load from its old cfs_rq. + */ + if (task_on_rq_migrating(p)) + detach_task_cfs_rq(p); } +static bool enqueue_throttled_task(struct task_struct *p) +{ + struct cfs_rq *cfs_rq = cfs_rq_of(&p->se); + + /* @p should have gone through dequeue_throttled_task() first */ + WARN_ON_ONCE(!list_empty(&p->throttle_node)); + + /* + * If the throttled task @p is enqueued to a throttled cfs_rq, + * take the fast path by directly putting the task on the + * target cfs_rq's limbo list. + * + * Do not do that when @p is current because the following race can + * cause @p's group_node to be incorectly re-insterted in its rq's + * cfs_tasks list, despite being throttled: + * + * cpuX cpuY + * p ret2user + * throttle_cfs_rq_work() sched_move_task(p) + * LOCK task_rq_lock + * dequeue_task_fair(p) + * UNLOCK task_rq_lock + * LOCK task_rq_lock + * task_current_donor(p) == true + * task_on_rq_queued(p) == true + * dequeue_task(p) + * put_prev_task(p) + * sched_change_group() + * enqueue_task(p) -> p's new cfs_rq + * is throttled, go + * fast path and skip + * actual enqueue + * set_next_task(p) + * list_move(&se->group_node, &rq->cfs_tasks); // bug + * schedule() + * + * In the above race case, @p current cfs_rq is in the same rq as + * its previous cfs_rq because sched_move_task() only moves a task + * to a different group from the same rq, so we can use its current + * cfs_rq to derive rq and test if the task is current. + */ + if (throttled_hierarchy(cfs_rq) && + !task_current_donor(rq_of(cfs_rq), p)) { + list_add(&p->throttle_node, &cfs_rq->throttled_limbo_list); + return true; + } + + /* we can't take the fast path, do an actual enqueue*/ + p->throttled = false; + return false; +} + +static void enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags); static int tg_unthrottle_up(struct task_group *tg, void *data) { struct rq *rq = data; struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; + struct task_struct *p, *tmp; + + if (--cfs_rq->throttle_count) + return 0; + + if (cfs_rq->pelt_clock_throttled) { + cfs_rq->throttled_clock_pelt_time += rq_clock_pelt(rq) - + cfs_rq->throttled_clock_pelt; + cfs_rq->pelt_clock_throttled = 0; + } + + if (cfs_rq->throttled_clock_self) { + u64 delta = rq_clock(rq) - cfs_rq->throttled_clock_self; + + cfs_rq->throttled_clock_self = 0; - cfs_rq->throttle_count--; - if (!cfs_rq->throttle_count) { - /* adjust cfs_rq_clock_task() */ - cfs_rq->throttled_clock_task_time += rq_clock_task(rq) - - cfs_rq->throttled_clock_task; + if (WARN_ON_ONCE((s64)delta < 0)) + delta = 0; + + cfs_rq->throttled_clock_self_time += delta; + } + + /* Re-enqueue the tasks that have been throttled at this level. */ + list_for_each_entry_safe(p, tmp, &cfs_rq->throttled_limbo_list, throttle_node) { + list_del_init(&p->throttle_node); + p->throttled = false; + enqueue_task_fair(rq_of(cfs_rq), p, ENQUEUE_WAKEUP); } + /* Add cfs_rq with load or one or more already running entities to the list */ + if (!cfs_rq_is_decayed(cfs_rq)) + list_add_leaf_cfs_rq(cfs_rq); + return 0; } +static inline bool task_has_throttle_work(struct task_struct *p) +{ + return p->sched_throttle_work.next != &p->sched_throttle_work; +} + +static inline void task_throttle_setup_work(struct task_struct *p) +{ + if (task_has_throttle_work(p)) + return; + + /* + * Kthreads and exiting tasks don't return to userspace, so adding the + * work is pointless + */ + if ((p->flags & (PF_EXITING | PF_KTHREAD))) + return; + + task_work_add(p, &p->sched_throttle_work, TWA_RESUME); +} + +static void record_throttle_clock(struct cfs_rq *cfs_rq) +{ + struct rq *rq = rq_of(cfs_rq); + + if (cfs_rq_throttled(cfs_rq) && !cfs_rq->throttled_clock) + cfs_rq->throttled_clock = rq_clock(rq); + + if (!cfs_rq->throttled_clock_self) + cfs_rq->throttled_clock_self = rq_clock(rq); +} + static int tg_throttle_down(struct task_group *tg, void *data) { struct rq *rq = data; struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; - /* group is entering throttled state, stop time */ - if (!cfs_rq->throttle_count) - cfs_rq->throttled_clock_task = rq_clock_task(rq); - cfs_rq->throttle_count++; + if (cfs_rq->throttle_count++) + return 0; + + /* + * For cfs_rqs that still have entities enqueued, PELT clock + * stop happens at dequeue time when all entities are dequeued. + */ + if (!cfs_rq->nr_queued) { + list_del_leaf_cfs_rq(cfs_rq); + cfs_rq->throttled_clock_pelt = rq_clock_pelt(rq); + cfs_rq->pelt_clock_throttled = 1; + } + WARN_ON_ONCE(cfs_rq->throttled_clock_self); + WARN_ON_ONCE(!list_empty(&cfs_rq->throttled_limbo_list)); return 0; } -static void throttle_cfs_rq(struct cfs_rq *cfs_rq) +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, dequeue = 1; - bool empty; + int dequeue = 1; - se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; + raw_spin_lock(&cfs_b->lock); + /* This will start the period timer if necessary */ + if (__assign_cfs_rq_runtime(cfs_b, cfs_rq, 1)) { + /* + * We have raced with bandwidth becoming available, and if we + * actually throttled the timer might not unthrottle us for an + * entire period. We additionally needed to make sure that any + * subsequent check_cfs_rq_runtime calls agree not to throttle + * us, as we may commit to do cfs put_prev+pick_next, so we ask + * for 1ns of runtime rather than just check cfs_b. + */ + dequeue = 0; + } else { + list_add_tail_rcu(&cfs_rq->throttled_list, + &cfs_b->throttled_cfs_rq); + } + raw_spin_unlock(&cfs_b->lock); + + if (!dequeue) + return false; /* Throttle no longer required. */ /* freeze hierarchy runnable averages while throttled */ rcu_read_lock(); walk_tg_tree_from(cfs_rq->tg, tg_throttle_down, tg_nop, (void *)rq); rcu_read_unlock(); - task_delta = cfs_rq->h_nr_running; - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); - /* throttled entity or throttle-on-deactivate */ - if (!se->on_rq) - break; - - if (dequeue) - dequeue_entity(qcfs_rq, se, DEQUEUE_SLEEP); - qcfs_rq->h_nr_running -= task_delta; - - if (qcfs_rq->load.weight) - dequeue = 0; - } - - if (!se) - sub_nr_running(rq, task_delta); - - cfs_rq->throttled = 1; - cfs_rq->throttled_clock = rq_clock(rq); - raw_spin_lock(&cfs_b->lock); - empty = list_empty(&cfs_b->throttled_cfs_rq); - /* - * Add to the _head_ of the list, so that an already-started - * distribute_cfs_runtime will not see us. If disribute_cfs_runtime is - * not running add to the tail so that later runqueues don't get starved. + * Note: distribution will already see us throttled via the + * throttled-list. rq->lock protects completion. */ - if (cfs_b->distribute_running) - list_add_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); - else - list_add_tail_rcu(&cfs_rq->throttled_list, &cfs_b->throttled_cfs_rq); - - /* - * If we're the first throttled task, make sure the bandwidth - * timer is running. - */ - if (empty) - start_cfs_bandwidth(cfs_b); - - raw_spin_unlock(&cfs_b->lock); + cfs_rq->throttled = 1; + WARN_ON_ONCE(cfs_rq->throttled_clock); + return true; } 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; - int enqueue = 1; - long task_delta; + struct sched_entity *se = cfs_rq->tg->se[cpu_of(rq)]; - se = cfs_rq->tg->se[cpu_of(rq)]; + /* + * It's possible we are called with runtime_remaining < 0 due to things + * like async unthrottled us with a positive runtime_remaining but other + * still running entities consumed those runtime before we reached here. + * + * We can't unthrottle this cfs_rq without any runtime remaining because + * any enqueue in tg_unthrottle_up() will immediately trigger a throttle, + * which is not supposed to happen on unthrottle path. + */ + if (cfs_rq->runtime_enabled && cfs_rq->runtime_remaining <= 0) + return; cfs_rq->throttled = 0; update_rq_clock(rq); raw_spin_lock(&cfs_b->lock); - cfs_b->throttled_time += rq_clock(rq) - cfs_rq->throttled_clock; + if (cfs_rq->throttled_clock) { + cfs_b->throttled_time += rq_clock(rq) - cfs_rq->throttled_clock; + cfs_rq->throttled_clock = 0; + } list_del_rcu(&cfs_rq->throttled_list); raw_spin_unlock(&cfs_b->lock); /* update hierarchical throttle state */ walk_tg_tree_from(cfs_rq->tg, tg_nop, tg_unthrottle_up, (void *)rq); - if (!cfs_rq->load.weight) - return; + if (!cfs_rq->load.weight) { + if (!cfs_rq->on_list) + return; + /* + * Nothing to run but something to decay (on_list)? + * Complete the branch. + */ + for_each_sched_entity(se) { + if (list_add_leaf_cfs_rq(cfs_rq_of(se))) + break; + } + } - task_delta = cfs_rq->h_nr_running; - for_each_sched_entity(se) { - if (se->on_rq) - enqueue = 0; + assert_list_leaf_cfs_rq(rq); - cfs_rq = cfs_rq_of(se); - if (enqueue) - enqueue_entity(cfs_rq, se, ENQUEUE_WAKEUP); - cfs_rq->h_nr_running += task_delta; + /* Determine whether we need to wake up potentially idle CPU: */ + if (rq->curr == rq->idle && rq->cfs.nr_queued) + resched_curr(rq); +} - if (cfs_rq_throttled(cfs_rq)) - break; +static void __cfsb_csd_unthrottle(void *arg) +{ + struct cfs_rq *cursor, *tmp; + struct rq *rq = arg; + struct rq_flags rf; + + rq_lock(rq, &rf); + + /* + * Iterating over the list can trigger several call to + * update_rq_clock() in unthrottle_cfs_rq(). + * Do it once and skip the potential next ones. + */ + update_rq_clock(rq); + rq_clock_start_loop_update(rq); + + /* + * Since we hold rq lock we're safe from concurrent manipulation of + * the CSD list. However, this RCU critical section annotates the + * fact that we pair with sched_free_group_rcu(), so that we cannot + * race with group being freed in the window between removing it + * from the list and advancing to the next entry in the list. + */ + rcu_read_lock(); + + list_for_each_entry_safe(cursor, tmp, &rq->cfsb_csd_list, + throttled_csd_list) { + list_del_init(&cursor->throttled_csd_list); + + if (cfs_rq_throttled(cursor)) + unthrottle_cfs_rq(cursor); } - if (!se) - add_nr_running(rq, task_delta); + rcu_read_unlock(); - /* Determine whether we need to wake up potentially idle CPU: */ - if (rq->curr == rq->idle && rq->cfs.nr_running) - resched_curr(rq); + rq_clock_stop_loop_update(rq); + rq_unlock(rq, &rf); } -static u64 distribute_cfs_runtime(struct cfs_bandwidth *cfs_b, - u64 remaining, u64 expires) +static inline void __unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) { - struct cfs_rq *cfs_rq; - u64 runtime; - u64 starting_runtime = remaining; + struct rq *rq = rq_of(cfs_rq); + bool first; + + if (rq == this_rq()) { + unthrottle_cfs_rq(cfs_rq); + return; + } + + /* Already enqueued */ + if (WARN_ON_ONCE(!list_empty(&cfs_rq->throttled_csd_list))) + return; + + first = list_empty(&rq->cfsb_csd_list); + list_add_tail(&cfs_rq->throttled_csd_list, &rq->cfsb_csd_list); + if (first) + smp_call_function_single_async(cpu_of(rq), &rq->cfsb_csd); +} + +static void unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) +{ + lockdep_assert_rq_held(rq_of(cfs_rq)); + + if (WARN_ON_ONCE(!cfs_rq_throttled(cfs_rq) || + cfs_rq->runtime_remaining <= 0)) + return; + + __unthrottle_cfs_rq_async(cfs_rq); +} + +static bool distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) +{ + int this_cpu = smp_processor_id(); + u64 runtime, remaining = 1; + bool throttled = false; + struct cfs_rq *cfs_rq, *tmp; + struct rq_flags rf; + struct rq *rq; + LIST_HEAD(local_unthrottle); rcu_read_lock(); list_for_each_entry_rcu(cfs_rq, &cfs_b->throttled_cfs_rq, throttled_list) { - struct rq *rq = rq_of(cfs_rq); - struct rq_flags rf; + rq = rq_of(cfs_rq); + + if (!remaining) { + throttled = true; + break; + } - rq_lock(rq, &rf); + rq_lock_irqsave(rq, &rf); if (!cfs_rq_throttled(cfs_rq)) goto next; + /* Already queued for async unthrottle */ + if (!list_empty(&cfs_rq->throttled_csd_list)) + goto next; + + /* By the above checks, this should never be true */ + WARN_ON_ONCE(cfs_rq->runtime_remaining > 0); + + raw_spin_lock(&cfs_b->lock); runtime = -cfs_rq->runtime_remaining + 1; - if (runtime > remaining) - runtime = remaining; - remaining -= runtime; + if (runtime > cfs_b->runtime) + runtime = cfs_b->runtime; + cfs_b->runtime -= runtime; + remaining = cfs_b->runtime; + raw_spin_unlock(&cfs_b->lock); cfs_rq->runtime_remaining += runtime; - cfs_rq->runtime_expires = expires; /* we check whether we're throttled above */ - if (cfs_rq->runtime_remaining > 0) - unthrottle_cfs_rq(cfs_rq); + if (cfs_rq->runtime_remaining > 0) { + if (cpu_of(rq) != this_cpu) { + unthrottle_cfs_rq_async(cfs_rq); + } else { + /* + * We currently only expect to be unthrottling + * a single cfs_rq locally. + */ + WARN_ON_ONCE(!list_empty(&local_unthrottle)); + list_add_tail(&cfs_rq->throttled_csd_list, + &local_unthrottle); + } + } else { + throttled = true; + } next: - rq_unlock(rq, &rf); + rq_unlock_irqrestore(rq, &rf); + } - if (!remaining) - break; + list_for_each_entry_safe(cfs_rq, tmp, &local_unthrottle, + throttled_csd_list) { + struct rq *rq = rq_of(cfs_rq); + + rq_lock_irqsave(rq, &rf); + + list_del_init(&cfs_rq->throttled_csd_list); + + if (cfs_rq_throttled(cfs_rq)) + unthrottle_cfs_rq(cfs_rq); + + rq_unlock_irqrestore(rq, &rf); } + WARN_ON_ONCE(!list_empty(&local_unthrottle)); + rcu_read_unlock(); - return starting_runtime - remaining; + return throttled; } /* @@ -4598,9 +6187,8 @@ next: * period the timer is deactivated until scheduling resumes; cfs_b->idle is * used to track this state. */ -static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) +static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun, unsigned long flags) { - u64 runtime, runtime_expires; int throttled; /* no need to continue the timer with no bandwidth constraint */ @@ -4610,6 +6198,9 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) throttled = !list_empty(&cfs_b->throttled_cfs_rq); cfs_b->nr_periods += overrun; + /* Refill extra burst quota even if cfs_b->idle */ + __refill_cfs_bandwidth_runtime(cfs_b); + /* * idle depends on !throttled (for the case of a large deficit), and if * we're going inactive then everything else can be deferred @@ -4617,8 +6208,6 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) if (cfs_b->idle && !throttled) goto out_deactivate; - __refill_cfs_bandwidth_runtime(cfs_b); - if (!throttled) { /* mark as potentially idle for the upcoming period */ cfs_b->idle = 1; @@ -4628,28 +6217,14 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun) /* account preceding periods in which throttling occurred */ cfs_b->nr_throttled += overrun; - runtime_expires = cfs_b->runtime_expires; - /* - * This check is repeated as we are holding onto the new bandwidth while - * we unthrottle. This can potentially race with an unthrottled group - * trying to acquire new bandwidth from the global pool. This can result - * in us over-using our runtime if it is all used during this loop, but - * only by limited amounts in that extreme case. + * This check is repeated as we release cfs_b->lock while we unthrottle. */ - while (throttled && cfs_b->runtime > 0 && !cfs_b->distribute_running) { - runtime = cfs_b->runtime; - cfs_b->distribute_running = 1; - raw_spin_unlock(&cfs_b->lock); + while (throttled && cfs_b->runtime > 0) { + raw_spin_unlock_irqrestore(&cfs_b->lock, flags); /* we can't nest cfs_b->lock while distributing bandwidth */ - runtime = distribute_cfs_runtime(cfs_b, runtime, - runtime_expires); - raw_spin_lock(&cfs_b->lock); - - cfs_b->distribute_running = 0; - throttled = !list_empty(&cfs_b->throttled_cfs_rq); - - lsub_positive(&cfs_b->runtime, runtime); + throttled = distribute_cfs_runtime(cfs_b); + raw_spin_lock_irqsave(&cfs_b->lock, flags); } /* @@ -4683,7 +6258,7 @@ static const u64 cfs_bandwidth_slack_period = 5 * NSEC_PER_MSEC; static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) { struct hrtimer *refresh_timer = &cfs_b->period_timer; - u64 remaining; + s64 remaining; /* if the call-back is running a quota refresh is already occurring */ if (hrtimer_callback_running(refresh_timer)) @@ -4691,7 +6266,7 @@ static int runtime_refresh_within(struct cfs_bandwidth *cfs_b, u64 min_expire) /* is a quota refresh about to occur? */ remaining = ktime_to_ns(hrtimer_expires_remaining(refresh_timer)); - if (remaining < min_expire) + if (remaining < (s64)min_expire) return 1; return 0; @@ -4705,6 +6280,11 @@ static void start_cfs_slack_bandwidth(struct cfs_bandwidth *cfs_b) if (runtime_refresh_within(cfs_b, min_left)) return; + /* don't push forwards an existing deferred unthrottle */ + if (cfs_b->slack_started) + return; + cfs_b->slack_started = true; + hrtimer_start(&cfs_b->slack_timer, ns_to_ktime(cfs_bandwidth_slack_period), HRTIMER_MODE_REL); @@ -4720,8 +6300,7 @@ static void __return_cfs_rq_runtime(struct cfs_rq *cfs_rq) return; raw_spin_lock(&cfs_b->lock); - if (cfs_b->quota != RUNTIME_INF && - cfs_rq->runtime_expires == cfs_b->runtime_expires) { + if (cfs_b->quota != RUNTIME_INF) { cfs_b->runtime += slack_runtime; /* we are under rq->lock, defer unthrottling using a timer */ @@ -4740,7 +6319,7 @@ 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); @@ -4753,45 +6332,32 @@ static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) { u64 runtime = 0, slice = sched_cfs_bandwidth_slice(); - u64 expires; + unsigned long flags; /* confirm we're still not at a refresh boundary */ - raw_spin_lock(&cfs_b->lock); - if (cfs_b->distribute_running) { - raw_spin_unlock(&cfs_b->lock); - return; - } + raw_spin_lock_irqsave(&cfs_b->lock, flags); + cfs_b->slack_started = false; if (runtime_refresh_within(cfs_b, min_bandwidth_expiration)) { - raw_spin_unlock(&cfs_b->lock); + raw_spin_unlock_irqrestore(&cfs_b->lock, flags); return; } if (cfs_b->quota != RUNTIME_INF && cfs_b->runtime > slice) runtime = cfs_b->runtime; - expires = cfs_b->runtime_expires; - if (runtime) - cfs_b->distribute_running = 1; - - raw_spin_unlock(&cfs_b->lock); + raw_spin_unlock_irqrestore(&cfs_b->lock, flags); if (!runtime) return; - runtime = distribute_cfs_runtime(cfs_b, runtime, expires); - - raw_spin_lock(&cfs_b->lock); - if (expires == cfs_b->runtime_expires) - lsub_positive(&cfs_b->runtime, runtime); - cfs_b->distribute_running = 0; - raw_spin_unlock(&cfs_b->lock); + distribute_cfs_runtime(cfs_b); } /* * When a group wakes up we want to make sure that its quota is not already * expired/exceeded, otherwise it may be allowed to steal additional ticks of - * runtime as update_curr() throttling can not not trigger until it's on-rq. + * runtime as update_curr() throttling can not trigger until it's on-rq. */ static void check_enqueue_throttle(struct cfs_rq *cfs_rq) { @@ -4826,7 +6392,17 @@ static void sync_throttle(struct task_group *tg, int cpu) pcfs_rq = tg->parent->cfs_rq[cpu]; cfs_rq->throttle_count = pcfs_rq->throttle_count; - cfs_rq->throttled_clock_task = rq_clock_task(cpu_rq(cpu)); + cfs_rq->throttled_clock_pelt = rq_clock_pelt(cpu_rq(cpu)); + + /* + * It is not enough to sync the "pelt_clock_throttled" indicator + * with the parent cfs_rq when the hierarchy is not queued. + * Always join a throttled hierarchy with PELT clock throttled + * and leaf it to the first enqueue, or distribution to + * unthrottle the PELT clock. + */ + if (cfs_rq->throttle_count) + cfs_rq->pelt_clock_throttled = 1; } /* conditionally throttle active cfs_rq's from put_prev_entity() */ @@ -4845,8 +6421,7 @@ static bool check_cfs_rq_runtime(struct cfs_rq *cfs_rq) if (cfs_rq_throttled(cfs_rq)) return true; - throttle_cfs_rq(cfs_rq); - return true; + return throttle_cfs_rq(cfs_rq); } static enum hrtimer_restart sched_cfs_slack_timer(struct hrtimer *timer) @@ -4863,69 +6438,130 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) { struct cfs_bandwidth *cfs_b = container_of(timer, struct cfs_bandwidth, period_timer); + unsigned long flags; int overrun; int idle = 0; + int count = 0; - raw_spin_lock(&cfs_b->lock); + raw_spin_lock_irqsave(&cfs_b->lock, flags); for (;;) { overrun = hrtimer_forward_now(timer, cfs_b->period); if (!overrun) break; - idle = do_sched_cfs_period_timer(cfs_b, overrun); + idle = do_sched_cfs_period_timer(cfs_b, overrun, flags); + + if (++count > 3) { + u64 new, old = ktime_to_ns(cfs_b->period); + + /* + * Grow period by a factor of 2 to avoid losing precision. + * Precision loss in the quota/period ratio can cause __cfs_schedulable + * to fail. + */ + new = old * 2; + if (new < max_bw_quota_period_us * NSEC_PER_USEC) { + cfs_b->period = ns_to_ktime(new); + cfs_b->quota *= 2; + cfs_b->burst *= 2; + + pr_warn_ratelimited( + "cfs_period_timer[cpu%d]: period too short, scaling up (new cfs_period_us = %lld, cfs_quota_us = %lld)\n", + smp_processor_id(), + div_u64(new, NSEC_PER_USEC), + div_u64(cfs_b->quota, NSEC_PER_USEC)); + } else { + pr_warn_ratelimited( + "cfs_period_timer[cpu%d]: period too short, but cannot scale up without losing precision (cfs_period_us = %lld, cfs_quota_us = %lld)\n", + smp_processor_id(), + div_u64(old, NSEC_PER_USEC), + div_u64(cfs_b->quota, NSEC_PER_USEC)); + } + + /* reset count so we don't come right back in here */ + count = 0; + } } if (idle) cfs_b->period_active = 0; - raw_spin_unlock(&cfs_b->lock); + raw_spin_unlock_irqrestore(&cfs_b->lock, flags); return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; } -void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) +void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *parent) { raw_spin_lock_init(&cfs_b->lock); cfs_b->runtime = 0; cfs_b->quota = RUNTIME_INF; - cfs_b->period = ns_to_ktime(default_cfs_period()); + cfs_b->period = us_to_ktime(default_bw_period_us()); + cfs_b->burst = 0; + cfs_b->hierarchical_quota = parent ? parent->hierarchical_quota : RUNTIME_INF; INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); - hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); - cfs_b->period_timer.function = sched_cfs_period_timer; - hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - cfs_b->slack_timer.function = sched_cfs_slack_timer; - cfs_b->distribute_running = 0; + hrtimer_setup(&cfs_b->period_timer, sched_cfs_period_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_ABS_PINNED); + + /* Add a random offset so that timers interleave */ + hrtimer_set_expires(&cfs_b->period_timer, + get_random_u32_below(cfs_b->period)); + hrtimer_setup(&cfs_b->slack_timer, sched_cfs_slack_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL); + cfs_b->slack_started = false; } static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) { cfs_rq->runtime_enabled = 0; INIT_LIST_HEAD(&cfs_rq->throttled_list); + INIT_LIST_HEAD(&cfs_rq->throttled_csd_list); + INIT_LIST_HEAD(&cfs_rq->throttled_limbo_list); } void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { - u64 overrun; - lockdep_assert_held(&cfs_b->lock); if (cfs_b->period_active) return; cfs_b->period_active = 1; - overrun = hrtimer_forward_now(&cfs_b->period_timer, cfs_b->period); - cfs_b->runtime_expires += (overrun + 1) * ktime_to_ns(cfs_b->period); - cfs_b->expires_seq++; + hrtimer_forward_now(&cfs_b->period_timer, cfs_b->period); hrtimer_start_expires(&cfs_b->period_timer, HRTIMER_MODE_ABS_PINNED); } static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { + int __maybe_unused i; + /* init_cfs_bandwidth() was not called */ if (!cfs_b->throttled_cfs_rq.next) return; hrtimer_cancel(&cfs_b->period_timer); hrtimer_cancel(&cfs_b->slack_timer); + + /* + * It is possible that we still have some cfs_rq's pending on a CSD + * list, though this race is very rare. In order for this to occur, we + * must have raced with the last task leaving the group while there + * exist throttled cfs_rq(s), and the period_timer must have queued the + * CSD item but the remote cpu has not yet processed it. To handle this, + * we can simply flush all pending CSD work inline here. We're + * guaranteed at this point that no additional cfs_rq of this group can + * join a CSD list. + */ + for_each_possible_cpu(i) { + struct rq *rq = cpu_rq(i); + unsigned long flags; + + if (list_empty(&rq->cfsb_csd_list)) + continue; + + local_irq_save(flags); + __cfsb_csd_unthrottle(rq); + local_irq_restore(flags); + } } /* @@ -4935,12 +6571,12 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) * bits doesn't do much. */ -/* cpu online calback */ +/* cpu online callback */ static void __maybe_unused update_runtime_enabled(struct rq *rq) { struct task_group *tg; - lockdep_assert_held(&rq->lock); + lockdep_assert_rq_held(rq); rcu_read_lock(); list_for_each_entry_rcu(tg, &task_groups, list) { @@ -4959,7 +6595,18 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) { struct task_group *tg; - lockdep_assert_held(&rq->lock); + 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 + * the rq clock again in unthrottle_cfs_rq(). + */ + rq_clock_start_loop_update(rq); rcu_read_lock(); list_for_each_entry_rcu(tg, &task_groups, list) { @@ -4969,53 +6616,101 @@ 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(); + + rq_clock_stop_loop_update(rq); +} + +bool cfs_task_bw_constrained(struct task_struct *p) +{ + struct cfs_rq *cfs_rq = task_cfs_rq(p); + + if (!cfs_bandwidth_used()) + return false; + + if (cfs_rq->runtime_enabled || + tg_cfs_bandwidth(cfs_rq->tg)->hierarchical_quota != RUNTIME_INF) + return true; + + return false; } -#else /* CONFIG_CFS_BANDWIDTH */ -static inline u64 cfs_rq_clock_task(struct cfs_rq *cfs_rq) +#ifdef CONFIG_NO_HZ_FULL +/* called from pick_next_task_fair() */ +static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) { - return rq_clock_task(rq_of(cfs_rq)); + int cpu = cpu_of(rq); + + if (!cfs_bandwidth_used()) + return; + + if (!tick_nohz_full_cpu(cpu)) + return; + + if (rq->nr_running != 1) + return; + + /* + * We know there is only one task runnable and we've just picked it. The + * normal enqueue path will have cleared TICK_DEP_BIT_SCHED if we will + * be otherwise able to stop the tick. Just need to check if we are using + * bandwidth control. + */ + if (cfs_task_bw_constrained(p)) + tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED); } +#endif /* CONFIG_NO_HZ_FULL */ + +#else /* !CONFIG_CFS_BANDWIDTH: */ 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) {} static inline void sync_throttle(struct task_group *tg, int cpu) {} static __always_inline void return_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} +static void task_throttle_setup_work(struct task_struct *p) {} +static bool task_is_throttled(struct task_struct *p) { return false; } +static void dequeue_throttled_task(struct task_struct *p, int flags) {} +static bool enqueue_throttled_task(struct task_struct *p) { return false; } +static void record_throttle_clock(struct cfs_rq *cfs_rq) {} static inline int cfs_rq_throttled(struct cfs_rq *cfs_rq) { return 0; } +static inline bool cfs_rq_pelt_clock_throttled(struct cfs_rq *cfs_rq) +{ + return false; +} + static inline int throttled_hierarchy(struct cfs_rq *cfs_rq) { return 0; } -static inline int throttled_lb_pair(struct task_group *tg, - int src_cpu, int dest_cpu) +static inline int lb_throttled_hierarchy(struct task_struct *p, int dst_cpu) { return 0; } -void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} - #ifdef CONFIG_FAIR_GROUP_SCHED +void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *parent) {} static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} #endif @@ -5026,8 +6721,17 @@ static inline struct cfs_bandwidth *tg_cfs_bandwidth(struct task_group *tg) static inline void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} static inline void update_runtime_enabled(struct rq *rq) {} static inline void unthrottle_offline_cfs_rqs(struct rq *rq) {} +#ifdef CONFIG_CGROUP_SCHED +bool cfs_task_bw_constrained(struct task_struct *p) +{ + return false; +} +#endif +#endif /* !CONFIG_CFS_BANDWIDTH */ -#endif /* CONFIG_CFS_BANDWIDTH */ +#if !defined(CONFIG_CFS_BANDWIDTH) || !defined(CONFIG_NO_HZ_FULL) +static inline void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) {} +#endif /************************************************** * CFS operations on tasks: @@ -5037,17 +6741,16 @@ static inline void unthrottle_offline_cfs_rqs(struct rq *rq) {} static void hrtick_start_fair(struct rq *rq, struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); - SCHED_WARN_ON(task_rq(p) != rq); + WARN_ON_ONCE(task_rq(p) != rq); - if (rq->cfs.h_nr_running > 1) { - u64 slice = sched_slice(cfs_rq, se); + 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; if (delta < 0) { - if (rq->curr == p) + if (task_current_donor(rq, p)) resched_curr(rq); return; } @@ -5062,15 +6765,14 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p) */ static void hrtick_update(struct rq *rq) { - struct task_struct *curr = rq->curr; + struct task_struct *donor = rq->donor; - if (!hrtick_enabled(rq) || curr->sched_class != &fair_sched_class) + if (!hrtick_enabled_fair(rq) || donor->sched_class != &fair_sched_class) return; - if (cfs_rq_of(&curr->se)->nr_running < sched_nr_latency) - hrtick_start_fair(rq, curr); + hrtick_start_fair(rq, donor); } -#else /* !CONFIG_SCHED_HRTICK */ +#else /* !CONFIG_SCHED_HRTICK: */ static inline void hrtick_start_fair(struct rq *rq, struct task_struct *p) { @@ -5079,25 +6781,92 @@ hrtick_start_fair(struct rq *rq, struct task_struct *p) static inline void hrtick_update(struct rq *rq) { } -#endif - -#ifdef CONFIG_SMP -static inline unsigned long cpu_util(int cpu); -static unsigned long capacity_of(int cpu); +#endif /* !CONFIG_SCHED_HRTICK */ static inline bool cpu_overutilized(int cpu) { - return (capacity_of(cpu) * 1024) < (cpu_util(cpu) * capacity_margin); + unsigned long rq_util_min, rq_util_max; + + if (!sched_energy_enabled()) + return false; + + rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN); + rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX); + + /* Return true only if the utilization doesn't fit CPU's capacity */ + return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu); } -static inline void update_overutilized_status(struct rq *rq) +/* + * overutilized value make sense only if EAS is enabled + */ +static inline bool is_rd_overutilized(struct root_domain *rd) { - if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu)) - WRITE_ONCE(rq->rd->overutilized, SG_OVERUTILIZED); + return !sched_energy_enabled() || READ_ONCE(rd->overutilized); +} + +static inline void set_rd_overutilized(struct root_domain *rd, bool flag) +{ + if (!sched_energy_enabled()) + return; + + WRITE_ONCE(rd->overutilized, flag); + trace_sched_overutilized_tp(rd, flag); +} + +static inline void check_update_overutilized_status(struct rq *rq) +{ + /* + * overutilized field is used for load balancing decisions only + * if energy aware scheduler is being used + */ + + if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu)) + set_rd_overutilized(rq->rd, 1); +} + +/* Runqueue only has SCHED_IDLE tasks enqueued */ +static int sched_idle_rq(struct rq *rq) +{ + return unlikely(rq->nr_running == rq->cfs.h_nr_idle && + rq->nr_running); +} + +static int sched_idle_cpu(int cpu) +{ + return sched_idle_rq(cpu_rq(cpu)); +} + +static void +requeue_delayed_entity(struct sched_entity *se) +{ + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + /* + * se->sched_delayed should imply: se->on_rq == 1. + * Because a delayed entity is one that is still on + * the runqueue competing until elegibility. + */ + WARN_ON_ONCE(!se->sched_delayed); + WARN_ON_ONCE(!se->on_rq); + + if (sched_feat(DELAY_ZERO)) { + update_entity_lag(cfs_rq, se); + if (se->vlag > 0) { + 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_queued++; + } + } + + update_load_avg(cfs_rq, se, 0); + clear_delayed(se); } -#else -static inline void update_overutilized_status(struct rq *rq) { } -#endif /* * The enqueue_task method is called before nr_running is @@ -5109,6 +6878,14 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) { struct cfs_rq *cfs_rq; struct sched_entity *se = &p->se; + int h_nr_idle = task_has_idle_policy(p); + int h_nr_runnable = 1; + int task_new = !(flags & ENQUEUE_WAKEUP); + int rq_h_nr_queued = rq->cfs.h_nr_queued; + u64 slice = 0; + + if (task_is_throttled(p) && enqueue_throttled_task(p)) + return; /* * The code below (indirectly) updates schedutil which looks at @@ -5116,7 +6893,13 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) * Let's add the task's estimated utilization to the cfs_rq's * estimated utilization, before we update schedutil. */ - util_est_enqueue(&rq->cfs, p); + if (!p->se.sched_delayed || (flags & ENQUEUE_DELAYED)) + util_est_enqueue(&rq->cfs, p); + + if (flags & ENQUEUE_DELAYED) { + requeue_delayed_entity(se); + return; + } /* * If in_iowait is set, the code below may not trigger any cpufreq @@ -5126,429 +6909,310 @@ 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->on_rq) { + if (se->sched_delayed) + requeue_delayed_entity(se); break; + } cfs_rq = cfs_rq_of(se); - enqueue_entity(cfs_rq, se, flags); /* - * end evaluation on encountering a throttled cfs_rq - * - * note: in the case of encountering a throttled cfs_rq we will - * post the final h_nr_running increment below. + * Basically set the slice of group entries to the min_slice of + * their respective cfs_rq. This ensures the group can service + * its entities in the desired time-frame. */ - if (cfs_rq_throttled(cfs_rq)) - break; - cfs_rq->h_nr_running++; + if (slice) { + se->slice = slice; + se->custom_slice = 1; + } + enqueue_entity(cfs_rq, se, flags); + slice = cfs_rq_min_slice(cfs_rq); + + 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)) + h_nr_idle = 1; flags = ENQUEUE_WAKEUP; } for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - cfs_rq->h_nr_running++; - - if (cfs_rq_throttled(cfs_rq)) - break; update_load_avg(cfs_rq, se, UPDATE_TG); + se_update_runnable(se); update_cfs_group(se); - } - if (!se) { - add_nr_running(rq, 1); - /* - * Since new tasks are assigned an initial util_avg equal to - * half of the spare capacity of their CPU, tiny tasks have the - * ability to cross the overutilized threshold, which will - * result in the load balancer ruining all the task placement - * done by EAS. As a way to mitigate that effect, do not account - * for the first enqueue operation of new tasks during the - * overutilized flag detection. - * - * A better way of solving this problem would be to wait for - * the PELT signals of tasks to converge before taking them - * into account, but that is not straightforward to implement, - * and the following generally works well enough in practice. - */ - if (flags & ENQUEUE_WAKEUP) - update_overutilized_status(rq); + se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); + slice = cfs_rq_min_slice(cfs_rq); + + 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)) + h_nr_idle = 1; } + 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, 1); + + /* + * Since new tasks are assigned an initial util_avg equal to + * half of the spare capacity of their CPU, tiny tasks have the + * ability to cross the overutilized threshold, which will + * result in the load balancer ruining all the task placement + * done by EAS. As a way to mitigate that effect, do not account + * for the first enqueue operation of new tasks during the + * overutilized flag detection. + * + * A better way of solving this problem would be to wait for + * the PELT signals of tasks to converge before taking them + * into account, but that is not straightforward to implement, + * and the following generally works well enough in practice. + */ + if (!task_new) + check_update_overutilized_status(rq); + + assert_list_leaf_cfs_rq(rq); + hrtick_update(rq); } -static void set_next_buddy(struct sched_entity *se); - /* - * The dequeue_task method is called before nr_running is - * decreased. We remove the task from the rbtree and - * update the fair scheduling stats: + * Basically dequeue_task_fair(), except it can deal with dequeue_entity() + * failing half-way through and resume the dequeue later. + * + * Returns: + * -1 - dequeue delayed + * 0 - dequeue throttled + * 1 - dequeue complete */ -static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) +static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) { + bool was_sched_idle = sched_idle_rq(rq); + bool task_sleep = flags & DEQUEUE_SLEEP; + bool task_delayed = flags & DEQUEUE_DELAYED; + bool task_throttled = flags & DEQUEUE_THROTTLE; + struct task_struct *p = NULL; + int h_nr_idle = 0; + int h_nr_queued = 0; + int h_nr_runnable = 0; struct cfs_rq *cfs_rq; - struct sched_entity *se = &p->se; - int task_sleep = flags & DEQUEUE_SLEEP; + u64 slice = 0; + + if (entity_is_task(se)) { + p = task_of(se); + h_nr_queued = 1; + h_nr_idle = task_has_idle_policy(p); + if (task_sleep || task_delayed || !se->sched_delayed) + h_nr_runnable = 1; + } for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - dequeue_entity(cfs_rq, se, flags); - /* - * end evaluation on encountering a throttled cfs_rq - * - * note: in the case of encountering a throttled cfs_rq we will - * post the final h_nr_running decrement below. - */ - if (cfs_rq_throttled(cfs_rq)) + if (!dequeue_entity(cfs_rq, se, flags)) { + if (p && &p->se == se) + return -1; + + slice = cfs_rq_min_slice(cfs_rq); break; - cfs_rq->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)) + h_nr_idle = h_nr_queued; + + if (throttled_hierarchy(cfs_rq) && task_throttled) + record_throttle_clock(cfs_rq); /* Don't dequeue parent if it has other entities besides us */ if (cfs_rq->load.weight) { + slice = cfs_rq_min_slice(cfs_rq); + /* Avoid re-evaluating load for this entity: */ se = parent_entity(se); /* * Bias pick_next to pick a task from this cfs_rq, as * p is sleeping when it is within its sched_slice. */ - if (task_sleep && se && !throttled_hierarchy(cfs_rq)) + if (task_sleep && se) set_next_buddy(se); break; } flags |= DEQUEUE_SLEEP; + flags &= ~(DEQUEUE_DELAYED | DEQUEUE_SPECIAL); } for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - cfs_rq->h_nr_running--; - - if (cfs_rq_throttled(cfs_rq)) - break; update_load_avg(cfs_rq, se, UPDATE_TG); + se_update_runnable(se); update_cfs_group(se); - } - if (!se) - sub_nr_running(rq, 1); + se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); + slice = cfs_rq_min_slice(cfs_rq); - util_est_dequeue(&rq->cfs, p, task_sleep); - hrtick_update(rq); -} + cfs_rq->h_nr_runnable -= h_nr_runnable; + cfs_rq->h_nr_queued -= h_nr_queued; + cfs_rq->h_nr_idle -= h_nr_idle; -#ifdef CONFIG_SMP + if (cfs_rq_is_idle(cfs_rq)) + h_nr_idle = h_nr_queued; -/* Working cpumask for: load_balance, load_balance_newidle. */ -DEFINE_PER_CPU(cpumask_var_t, load_balance_mask); -DEFINE_PER_CPU(cpumask_var_t, select_idle_mask); + if (throttled_hierarchy(cfs_rq) && task_throttled) + record_throttle_clock(cfs_rq); + } -#ifdef CONFIG_NO_HZ_COMMON -/* - * per rq 'load' arrray crap; XXX kill this. - */ + sub_nr_running(rq, h_nr_queued); -/* - * The exact cpuload calculated at every tick would be: - * - * load' = (1 - 1/2^i) * load + (1/2^i) * cur_load - * - * If a CPU misses updates for n ticks (as it was idle) and update gets - * called on the n+1-th tick when CPU may be busy, then we have: - * - * load_n = (1 - 1/2^i)^n * load_0 - * load_n+1 = (1 - 1/2^i) * load_n + (1/2^i) * cur_load - * - * decay_load_missed() below does efficient calculation of - * - * load' = (1 - 1/2^i)^n * load - * - * Because x^(n+m) := x^n * x^m we can decompose any x^n in power-of-2 factors. - * This allows us to precompute the above in said factors, thereby allowing the - * reduction of an arbitrary n in O(log_2 n) steps. (See also - * fixed_power_int()) - * - * The calculation is approximated on a 128 point scale. - */ -#define DEGRADE_SHIFT 7 + /* balance early to pull high priority tasks */ + if (unlikely(!was_sched_idle && sched_idle_rq(rq))) + rq->next_balance = jiffies; -static const u8 degrade_zero_ticks[CPU_LOAD_IDX_MAX] = {0, 8, 32, 64, 128}; -static const u8 degrade_factor[CPU_LOAD_IDX_MAX][DEGRADE_SHIFT + 1] = { - { 0, 0, 0, 0, 0, 0, 0, 0 }, - { 64, 32, 8, 0, 0, 0, 0, 0 }, - { 96, 72, 40, 12, 1, 0, 0, 0 }, - { 112, 98, 75, 43, 15, 1, 0, 0 }, - { 120, 112, 98, 76, 45, 16, 2, 0 } -}; + if (p && task_delayed) { + WARN_ON_ONCE(!task_sleep); + WARN_ON_ONCE(p->on_rq != 1); + + /* Fix-up what dequeue_task_fair() skipped */ + hrtick_update(rq); + + /* + * Fix-up what block_task() skipped. + * + * Must be last, @p might not be valid after this. + */ + __block_task(rq, p); + } + + return 1; +} /* - * Update cpu_load for any missed ticks, due to tickless idle. The backlog - * would be when CPU is idle and so we just decay the old load without - * adding any new load. + * The dequeue_task method is called before nr_running is + * decreased. We remove the task from the rbtree and + * update the fair scheduling stats: */ -static unsigned long -decay_load_missed(unsigned long load, unsigned long missed_updates, int idx) +static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) { - int j = 0; + if (task_is_throttled(p)) { + dequeue_throttled_task(p, flags); + return true; + } - if (!missed_updates) - return load; + if (!p->se.sched_delayed) + util_est_dequeue(&rq->cfs, p); - if (missed_updates >= degrade_zero_ticks[idx]) - return 0; + util_est_update(&rq->cfs, p, flags & DEQUEUE_SLEEP); + if (dequeue_entities(rq, &p->se, flags) < 0) + return false; - if (idx == 1) - return load >> missed_updates; + /* + * Must not reference @p after dequeue_entities(DEQUEUE_DELAYED). + */ - while (missed_updates) { - if (missed_updates % 2) - load = (load * degrade_factor[idx][j]) >> DEGRADE_SHIFT; + hrtick_update(rq); + return true; +} - missed_updates >>= 1; - j++; - } - return load; +static inline unsigned int cfs_h_nr_delayed(struct rq *rq) +{ + return (rq->cfs.h_nr_queued - rq->cfs.h_nr_runnable); } +/* Working cpumask for: sched_balance_rq(), sched_balance_newidle(). */ +static DEFINE_PER_CPU(cpumask_var_t, load_balance_mask); +static DEFINE_PER_CPU(cpumask_var_t, select_rq_mask); +static DEFINE_PER_CPU(cpumask_var_t, should_we_balance_tmpmask); + +#ifdef CONFIG_NO_HZ_COMMON + static struct { cpumask_var_t idle_cpus_mask; atomic_t nr_cpus; int has_blocked; /* Idle CPUS has blocked load */ + int needs_update; /* Newly idle CPUs need their next_balance collated */ unsigned long next_balance; /* in jiffy units */ unsigned long next_blocked; /* Next update of blocked load in jiffies */ } nohz ____cacheline_aligned; #endif /* CONFIG_NO_HZ_COMMON */ -/** - * __cpu_load_update - update the rq->cpu_load[] statistics - * @this_rq: The rq to update statistics for - * @this_load: The current load - * @pending_updates: The number of missed updates - * - * Update rq->cpu_load[] statistics. This function is usually called every - * scheduler tick (TICK_NSEC). - * - * This function computes a decaying average: - * - * load[i]' = (1 - 1/2^i) * load[i] + (1/2^i) * load - * - * Because of NOHZ it might not get called on every tick which gives need for - * the @pending_updates argument. - * - * load[i]_n = (1 - 1/2^i) * load[i]_n-1 + (1/2^i) * load_n-1 - * = A * load[i]_n-1 + B ; A := (1 - 1/2^i), B := (1/2^i) * load - * = A * (A * load[i]_n-2 + B) + B - * = A * (A * (A * load[i]_n-3 + B) + B) + B - * = A^3 * load[i]_n-3 + (A^2 + A + 1) * B - * = A^n * load[i]_0 + (A^(n-1) + A^(n-2) + ... + 1) * B - * = A^n * load[i]_0 + ((1 - A^n) / (1 - A)) * B - * = (1 - 1/2^i)^n * (load[i]_0 - load) + load - * - * In the above we've assumed load_n := load, which is true for NOHZ_FULL as - * any change in load would have resulted in the tick being turned back on. - * - * For regular NOHZ, this reduces to: - * - * load[i]_n = (1 - 1/2^i)^n * load[i]_0 - * - * see decay_load_misses(). For NOHZ_FULL we get to subtract and add the extra - * term. - */ -static void cpu_load_update(struct rq *this_rq, unsigned long this_load, - unsigned long pending_updates) +static unsigned long cpu_load(struct rq *rq) { - unsigned long __maybe_unused tickless_load = this_rq->cpu_load[0]; - int i, scale; - - this_rq->nr_load_updates++; - - /* Update our load: */ - this_rq->cpu_load[0] = this_load; /* Fasttrack for idx 0 */ - for (i = 1, scale = 2; i < CPU_LOAD_IDX_MAX; i++, scale += scale) { - unsigned long old_load, new_load; - - /* scale is effectively 1 << i now, and >> i divides by scale */ - - old_load = this_rq->cpu_load[i]; -#ifdef CONFIG_NO_HZ_COMMON - old_load = decay_load_missed(old_load, pending_updates - 1, i); - if (tickless_load) { - old_load -= decay_load_missed(tickless_load, pending_updates - 1, i); - /* - * old_load can never be a negative value because a - * decayed tickless_load cannot be greater than the - * original tickless_load. - */ - old_load += tickless_load; - } -#endif - new_load = this_load; - /* - * Round up the averaging division if load is increasing. This - * prevents us from getting stuck on 9 if the load is 10, for - * example. - */ - if (new_load > old_load) - new_load += scale - 1; - - this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i; - } + return cfs_rq_load_avg(&rq->cfs); } -/* Used instead of source_load when we know the type == 0 */ -static unsigned long weighted_cpuload(struct rq *rq) -{ - return cfs_rq_runnable_load_avg(&rq->cfs); -} - -#ifdef CONFIG_NO_HZ_COMMON /* - * There is no sane way to deal with nohz on smp when using jiffies because the - * CPU doing the jiffies update might drift wrt the CPU doing the jiffy reading - * causing off-by-one errors in observed deltas; {0,2} instead of {1,1}. + * cpu_load_without - compute CPU load without any contributions from *p + * @cpu: the CPU which load is requested + * @p: the task which load should be discounted * - * Therefore we need to avoid the delta approach from the regular tick when - * possible since that would seriously skew the load calculation. This is why we - * use cpu_load_update_periodic() for CPUs out of nohz. However we'll rely on - * jiffies deltas for updates happening while in nohz mode (idle ticks, idle - * loop exit, nohz_idle_balance, nohz full exit...) + * The load of a CPU is defined by the load of tasks currently enqueued on that + * CPU as well as tasks which are currently sleeping after an execution on that + * CPU. * - * This means we might still be one tick off for nohz periods. - */ - -static void cpu_load_update_nohz(struct rq *this_rq, - unsigned long curr_jiffies, - unsigned long load) -{ - unsigned long pending_updates; - - pending_updates = curr_jiffies - this_rq->last_load_update_tick; - if (pending_updates) { - this_rq->last_load_update_tick = curr_jiffies; - /* - * In the regular NOHZ case, we were idle, this means load 0. - * In the NOHZ_FULL case, we were non-idle, we should consider - * its weighted load. - */ - cpu_load_update(this_rq, load, pending_updates); - } -} - -/* - * Called from nohz_idle_balance() to update the load ratings before doing the - * idle balance. - */ -static void cpu_load_update_idle(struct rq *this_rq) -{ - /* - * bail if there's load or we're actually up-to-date. - */ - if (weighted_cpuload(this_rq)) - return; - - cpu_load_update_nohz(this_rq, READ_ONCE(jiffies), 0); -} - -/* - * Record CPU load on nohz entry so we know the tickless load to account - * on nohz exit. cpu_load[0] happens then to be updated more frequently - * than other cpu_load[idx] but it should be fine as cpu_load readers - * shouldn't rely into synchronized cpu_load[*] updates. + * This method returns the load of the specified CPU by discounting the load of + * the specified task, whenever the task is currently contributing to the CPU + * load. */ -void cpu_load_update_nohz_start(void) +static unsigned long cpu_load_without(struct rq *rq, struct task_struct *p) { - struct rq *this_rq = this_rq(); - - /* - * This is all lockless but should be fine. If weighted_cpuload changes - * concurrently we'll exit nohz. And cpu_load write can race with - * cpu_load_update_idle() but both updater would be writing the same. - */ - this_rq->cpu_load[0] = weighted_cpuload(this_rq); -} + struct cfs_rq *cfs_rq; + unsigned int load; -/* - * Account the tickless load in the end of a nohz frame. - */ -void cpu_load_update_nohz_stop(void) -{ - unsigned long curr_jiffies = READ_ONCE(jiffies); - struct rq *this_rq = this_rq(); - unsigned long load; - struct rq_flags rf; + /* Task has no contribution or is new */ + if (cpu_of(rq) != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) + return cpu_load(rq); - if (curr_jiffies == this_rq->last_load_update_tick) - return; + cfs_rq = &rq->cfs; + load = READ_ONCE(cfs_rq->avg.load_avg); - load = weighted_cpuload(this_rq); - rq_lock(this_rq, &rf); - update_rq_clock(this_rq); - cpu_load_update_nohz(this_rq, curr_jiffies, load); - rq_unlock(this_rq, &rf); -} -#else /* !CONFIG_NO_HZ_COMMON */ -static inline void cpu_load_update_nohz(struct rq *this_rq, - unsigned long curr_jiffies, - unsigned long load) { } -#endif /* CONFIG_NO_HZ_COMMON */ + /* Discount task's util from CPU's util */ + lsub_positive(&load, task_h_load(p)); -static void cpu_load_update_periodic(struct rq *this_rq, unsigned long load) -{ -#ifdef CONFIG_NO_HZ_COMMON - /* See the mess around cpu_load_update_nohz(). */ - this_rq->last_load_update_tick = READ_ONCE(jiffies); -#endif - cpu_load_update(this_rq, load, 1); + return load; } -/* - * Called from scheduler_tick() - */ -void cpu_load_update_active(struct rq *this_rq) +static unsigned long cpu_runnable(struct rq *rq) { - unsigned long load = weighted_cpuload(this_rq); - - if (tick_nohz_tick_stopped()) - cpu_load_update_nohz(this_rq, READ_ONCE(jiffies), load); - else - cpu_load_update_periodic(this_rq, load); + return cfs_rq_runnable_avg(&rq->cfs); } -/* - * Return a low guess at the load of a migration-source CPU weighted - * according to the scheduling class and "nice" value. - * - * We want to under-estimate the load of migration sources, to - * balance conservatively. - */ -static unsigned long source_load(int cpu, int type) +static unsigned long cpu_runnable_without(struct rq *rq, struct task_struct *p) { - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(rq); - - if (type == 0 || !sched_feat(LB_BIAS)) - return total; + struct cfs_rq *cfs_rq; + unsigned int runnable; - return min(rq->cpu_load[type-1], total); -} + /* Task has no contribution or is new */ + if (cpu_of(rq) != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) + return cpu_runnable(rq); -/* - * Return a high guess at the load of a migration-target CPU weighted - * according to the scheduling class and "nice" value. - */ -static unsigned long target_load(int cpu, int type) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long total = weighted_cpuload(rq); + cfs_rq = &rq->cfs; + runnable = READ_ONCE(cfs_rq->avg.runnable_avg); - if (type == 0 || !sched_feat(LB_BIAS)) - return total; + /* Discount task's runnable from CPU's runnable */ + lsub_positive(&runnable, p->se.avg.runnable_avg); - return max(rq->cpu_load[type-1], total); + return runnable; } static unsigned long capacity_of(int cpu) @@ -5556,23 +7220,6 @@ static unsigned long capacity_of(int cpu) return cpu_rq(cpu)->cpu_capacity; } -static unsigned long capacity_orig_of(int cpu) -{ - return cpu_rq(cpu)->cpu_capacity_orig; -} - -static unsigned long cpu_avg_load_per_task(int cpu) -{ - struct rq *rq = cpu_rq(cpu); - unsigned long nr_running = READ_ONCE(rq->cfs.h_nr_running); - unsigned long load_avg = weighted_cpuload(rq); - - if (nr_running) - return load_avg / nr_running; - - return 0; -} - static void record_wakee(struct task_struct *p) { /* @@ -5611,7 +7258,7 @@ static int wake_wide(struct task_struct *p) { unsigned int master = current->wakee_flips; unsigned int slave = p->wakee_flips; - int factor = this_cpu_read(sd_llc_size); + int factor = __this_cpu_read(sd_llc_size); if (master < slave) swap(master, slave); @@ -5650,8 +7297,15 @@ wake_affine_idle(int this_cpu, int prev_cpu, int sync) if (available_idle_cpu(this_cpu) && cpus_share_cache(this_cpu, prev_cpu)) return available_idle_cpu(prev_cpu) ? prev_cpu : this_cpu; - if (sync && cpu_rq(this_cpu)->nr_running == 1) - return this_cpu; + if (sync) { + struct rq *rq = cpu_rq(this_cpu); + + if ((rq->nr_running - cfs_h_nr_delayed(rq)) == 1) + return this_cpu; + } + + if (available_idle_cpu(prev_cpu)) + return prev_cpu; return nr_cpumask_bits; } @@ -5663,7 +7317,7 @@ wake_affine_weight(struct sched_domain *sd, struct task_struct *p, s64 this_eff_load, prev_eff_load; unsigned long task_load; - this_eff_load = target_load(this_cpu, sd->wake_idx); + this_eff_load = cpu_load(cpu_rq(this_cpu)); if (sync) { unsigned long current_load = task_h_load(current); @@ -5681,7 +7335,7 @@ wake_affine_weight(struct sched_domain *sd, struct task_struct *p, this_eff_load *= 100; this_eff_load *= capacity_of(prev_cpu); - prev_eff_load = source_load(prev_cpu, sd->wake_idx); + prev_eff_load = cpu_load(cpu_rq(prev_cpu)); prev_eff_load -= task_load; if (sched_feat(WA_BIAS)) prev_eff_load *= 100 + (sd->imbalance_pct - 100) / 2; @@ -5710,173 +7364,23 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, if (sched_feat(WA_WEIGHT) && target == nr_cpumask_bits) target = wake_affine_weight(sd, p, this_cpu, prev_cpu, sync); - schedstat_inc(p->se.statistics.nr_wakeups_affine_attempts); - if (target == nr_cpumask_bits) + schedstat_inc(p->stats.nr_wakeups_affine_attempts); + if (target != this_cpu) return prev_cpu; schedstat_inc(sd->ttwu_move_affine); - schedstat_inc(p->se.statistics.nr_wakeups_affine); + schedstat_inc(p->stats.nr_wakeups_affine); return target; } -static unsigned long cpu_util_without(int cpu, struct task_struct *p); - -static unsigned long capacity_spare_without(int cpu, struct task_struct *p) -{ - return max_t(long, capacity_of(cpu) - cpu_util_without(cpu, p), 0); -} - -/* - * find_idlest_group finds and returns the least busy CPU group within the - * domain. - * - * Assumes p is allowed on at least one CPU in sd. - */ static struct sched_group * -find_idlest_group(struct sched_domain *sd, struct task_struct *p, - int this_cpu, int sd_flag) -{ - struct sched_group *idlest = NULL, *group = sd->groups; - struct sched_group *most_spare_sg = NULL; - unsigned long min_runnable_load = ULONG_MAX; - unsigned long this_runnable_load = ULONG_MAX; - unsigned long min_avg_load = ULONG_MAX, this_avg_load = ULONG_MAX; - unsigned long most_spare = 0, this_spare = 0; - int load_idx = sd->forkexec_idx; - int imbalance_scale = 100 + (sd->imbalance_pct-100)/2; - unsigned long imbalance = scale_load_down(NICE_0_LOAD) * - (sd->imbalance_pct-100) / 100; - - if (sd_flag & SD_BALANCE_WAKE) - load_idx = sd->wake_idx; - - do { - unsigned long load, avg_load, runnable_load; - unsigned long spare_cap, max_spare_cap; - int local_group; - int i; - - /* Skip over this group if it has no CPUs allowed */ - if (!cpumask_intersects(sched_group_span(group), - &p->cpus_allowed)) - continue; - - local_group = cpumask_test_cpu(this_cpu, - sched_group_span(group)); - - /* - * Tally up the load of all CPUs in the group and find - * the group containing the CPU with most spare capacity. - */ - avg_load = 0; - runnable_load = 0; - max_spare_cap = 0; - - for_each_cpu(i, sched_group_span(group)) { - /* Bias balancing toward CPUs of our domain */ - if (local_group) - load = source_load(i, load_idx); - else - load = target_load(i, load_idx); - - runnable_load += load; - - avg_load += cfs_rq_load_avg(&cpu_rq(i)->cfs); - - spare_cap = capacity_spare_without(i, p); - - if (spare_cap > max_spare_cap) - max_spare_cap = spare_cap; - } - - /* Adjust by relative CPU capacity of the group */ - avg_load = (avg_load * SCHED_CAPACITY_SCALE) / - group->sgc->capacity; - runnable_load = (runnable_load * SCHED_CAPACITY_SCALE) / - group->sgc->capacity; - - if (local_group) { - this_runnable_load = runnable_load; - this_avg_load = avg_load; - this_spare = max_spare_cap; - } else { - if (min_runnable_load > (runnable_load + imbalance)) { - /* - * The runnable load is significantly smaller - * so we can pick this new CPU: - */ - min_runnable_load = runnable_load; - min_avg_load = avg_load; - idlest = group; - } else if ((runnable_load < (min_runnable_load + imbalance)) && - (100*min_avg_load > imbalance_scale*avg_load)) { - /* - * The runnable loads are close so take the - * blocked load into account through avg_load: - */ - min_avg_load = avg_load; - idlest = group; - } - - if (most_spare < max_spare_cap) { - most_spare = max_spare_cap; - most_spare_sg = group; - } - } - } while (group = group->next, group != sd->groups); - - /* - * The cross-over point between using spare capacity or least load - * is too conservative for high utilization tasks on partially - * utilized systems if we require spare_capacity > task_util(p), - * so we allow for some task stuffing by using - * spare_capacity > task_util(p)/2. - * - * Spare capacity can't be used for fork because the utilization has - * not been set yet, we must first select a rq to compute the initial - * utilization. - */ - if (sd_flag & SD_BALANCE_FORK) - goto skip_spare; - - if (this_spare > task_util(p) / 2 && - imbalance_scale*this_spare > 100*most_spare) - return NULL; - - if (most_spare > task_util(p) / 2) - return most_spare_sg; - -skip_spare: - if (!idlest) - return NULL; - - /* - * When comparing groups across NUMA domains, it's possible for the - * local domain to be very lightly loaded relative to the remote - * domains but "imbalance" skews the comparison making remote CPUs - * look much more favourable. When considering cross-domain, add - * imbalance to the runnable load on the remote node and consider - * staying local. - */ - if ((sd->flags & SD_NUMA) && - min_runnable_load + imbalance >= this_runnable_load) - return NULL; - - if (min_runnable_load > (this_runnable_load + imbalance)) - return NULL; - - if ((this_runnable_load < (min_runnable_load + imbalance)) && - (100*this_avg_load < imbalance_scale*min_avg_load)) - return NULL; - - return idlest; -} +sched_balance_find_dst_group(struct sched_domain *sd, struct task_struct *p, int this_cpu); /* - * find_idlest_group_cpu - find the idlest CPU among the CPUs in the group. + * sched_balance_find_dst_group_cpu - find the idlest CPU among the CPUs in the group. */ static int -find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) +sched_balance_find_dst_group_cpu(struct sched_group *group, struct task_struct *p, int this_cpu) { unsigned long load, min_load = ULONG_MAX; unsigned int min_exit_latency = UINT_MAX; @@ -5890,9 +7394,16 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this return cpumask_first(sched_group_span(group)); /* Traverse only the allowed CPUs */ - for_each_cpu_and(i, sched_group_span(group), &p->cpus_allowed) { + for_each_cpu_and(i, sched_group_span(group), p->cpus_ptr) { + struct rq *rq = cpu_rq(i); + + if (!sched_core_cookie_match(rq, p)) + continue; + + if (sched_idle_cpu(i)) + return i; + if (available_idle_cpu(i)) { - struct rq *rq = cpu_rq(i); struct cpuidle_state *idle = idle_get_state(rq); if (idle && idle->exit_latency < min_exit_latency) { /* @@ -5914,7 +7425,7 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this shallowest_idle_cpu = i; } } else if (shallowest_idle_cpu == -1) { - load = weighted_cpuload(cpu_rq(i)); + load = cpu_load(cpu_rq(i)); if (load < min_load) { min_load = load; least_loaded_cpu = i; @@ -5925,16 +7436,16 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this return shallowest_idle_cpu != -1 ? shallowest_idle_cpu : least_loaded_cpu; } -static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p, +static inline int sched_balance_find_dst_cpu(struct sched_domain *sd, struct task_struct *p, int cpu, int prev_cpu, int sd_flag) { int new_cpu = cpu; - if (!cpumask_intersects(sched_domain_span(sd), &p->cpus_allowed)) + if (!cpumask_intersects(sched_domain_span(sd), p->cpus_ptr)) return prev_cpu; /* - * We need task's util for capacity_spare_without, sync it up to + * We need task's util for cpu_util_without, sync it up to * prev_cpu's last_update_time. */ if (!(sd_flag & SD_BALANCE_FORK)) @@ -5950,13 +7461,13 @@ static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p continue; } - group = find_idlest_group(sd, p, cpu, sd_flag); + group = sched_balance_find_dst_group(sd, p, cpu); if (!group) { sd = sd->child; continue; } - new_cpu = find_idlest_group_cpu(group, p, cpu); + new_cpu = sched_balance_find_dst_group_cpu(group, p, cpu); if (new_cpu == cpu) { /* Now try balancing at a lower domain level of 'cpu': */ sd = sd->child; @@ -5978,8 +7489,18 @@ static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p return new_cpu; } +static inline int __select_idle_cpu(int cpu, struct task_struct *p) +{ + if ((available_idle_cpu(cpu) || sched_idle_cpu(cpu)) && + sched_cpu_cookie_match(cpu_rq(cpu), p)) + return cpu; + + return -1; +} + #ifdef CONFIG_SCHED_SMT DEFINE_STATIC_KEY_FALSE(sched_smt_present); +EXPORT_SYMBOL_GPL(sched_smt_present); static inline void set_idle_cores(int cpu, int val) { @@ -5990,7 +7511,7 @@ static inline void set_idle_cores(int cpu, int val) WRITE_ONCE(sds->has_idle_cores, val); } -static inline bool test_idle_cores(int cpu, bool def) +static inline bool test_idle_cores(int cpu) { struct sched_domain_shared *sds; @@ -5998,7 +7519,7 @@ static inline bool test_idle_cores(int cpu, bool def) if (sds) return READ_ONCE(sds->has_idle_cores); - return def; + return false; } /* @@ -6014,7 +7535,7 @@ void __update_idle_core(struct rq *rq) int cpu; rcu_read_lock(); - if (test_idle_cores(core, true)) + if (test_idle_cores(core)) goto unlock; for_each_cpu(cpu, cpu_smt_mask(core)) { @@ -6035,37 +7556,31 @@ unlock: * there are no idle cores left in the system; tracked through * sd_llc->shared->has_idle_cores and enabled through update_idle_core() above. */ -static int select_idle_core(struct task_struct *p, struct sched_domain *sd, int target) +static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpus, int *idle_cpu) { - struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask); - int core, cpu; - - if (!static_branch_likely(&sched_smt_present)) - return -1; - - if (!test_idle_cores(target, false)) - return -1; - - cpumask_and(cpus, sched_domain_span(sd), &p->cpus_allowed); - - for_each_cpu_wrap(core, cpus, target) { - bool idle = true; + bool idle = true; + int cpu; - for_each_cpu(cpu, cpu_smt_mask(core)) { - cpumask_clear_cpu(cpu, cpus); - if (!available_idle_cpu(cpu)) - idle = false; + for_each_cpu(cpu, cpu_smt_mask(core)) { + if (!available_idle_cpu(cpu)) { + idle = false; + if (*idle_cpu == -1) { + if (sched_idle_cpu(cpu) && cpumask_test_cpu(cpu, cpus)) { + *idle_cpu = cpu; + break; + } + continue; + } + break; } - - if (idle) - return core; + if (*idle_cpu == -1 && cpumask_test_cpu(cpu, cpus)) + *idle_cpu = cpu; } - /* - * Failed to find an idle core; stop looking for one. - */ - set_idle_cores(target, 0); + if (idle) + return core; + cpumask_andnot(cpus, cpus, cpu_smt_mask(core)); return -1; } @@ -6076,24 +7591,36 @@ static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int t { int cpu; - if (!static_branch_likely(&sched_smt_present)) - return -1; - - for_each_cpu(cpu, cpu_smt_mask(target)) { - if (!cpumask_test_cpu(cpu, &p->cpus_allowed)) + for_each_cpu_and(cpu, cpu_smt_mask(target), p->cpus_ptr) { + if (cpu == target) + continue; + /* + * Check if the CPU is in the LLC scheduling domain of @target. + * Due to isolcpus, there is no guarantee that all the siblings are in the domain. + */ + if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) continue; - if (available_idle_cpu(cpu)) + if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) return cpu; } return -1; } -#else /* CONFIG_SCHED_SMT */ +#else /* !CONFIG_SCHED_SMT: */ -static inline int select_idle_core(struct task_struct *p, struct sched_domain *sd, int target) +static inline void set_idle_cores(int cpu, int val) { - return -1; +} + +static inline bool test_idle_cores(int cpu) +{ + return false; +} + +static inline int select_idle_core(struct task_struct *p, int core, struct cpumask *cpus, int *idle_cpu) +{ + return __select_idle_cpu(core, p); } static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) @@ -6101,60 +7628,143 @@ static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd return -1; } -#endif /* CONFIG_SCHED_SMT */ +#endif /* !CONFIG_SCHED_SMT */ /* * Scan the LLC domain for idle CPUs; this is dynamically regulated by * comparing the average scan cost (tracked in sd->avg_scan_cost) against the * average idle time for this rq (as found in rq->avg_idle). */ -static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target) +static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool has_idle_core, int target) { - struct sched_domain *this_sd; - u64 avg_cost, avg_idle; - u64 time, cost; - s64 delta; - int cpu, nr = INT_MAX; + struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask); + int i, cpu, idle_cpu = -1, nr = INT_MAX; + struct sched_domain_shared *sd_share; - this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc)); - if (!this_sd) - return -1; + cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); - /* - * Due to large variance we need a large fuzz factor; hackbench in - * particularly is sensitive here. - */ - avg_idle = this_rq()->avg_idle / 512; - avg_cost = this_sd->avg_scan_cost + 1; + if (sched_feat(SIS_UTIL)) { + sd_share = rcu_dereference(per_cpu(sd_llc_shared, target)); + if (sd_share) { + /* because !--nr is the condition to stop scan */ + nr = READ_ONCE(sd_share->nr_idle_scan) + 1; + /* overloaded LLC is unlikely to have idle cpu/core */ + if (nr == 1) + return -1; + } + } - if (sched_feat(SIS_AVG_CPU) && avg_idle < avg_cost) - return -1; + if (static_branch_unlikely(&sched_cluster_active)) { + struct sched_group *sg = sd->groups; + + if (sg->flags & SD_CLUSTER) { + for_each_cpu_wrap(cpu, sched_group_span(sg), target + 1) { + if (!cpumask_test_cpu(cpu, cpus)) + continue; + + if (has_idle_core) { + i = select_idle_core(p, cpu, cpus, &idle_cpu); + if ((unsigned int)i < nr_cpumask_bits) + return i; + } else { + if (--nr <= 0) + return -1; + idle_cpu = __select_idle_cpu(cpu, p); + if ((unsigned int)idle_cpu < nr_cpumask_bits) + return idle_cpu; + } + } + cpumask_andnot(cpus, cpus, sched_group_span(sg)); + } + } - if (sched_feat(SIS_PROP)) { - u64 span_avg = sd->span_weight * avg_idle; - if (span_avg > 4*avg_cost) - nr = div_u64(span_avg, avg_cost); - else - nr = 4; + for_each_cpu_wrap(cpu, cpus, target + 1) { + if (has_idle_core) { + i = select_idle_core(p, cpu, cpus, &idle_cpu); + if ((unsigned int)i < nr_cpumask_bits) + return i; + + } else { + if (--nr <= 0) + return -1; + idle_cpu = __select_idle_cpu(cpu, p); + if ((unsigned int)idle_cpu < nr_cpumask_bits) + break; + } } - time = local_clock(); + if (has_idle_core) + set_idle_cores(target, false); - for_each_cpu_wrap(cpu, sched_domain_span(sd), target) { - if (!--nr) - return -1; - if (!cpumask_test_cpu(cpu, &p->cpus_allowed)) + return idle_cpu; +} + +/* + * Scan the asym_capacity domain for idle CPUs; pick the first idle one on which + * the task fits. If no CPU is big enough, but there are idle ones, try to + * maximize capacity. + */ +static int +select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target) +{ + unsigned long task_util, util_min, util_max, best_cap = 0; + int fits, best_fits = 0; + int cpu, best_cpu = -1; + struct cpumask *cpus; + + cpus = this_cpu_cpumask_var_ptr(select_rq_mask); + cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); + + task_util = task_util_est(p); + util_min = uclamp_eff_value(p, UCLAMP_MIN); + util_max = uclamp_eff_value(p, UCLAMP_MAX); + + for_each_cpu_wrap(cpu, cpus, target) { + unsigned long cpu_cap = capacity_of(cpu); + + if (!available_idle_cpu(cpu) && !sched_idle_cpu(cpu)) continue; - if (available_idle_cpu(cpu)) - break; + + fits = util_fits_cpu(task_util, util_min, util_max, cpu); + + /* This CPU fits with all requirements */ + if (fits > 0) + return cpu; + /* + * Only the min performance hint (i.e. uclamp_min) doesn't fit. + * Look for the CPU with best capacity. + */ + else if (fits < 0) + cpu_cap = get_actual_cpu_capacity(cpu); + + /* + * First, select CPU which fits better (-1 being better than 0). + * Then, select the one with best capacity at same level. + */ + if ((fits < best_fits) || + ((fits == best_fits) && (cpu_cap > best_cap))) { + best_cap = cpu_cap; + best_cpu = cpu; + best_fits = fits; + } } - time = local_clock() - time; - cost = this_sd->avg_scan_cost; - delta = (s64)(time - cost) / 8; - this_sd->avg_scan_cost += delta; + return best_cpu; +} - return cpu; +static inline bool asym_fits_cpu(unsigned long util, + unsigned long util_min, + unsigned long util_max, + int cpu) +{ + if (sched_asym_cpucap_active()) + /* + * Return true only if the cpu fully fits the task requirements + * which include the utilization and the performance hints. + */ + return (util_fits_cpu(util, util_min, util_max, cpu) > 0); + + return true; } /* @@ -6162,102 +7772,245 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t */ static int select_idle_sibling(struct task_struct *p, int prev, int target) { + bool has_idle_core = false; struct sched_domain *sd; - int i, recent_used_cpu; + unsigned long task_util, util_min, util_max; + int i, recent_used_cpu, prev_aff = -1; + + /* + * On asymmetric system, update task utilization because we will check + * that the task fits with CPU's capacity. + */ + if (sched_asym_cpucap_active()) { + sync_entity_load_avg(&p->se); + task_util = task_util_est(p); + util_min = uclamp_eff_value(p, UCLAMP_MIN); + util_max = uclamp_eff_value(p, UCLAMP_MAX); + } - if (available_idle_cpu(target)) + /* + * per-cpu select_rq_mask usage + */ + lockdep_assert_irqs_disabled(); + + if ((available_idle_cpu(target) || sched_idle_cpu(target)) && + asym_fits_cpu(task_util, util_min, util_max, target)) return target; /* * If the previous CPU is cache affine and idle, don't be stupid: */ - if (prev != target && cpus_share_cache(prev, target) && available_idle_cpu(prev)) + if (prev != target && cpus_share_cache(prev, target) && + (available_idle_cpu(prev) || sched_idle_cpu(prev)) && + asym_fits_cpu(task_util, util_min, util_max, prev)) { + + if (!static_branch_unlikely(&sched_cluster_active) || + cpus_share_resources(prev, target)) + return prev; + + prev_aff = prev; + } + + /* + * Allow a per-cpu kthread to stack with the wakee if the + * kworker thread and the tasks previous CPUs are the same. + * The assumption is that the wakee queued work for the + * per-cpu kthread that is now complete and the wakeup is + * essentially a sync wakeup. An obvious example of this + * pattern is IO completions. + */ + if (is_per_cpu_kthread(current) && + in_task() && + prev == smp_processor_id() && + this_rq()->nr_running <= 1 && + asym_fits_cpu(task_util, util_min, util_max, prev)) { return prev; + } /* Check a recently used CPU as a potential idle candidate: */ recent_used_cpu = p->recent_used_cpu; + p->recent_used_cpu = prev; if (recent_used_cpu != prev && recent_used_cpu != target && cpus_share_cache(recent_used_cpu, target) && - available_idle_cpu(recent_used_cpu) && - cpumask_test_cpu(p->recent_used_cpu, &p->cpus_allowed)) { + (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) && + cpumask_test_cpu(recent_used_cpu, p->cpus_ptr) && + asym_fits_cpu(task_util, util_min, util_max, recent_used_cpu)) { + + if (!static_branch_unlikely(&sched_cluster_active) || + cpus_share_resources(recent_used_cpu, target)) + return recent_used_cpu; + + } else { + recent_used_cpu = -1; + } + + /* + * For asymmetric CPU capacity systems, our domain of interest is + * sd_asym_cpucapacity rather than sd_llc. + */ + if (sched_asym_cpucap_active()) { + sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target)); /* - * Replace recent_used_cpu with prev as it is a potential - * candidate for the next wake: + * On an asymmetric CPU capacity system where an exclusive + * cpuset defines a symmetric island (i.e. one unique + * capacity_orig value through the cpuset), the key will be set + * but the CPUs within that cpuset will not have a domain with + * SD_ASYM_CPUCAPACITY. These should follow the usual symmetric + * capacity path. */ - p->recent_used_cpu = prev; - return recent_used_cpu; + if (sd) { + i = select_idle_capacity(p, sd, target); + return ((unsigned)i < nr_cpumask_bits) ? i : target; + } } sd = rcu_dereference(per_cpu(sd_llc, target)); if (!sd) return target; - i = select_idle_core(p, sd, target); - if ((unsigned)i < nr_cpumask_bits) - return i; + if (sched_smt_active()) { + has_idle_core = test_idle_cores(target); - i = select_idle_cpu(p, sd, target); - if ((unsigned)i < nr_cpumask_bits) - return i; + if (!has_idle_core && cpus_share_cache(prev, target)) { + i = select_idle_smt(p, sd, prev); + if ((unsigned int)i < nr_cpumask_bits) + return i; + } + } - i = select_idle_smt(p, sd, target); + i = select_idle_cpu(p, sd, has_idle_core, target); if ((unsigned)i < nr_cpumask_bits) return i; + /* + * For cluster machines which have lower sharing cache like L2 or + * LLC Tag, we tend to find an idle CPU in the target's cluster + * first. But prev_cpu or recent_used_cpu may also be a good candidate, + * use them if possible when no idle CPU found in select_idle_cpu(). + */ + if ((unsigned int)prev_aff < nr_cpumask_bits) + return prev_aff; + if ((unsigned int)recent_used_cpu < nr_cpumask_bits) + return recent_used_cpu; + return target; } /** - * Amount of capacity of a CPU that is (estimated to be) used by CFS tasks - * @cpu: the CPU to get the utilization of - * - * The unit of the return value must be the one of capacity so we can compare - * the utilization with the capacity of the CPU that is available for CFS task - * (ie cpu_capacity). - * - * cfs_rq.avg.util_avg is the sum of running time of runnable tasks plus the - * recent utilization of currently non-runnable tasks on a CPU. It represents - * the amount of utilization of a CPU in the range [0..capacity_orig] where - * capacity_orig is the cpu_capacity available at the highest frequency - * (arch_scale_freq_capacity()). - * The utilization of a CPU converges towards a sum equal to or less than the - * current capacity (capacity_curr <= capacity_orig) of the CPU because it is - * the running time on this CPU scaled by capacity_curr. - * - * The estimated utilization of a CPU is defined to be the maximum between its - * cfs_rq.avg.util_avg and the sum of the estimated utilization of the tasks - * currently RUNNABLE on that CPU. - * This allows to properly represent the expected utilization of a CPU which - * has just got a big task running since a long sleep period. At the same time - * however it preserves the benefits of the "blocked utilization" in - * describing the potential for other tasks waking up on the same CPU. - * - * Nevertheless, cfs_rq.avg.util_avg can be higher than capacity_curr or even - * higher than capacity_orig because of unfortunate rounding in - * cfs.avg.util_avg or just after migrating tasks and new task wakeups until - * the average stabilizes with the new running time. We need to check that the - * utilization stays within the range of [0..capacity_orig] and cap it if - * necessary. Without utilization capping, a group could be seen as overloaded - * (CPU0 utilization at 121% + CPU1 utilization at 80%) whereas CPU1 has 20% of - * available capacity. We allow utilization to overshoot capacity_curr (but not - * capacity_orig) as it useful for predicting the capacity required after task - * migrations (scheduler-driven DVFS). - * - * Return: the (estimated) utilization for the specified CPU - */ -static inline unsigned long cpu_util(int cpu) + * cpu_util() - Estimates the amount of CPU capacity used by CFS tasks. + * @cpu: the CPU to get the utilization for + * @p: task for which the CPU utilization should be predicted or NULL + * @dst_cpu: CPU @p migrates to, -1 if @p moves from @cpu or @p == NULL + * @boost: 1 to enable boosting, otherwise 0 + * + * The unit of the return value must be the same as the one of CPU capacity + * so that CPU utilization can be compared with CPU capacity. + * + * CPU utilization is the sum of running time of runnable tasks plus the + * recent utilization of currently non-runnable tasks on that CPU. + * It represents the amount of CPU capacity currently used by CFS tasks in + * the range [0..max CPU capacity] with max CPU capacity being the CPU + * capacity at f_max. + * + * The estimated CPU utilization is defined as the maximum between CPU + * utilization and sum of the estimated utilization of the currently + * runnable tasks on that CPU. It preserves a utilization "snapshot" of + * previously-executed tasks, which helps better deduce how busy a CPU will + * be when a long-sleeping task wakes up. The contribution to CPU utilization + * of such a task would be significantly decayed at this point of time. + * + * Boosted CPU utilization is defined as max(CPU runnable, CPU utilization). + * CPU contention for CFS tasks can be detected by CPU runnable > CPU + * utilization. Boosting is implemented in cpu_util() so that internal + * users (e.g. EAS) can use it next to external users (e.g. schedutil), + * latter via cpu_util_cfs_boost(). + * + * CPU utilization can be higher than the current CPU capacity + * (f_curr/f_max * max CPU capacity) or even the max CPU capacity because + * of rounding errors as well as task migrations or wakeups of new tasks. + * CPU utilization has to be capped to fit into the [0..max CPU capacity] + * range. Otherwise a group of CPUs (CPU0 util = 121% + CPU1 util = 80%) + * could be seen as over-utilized even though CPU1 has 20% of spare CPU + * capacity. CPU utilization is allowed to overshoot current CPU capacity + * though since this is useful for predicting the CPU capacity required + * after task migrations (scheduler-driven DVFS). + * + * Return: (Boosted) (estimated) utilization for the specified CPU. + */ +static unsigned long +cpu_util(int cpu, struct task_struct *p, int dst_cpu, int boost) { - struct cfs_rq *cfs_rq; - unsigned int util; + struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs; + unsigned long util = READ_ONCE(cfs_rq->avg.util_avg); + unsigned long runnable; - cfs_rq = &cpu_rq(cpu)->cfs; - util = READ_ONCE(cfs_rq->avg.util_avg); + if (boost) { + runnable = READ_ONCE(cfs_rq->avg.runnable_avg); + util = max(util, runnable); + } - if (sched_feat(UTIL_EST)) - util = max(util, READ_ONCE(cfs_rq->avg.util_est.enqueued)); + /* + * If @dst_cpu is -1 or @p migrates from @cpu to @dst_cpu remove its + * contribution. If @p migrates from another CPU to @cpu add its + * contribution. In all the other cases @cpu is not impacted by the + * migration so its util_avg is already correct. + */ + if (p && task_cpu(p) == cpu && dst_cpu != cpu) + lsub_positive(&util, task_util(p)); + else if (p && task_cpu(p) != cpu && dst_cpu == cpu) + util += task_util(p); - return min_t(unsigned long, util, capacity_orig_of(cpu)); + if (sched_feat(UTIL_EST)) { + unsigned long util_est; + + util_est = READ_ONCE(cfs_rq->avg.util_est); + + /* + * During wake-up @p isn't enqueued yet and doesn't contribute + * to any cpu_rq(cpu)->cfs.avg.util_est. + * If @dst_cpu == @cpu add it to "simulate" cpu_util after @p + * has been enqueued. + * + * During exec (@dst_cpu = -1) @p is enqueued and does + * contribute to cpu_rq(cpu)->cfs.util_est. + * Remove it to "simulate" cpu_util without @p's contribution. + * + * Despite the task_on_rq_queued(@p) check there is still a + * small window for a possible race when an exec + * select_task_rq_fair() races with LB's detach_task(). + * + * detach_task() + * deactivate_task() + * p->on_rq = TASK_ON_RQ_MIGRATING; + * -------------------------------- A + * dequeue_task() \ + * dequeue_task_fair() + Race Time + * util_est_dequeue() / + * -------------------------------- B + * + * The additional check "current == p" is required to further + * reduce the race window. + */ + if (dst_cpu == cpu) + util_est += _task_util_est(p); + else if (p && unlikely(task_on_rq_queued(p) || current == p)) + lsub_positive(&util_est, _task_util_est(p)); + + util = max(util, util_est); + } + + return min(util, arch_scale_cpu_capacity(cpu)); +} + +unsigned long cpu_util_cfs(int cpu) +{ + return cpu_util(cpu, NULL, -1, 0); +} + +unsigned long cpu_util_cfs_boost(int cpu) +{ + return cpu_util(cpu, NULL, -1, 1); } /* @@ -6275,179 +8028,253 @@ static inline unsigned long cpu_util(int cpu) */ static unsigned long cpu_util_without(int cpu, struct task_struct *p) { - struct cfs_rq *cfs_rq; - unsigned int util; - /* Task has no contribution or is new */ if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) - return cpu_util(cpu); + p = NULL; - cfs_rq = &cpu_rq(cpu)->cfs; - util = READ_ONCE(cfs_rq->avg.util_avg); + return cpu_util(cpu, p, -1, 0); +} - /* Discount task's util from CPU's util */ - lsub_positive(&util, task_util(p)); +/* + * This function computes an effective utilization for the given CPU, to be + * used for frequency selection given the linear relation: f = u * f_max. + * + * The scheduler tracks the following metrics: + * + * cpu_util_{cfs,rt,dl,irq}() + * cpu_bw_dl() + * + * Where the cfs,rt and dl util numbers are tracked with the same metric and + * synchronized windows and are thus directly comparable. + * + * The cfs,rt,dl utilization are the running times measured with rq->clock_task + * which excludes things like IRQ and steal-time. These latter are then accrued + * in the IRQ utilization. + * + * The DL bandwidth number OTOH is not a measured metric but a value computed + * based on the task model parameters and gives the minimal utilization + * required to meet deadlines. + */ +unsigned long effective_cpu_util(int cpu, unsigned long util_cfs, + unsigned long *min, + unsigned long *max) +{ + unsigned long util, irq, scale; + struct rq *rq = cpu_rq(cpu); + + scale = arch_scale_cpu_capacity(cpu); /* - * Covered cases: - * - * a) if *p is the only task sleeping on this CPU, then: - * cpu_util (== task_util) > util_est (== 0) - * and thus we return: - * cpu_util_without = (cpu_util - task_util) = 0 - * - * b) if other tasks are SLEEPING on this CPU, which is now exiting - * IDLE, then: - * cpu_util >= task_util - * cpu_util > util_est (== 0) - * and thus we discount *p's blocked utilization to return: - * cpu_util_without = (cpu_util - task_util) >= 0 - * - * c) if other tasks are RUNNABLE on that CPU and - * util_est > cpu_util - * then we use util_est since it returns a more restrictive - * estimation of the spare capacity on that CPU, by just - * considering the expected utilization of tasks already - * runnable on that CPU. - * - * Cases a) and b) are covered by the above code, while case c) is - * covered by the following code when estimated utilization is - * enabled. + * Early check to see if IRQ/steal time saturates the CPU, can be + * because of inaccuracies in how we track these -- see + * update_irq_load_avg(). */ - if (sched_feat(UTIL_EST)) { - unsigned int estimated = - READ_ONCE(cfs_rq->avg.util_est.enqueued); + irq = cpu_util_irq(rq); + if (unlikely(irq >= scale)) { + if (min) + *min = scale; + if (max) + *max = scale; + return scale; + } + if (min) { /* - * Despite the following checks we still have a small window - * for a possible race, when an execl's select_task_rq_fair() - * races with LB's detach_task(): - * - * detach_task() - * p->on_rq = TASK_ON_RQ_MIGRATING; - * ---------------------------------- A - * deactivate_task() \ - * dequeue_task() + RaceTime - * util_est_dequeue() / - * ---------------------------------- B - * - * The additional check on "current == p" it's required to - * properly fix the execl regression and it helps in further - * reducing the chances for the above race. + * The minimum utilization returns the highest level between: + * - the computed DL bandwidth needed with the IRQ pressure which + * steals time to the deadline task. + * - The minimum performance requirement for CFS and/or RT. */ - if (unlikely(task_on_rq_queued(p) || current == p)) - lsub_positive(&estimated, _task_util_est(p)); + *min = max(irq + cpu_bw_dl(rq), uclamp_rq_get(rq, UCLAMP_MIN)); - util = max(util, estimated); + /* + * When an RT task is runnable and uclamp is not used, we must + * ensure that the task will run at maximum compute capacity. + */ + if (!uclamp_is_used() && rt_rq_is_runnable(&rq->rt)) + *min = max(*min, scale); } /* - * Utilization (estimated) can exceed the CPU capacity, thus let's - * clamp to the maximum CPU capacity to ensure consistency with - * the cpu_util call. + * Because the time spend on RT/DL tasks is visible as 'lost' time to + * CFS tasks and we use the same metric to track the effective + * utilization (PELT windows are synchronized) we can directly add them + * to obtain the CPU's actual utilization. + */ + util = util_cfs + cpu_util_rt(rq); + util += cpu_util_dl(rq); + + /* + * The maximum hint is a soft bandwidth requirement, which can be lower + * than the actual utilization because of uclamp_max requirements. */ - return min_t(unsigned long, util, capacity_orig_of(cpu)); + if (max) + *max = min(scale, uclamp_rq_get(rq, UCLAMP_MAX)); + + if (util >= scale) + return scale; + + /* + * There is still idle time; further improve the number by using the + * IRQ metric. Because IRQ/steal time is hidden from the task clock we + * need to scale the task numbers: + * + * max - irq + * U' = irq + --------- * U + * max + */ + util = scale_irq_capacity(util, irq, scale); + util += irq; + + return min(scale, util); +} + +unsigned long sched_cpu_util(int cpu) +{ + return effective_cpu_util(cpu, cpu_util_cfs(cpu), NULL, NULL); } /* - * Disable WAKE_AFFINE in the case where task @p doesn't fit in the - * capacity of either the waking CPU @cpu or the previous CPU @prev_cpu. - * - * In that case WAKE_AFFINE doesn't make sense and we'll let - * BALANCE_WAKE sort things out. + * energy_env - Utilization landscape for energy estimation. + * @task_busy_time: Utilization contribution by the task for which we test the + * placement. Given by eenv_task_busy_time(). + * @pd_busy_time: Utilization of the whole perf domain without the task + * contribution. Given by eenv_pd_busy_time(). + * @cpu_cap: Maximum CPU capacity for the perf domain. + * @pd_cap: Entire perf domain capacity. (pd->nr_cpus * cpu_cap). */ -static int wake_cap(struct task_struct *p, int cpu, int prev_cpu) +struct energy_env { + unsigned long task_busy_time; + unsigned long pd_busy_time; + unsigned long cpu_cap; + unsigned long pd_cap; +}; + +/* + * Compute the task busy time for compute_energy(). This time cannot be + * injected directly into effective_cpu_util() because of the IRQ scaling. + * The latter only makes sense with the most recent CPUs where the task has + * run. + */ +static inline void eenv_task_busy_time(struct energy_env *eenv, + struct task_struct *p, int prev_cpu) { - long min_cap, max_cap; + unsigned long busy_time, max_cap = arch_scale_cpu_capacity(prev_cpu); + unsigned long irq = cpu_util_irq(cpu_rq(prev_cpu)); - if (!static_branch_unlikely(&sched_asym_cpucapacity)) - return 0; + if (unlikely(irq >= max_cap)) + busy_time = max_cap; + else + busy_time = scale_irq_capacity(task_util_est(p), irq, max_cap); - min_cap = min(capacity_orig_of(prev_cpu), capacity_orig_of(cpu)); - max_cap = cpu_rq(cpu)->rd->max_cpu_capacity; + eenv->task_busy_time = busy_time; +} - /* Minimum capacity is close to max, no need to abort wake_affine */ - if (max_cap - min_cap < max_cap >> 3) - return 0; +/* + * Compute the perf_domain (PD) busy time for compute_energy(). Based on the + * utilization for each @pd_cpus, it however doesn't take into account + * clamping since the ratio (utilization / cpu_capacity) is already enough to + * scale the EM reported power consumption at the (eventually clamped) + * cpu_capacity. + * + * The contribution of the task @p for which we want to estimate the + * energy cost is removed (by cpu_util()) and must be calculated + * separately (see eenv_task_busy_time). This ensures: + * + * - A stable PD utilization, no matter which CPU of that PD we want to place + * the task on. + * + * - A fair comparison between CPUs as the task contribution (task_util()) + * will always be the same no matter which CPU utilization we rely on + * (util_avg or util_est). + * + * Set @eenv busy time for the PD that spans @pd_cpus. This busy time can't + * exceed @eenv->pd_cap. + */ +static inline void eenv_pd_busy_time(struct energy_env *eenv, + struct cpumask *pd_cpus, + struct task_struct *p) +{ + unsigned long busy_time = 0; + int cpu; - /* Bring task utilization in sync with prev_cpu */ - sync_entity_load_avg(&p->se); + for_each_cpu(cpu, pd_cpus) { + unsigned long util = cpu_util(cpu, p, -1, 0); + + busy_time += effective_cpu_util(cpu, util, NULL, NULL); + } - return !task_fits_capacity(p, min_cap); + eenv->pd_busy_time = min(eenv->pd_cap, busy_time); } /* - * Predicts what cpu_util(@cpu) would return if @p was migrated (and enqueued) - * to @dst_cpu. + * Compute the maximum utilization for compute_energy() when the task @p + * is placed on the cpu @dst_cpu. + * + * Returns the maximum utilization among @eenv->cpus. This utilization can't + * exceed @eenv->cpu_cap. */ -static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) +static inline unsigned long +eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, + struct task_struct *p, int dst_cpu) { - struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs; - unsigned long util_est, util = READ_ONCE(cfs_rq->avg.util_avg); - - /* - * If @p migrates from @cpu to another, remove its contribution. Or, - * if @p migrates from another CPU to @cpu, add its contribution. In - * the other cases, @cpu is not impacted by the migration, so the - * util_avg should already be correct. - */ - if (task_cpu(p) == cpu && dst_cpu != cpu) - sub_positive(&util, task_util(p)); - else if (task_cpu(p) != cpu && dst_cpu == cpu) - util += task_util(p); + unsigned long max_util = 0; + int cpu; - if (sched_feat(UTIL_EST)) { - util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued); + for_each_cpu(cpu, pd_cpus) { + struct task_struct *tsk = (cpu == dst_cpu) ? p : NULL; + unsigned long util = cpu_util(cpu, p, dst_cpu, 1); + unsigned long eff_util, min, max; /* - * During wake-up, the task isn't enqueued yet and doesn't - * appear in the cfs_rq->avg.util_est.enqueued of any rq, - * so just add it (if needed) to "simulate" what will be - * cpu_util() after the task has been enqueued. + * Performance domain frequency: utilization clamping + * must be considered since it affects the selection + * of the performance domain frequency. + * NOTE: in case RT tasks are running, by default the min + * utilization can be max OPP. */ - if (dst_cpu == cpu) - util_est += _task_util_est(p); + eff_util = effective_cpu_util(cpu, util, &min, &max); - util = max(util, util_est); + /* Task's uclamp can modify min and max value */ + if (tsk && uclamp_is_used()) { + min = max(min, uclamp_eff_value(p, UCLAMP_MIN)); + + /* + * If there is no active max uclamp constraint, + * directly use task's one, otherwise keep max. + */ + if (uclamp_rq_is_idle(cpu_rq(cpu))) + max = uclamp_eff_value(p, UCLAMP_MAX); + else + max = max(max, uclamp_eff_value(p, UCLAMP_MAX)); + } + + eff_util = sugov_effective_cpu_perf(cpu, eff_util, min, max); + max_util = max(max_util, eff_util); } - return min(util, capacity_orig_of(cpu)); + return min(max_util, eenv->cpu_cap); } /* - * compute_energy(): Estimates the energy that would be consumed if @p was - * migrated to @dst_cpu. compute_energy() predicts what will be the utilization - * landscape of the * CPUs after the task migration, and uses the Energy Model - * to compute what would be the energy if we decided to actually migrate that - * task. + * compute_energy(): Use the Energy Model to estimate the energy that @pd would + * consume for a given utilization landscape @eenv. When @dst_cpu < 0, the task + * contribution is ignored. */ -static long -compute_energy(struct task_struct *p, int dst_cpu, struct perf_domain *pd) +static inline unsigned long +compute_energy(struct energy_env *eenv, struct perf_domain *pd, + struct cpumask *pd_cpus, struct task_struct *p, int dst_cpu) { - long util, max_util, sum_util, energy = 0; - int cpu; + unsigned long max_util = eenv_pd_max_util(eenv, pd_cpus, p, dst_cpu); + unsigned long busy_time = eenv->pd_busy_time; + unsigned long energy; - for (; pd; pd = pd->next) { - max_util = sum_util = 0; - /* - * The capacity state of CPUs of the current rd can be driven by - * CPUs of another rd if they belong to the same performance - * domain. So, account for the utilization of these CPUs too - * by masking pd with cpu_online_mask instead of the rd span. - * - * If an entire performance domain is outside of the current rd, - * it will not appear in its pd list and will not be accounted - * by compute_energy(). - */ - for_each_cpu_and(cpu, perf_domain_span(pd), cpu_online_mask) { - util = cpu_util_next(cpu, p, dst_cpu); - util = schedutil_energy_util(cpu, util); - max_util = max(util, max_util); - sum_util += util; - } + if (dst_cpu >= 0) + busy_time = min(eenv->pd_cap, busy_time + eenv->task_busy_time); - energy += em_pd_energy(pd->em_pd, max_util, sum_util); - } + energy = em_cpu_energy(pd->em_pd, max_util, busy_time, eenv->cpu_cap); + + trace_sched_compute_energy_tp(p, dst_cpu, energy, max_util, busy_time); return energy; } @@ -6484,28 +8311,32 @@ compute_energy(struct task_struct *p, int dst_cpu, struct perf_domain *pd) * NOTE: Forkees are not accepted in the energy-aware wake-up path because * they don't have any useful utilization data yet and it's not possible to * forecast their impact on energy consumption. Consequently, they will be - * placed by find_idlest_cpu() on the least loaded CPU, which might turn out + * placed by sched_balance_find_dst_cpu() on the least loaded CPU, which might turn out * to be energy-inefficient in some use-cases. The alternative would be to * bias new tasks towards specific types of CPUs first, or to try to infer * their util_avg from the parent task, but those heuristics could hurt * other use-cases too. So, until someone finds a better way to solve this, * let's keep things simple by re-using the existing slow path. */ - static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) { - unsigned long prev_energy = ULONG_MAX, best_energy = ULONG_MAX; - struct root_domain *rd = cpu_rq(smp_processor_id())->rd; - int cpu, best_energy_cpu = prev_cpu; - struct perf_domain *head, *pd; - unsigned long cpu_cap, util; + struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask); + unsigned long prev_delta = ULONG_MAX, best_delta = ULONG_MAX; + unsigned long p_util_min = uclamp_is_used() ? uclamp_eff_value(p, UCLAMP_MIN) : 0; + unsigned long p_util_max = uclamp_is_used() ? uclamp_eff_value(p, UCLAMP_MAX) : 1024; + struct root_domain *rd = this_rq()->rd; + int cpu, best_energy_cpu, target = -1; + int prev_fits = -1, best_fits = -1; + unsigned long best_actual_cap = 0; + unsigned long prev_actual_cap = 0; struct sched_domain *sd; + struct perf_domain *pd; + struct energy_env eenv; rcu_read_lock(); pd = rcu_dereference(rd->pd); - if (!pd || READ_ONCE(rd->overutilized)) - goto fail; - head = pd; + if (!pd) + goto unlock; /* * Energy-aware wake-up happens on the lowest sched_domain starting @@ -6515,114 +8346,209 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) while (sd && !cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) sd = sd->parent; if (!sd) - goto fail; + goto unlock; + + target = prev_cpu; sync_entity_load_avg(&p->se); - if (!task_util_est(p)) + if (!task_util_est(p) && p_util_min == 0) goto unlock; + eenv_task_busy_time(&eenv, p, prev_cpu); + for (; pd; pd = pd->next) { - unsigned long cur_energy, spare_cap, max_spare_cap = 0; + unsigned long util_min = p_util_min, util_max = p_util_max; + unsigned long cpu_cap, cpu_actual_cap, util; + long prev_spare_cap = -1, max_spare_cap = -1; + unsigned long rq_util_min, rq_util_max; + unsigned long cur_delta, base_energy; int max_spare_cap_cpu = -1; + int fits, max_fits = -1; - for_each_cpu_and(cpu, perf_domain_span(pd), sched_domain_span(sd)) { - if (!cpumask_test_cpu(cpu, &p->cpus_allowed)) - continue; + cpumask_and(cpus, perf_domain_span(pd), cpu_online_mask); - /* Skip CPUs that will be overutilized. */ - util = cpu_util_next(cpu, p, cpu); - cpu_cap = capacity_of(cpu); - if (cpu_cap * 1024 < util * capacity_margin) + if (cpumask_empty(cpus)) + continue; + + /* Account external pressure for the energy estimation */ + cpu = cpumask_first(cpus); + cpu_actual_cap = get_actual_cpu_capacity(cpu); + + eenv.cpu_cap = cpu_actual_cap; + eenv.pd_cap = 0; + + for_each_cpu(cpu, cpus) { + struct rq *rq = cpu_rq(cpu); + + eenv.pd_cap += cpu_actual_cap; + + if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) continue; - /* Always use prev_cpu as a candidate. */ - if (cpu == prev_cpu) { - prev_energy = compute_energy(p, prev_cpu, head); - best_energy = min(best_energy, prev_energy); + if (!cpumask_test_cpu(cpu, p->cpus_ptr)) continue; - } + + util = cpu_util(cpu, p, cpu, 0); + cpu_cap = capacity_of(cpu); /* - * Find the CPU with the maximum spare capacity in - * the performance domain + * Skip CPUs that cannot satisfy the capacity request. + * IOW, placing the task there would make the CPU + * overutilized. Take uclamp into account to see how + * much capacity we can get out of the CPU; this is + * aligned with sched_cpu_util(). */ - spare_cap = cpu_cap - util; - if (spare_cap > max_spare_cap) { - max_spare_cap = spare_cap; + if (uclamp_is_used() && !uclamp_rq_is_idle(rq)) { + /* + * Open code uclamp_rq_util_with() except for + * the clamp() part. I.e.: apply max aggregation + * only. util_fits_cpu() logic requires to + * operate on non clamped util but must use the + * max-aggregated uclamp_{min, max}. + */ + rq_util_min = uclamp_rq_get(rq, UCLAMP_MIN); + rq_util_max = uclamp_rq_get(rq, UCLAMP_MAX); + + util_min = max(rq_util_min, p_util_min); + util_max = max(rq_util_max, p_util_max); + } + + fits = util_fits_cpu(util, util_min, util_max, cpu); + if (!fits) + continue; + + lsub_positive(&cpu_cap, util); + + if (cpu == prev_cpu) { + /* Always use prev_cpu as a candidate. */ + prev_spare_cap = cpu_cap; + prev_fits = fits; + } else if ((fits > max_fits) || + ((fits == max_fits) && ((long)cpu_cap > max_spare_cap))) { + /* + * Find the CPU with the maximum spare capacity + * among the remaining CPUs in the performance + * domain. + */ + max_spare_cap = cpu_cap; max_spare_cap_cpu = cpu; + max_fits = fits; } } - /* Evaluate the energy impact of using this CPU. */ - if (max_spare_cap_cpu >= 0) { - cur_energy = compute_energy(p, max_spare_cap_cpu, head); - if (cur_energy < best_energy) { - best_energy = cur_energy; - best_energy_cpu = max_spare_cap_cpu; - } + if (max_spare_cap_cpu < 0 && prev_spare_cap < 0) + continue; + + eenv_pd_busy_time(&eenv, cpus, p); + /* Compute the 'base' energy of the pd, without @p */ + base_energy = compute_energy(&eenv, pd, cpus, p, -1); + + /* Evaluate the energy impact of using prev_cpu. */ + if (prev_spare_cap > -1) { + prev_delta = compute_energy(&eenv, pd, cpus, p, + prev_cpu); + /* CPU utilization has changed */ + if (prev_delta < base_energy) + goto unlock; + prev_delta -= base_energy; + prev_actual_cap = cpu_actual_cap; + best_delta = min(best_delta, prev_delta); + } + + /* Evaluate the energy impact of using max_spare_cap_cpu. */ + if (max_spare_cap_cpu >= 0 && max_spare_cap > prev_spare_cap) { + /* Current best energy cpu fits better */ + if (max_fits < best_fits) + continue; + + /* + * Both don't fit performance hint (i.e. uclamp_min) + * but best energy cpu has better capacity. + */ + if ((max_fits < 0) && + (cpu_actual_cap <= best_actual_cap)) + continue; + + cur_delta = compute_energy(&eenv, pd, cpus, p, + max_spare_cap_cpu); + /* CPU utilization has changed */ + if (cur_delta < base_energy) + goto unlock; + cur_delta -= base_energy; + + /* + * Both fit for the task but best energy cpu has lower + * energy impact. + */ + if ((max_fits > 0) && (best_fits > 0) && + (cur_delta >= best_delta)) + continue; + + best_delta = cur_delta; + best_energy_cpu = max_spare_cap_cpu; + best_fits = max_fits; + best_actual_cap = cpu_actual_cap; } } -unlock: rcu_read_unlock(); - /* - * Pick the best CPU if prev_cpu cannot be used, or if it saves at - * least 6% of the energy used by prev_cpu. - */ - if (prev_energy == ULONG_MAX) - return best_energy_cpu; - - if ((prev_energy - best_energy) > (prev_energy >> 4)) - return best_energy_cpu; + if ((best_fits > prev_fits) || + ((best_fits > 0) && (best_delta < prev_delta)) || + ((best_fits < 0) && (best_actual_cap > prev_actual_cap))) + target = best_energy_cpu; - return prev_cpu; + return target; -fail: +unlock: rcu_read_unlock(); - return -1; + return target; } /* * select_task_rq_fair: Select target runqueue for the waking task in domains - * that have the 'sd_flag' flag set. In practice, this is SD_BALANCE_WAKE, + * that have the relevant SD flag set. In practice, this is SD_BALANCE_WAKE, * SD_BALANCE_FORK, or SD_BALANCE_EXEC. * * Balances load by selecting the idlest CPU in the idlest group, or under * certain conditions an idle sibling CPU if the domain has SD_WAKE_AFFINE set. * * Returns the target CPU number. - * - * preempt must be disabled. */ static int -select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_flags) +select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) { + int sync = (wake_flags & WF_SYNC) && !(current->flags & PF_EXITING); struct sched_domain *tmp, *sd = NULL; int cpu = smp_processor_id(); int new_cpu = prev_cpu; int want_affine = 0; - int sync = (wake_flags & WF_SYNC) && !(current->flags & PF_EXITING); + /* SD_flags and WF_flags share the first nibble */ + int sd_flag = wake_flags & 0xF; - if (sd_flag & SD_BALANCE_WAKE) { + /* + * required for stable ->cpus_allowed + */ + lockdep_assert_held(&p->pi_lock); + if (wake_flags & WF_TTWU) { record_wakee(p); - if (static_branch_unlikely(&sched_energy_present)) { + if ((wake_flags & WF_CURRENT_CPU) && + cpumask_test_cpu(cpu, p->cpus_ptr)) + return cpu; + + if (!is_rd_overutilized(this_rq()->rd)) { new_cpu = find_energy_efficient_cpu(p, prev_cpu); if (new_cpu >= 0) return new_cpu; new_cpu = prev_cpu; } - want_affine = !wake_wide(p) && !wake_cap(p, cpu, prev_cpu) && - cpumask_test_cpu(cpu, &p->cpus_allowed); + want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, p->cpus_ptr); } rcu_read_lock(); for_each_domain(cpu, tmp) { - if (!(tmp->flags & SD_LOAD_BALANCE)) - break; - /* * If both 'cpu' and 'prev_cpu' are part of this domain, * cpu is a valid SD_WAKE_AFFINE target. @@ -6636,6 +8562,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f break; } + /* + * Usually only true for WF_EXEC and WF_FORK, as sched_domains + * usually do not have SD_BALANCE_WAKE set. That means wakeup + * will usually go to the fast path. + */ if (tmp->flags & sd_flag) sd = tmp; else if (!want_affine) @@ -6644,22 +8575,16 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f if (unlikely(sd)) { /* Slow path */ - new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag); - } else if (sd_flag & SD_BALANCE_WAKE) { /* XXX always ? */ + new_cpu = sched_balance_find_dst_cpu(sd, p, cpu, prev_cpu, sd_flag); + } else if (wake_flags & WF_TTWU) { /* XXX always ? */ /* Fast path */ - new_cpu = select_idle_sibling(p, prev_cpu, new_cpu); - - if (want_affine) - current->recent_used_cpu = cpu; } rcu_read_unlock(); return new_cpu; } -static void detach_entity_cfs_rq(struct sched_entity *se); - /* * Called immediately before a task is migrated to a new CPU; task_cpu(p) and * cfs_rq_of(p) references at time of call are still valid and identify the @@ -6667,174 +8592,178 @@ static void detach_entity_cfs_rq(struct sched_entity *se); */ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu) { - /* - * As blocked tasks retain absolute vruntime the migration needs to - * deal with this by subtracting the old and adding the new - * min_vruntime -- the latter is done by enqueue_entity() when placing - * the task on the new runqueue. - */ - if (p->state == TASK_WAKING) { - struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); - u64 min_vruntime; - -#ifndef CONFIG_64BIT - u64 min_vruntime_copy; - - do { - min_vruntime_copy = cfs_rq->min_vruntime_copy; - smp_rmb(); - min_vruntime = cfs_rq->min_vruntime; - } while (min_vruntime != min_vruntime_copy); -#else - min_vruntime = cfs_rq->min_vruntime; -#endif + struct sched_entity *se = &p->se; - se->vruntime -= min_vruntime; - } + if (!task_on_rq_migrating(p)) { + remove_entity_load_avg(se); - if (p->on_rq == TASK_ON_RQ_MIGRATING) { /* - * In case of TASK_ON_RQ_MIGRATING we in fact hold the 'old' - * rq->lock and can modify state directly. - */ - lockdep_assert_held(&task_rq(p)->lock); - detach_entity_cfs_rq(&p->se); - - } else { - /* - * We are supposed to update the task to "current" time, then - * its up to date and ready to go to new CPU/cfs_rq. But we - * have difficulty in getting what current time is, so simply - * throw away the out-of-date time. This will result in the - * wakee task is less decayed, but giving the wakee more load - * sounds not bad. + * Here, the task's PELT values have been updated according to + * the current rq's clock. But if that clock hasn't been + * updated in a while, a substantial idle time will be missed, + * leading to an inflation after wake-up on the new rq. + * + * Estimate the missing time from the cfs_rq last_update_time + * and update sched_avg to improve the PELT continuity after + * migration. */ - remove_entity_load_avg(&p->se); + migrate_se_pelt_lag(se); } /* Tell new CPU we are migrated */ - p->se.avg.last_update_time = 0; - - /* We have migrated, no longer consider this task hot */ - p->se.exec_start = 0; + se->avg.last_update_time = 0; update_scan_period(p, new_cpu); } static void task_dead_fair(struct task_struct *p) { - remove_entity_load_avg(&p->se); -} -#endif /* CONFIG_SMP */ + struct sched_entity *se = &p->se; -static unsigned long wakeup_gran(struct sched_entity *se) -{ - unsigned long gran = sysctl_sched_wakeup_granularity; + if (se->sched_delayed) { + struct rq_flags rf; + struct rq *rq; - /* - * Since its curr running now, convert the gran from real-time - * to virtual-time in his units. - * - * By using 'se' instead of 'curr' we penalize light tasks, so - * they get preempted easier. That is, if 'se' < 'curr' then - * the resulting gran will be larger, therefore penalizing the - * lighter, if otoh 'se' > 'curr' then the resulting gran will - * be smaller, again penalizing the lighter task. - * - * This is especially important for buddies when the leftmost - * task is higher priority than the buddy. - */ - return calc_delta_fair(gran, se); + rq = task_rq_lock(p, &rf); + if (se->sched_delayed) { + update_rq_clock(rq); + dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); + } + task_rq_unlock(rq, p, &rf); + } + + remove_entity_load_avg(se); } /* - * Should 'se' preempt 'curr'. - * - * |s1 - * |s2 - * |s3 - * g - * |<--->|c - * - * w(c, s1) = -1 - * w(c, s2) = 0 - * w(c, s3) = 1 - * + * Set the max capacity the task is allowed to run at for misfit detection. */ -static int -wakeup_preempt_entity(struct sched_entity *curr, struct sched_entity *se) +static void set_task_max_allowed_capacity(struct task_struct *p) { - s64 gran, vdiff = curr->vruntime - se->vruntime; + struct asym_cap_data *entry; - if (vdiff <= 0) - return -1; + if (!sched_asym_cpucap_active()) + return; - gran = wakeup_gran(se); - if (vdiff > gran) - return 1; + rcu_read_lock(); + list_for_each_entry_rcu(entry, &asym_cap_list, link) { + cpumask_t *cpumask; - return 0; + cpumask = cpu_capacity_span(entry); + if (!cpumask_intersects(p->cpus_ptr, cpumask)) + continue; + + p->max_allowed_capacity = entry->capacity; + break; + } + rcu_read_unlock(); } -static void set_last_buddy(struct sched_entity *se) +static void set_cpus_allowed_fair(struct task_struct *p, struct affinity_context *ctx) { - if (entity_is_task(se) && unlikely(task_has_idle_policy(task_of(se)))) - return; - - for_each_sched_entity(se) { - if (SCHED_WARN_ON(!se->on_rq)) - return; - cfs_rq_of(se)->last = se; - } + set_cpus_allowed_common(p, ctx); + set_task_max_allowed_capacity(p); } static void set_next_buddy(struct sched_entity *se) { - if (entity_is_task(se) && unlikely(task_has_idle_policy(task_of(se)))) - return; - for_each_sched_entity(se) { - if (SCHED_WARN_ON(!se->on_rq)) + if (WARN_ON_ONCE(!se->on_rq)) + return; + if (se_is_idle(se)) return; cfs_rq_of(se)->next = se; } } -static void set_skip_buddy(struct sched_entity *se) +enum preempt_wakeup_action { + PREEMPT_WAKEUP_NONE, /* No preemption. */ + PREEMPT_WAKEUP_SHORT, /* Ignore slice protection. */ + PREEMPT_WAKEUP_PICK, /* Let __pick_eevdf() decide. */ + PREEMPT_WAKEUP_RESCHED, /* Force reschedule. */ +}; + +static inline bool +set_preempt_buddy(struct cfs_rq *cfs_rq, int wake_flags, + struct sched_entity *pse, struct sched_entity *se) +{ + /* + * Keep existing buddy if the deadline is sooner than pse. + * The older buddy may be cache cold and completely unrelated + * to the current wakeup but that is unpredictable where as + * obeying the deadline is more in line with EEVDF objectives. + */ + if (cfs_rq->next && entity_before(cfs_rq->next, pse)) + return false; + + set_next_buddy(pse); + return true; +} + +/* + * WF_SYNC|WF_TTWU indicates the waker expects to sleep but it is not + * strictly enforced because the hint is either misunderstood or + * multiple tasks must be woken up. + */ +static inline enum preempt_wakeup_action +preempt_sync(struct rq *rq, int wake_flags, + struct sched_entity *pse, struct sched_entity *se) { - for_each_sched_entity(se) - cfs_rq_of(se)->skip = se; + u64 threshold, delta; + + /* + * WF_SYNC without WF_TTWU is not expected so warn if it happens even + * though it is likely harmless. + */ + WARN_ON_ONCE(!(wake_flags & WF_TTWU)); + + threshold = sysctl_sched_migration_cost; + delta = rq_clock_task(rq) - se->exec_start; + if ((s64)delta < 0) + delta = 0; + + /* + * WF_RQ_SELECTED implies the tasks are stacking on a CPU when they + * could run on other CPUs. Reduce the threshold before preemption is + * allowed to an arbitrary lower value as it is more likely (but not + * guaranteed) the waker requires the wakee to finish. + */ + if (wake_flags & WF_RQ_SELECTED) + threshold >>= 2; + + /* + * As WF_SYNC is not strictly obeyed, allow some runtime for batch + * wakeups to be issued. + */ + if (entity_before(pse, se) && delta >= threshold) + return PREEMPT_WAKEUP_RESCHED; + + return PREEMPT_WAKEUP_NONE; } /* * Preempt the current task with a newly woken task if needed: */ -static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_flags) +static void check_preempt_wakeup_fair(struct rq *rq, struct task_struct *p, int wake_flags) { - struct task_struct *curr = rq->curr; - struct sched_entity *se = &curr->se, *pse = &p->se; - struct cfs_rq *cfs_rq = task_cfs_rq(curr); - int scale = cfs_rq->nr_running >= sched_nr_latency; - int next_buddy_marked = 0; + enum preempt_wakeup_action preempt_action = PREEMPT_WAKEUP_PICK; + struct task_struct *donor = rq->donor; + struct sched_entity *se = &donor->se, *pse = &p->se; + struct cfs_rq *cfs_rq = task_cfs_rq(donor); + int cse_is_idle, pse_is_idle; if (unlikely(se == pse)) return; /* * This is possible from callers such as attach_tasks(), in which we - * unconditionally check_prempt_curr() after an enqueue (which may have + * unconditionally wakeup_preempt() after an enqueue (which may have * lead to a throttle). This both saves work and prevents false * next-buddy nomination below. */ - if (unlikely(throttled_hierarchy(cfs_rq_of(pse)))) + if (task_is_throttled(p)) return; - if (sched_feat(NEXT_BUDDY) && scale && !(wake_flags & WF_FORK)) { - set_next_buddy(pse); - next_buddy_marked = 1; - } - /* * We can come here with TIF_NEED_RESCHED already set from new task * wake up path. @@ -6845,122 +8774,182 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ * prevents us from potentially nominating it as a false LAST_BUDDY * below. */ - if (test_tsk_need_resched(curr)) + if (test_tsk_need_resched(rq->curr)) + return; + + if (!sched_feat(WAKEUP_PREEMPTION)) return; - /* Idle tasks are by definition preempted by non-idle tasks. */ - if (unlikely(task_has_idle_policy(curr)) && - likely(!task_has_idle_policy(p))) + find_matching_se(&se, &pse); + WARN_ON_ONCE(!pse); + + cse_is_idle = se_is_idle(se); + pse_is_idle = se_is_idle(pse); + + /* + * Preempt an idle entity in favor of a non-idle entity (and don't preempt + * in the inverse case). + */ + if (cse_is_idle && !pse_is_idle) { + /* + * When non-idle entity preempt an idle entity, + * don't give idle entity slice protection. + */ + preempt_action = PREEMPT_WAKEUP_SHORT; goto preempt; + } + + if (cse_is_idle != pse_is_idle) + return; /* - * Batch and idle tasks do not preempt non-idle tasks (their preemption - * is driven by the tick): + * BATCH and IDLE tasks do not preempt others. */ - if (unlikely(p->policy != SCHED_NORMAL) || !sched_feat(WAKEUP_PREEMPTION)) + if (unlikely(!normal_policy(p->policy))) return; - find_matching_se(&se, &pse); - update_curr(cfs_rq_of(se)); - BUG_ON(!pse); - if (wakeup_preempt_entity(se, pse) == 1) { + cfs_rq = cfs_rq_of(se); + update_curr(cfs_rq); + /* + * If @p has a shorter slice than current and @p is eligible, override + * current's slice protection in order to allow preemption. + */ + if (sched_feat(PREEMPT_SHORT) && (pse->slice < se->slice)) { + preempt_action = PREEMPT_WAKEUP_SHORT; + goto pick; + } + + /* + * Ignore wakee preemption on WF_FORK as it is less likely that + * there is shared data as exec often follow fork. Do not + * preempt for tasks that are sched_delayed as it would violate + * EEVDF to forcibly queue an ineligible task. + */ + if ((wake_flags & WF_FORK) || pse->sched_delayed) + return; + + /* + * If @p potentially is completing work required by current then + * consider preemption. + * + * Reschedule if waker is no longer eligible. */ + if (in_task() && !entity_eligible(cfs_rq, se)) { + preempt_action = PREEMPT_WAKEUP_RESCHED; + goto preempt; + } + + /* Prefer picking wakee soon if appropriate. */ + if (sched_feat(NEXT_BUDDY) && + set_preempt_buddy(cfs_rq, wake_flags, pse, se)) { + /* - * Bias pick_next to pick the sched entity that is - * triggering this preemption. + * Decide whether to obey WF_SYNC hint for a new buddy. Old + * buddies are ignored as they may not be relevant to the + * waker and less likely to be cache hot. */ - if (!next_buddy_marked) - set_next_buddy(pse); + if (wake_flags & WF_SYNC) + preempt_action = preempt_sync(rq, wake_flags, pse, se); + } + + switch (preempt_action) { + case PREEMPT_WAKEUP_NONE: + return; + case PREEMPT_WAKEUP_RESCHED: goto preempt; + case PREEMPT_WAKEUP_SHORT: + fallthrough; + case PREEMPT_WAKEUP_PICK: + break; } +pick: + /* + * If @p has become the most eligible task, force preemption. + */ + if (__pick_eevdf(cfs_rq, preempt_action != PREEMPT_WAKEUP_SHORT) == pse) + goto preempt; + + if (sched_feat(RUN_TO_PARITY)) + update_protect_slice(cfs_rq, se); + return; preempt: - resched_curr(rq); - /* - * Only set the backward buddy when the current task is still - * on the rq. This can happen when a wakeup gets interleaved - * with schedule on the ->pre_schedule() or idle_balance() - * point, either of which can * drop the rq lock. - * - * Also, during early boot the idle thread is in the fair class, - * for obvious reasons its a bad idea to schedule back to it. - */ - if (unlikely(!se->on_rq || curr == rq->idle)) - return; + if (preempt_action == PREEMPT_WAKEUP_SHORT) + cancel_protect_slice(se); - if (sched_feat(LAST_BUDDY) && scale && entity_is_task(se)) - set_last_buddy(se); + resched_curr_lazy(rq); } -static struct task_struct * +static struct task_struct *pick_task_fair(struct rq *rq, struct rq_flags *rf) +{ + struct sched_entity *se; + struct cfs_rq *cfs_rq; + struct task_struct *p; + bool throttled; + +again: + cfs_rq = &rq->cfs; + if (!cfs_rq->nr_queued) + return NULL; + + throttled = false; + + do { + /* Might not have done put_prev_entity() */ + if (cfs_rq->curr && cfs_rq->curr->on_rq) + update_curr(cfs_rq); + + throttled |= check_cfs_rq_runtime(cfs_rq); + + se = pick_next_entity(rq, cfs_rq); + if (!se) + goto again; + cfs_rq = group_cfs_rq(se); + } while (cfs_rq); + + p = task_of(se); + if (unlikely(throttled)) + task_throttle_setup_work(p); + return p; +} + +static void __set_next_task_fair(struct rq *rq, struct task_struct *p, bool first); +static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first); + +struct task_struct * pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { - struct cfs_rq *cfs_rq = &rq->cfs; struct sched_entity *se; struct task_struct *p; int new_tasks; again: - if (!cfs_rq->nr_running) + p = pick_task_fair(rq, rf); + if (!p) goto idle; + se = &p->se; #ifdef CONFIG_FAIR_GROUP_SCHED if (prev->sched_class != &fair_sched_class) goto simple; + __put_prev_set_next_dl_server(rq, prev, p); + /* * Because of the set_next_buddy() in dequeue_task_fair() it is rather * likely that a next task is from the same cgroup as the current. * * Therefore attempt to avoid putting and setting the entire cgroup * hierarchy, only change the part that actually changes. - */ - - do { - struct sched_entity *curr = cfs_rq->curr; - - /* - * Since we got here without doing put_prev_entity() we also - * have to consider cfs_rq->curr. If it is still a runnable - * entity, update_curr() will update its vruntime, otherwise - * forget we've ever seen it. - */ - if (curr) { - if (curr->on_rq) - update_curr(cfs_rq); - else - curr = NULL; - - /* - * This call to check_cfs_rq_runtime() will do the - * throttle and dequeue its entity in the parent(s). - * Therefore the nr_running test will indeed - * be correct. - */ - if (unlikely(check_cfs_rq_runtime(cfs_rq))) { - cfs_rq = &rq->cfs; - - if (!cfs_rq->nr_running) - goto idle; - - goto simple; - } - } - - se = pick_next_entity(cfs_rq, curr); - cfs_rq = group_cfs_rq(se); - } while (cfs_rq); - - p = task_of(se); - - /* + * * Since we haven't yet done put_prev_entity and if the selected task * is a different task than we started out with, try and touch the * least amount of cfs_rqs. */ if (prev != p) { struct sched_entity *pse = &prev->se; + struct cfs_rq *cfs_rq; while (!(cfs_rq = is_same_group(se, pse))) { int se_depth = se->depth; @@ -6978,61 +8967,62 @@ again: put_prev_entity(cfs_rq, pse); set_next_entity(cfs_rq, se); + + __set_next_task_fair(rq, p, true); } - goto done; + return p; + simple: -#endif +#endif /* CONFIG_FAIR_GROUP_SCHED */ + put_prev_set_next_task(rq, prev, p); + return p; - put_prev_task(rq, prev); +idle: + if (rf) { + new_tasks = sched_balance_newidle(rq, rf); - do { - se = pick_next_entity(cfs_rq, NULL); - set_next_entity(cfs_rq, se); - cfs_rq = group_cfs_rq(se); - } while (cfs_rq); + /* + * Because sched_balance_newidle() releases (and re-acquires) + * rq->lock, it is possible for any higher priority task to + * appear. In that case we must re-start the pick_next_entity() + * loop. + */ + if (new_tasks < 0) + return RETRY_TASK; - p = task_of(se); + if (new_tasks > 0) + goto again; + } -done: __maybe_unused; -#ifdef CONFIG_SMP /* - * Move the next running task to the front of - * the list, so our cfs_tasks list becomes MRU - * one. + * rq is about to be idle, check if we need to update the + * lost_idle_time of clock_pelt */ - list_move(&p->se.group_node, &rq->cfs_tasks); -#endif - - if (hrtick_enabled(rq)) - hrtick_start_fair(rq, p); - - update_misfit_status(p, rq); + update_idle_rq_clock_pelt(rq); - return p; + return NULL; +} -idle: - update_misfit_status(NULL, rq); - new_tasks = idle_balance(rq, rf); +static struct task_struct * +fair_server_pick_task(struct sched_dl_entity *dl_se, struct rq_flags *rf) +{ + return pick_task_fair(dl_se->rq, rf); +} - /* - * Because idle_balance() releases (and re-acquires) rq->lock, it is - * possible for any higher priority task to appear. In that case we - * must re-start the pick_next_entity() loop. - */ - if (new_tasks < 0) - return RETRY_TASK; +void fair_server_init(struct rq *rq) +{ + struct sched_dl_entity *dl_se = &rq->fair_server; - if (new_tasks > 0) - goto again; + init_dl_entity(dl_se); - return NULL; + dl_server_init(dl_se, rq, fair_server_pick_task); } /* * Account for a descheduled task: */ -static void put_prev_task_fair(struct rq *rq, struct task_struct *prev) +static void put_prev_task_fair(struct rq *rq, struct task_struct *prev, struct task_struct *next) { struct sched_entity *se = &prev->se; struct cfs_rq *cfs_rq; @@ -7045,12 +9035,10 @@ static void put_prev_task_fair(struct rq *rq, struct task_struct *prev) /* * sched_yield() is very simple - * - * The magic of dealing with the ->skip buddy is in pick_next_entity. */ static void yield_task_fair(struct rq *rq) { - struct task_struct *curr = rq->curr; + struct task_struct *curr = rq->donor; struct cfs_rq *cfs_rq = task_cfs_rq(curr); struct sched_entity *se = &curr->se; @@ -7062,32 +9050,41 @@ static void yield_task_fair(struct rq *rq) clear_buddies(cfs_rq, se); - if (curr->policy != SCHED_BATCH) { - update_rq_clock(rq); - /* - * Update run-time statistics of the 'current'. - */ - update_curr(cfs_rq); - /* - * Tell update_rq_clock() that we've just updated, - * so we don't do microscopic update in schedule() - * and double the fastpath cost. - */ - rq_clock_skip_update(rq); - } + update_rq_clock(rq); + /* + * Update run-time statistics of the 'current'. + */ + update_curr(cfs_rq); + /* + * Tell update_rq_clock() that we've just updated, + * so we don't do microscopic update in schedule() + * and double the fastpath cost. + */ + rq_clock_skip_update(rq); - set_skip_buddy(se); + /* + * Forfeit the remaining vruntime, only if the entity is eligible. This + * condition is necessary because in core scheduling we prefer to run + * ineligible tasks rather than force idling. If this happens we may + * end up in a loop where the core scheduler picks the yielding task, + * which yields immediately again; without the condition the vruntime + * ends up quickly running away. + */ + if (entity_eligible(cfs_rq, se)) { + se->vruntime = se->deadline; + se->deadline += calc_delta_fair(se->slice, se); + } } -static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preempt) +static bool yield_to_task_fair(struct rq *rq, struct task_struct *p) { struct sched_entity *se = &p->se; - /* throttled hierarchies are not runnable */ - if (!se->on_rq || throttled_hierarchy(cfs_rq_of(se))) + /* !se->on_rq also covers throttled task */ + if (!se->on_rq) return false; - /* Tell the scheduler that we'd really like pse to run next. */ + /* Tell the scheduler that we'd really like se to run next. */ set_next_buddy(se); yield_task_fair(rq); @@ -7095,7 +9092,6 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp return true; } -#ifdef CONFIG_SMP /************************************************** * Fair scheduling class load-balancing methods. * @@ -7145,7 +9141,7 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p, bool preemp * topology where each level pairs two lower groups (or better). This results * in O(log n) layers. Furthermore we reduce the number of CPUs going up the * tree to only the first of the previous level and we decrease the frequency - * of load-balance at each level inv. proportional to the number of CPUs in + * of load-balance at each level inversely proportional to the number of CPUs in * the groups. * * This yields: @@ -7218,19 +9214,61 @@ static unsigned long __read_mostly max_load_balance_interval = HZ/10; enum fbq_type { regular, remote, all }; +/* + * 'group_type' describes the group of CPUs at the moment of load balancing. + * + * The enum is ordered by pulling priority, with the group with lowest priority + * first so the group_type can simply be compared when selecting the busiest + * group. See update_sd_pick_busiest(). + */ enum group_type { - group_other = 0, + /* The group has spare capacity that can be used to run more tasks. */ + group_has_spare = 0, + /* + * The group is fully used and the tasks don't compete for more CPU + * cycles. Nevertheless, some tasks might wait before running. + */ + group_fully_busy, + /* + * One task doesn't fit with CPU's capacity and must be migrated to a + * more powerful CPU. + */ group_misfit_task, + /* + * Balance SMT group that's fully busy. Can benefit from migration + * a task on SMT with busy sibling to another CPU on idle core. + */ + group_smt_balance, + /* + * SD_ASYM_PACKING only: One local CPU with higher capacity is available, + * and the task should be migrated to it instead of running on the + * current CPU. + */ + group_asym_packing, + /* + * The tasks' affinity constraints previously prevented the scheduler + * from balancing the load across the system. + */ group_imbalanced, - group_overloaded, + /* + * The CPU is overloaded and can't provide expected CPU cycles to all + * tasks. + */ + group_overloaded +}; + +enum migration_type { + migrate_load = 0, + migrate_util, + migrate_task, + migrate_misfit }; #define LBF_ALL_PINNED 0x01 #define LBF_NEED_BREAK 0x02 #define LBF_DST_PINNED 0x04 #define LBF_SOME_PINNED 0x08 -#define LBF_NOHZ_STATS 0x10 -#define LBF_NOHZ_AGAIN 0x20 +#define LBF_ACTIVE_LB 0x10 struct lb_env { struct sched_domain *sd; @@ -7255,7 +9293,7 @@ struct lb_env { unsigned int loop_max; enum fbq_type fbq_type; - enum group_type src_grp_type; + enum migration_type migration_type; struct list_head tasks; }; @@ -7266,7 +9304,7 @@ static int task_hot(struct task_struct *p, struct lb_env *env) { s64 delta; - lockdep_assert_held(&env->src_rq->lock); + lockdep_assert_rq_held(env->src_rq); if (p->sched_class != &fair_sched_class) return 0; @@ -7274,16 +9312,27 @@ static int task_hot(struct task_struct *p, struct lb_env *env) if (unlikely(task_has_idle_policy(p))) return 0; + /* SMT siblings share cache */ + if (env->sd->flags & SD_SHARE_CPUCAPACITY) + return 0; + /* * Buddy candidates are cache hot: */ if (sched_feat(CACHE_HOT_BUDDY) && env->dst_rq->nr_running && - (&p->se == cfs_rq_of(&p->se)->next || - &p->se == cfs_rq_of(&p->se)->last)) + (&p->se == cfs_rq_of(&p->se)->next)) return 1; if (sysctl_sched_migration_cost == -1) return 1; + + /* + * Don't migrate task if the task's cookie does not match + * with the destination CPU's core cookie. + */ + if (!sched_core_cookie_match(cpu_rq(env->dst_cpu), p)) + return 1; + if (sysctl_sched_migration_cost == 0) return 0; @@ -7294,43 +9343,43 @@ 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) { @@ -7341,16 +9390,40 @@ 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, +#else /* !CONFIG_NUMA_BALANCING: */ +static inline long migrate_degrades_locality(struct task_struct *p, struct lb_env *env) { - return -1; + return 0; } +#endif /* !CONFIG_NUMA_BALANCING */ + +/* + * 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? @@ -7358,24 +9431,48 @@ static inline int migrate_degrades_locality(struct task_struct *p, static int can_migrate_task(struct task_struct *p, struct lb_env *env) { - int tsk_cache_hot; + long degrades, hot; - lockdep_assert_held(&env->src_rq->lock); + 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_allowed, or - * 3) running (obviously), or - * 4) are cache-hot on their current CPU. + * 1) delayed dequeued unless we migrate load, or + * 2) target cfs_rq is in throttled hierarchy, 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, or + * 6) are blocked on mutexes (if SCHED_PROXY_EXEC is enabled) + */ + if ((p->se.sched_delayed) && (env->migration_type != migrate_load)) + return 0; + + if (lb_throttled_hierarchy(p, 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 (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu)) + 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; + + if (task_is_blocked(p)) return 0; - if (!cpumask_test_cpu(env->dst_cpu, &p->cpus_allowed)) { + if (!cpumask_test_cpu(env->dst_cpu, p->cpus_ptr)) { int cpu; - schedstat_inc(p->se.statistics.nr_failed_migrations_affine); + schedstat_inc(p->stats.nr_failed_migrations_affine); env->flags |= LBF_SOME_PINNED; @@ -7384,52 +9481,58 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) * our sched_group. We may want to revisit it if we couldn't * meet load balance goals by pulling other tasks on src_cpu. * - * Avoid computing new_dst_cpu for NEWLY_IDLE or if we have - * already computed one in current iteration. + * Avoid computing new_dst_cpu + * - for NEWLY_IDLE + * - if we have already computed one in current iteration + * - if it's an active balance */ - if (env->idle == CPU_NEWLY_IDLE || (env->flags & LBF_DST_PINNED)) + if (env->idle == CPU_NEWLY_IDLE || + env->flags & (LBF_DST_PINNED | LBF_ACTIVE_LB)) return 0; /* Prevent to re-select dst_cpu via env's CPUs: */ - for_each_cpu_and(cpu, env->dst_grpmask, env->cpus) { - if (cpumask_test_cpu(cpu, &p->cpus_allowed)) { - env->flags |= LBF_DST_PINNED; - env->new_dst_cpu = cpu; - break; - } + cpu = cpumask_first_and_and(env->dst_grpmask, env->cpus, p->cpus_ptr); + + if (cpu < nr_cpu_ids) { + env->flags |= LBF_DST_PINNED; + env->new_dst_cpu = cpu; } return 0; } - /* Record that we found atleast one task that could run on dst_cpu */ + /* Record that we found at least one task that could run on dst_cpu */ env->flags &= ~LBF_ALL_PINNED; - if (task_running(env->src_rq, p)) { - schedstat_inc(p->se.statistics.nr_failed_migrations_running); + if (task_on_cpu(env->src_rq, p) || + task_current_donor(env->src_rq, p)) { + schedstat_inc(p->stats.nr_failed_migrations_running); return 0; } /* * Aggressive migration if: - * 1) destination numa is preferred - * 2) task is cache cold, or - * 3) too many balance attempts have failed. - */ - tsk_cache_hot = migrate_degrades_locality(p, env); - if (tsk_cache_hot == -1) - tsk_cache_hot = task_hot(p, env); - - 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->se.statistics.nr_forced_migrations); - } + * 1) active balance + * 2) destination numa is preferred + * 3) task is cache cold, or + * 4) too many balance attempts have failed. + */ + if (env->flags & LBF_ACTIVE_LB) + return 1; + + degrades = migrate_degrades_locality(p, env); + if (!degrades) + hot = task_hot(p, env); + else + hot = degrades > 0; + + if (!hot || env->sd->nr_balance_failed > env->sd->cache_nice_tries) { + if (hot) + p->sched_task_hot = 1; return 1; } - schedstat_inc(p->se.statistics.nr_failed_migrations_hot); + schedstat_inc(p->stats.nr_failed_migrations_hot); return 0; } @@ -7438,9 +9541,17 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) */ static void detach_task(struct task_struct *p, struct lb_env *env) { - lockdep_assert_held(&env->src_rq->lock); + 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); + } + + WARN_ON(task_current(env->src_rq, p)); + WARN_ON(task_current_donor(env->src_rq, p)); - p->on_rq = TASK_ON_RQ_MIGRATING; deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK); set_task_cpu(p, env->dst_cpu); } @@ -7455,7 +9566,7 @@ static struct task_struct *detach_one_task(struct lb_env *env) { struct task_struct *p; - lockdep_assert_held(&env->src_rq->lock); + lockdep_assert_rq_held(env->src_rq); list_for_each_entry_reverse(p, &env->src_rq->cfs_tasks, se.group_node) { @@ -7476,10 +9587,8 @@ static struct task_struct *detach_one_task(struct lb_env *env) return NULL; } -static const unsigned int sched_nr_migrate_break = 32; - /* - * detach_tasks() -- tries to detach up to imbalance weighted load from + * detach_tasks() -- tries to detach up to imbalance load/util/tasks from * busiest_rq, as part of a balancing operation within domain "sd". * * Returns number of detached tasks if successful and 0 otherwise. @@ -7487,11 +9596,20 @@ static const unsigned int sched_nr_migrate_break = 32; static int detach_tasks(struct lb_env *env) { struct list_head *tasks = &env->src_rq->cfs_tasks; + unsigned long util, load; struct task_struct *p; - unsigned long load; int detached = 0; - lockdep_assert_held(&env->src_rq->lock); + lockdep_assert_rq_held(env->src_rq); + + /* + * Source run queue has been emptied by another CPU, clear + * LBF_ALL_PINNED flag as we will not test any task. + */ + if (env->src_rq->nr_running <= 1) { + env->flags &= ~LBF_ALL_PINNED; + return 0; + } if (env->imbalance <= 0) return 0; @@ -7501,11 +9619,9 @@ static int detach_tasks(struct lb_env *env) * We don't want to steal all, otherwise we may be treated likewise, * which could at worst lead to a livelock crash. */ - if (env->idle != CPU_NOT_IDLE && env->src_rq->nr_running <= 1) + if (env->idle && env->src_rq->nr_running <= 1) break; - p = list_last_entry(tasks, struct task_struct, se.group_node); - env->loop++; /* We've more or less seen every task there is, call it quits */ if (env->loop > env->loop_max) @@ -7513,29 +9629,71 @@ static int detach_tasks(struct lb_env *env) /* take a breather every nr_migrate tasks */ if (env->loop > env->loop_break) { - env->loop_break += sched_nr_migrate_break; + env->loop_break += SCHED_NR_MIGRATE_BREAK; env->flags |= LBF_NEED_BREAK; break; } + p = list_last_entry(tasks, struct task_struct, se.group_node); + if (!can_migrate_task(p, env)) goto next; - load = task_h_load(p); + switch (env->migration_type) { + case migrate_load: + /* + * Depending of the number of CPUs and tasks and the + * cgroup hierarchy, task_h_load() can return a null + * value. Make sure that env->imbalance decreases + * otherwise detach_tasks() will stop only after + * detaching up to loop_max tasks. + */ + load = max_t(unsigned long, task_h_load(p), 1); + + if (sched_feat(LB_MIN) && + load < 16 && !env->sd->nr_balance_failed) + goto next; - if (sched_feat(LB_MIN) && load < 16 && !env->sd->nr_balance_failed) - goto next; + /* + * Make sure that we don't migrate too much load. + * Nevertheless, let relax the constraint if + * scheduler fails to find a good waiting task to + * migrate. + */ + if (shr_bound(load, env->sd->nr_balance_failed) > env->imbalance) + goto next; - if ((load / 2) > env->imbalance) - goto next; + env->imbalance -= load; + break; + + case migrate_util: + util = task_util_est(p); + + if (shr_bound(util, env->sd->nr_balance_failed) > env->imbalance) + goto next; + + env->imbalance -= util; + break; + + case migrate_task: + env->imbalance--; + break; + + case migrate_misfit: + /* This is not a misfit task */ + if (task_fits_cpu(p, env->src_cpu)) + goto next; + + env->imbalance = 0; + break; + } detach_task(p, env); list_add(&p->se.group_node, &env->tasks); detached++; - env->imbalance -= load; -#ifdef CONFIG_PREEMPT +#ifdef CONFIG_PREEMPTION /* * NEWIDLE balancing is a source of latency, so preemptible * kernels will stop after the first task is detached to minimize @@ -7547,13 +9705,16 @@ static int detach_tasks(struct lb_env *env) /* * We only want to steal up to the prescribed amount of - * weighted load. + * load/util/tasks. */ if (env->imbalance <= 0) break; continue; next: + if (p->sched_task_hot) + schedstat_inc(p->stats.nr_failed_migrations_hot); + list_move(&p->se.group_node, tasks); } @@ -7572,12 +9733,11 @@ next: */ static void attach_task(struct rq *rq, struct task_struct *p) { - lockdep_assert_held(&rq->lock); + lockdep_assert_rq_held(rq); - BUG_ON(task_rq(p) != rq); + WARN_ON_ONCE(task_rq(p) != rq); activate_task(rq, p, ENQUEUE_NOCLOCK); - p->on_rq = TASK_ON_RQ_QUEUED; - check_preempt_curr(rq, p, 0); + wakeup_preempt(rq, p, 0); } /* @@ -7617,6 +9777,7 @@ static void attach_tasks(struct lb_env *env) rq_unlock(env->dst_rq, &rf); } +#ifdef CONFIG_NO_HZ_COMMON static inline bool cfs_rq_has_blocked(struct cfs_rq *cfs_rq) { if (cfs_rq->avg.load_avg) @@ -7630,71 +9791,97 @@ static inline bool cfs_rq_has_blocked(struct cfs_rq *cfs_rq) static inline bool others_have_blocked(struct rq *rq) { - if (READ_ONCE(rq->avg_rt.util_avg)) + if (cpu_util_rt(rq)) return true; - if (READ_ONCE(rq->avg_dl.util_avg)) + if (cpu_util_dl(rq)) return true; -#ifdef CONFIG_HAVE_SCHED_AVG_IRQ - if (READ_ONCE(rq->avg_irq.util_avg)) + if (hw_load_avg(rq)) + return true; + + if (cpu_util_irq(rq)) return true; -#endif return false; } -#ifdef CONFIG_FAIR_GROUP_SCHED +static inline void update_blocked_load_tick(struct rq *rq) +{ + WRITE_ONCE(rq->last_blocked_load_update_tick, jiffies); +} -static void update_blocked_averages(int cpu) +static inline void update_blocked_load_status(struct rq *rq, bool has_blocked) { - struct rq *rq = cpu_rq(cpu); - struct cfs_rq *cfs_rq; - const struct sched_class *curr_class; - struct rq_flags rf; - bool done = true; + if (!has_blocked) + rq->has_blocked_load = 0; +} +#else /* !CONFIG_NO_HZ_COMMON: */ +static inline bool cfs_rq_has_blocked(struct cfs_rq *cfs_rq) { return false; } +static inline bool others_have_blocked(struct rq *rq) { return false; } +static inline void update_blocked_load_tick(struct rq *rq) {} +static inline void update_blocked_load_status(struct rq *rq, bool has_blocked) {} +#endif /* !CONFIG_NO_HZ_COMMON */ - rq_lock_irqsave(rq, &rf); - update_rq_clock(rq); +static bool __update_blocked_others(struct rq *rq, bool *done) +{ + bool updated; + + /* + * update_load_avg() can call cpufreq_update_util(). Make sure that RT, + * DL and IRQ signals have been updated before updating CFS. + */ + updated = update_other_load_avgs(rq); + + if (others_have_blocked(rq)) + *done = false; + + return updated; +} + +#ifdef CONFIG_FAIR_GROUP_SCHED + +static bool __update_blocked_fair(struct rq *rq, bool *done) +{ + struct cfs_rq *cfs_rq, *pos; + bool decayed = false; + int cpu = cpu_of(rq); /* * Iterates the task_group tree in a bottom up fashion, see * list_add_leaf_cfs_rq() for details. */ - for_each_leaf_cfs_rq(rq, cfs_rq) { + for_each_leaf_cfs_rq_safe(rq, cfs_rq, pos) { struct sched_entity *se; - /* throttled entities do not contribute to load */ - if (throttled_hierarchy(cfs_rq)) - continue; + if (update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq)) { + update_tg_load_avg(cfs_rq); - if (update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq)) - update_tg_load_avg(cfs_rq, 0); + if (cfs_rq->nr_queued == 0) + update_idle_cfs_rq_clock_pelt(cfs_rq); + + if (cfs_rq == &rq->cfs) + decayed = true; + } /* Propagate pending load changes to the parent, if any: */ se = cfs_rq->tg->se[cpu]; if (se && !skip_blocked_update(se)) - update_load_avg(cfs_rq_of(se), se, 0); + update_load_avg(cfs_rq_of(se), se, UPDATE_TG); + + /* + * There can be a lot of idle CPU cgroups. Don't let fully + * decayed cfs_rqs linger on the list. + */ + if (cfs_rq_is_decayed(cfs_rq)) + list_del_leaf_cfs_rq(cfs_rq); /* Don't need periodic decay once load/util_avg are null */ if (cfs_rq_has_blocked(cfs_rq)) - done = false; + *done = false; } - curr_class = rq->curr->sched_class; - update_rt_rq_load_avg(rq_clock_task(rq), rq, curr_class == &rt_sched_class); - update_dl_rq_load_avg(rq_clock_task(rq), rq, curr_class == &dl_sched_class); - update_irq_load_avg(rq, 0); - /* Don't need periodic decay once load/util_avg are null */ - if (others_have_blocked(rq)) - done = false; - -#ifdef CONFIG_NO_HZ_COMMON - rq->last_blocked_load_update_tick = jiffies; - if (done) - rq->has_blocked_load = 0; -#endif - rq_unlock_irqrestore(rq, &rf); + return decayed; } /* @@ -7712,10 +9899,10 @@ static void update_cfs_rq_h_load(struct cfs_rq *cfs_rq) if (cfs_rq->last_h_load_update == now) return; - cfs_rq->h_load_next = NULL; + WRITE_ONCE(cfs_rq->h_load_next, NULL); for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - cfs_rq->h_load_next = se; + WRITE_ONCE(cfs_rq->h_load_next, se); if (cfs_rq->last_h_load_update == now) break; } @@ -7725,7 +9912,7 @@ static void update_cfs_rq_h_load(struct cfs_rq *cfs_rq) cfs_rq->last_h_load_update = now; } - while ((se = cfs_rq->h_load_next) != NULL) { + while ((se = READ_ONCE(cfs_rq->h_load_next)) != NULL) { load = cfs_rq->h_load; load = div64_ul(load * se->avg.load_avg, cfs_rq_load_avg(cfs_rq) + 1); @@ -7743,54 +9930,63 @@ static unsigned long task_h_load(struct task_struct *p) return div64_ul(p->se.avg.load_avg * cfs_rq->h_load, cfs_rq_load_avg(cfs_rq) + 1); } -#else -static inline void update_blocked_averages(int cpu) +#else /* !CONFIG_FAIR_GROUP_SCHED: */ +static bool __update_blocked_fair(struct rq *rq, bool *done) { - struct rq *rq = cpu_rq(cpu); struct cfs_rq *cfs_rq = &rq->cfs; - const struct sched_class *curr_class; - struct rq_flags rf; + bool decayed; - rq_lock_irqsave(rq, &rf); - update_rq_clock(rq); - update_cfs_rq_load_avg(cfs_rq_clock_task(cfs_rq), cfs_rq); + decayed = update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq); + if (cfs_rq_has_blocked(cfs_rq)) + *done = false; - curr_class = rq->curr->sched_class; - update_rt_rq_load_avg(rq_clock_task(rq), rq, curr_class == &rt_sched_class); - update_dl_rq_load_avg(rq_clock_task(rq), rq, curr_class == &dl_sched_class); - update_irq_load_avg(rq, 0); -#ifdef CONFIG_NO_HZ_COMMON - rq->last_blocked_load_update_tick = jiffies; - if (!cfs_rq_has_blocked(cfs_rq) && !others_have_blocked(rq)) - rq->has_blocked_load = 0; -#endif - rq_unlock_irqrestore(rq, &rf); + return decayed; } static unsigned long task_h_load(struct task_struct *p) { return p->se.avg.load_avg; } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ + +static void sched_balance_update_blocked_averages(int cpu) +{ + bool decayed = false, done = true; + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; + + rq_lock_irqsave(rq, &rf); + update_blocked_load_tick(rq); + update_rq_clock(rq); -/********** Helpers for find_busiest_group ************************/ + decayed |= __update_blocked_others(rq, &done); + decayed |= __update_blocked_fair(rq, &done); + + update_blocked_load_status(rq, !done); + if (decayed) + cpufreq_update_util(rq, 0); + rq_unlock_irqrestore(rq, &rf); +} + +/********** Helpers for sched_balance_find_src_group ************************/ /* - * sg_lb_stats - stats of a sched_group required for load_balancing + * sg_lb_stats - stats of a sched_group required for load-balancing: */ struct sg_lb_stats { - unsigned long avg_load; /*Avg load across the CPUs of the group */ - unsigned long group_load; /* Total load over the CPUs of the group */ - unsigned long sum_weighted_load; /* Weighted load of group's tasks */ - unsigned long load_per_task; - unsigned long group_capacity; - unsigned long group_util; /* Total utilization of the group */ - unsigned int sum_nr_running; /* Nr tasks running in the group */ - unsigned int idle_cpus; + unsigned long avg_load; /* Avg load over the CPUs of the group */ + unsigned long group_load; /* Total load over the CPUs of the group */ + unsigned long group_capacity; /* Capacity over the CPUs of the group */ + unsigned long group_util; /* Total utilization over the CPUs of the group */ + unsigned long group_runnable; /* Total runnable time over the CPUs of the group */ + unsigned int sum_nr_running; /* Nr of all tasks running in the group */ + unsigned int sum_h_nr_running; /* Nr of CFS tasks running in the group */ + unsigned int idle_cpus; /* Nr of idle CPUs in the group */ unsigned int group_weight; enum group_type group_type; - int group_no_capacity; - unsigned long group_misfit_task_load; /* A CPU has a task too big for its capacity */ + unsigned int group_asym_packing; /* Tasks should be moved to preferred CPU */ + unsigned int group_smt_balance; /* Task on busy SMT be moved */ + unsigned long group_misfit_task_load; /* A CPU has a task too big for its capacity */ #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running; @@ -7798,19 +9994,18 @@ struct sg_lb_stats { }; /* - * sd_lb_stats - Structure to store the statistics of a sched_domain - * during load balancing. + * sd_lb_stats - stats of a sched_domain required for load-balancing: */ struct sd_lb_stats { - struct sched_group *busiest; /* Busiest group in this sd */ - struct sched_group *local; /* Local group in this sd */ - unsigned long total_running; - unsigned long total_load; /* Total load of all groups in sd */ - unsigned long total_capacity; /* Total capacity of all groups in sd */ - unsigned long avg_load; /* Average load across all groups in sd */ - - struct sg_lb_stats busiest_stat;/* Statistics of the busiest group */ - struct sg_lb_stats local_stat; /* Statistics of the local group */ + struct sched_group *busiest; /* Busiest group in this sd */ + struct sched_group *local; /* Local group in this sd */ + unsigned long total_load; /* Total load of all groups in sd */ + unsigned long total_capacity; /* Total capacity of all groups in sd */ + unsigned long avg_load; /* Average load across all groups in sd */ + unsigned int prefer_sibling; /* Tasks should go to sibling first */ + + struct sg_lb_stats busiest_stat; /* Statistics of the busiest group */ + struct sg_lb_stats local_stat; /* Statistics of the local group */ }; static inline void init_sd_lb_stats(struct sd_lb_stats *sds) @@ -7818,55 +10013,26 @@ static inline void init_sd_lb_stats(struct sd_lb_stats *sds) /* * Skimp on the clearing to avoid duplicate work. We can avoid clearing * local_stat because update_sg_lb_stats() does a full clear/assignment. - * We must however clear busiest_stat::avg_load because - * update_sd_pick_busiest() reads this before assignment. + * We must however set busiest_stat::group_type and + * busiest_stat::idle_cpus to the worst busiest group because + * update_sd_pick_busiest() reads these before assignment. */ *sds = (struct sd_lb_stats){ .busiest = NULL, .local = NULL, - .total_running = 0UL, .total_load = 0UL, .total_capacity = 0UL, .busiest_stat = { - .avg_load = 0UL, - .sum_nr_running = 0, - .group_type = group_other, + .idle_cpus = UINT_MAX, + .group_type = group_has_spare, }, }; } -/** - * get_sd_load_idx - Obtain the load index for a given sched domain. - * @sd: The sched_domain whose load_idx is to be obtained. - * @idle: The idle status of the CPU for whose sd load_idx is obtained. - * - * Return: The load index. - */ -static inline int get_sd_load_idx(struct sched_domain *sd, - enum cpu_idle_type idle) -{ - int load_idx; - - switch (idle) { - case CPU_NOT_IDLE: - load_idx = sd->busy_idx; - break; - - case CPU_NEWLY_IDLE: - load_idx = sd->newidle_idx; - break; - default: - load_idx = sd->idle_idx; - break; - } - - return load_idx; -} - -static unsigned long scale_rt_capacity(struct sched_domain *sd, int cpu) +static unsigned long scale_rt_capacity(int cpu) { + unsigned long max = get_actual_cpu_capacity(cpu); struct rq *rq = cpu_rq(cpu); - unsigned long max = arch_scale_cpu_capacity(sd, cpu); unsigned long used, free; unsigned long irq; @@ -7875,8 +10041,12 @@ static unsigned long scale_rt_capacity(struct sched_domain *sd, int cpu) if (unlikely(irq >= max)) return 1; - used = READ_ONCE(rq->avg_rt.util_avg); - used += READ_ONCE(rq->avg_dl.util_avg); + /* + * avg_rt.util_avg and avg_dl.util_avg track binary signals + * (running and not running) with weights 0 and 1024 respectively. + */ + used = cpu_util_rt(rq); + used += cpu_util_dl(rq); if (unlikely(used >= max)) return 1; @@ -7888,15 +10058,15 @@ static unsigned long scale_rt_capacity(struct sched_domain *sd, int cpu) static void update_cpu_capacity(struct sched_domain *sd, int cpu) { - unsigned long capacity = scale_rt_capacity(sd, cpu); + unsigned long capacity = scale_rt_capacity(cpu); struct sched_group *sdg = sd->groups; - cpu_rq(cpu)->cpu_capacity_orig = arch_scale_cpu_capacity(sd, cpu); - if (!capacity) capacity = 1; cpu_rq(cpu)->cpu_capacity = capacity; + trace_sched_cpu_capacity_tp(cpu_rq(cpu)); + sdg->sgc->capacity = capacity; sdg->sgc->min_capacity = capacity; sdg->sgc->max_capacity = capacity; @@ -7922,40 +10092,22 @@ void update_group_capacity(struct sched_domain *sd, int cpu) min_capacity = ULONG_MAX; max_capacity = 0; - if (child->flags & SD_OVERLAP) { + if (child->flags & SD_NUMA) { /* - * SD_OVERLAP domains cannot assume that child groups + * SD_NUMA domains cannot assume that child groups * span the current group. */ for_each_cpu(cpu, sched_group_span(sdg)) { - struct sched_group_capacity *sgc; - struct rq *rq = cpu_rq(cpu); - - /* - * build_sched_domains() -> init_sched_groups_capacity() - * gets here before we've attached the domains to the - * runqueues. - * - * Use capacity_of(), which is set irrespective of domains - * in update_cpu_capacity(). - * - * This avoids capacity from being 0 and - * causing divide-by-zero issues on boot. - */ - if (unlikely(!rq->sd)) { - capacity += capacity_of(cpu); - } else { - sgc = rq->sd->groups->sgc; - capacity += sgc->capacity; - } + unsigned long cpu_cap = capacity_of(cpu); - min_capacity = min(capacity, min_capacity); - max_capacity = max(capacity, max_capacity); + capacity += cpu_cap; + min_capacity = min(cpu_cap, min_capacity); + max_capacity = max(cpu_cap, max_capacity); } } else { /* - * !SD_OVERLAP domains can assume that child groups + * !SD_NUMA domains can assume that child groups * span the current group. */ @@ -7984,12 +10136,18 @@ static inline int check_cpu_capacity(struct rq *rq, struct sched_domain *sd) { return ((rq->cpu_capacity * sd->imbalance_pct) < - (rq->cpu_capacity_orig * 100)); + (arch_scale_cpu_capacity(cpu_of(rq)) * 100)); +} + +/* Check if the rq has a misfit task */ +static inline bool check_misfit_status(struct rq *rq) +{ + return rq->misfit_task_load; } /* * Group imbalance indicates (and tries to solve) the problem where balancing - * groups is inadequate due to ->cpus_allowed constraints. + * groups is inadequate due to ->cpus_ptr constraints. * * Imagine a situation of two groups of 4 CPUs each and 4 tasks each with a * cpumask covering 1 CPU of the first group and 3 CPUs of the second group. @@ -8008,7 +10166,7 @@ check_cpu_capacity(struct rq *rq, struct sched_domain *sd) * * When this is so detected; this group becomes a candidate for busiest; see * update_sd_pick_busiest(). And calculate_imbalance() and - * find_busiest_group() avoid some of the usual balance conditions to allow it + * sched_balance_find_src_group() avoid some of the usual balance conditions to allow it * to create an effective group imbalance. * * This is a somewhat tricky proposition since the next run might not find the @@ -8024,7 +10182,7 @@ static inline int sg_imbalanced(struct sched_group *group) /* * group_has_capacity returns true if the group has spare capacity that could * be used by some tasks. - * We consider that a group has spare capacity if the * number of task is + * We consider that a group has spare capacity if the number of task is * smaller than the number of CPUs or if the utilization is lower than the * available capacity for CFS tasks. * For the latter, we use a threshold to stabilize the state, to take into @@ -8034,13 +10192,17 @@ static inline int sg_imbalanced(struct sched_group *group) * any benefit for the load balance. */ static inline bool -group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs) +group_has_capacity(unsigned int imbalance_pct, struct sg_lb_stats *sgs) { if (sgs->sum_nr_running < sgs->group_weight) return true; + if ((sgs->group_capacity * imbalance_pct) < + (sgs->group_runnable * 100)) + return false; + if ((sgs->group_capacity * 100) > - (sgs->group_util * env->sd->imbalance_pct)) + (sgs->group_util * imbalance_pct)) return true; return false; @@ -8055,149 +10217,277 @@ group_has_capacity(struct lb_env *env, struct sg_lb_stats *sgs) * false. */ static inline bool -group_is_overloaded(struct lb_env *env, struct sg_lb_stats *sgs) +group_is_overloaded(unsigned int imbalance_pct, struct sg_lb_stats *sgs) { if (sgs->sum_nr_running <= sgs->group_weight) return false; if ((sgs->group_capacity * 100) < - (sgs->group_util * env->sd->imbalance_pct)) + (sgs->group_util * imbalance_pct)) return true; - return false; -} - -/* - * group_smaller_min_cpu_capacity: Returns true if sched_group sg has smaller - * per-CPU capacity than sched_group ref. - */ -static inline bool -group_smaller_min_cpu_capacity(struct sched_group *sg, struct sched_group *ref) -{ - return sg->sgc->min_capacity * capacity_margin < - ref->sgc->min_capacity * 1024; -} + if ((sgs->group_capacity * imbalance_pct) < + (sgs->group_runnable * 100)) + return true; -/* - * group_smaller_max_cpu_capacity: Returns true if sched_group sg has smaller - * per-CPU capacity_orig than sched_group ref. - */ -static inline bool -group_smaller_max_cpu_capacity(struct sched_group *sg, struct sched_group *ref) -{ - return sg->sgc->max_capacity * capacity_margin < - ref->sgc->max_capacity * 1024; + return false; } static inline enum -group_type group_classify(struct sched_group *group, +group_type group_classify(unsigned int imbalance_pct, + struct sched_group *group, struct sg_lb_stats *sgs) { - if (sgs->group_no_capacity) + if (group_is_overloaded(imbalance_pct, sgs)) return group_overloaded; if (sg_imbalanced(group)) return group_imbalanced; + if (sgs->group_asym_packing) + return group_asym_packing; + + if (sgs->group_smt_balance) + return group_smt_balance; + if (sgs->group_misfit_task_load) return group_misfit_task; - return group_other; + if (!group_has_capacity(imbalance_pct, sgs)) + return group_fully_busy; + + return group_has_spare; } -static bool update_nohz_stats(struct rq *rq, bool force) +/** + * sched_use_asym_prio - Check whether asym_packing priority must be used + * @sd: The scheduling domain of the load balancing + * @cpu: A CPU + * + * Always use CPU priority when balancing load between SMT siblings. When + * balancing load between cores, it is not sufficient that @cpu is idle. Only + * use CPU priority if the whole core is idle. + * + * Returns: True if the priority of @cpu must be followed. False otherwise. + */ +static bool sched_use_asym_prio(struct sched_domain *sd, int cpu) { -#ifdef CONFIG_NO_HZ_COMMON - unsigned int cpu = rq->cpu; + if (!(sd->flags & SD_ASYM_PACKING)) + return false; - if (!rq->has_blocked_load) + if (!sched_smt_active()) + return true; + + return sd->flags & SD_SHARE_CPUCAPACITY || is_core_idle(cpu); +} + +static inline bool sched_asym(struct sched_domain *sd, int dst_cpu, int src_cpu) +{ + /* + * First check if @dst_cpu can do asym_packing load balance. Only do it + * if it has higher priority than @src_cpu. + */ + return sched_use_asym_prio(sd, dst_cpu) && + sched_asym_prefer(dst_cpu, src_cpu); +} + +/** + * sched_group_asym - Check if the destination CPU can do asym_packing balance + * @env: The load balancing environment + * @sgs: Load-balancing statistics of the candidate busiest group + * @group: The candidate busiest group + * + * @env::dst_cpu can do asym_packing if it has higher priority than the + * preferred CPU of @group. + * + * Return: true if @env::dst_cpu can do with asym_packing load balance. False + * otherwise. + */ +static inline bool +sched_group_asym(struct lb_env *env, struct sg_lb_stats *sgs, struct sched_group *group) +{ + /* + * CPU priorities do not make sense for SMT cores with more than one + * busy sibling. + */ + if ((group->flags & SD_SHARE_CPUCAPACITY) && + (sgs->group_weight - sgs->idle_cpus != 1)) return false; - if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask)) + return sched_asym(env->sd, env->dst_cpu, READ_ONCE(group->asym_prefer_cpu)); +} + +/* One group has more than one SMT CPU while the other group does not */ +static inline bool smt_vs_nonsmt_groups(struct sched_group *sg1, + struct sched_group *sg2) +{ + if (!sg1 || !sg2) return false; - if (!force && !time_after(jiffies, rq->last_blocked_load_update_tick)) - return true; + return (sg1->flags & SD_SHARE_CPUCAPACITY) != + (sg2->flags & SD_SHARE_CPUCAPACITY); +} - update_blocked_averages(cpu); +static inline bool smt_balance(struct lb_env *env, struct sg_lb_stats *sgs, + struct sched_group *group) +{ + if (!env->idle) + return false; + + /* + * For SMT source group, it is better to move a task + * to a CPU that doesn't have multiple tasks sharing its CPU capacity. + * Note that if a group has a single SMT, SD_SHARE_CPUCAPACITY + * will not be on. + */ + if (group->flags & SD_SHARE_CPUCAPACITY && + sgs->sum_h_nr_running > 1) + return true; - return rq->has_blocked_load; -#else return false; -#endif +} + +static inline long sibling_imbalance(struct lb_env *env, + struct sd_lb_stats *sds, + struct sg_lb_stats *busiest, + struct sg_lb_stats *local) +{ + int ncores_busiest, ncores_local; + long imbalance; + + if (!env->idle || !busiest->sum_nr_running) + return 0; + + ncores_busiest = sds->busiest->cores; + ncores_local = sds->local->cores; + + if (ncores_busiest == ncores_local) { + imbalance = busiest->sum_nr_running; + lsub_positive(&imbalance, local->sum_nr_running); + return imbalance; + } + + /* Balance such that nr_running/ncores ratio are same on both groups */ + imbalance = ncores_local * busiest->sum_nr_running; + lsub_positive(&imbalance, ncores_busiest * local->sum_nr_running); + /* Normalize imbalance and do rounding on normalization */ + imbalance = 2 * imbalance + ncores_local + ncores_busiest; + imbalance /= ncores_local + ncores_busiest; + + /* Take advantage of resource in an empty sched group */ + if (imbalance <= 1 && local->sum_nr_running == 0 && + busiest->sum_nr_running > 1) + imbalance = 2; + + return imbalance; +} + +static inline bool +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_runnable != 1) + return false; + + return check_cpu_capacity(rq, sd); } /** * update_sg_lb_stats - Update sched_group's statistics for load balancing. * @env: The load balancing environment. + * @sds: Load-balancing data with statistics of the local group. * @group: sched_group whose statistics are to be updated. * @sgs: variable to hold the statistics for this group. - * @sg_status: Holds flag indicating the status of the sched_group + * @sg_overloaded: sched_group is overloaded + * @sg_overutilized: sched_group is overutilized */ static inline void update_sg_lb_stats(struct lb_env *env, + struct sd_lb_stats *sds, struct sched_group *group, struct sg_lb_stats *sgs, - int *sg_status) + bool *sg_overloaded, + bool *sg_overutilized) { - int local_group = cpumask_test_cpu(env->dst_cpu, sched_group_span(group)); - int load_idx = get_sd_load_idx(env->sd, env->idle); - unsigned long load; - int i, nr_running; + int i, nr_running, local_group, sd_flags = env->sd->flags; + bool balancing_at_rd = !env->sd->parent; memset(sgs, 0, sizeof(*sgs)); + local_group = group == sds->local; + for_each_cpu_and(i, sched_group_span(group), env->cpus) { struct rq *rq = cpu_rq(i); - - if ((env->flags & LBF_NOHZ_STATS) && update_nohz_stats(rq, false)) - env->flags |= LBF_NOHZ_AGAIN; - - /* Bias balancing toward CPUs of our domain: */ - if (local_group) - load = target_load(i, load_idx); - else - load = source_load(i, load_idx); + unsigned long load = cpu_load(rq); sgs->group_load += load; - sgs->group_util += cpu_util(i); - sgs->sum_nr_running += rq->cfs.h_nr_running; + sgs->group_util += cpu_util_cfs(i); + sgs->group_runnable += cpu_runnable(rq); + sgs->sum_h_nr_running += rq->cfs.h_nr_runnable; nr_running = rq->nr_running; - if (nr_running > 1) - *sg_status |= SG_OVERLOAD; + sgs->sum_nr_running += nr_running; if (cpu_overutilized(i)) - *sg_status |= SG_OVERUTILIZED; + *sg_overutilized = 1; -#ifdef CONFIG_NUMA_BALANCING - sgs->nr_numa_running += rq->nr_numa_running; - sgs->nr_preferred_running += rq->nr_preferred_running; -#endif - sgs->sum_weighted_load += weighted_cpuload(rq); /* * No need to call idle_cpu() if nr_running is not 0 */ - if (!nr_running && idle_cpu(i)) + if (!nr_running && idle_cpu(i)) { sgs->idle_cpus++; + /* Idle cpu can't have misfit task */ + continue; + } - if (env->sd->flags & SD_ASYM_CPUCAPACITY && - sgs->group_misfit_task_load < rq->misfit_task_load) { - sgs->group_misfit_task_load = rq->misfit_task_load; - *sg_status |= SG_OVERLOAD; + /* 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 (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; + *sg_overloaded = 1; + } + } else if (env->idle && sched_reduced_capacity(rq, env->sd)) { + /* Check for a task running on a CPU with reduced capacity */ + if (sgs->group_misfit_task_load < load) + sgs->group_misfit_task_load = load; } } - /* Adjust by relative CPU capacity of the group */ sgs->group_capacity = group->sgc->capacity; - sgs->avg_load = (sgs->group_load*SCHED_CAPACITY_SCALE) / sgs->group_capacity; - - if (sgs->sum_nr_running) - sgs->load_per_task = sgs->sum_weighted_load / sgs->sum_nr_running; sgs->group_weight = group->group_weight; - sgs->group_no_capacity = group_is_overloaded(env, sgs); - sgs->group_type = group_classify(group, sgs); + /* Check if dst CPU is idle and preferred to this group */ + if (!local_group && env->idle && sgs->sum_h_nr_running && + sched_group_asym(env, sgs, group)) + sgs->group_asym_packing = 1; + + /* Check for loaded SMT group to be balanced to dst CPU */ + if (!local_group && smt_balance(env, sgs, group)) + sgs->group_smt_balance = 1; + + sgs->group_type = group_classify(env->sd->imbalance_pct, group, sgs); + + /* Computing avg_load makes sense only when group is overloaded */ + if (sgs->group_type == group_overloaded) + sgs->avg_load = (sgs->group_load * SCHED_CAPACITY_SCALE) / + sgs->group_capacity; } /** @@ -8220,15 +10510,20 @@ static bool update_sd_pick_busiest(struct lb_env *env, { struct sg_lb_stats *busiest = &sds->busiest_stat; + /* Make sure that there is at least one task to pull */ + if (!sgs->sum_h_nr_running) + return false; + /* * Don't try to pull misfit tasks we can't help. * We can use max_capacity here as reduction in capacity on some * CPUs in the group should either be possible to resolve * internally or be covered by avg_load imbalance (eventually). */ - if (sgs->group_type == group_misfit_task && - (!group_smaller_max_cpu_capacity(sg, sds->local) || - !group_has_capacity(env, &sds->local_stat))) + if ((env->sd->flags & SD_ASYM_CPUCAPACITY) && + (sgs->group_type == group_misfit_task) && + (!capacity_greater(capacity_of(env->dst_cpu), sg->sgc->max_capacity) || + sds->local_stat.group_type != group_has_spare)) return false; if (sgs->group_type > busiest->group_type) @@ -8237,62 +10532,122 @@ static bool update_sd_pick_busiest(struct lb_env *env, if (sgs->group_type < busiest->group_type) return false; - if (sgs->avg_load <= busiest->avg_load) - return false; - - if (!(env->sd->flags & SD_ASYM_CPUCAPACITY)) - goto asym_packing; - /* - * Candidate sg has no more than one task per CPU and - * has higher per-CPU capacity. Migrating tasks to less - * capable CPUs may harm throughput. Maximize throughput, - * power/energy consequences are not considered. + * The candidate and the current busiest group are the same type of + * group. Let check which one is the busiest according to the type. */ - if (sgs->sum_nr_running <= sgs->group_weight && - group_smaller_min_cpu_capacity(sds->local, sg)) - return false; - /* - * If we have more than one misfit sg go with the biggest misfit. - */ - if (sgs->group_type == group_misfit_task && - sgs->group_misfit_task_load < busiest->group_misfit_task_load) + switch (sgs->group_type) { + case group_overloaded: + /* Select the overloaded group with highest avg_load. */ + return sgs->avg_load > busiest->avg_load; + + case group_imbalanced: + /* + * Select the 1st imbalanced group as we don't have any way to + * choose one more than another. + */ return false; -asym_packing: - /* This is the busiest node in its class. */ - if (!(env->sd->flags & SD_ASYM_PACKING)) - return true; + case group_asym_packing: + /* Prefer to move from lowest priority CPU's work */ + return sched_asym_prefer(READ_ONCE(sds->busiest->asym_prefer_cpu), + READ_ONCE(sg->asym_prefer_cpu)); - /* No ASYM_PACKING if target CPU is already busy */ - if (env->idle == CPU_NOT_IDLE) - return true; - /* - * ASYM_PACKING needs to move all the work to the highest - * prority CPUs in the group, therefore mark all groups - * of lower priority than ourself as busy. - */ - if (sgs->sum_nr_running && - sched_asym_prefer(env->dst_cpu, sg->asym_prefer_cpu)) { - if (!sds->busiest) - return true; + case group_misfit_task: + /* + * If we have more than one misfit sg go with the biggest + * misfit. + */ + return sgs->group_misfit_task_load > busiest->group_misfit_task_load; - /* Prefer to move from lowest priority CPU's work */ - if (sched_asym_prefer(sds->busiest->asym_prefer_cpu, - sg->asym_prefer_cpu)) - return true; + case group_smt_balance: + /* + * Check if we have spare CPUs on either SMT group to + * choose has spare or fully busy handling. + */ + if (sgs->idle_cpus != 0 || busiest->idle_cpus != 0) + goto has_spare; + + fallthrough; + + case group_fully_busy: + /* + * Select the fully busy group with highest avg_load. In + * theory, there is no need to pull task from such kind of + * group because tasks have all compute capacity that they need + * but we can still improve the overall throughput by reducing + * contention when accessing shared HW resources. + * + * XXX for now avg_load is not computed and always 0 so we + * select the 1st one, except if @sg is composed of SMT + * siblings. + */ + + if (sgs->avg_load < busiest->avg_load) + return false; + + if (sgs->avg_load == busiest->avg_load) { + /* + * SMT sched groups need more help than non-SMT groups. + * If @sg happens to also be SMT, either choice is good. + */ + if (sds->busiest->flags & SD_SHARE_CPUCAPACITY) + return false; + } + + break; + + case group_has_spare: + /* + * Do not pick sg with SMT CPUs over sg with pure CPUs, + * as we do not want to pull task off SMT core with one task + * and make the core idle. + */ + if (smt_vs_nonsmt_groups(sds->busiest, sg)) { + if (sg->flags & SD_SHARE_CPUCAPACITY && sgs->sum_h_nr_running <= 1) + return false; + else + return true; + } +has_spare: + + /* + * Select not overloaded group with lowest number of idle CPUs + * and highest number of running tasks. We could also compare + * the spare capacity which is more stable but it can end up + * that the group has less spare capacity but finally more idle + * CPUs which means less opportunity to pull tasks. + */ + if (sgs->idle_cpus > busiest->idle_cpus) + return false; + else if ((sgs->idle_cpus == busiest->idle_cpus) && + (sgs->sum_nr_running <= busiest->sum_nr_running)) + return false; + + break; } - return false; + /* + * Candidate sg has no more than one task per CPU and has higher + * per-CPU capacity. Migrating tasks to less capable CPUs may harm + * throughput. Maximize throughput, power/energy consequences are not + * considered. + */ + if ((env->sd->flags & SD_ASYM_CPUCAPACITY) && + (sgs->group_type <= group_fully_busy) && + (capacity_greater(sg->sgc->min_capacity, capacity_of(env->dst_cpu)))) + return false; + + return true; } #ifdef CONFIG_NUMA_BALANCING static inline enum fbq_type fbq_classify_group(struct sg_lb_stats *sgs) { - if (sgs->sum_nr_running > sgs->nr_numa_running) + if (sgs->sum_h_nr_running > sgs->nr_numa_running) return regular; - if (sgs->sum_nr_running > sgs->nr_preferred_running) + if (sgs->sum_h_nr_running > sgs->nr_preferred_running) return remote; return all; } @@ -8305,7 +10660,7 @@ static inline enum fbq_type fbq_classify_rq(struct rq *rq) return remote; return all; } -#else +#else /* !CONFIG_NUMA_BALANCING: */ static inline enum fbq_type fbq_classify_group(struct sg_lb_stats *sgs) { return all; @@ -8315,217 +10670,477 @@ static inline enum fbq_type fbq_classify_rq(struct rq *rq) { return regular; } -#endif /* CONFIG_NUMA_BALANCING */ +#endif /* !CONFIG_NUMA_BALANCING */ + + +struct sg_lb_stats; + +/* + * task_running_on_cpu - return 1 if @p is running on @cpu. + */ + +static unsigned int task_running_on_cpu(int cpu, struct task_struct *p) +{ + /* Task has no contribution or is new */ + if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) + return 0; + + if (task_on_rq_queued(p)) + return 1; + + return 0; +} /** - * update_sd_lb_stats - Update sched_domain's statistics for load balancing. - * @env: The load balancing environment. - * @sds: variable to hold the statistics for this sched_domain. + * idle_cpu_without - would a given CPU be idle without p ? + * @cpu: the processor on which idleness is tested. + * @p: task which should be ignored. + * + * Return: 1 if the CPU would be idle. 0 otherwise. */ -static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sds) +static int idle_cpu_without(int cpu, struct task_struct *p) { - struct sched_domain *child = env->sd->child; - struct sched_group *sg = env->sd->groups; - struct sg_lb_stats *local = &sds->local_stat; - struct sg_lb_stats tmp_sgs; - bool prefer_sibling = child && child->flags & SD_PREFER_SIBLING; - int sg_status = 0; + struct rq *rq = cpu_rq(cpu); -#ifdef CONFIG_NO_HZ_COMMON - if (env->idle == CPU_NEWLY_IDLE && READ_ONCE(nohz.has_blocked)) - env->flags |= LBF_NOHZ_STATS; -#endif + if (rq->curr != rq->idle && rq->curr != p) + return 0; + + /* + * rq->nr_running can't be used but an updated version without the + * impact of p on cpu must be used instead. The updated nr_running + * be computed and tested before calling idle_cpu_without(). + */ + + if (rq->ttwu_pending) + return 0; + + return 1; +} + +/* + * update_sg_wakeup_stats - Update sched_group's statistics for wakeup. + * @sd: The sched_domain level to look for idlest group. + * @group: sched_group whose statistics are to be updated. + * @sgs: variable to hold the statistics for this group. + * @p: The task for which we look for the idlest group/CPU. + */ +static inline void update_sg_wakeup_stats(struct sched_domain *sd, + struct sched_group *group, + struct sg_lb_stats *sgs, + struct task_struct *p) +{ + int i, nr_running; + + memset(sgs, 0, sizeof(*sgs)); + + /* Assume that task can't fit any CPU of the group */ + if (sd->flags & SD_ASYM_CPUCAPACITY) + sgs->group_misfit_task_load = 1; + + for_each_cpu_and(i, sched_group_span(group), p->cpus_ptr) { + struct rq *rq = cpu_rq(i); + unsigned int local; + + sgs->group_load += cpu_load_without(rq, p); + 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_runnable - local; + + nr_running = rq->nr_running - local; + sgs->sum_nr_running += nr_running; + + /* + * No need to call idle_cpu_without() if nr_running is not 0 + */ + if (!nr_running && idle_cpu_without(i, p)) + sgs->idle_cpus++; + + /* Check if task fits in the CPU */ + if (sd->flags & SD_ASYM_CPUCAPACITY && + sgs->group_misfit_task_load && + task_fits_cpu(p, i)) + sgs->group_misfit_task_load = 0; + + } + + sgs->group_capacity = group->sgc->capacity; + + sgs->group_weight = group->group_weight; + + sgs->group_type = group_classify(sd->imbalance_pct, group, sgs); + + /* + * Computing avg_load makes sense only when group is fully busy or + * overloaded + */ + if (sgs->group_type == group_fully_busy || + sgs->group_type == group_overloaded) + sgs->avg_load = (sgs->group_load * SCHED_CAPACITY_SCALE) / + sgs->group_capacity; +} + +static bool update_pick_idlest(struct sched_group *idlest, + struct sg_lb_stats *idlest_sgs, + struct sched_group *group, + struct sg_lb_stats *sgs) +{ + if (sgs->group_type < idlest_sgs->group_type) + return true; + + if (sgs->group_type > idlest_sgs->group_type) + return false; + + /* + * The candidate and the current idlest group are the same type of + * group. Let check which one is the idlest according to the type. + */ + + switch (sgs->group_type) { + case group_overloaded: + case group_fully_busy: + /* Select the group with lowest avg_load. */ + if (idlest_sgs->avg_load <= sgs->avg_load) + return false; + break; + + case group_imbalanced: + case group_asym_packing: + case group_smt_balance: + /* Those types are not used in the slow wakeup path */ + return false; + + case group_misfit_task: + /* Select group with the highest max capacity */ + if (idlest->sgc->max_capacity >= group->sgc->max_capacity) + return false; + break; + + case group_has_spare: + /* Select group with most idle CPUs */ + if (idlest_sgs->idle_cpus > sgs->idle_cpus) + return false; + + /* Select group with lowest group_util */ + if (idlest_sgs->idle_cpus == sgs->idle_cpus && + idlest_sgs->group_util <= sgs->group_util) + return false; + + break; + } + + return true; +} + +/* + * sched_balance_find_dst_group() finds and returns the least busy CPU group within the + * domain. + * + * Assumes p is allowed on at least one CPU in sd. + */ +static struct sched_group * +sched_balance_find_dst_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) +{ + struct sched_group *idlest = NULL, *local = NULL, *group = sd->groups; + struct sg_lb_stats local_sgs, tmp_sgs; + struct sg_lb_stats *sgs; + unsigned long imbalance; + struct sg_lb_stats idlest_sgs = { + .avg_load = UINT_MAX, + .group_type = group_overloaded, + }; do { - struct sg_lb_stats *sgs = &tmp_sgs; int local_group; - local_group = cpumask_test_cpu(env->dst_cpu, sched_group_span(sg)); + /* Skip over this group if it has no CPUs allowed */ + if (!cpumask_intersects(sched_group_span(group), + p->cpus_ptr)) + continue; + + /* Skip over this group if no cookie matched */ + if (!sched_group_cookie_match(cpu_rq(this_cpu), p, group)) + continue; + + local_group = cpumask_test_cpu(this_cpu, + sched_group_span(group)); + if (local_group) { - sds->local = sg; - sgs = local; + sgs = &local_sgs; + local = group; + } else { + sgs = &tmp_sgs; + } - if (env->idle != CPU_NEWLY_IDLE || - time_after_eq(jiffies, sg->sgc->next_update)) - update_group_capacity(env->sd, env->dst_cpu); + update_sg_wakeup_stats(sd, group, sgs, p); + + if (!local_group && update_pick_idlest(idlest, &idlest_sgs, group, sgs)) { + idlest = group; + idlest_sgs = *sgs; } - update_sg_lb_stats(env, sg, sgs, &sg_status); + } while (group = group->next, group != sd->groups); - if (local_group) - goto next_group; + + /* There is no idlest group to push tasks to */ + if (!idlest) + return NULL; + + /* The local group has been skipped because of CPU affinity */ + if (!local) + return idlest; + + /* + * If the local group is idler than the selected idlest group + * don't try and push the task. + */ + if (local_sgs.group_type < idlest_sgs.group_type) + return NULL; + + /* + * If the local group is busier than the selected idlest group + * try and push the task. + */ + if (local_sgs.group_type > idlest_sgs.group_type) + return idlest; + + switch (local_sgs.group_type) { + case group_overloaded: + case group_fully_busy: + + /* Calculate allowed imbalance based on load */ + imbalance = scale_load_down(NICE_0_LOAD) * + (sd->imbalance_pct-100) / 100; /* - * In case the child domain prefers tasks go to siblings - * first, lower the sg capacity so that we'll try - * and move all the excess tasks away. We lower the capacity - * of a group only if the local group has the capacity to fit - * these excess tasks. The extra check prevents the case where - * you always pull from the heaviest group when it is already - * under-utilized (possible with a large weight task outweighs - * the tasks on the system). + * When comparing groups across NUMA domains, it's possible for + * the local domain to be very lightly loaded relative to the + * remote domains but "imbalance" skews the comparison making + * remote CPUs look much more favourable. When considering + * cross-domain, add imbalance to the load on the remote node + * and consider staying local. */ - if (prefer_sibling && sds->local && - group_has_capacity(env, local) && - (sgs->sum_nr_running > local->sum_nr_running + 1)) { - sgs->group_no_capacity = 1; - sgs->group_type = group_classify(sg, sgs); - } - if (update_sd_pick_busiest(env, sds, sg, sgs)) { - sds->busiest = sg; - sds->busiest_stat = *sgs; - } + if ((sd->flags & SD_NUMA) && + ((idlest_sgs.avg_load + imbalance) >= local_sgs.avg_load)) + return NULL; -next_group: - /* Now, start updating sd_lb_stats */ - sds->total_running += sgs->sum_nr_running; - sds->total_load += sgs->group_load; - sds->total_capacity += sgs->group_capacity; + /* + * If the local group is less loaded than the selected + * idlest group don't try and push any tasks. + */ + if (idlest_sgs.avg_load >= (local_sgs.avg_load + imbalance)) + return NULL; - sg = sg->next; - } while (sg != env->sd->groups); + if (100 * local_sgs.avg_load <= sd->imbalance_pct * idlest_sgs.avg_load) + return NULL; + break; -#ifdef CONFIG_NO_HZ_COMMON - if ((env->flags & LBF_NOHZ_AGAIN) && - cpumask_subset(nohz.idle_cpus_mask, sched_domain_span(env->sd))) { + case group_imbalanced: + case group_asym_packing: + case group_smt_balance: + /* Those type are not used in the slow wakeup path */ + return NULL; - WRITE_ONCE(nohz.next_blocked, - jiffies + msecs_to_jiffies(LOAD_AVG_PERIOD)); - } -#endif + case group_misfit_task: + /* Select group with the highest max capacity */ + if (local->sgc->max_capacity >= idlest->sgc->max_capacity) + return NULL; + break; - if (env->sd->flags & SD_NUMA) - env->fbq_type = fbq_classify_group(&sds->busiest_stat); + case group_has_spare: +#ifdef CONFIG_NUMA + if (sd->flags & SD_NUMA) { + int imb_numa_nr = sd->imb_numa_nr; +#ifdef CONFIG_NUMA_BALANCING + int idlest_cpu; + /* + * If there is spare capacity at NUMA, try to select + * the preferred node + */ + if (cpu_to_node(this_cpu) == p->numa_preferred_nid) + return NULL; - if (!env->sd->parent) { - struct root_domain *rd = env->dst_rq->rd; + idlest_cpu = cpumask_first(sched_group_span(idlest)); + if (cpu_to_node(idlest_cpu) == p->numa_preferred_nid) + return idlest; +#endif /* CONFIG_NUMA_BALANCING */ + /* + * Otherwise, keep the task close to the wakeup source + * and improve locality if the number of running tasks + * would remain below threshold where an imbalance is + * allowed while accounting for the possibility the + * task is pinned to a subset of CPUs. If there is a + * real need of migration, periodic load balance will + * take care of it. + */ + if (p->nr_cpus_allowed != NR_CPUS) { + struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask); - /* update overload indicator if we are at root domain */ - WRITE_ONCE(rd->overload, sg_status & SG_OVERLOAD); + cpumask_and(cpus, sched_group_span(local), p->cpus_ptr); + imb_numa_nr = min(cpumask_weight(cpus), sd->imb_numa_nr); + } - /* Update over-utilization (tipping point, U >= 0) indicator */ - WRITE_ONCE(rd->overutilized, sg_status & SG_OVERUTILIZED); - } else if (sg_status & SG_OVERUTILIZED) { - WRITE_ONCE(env->dst_rq->rd->overutilized, SG_OVERUTILIZED); + imbalance = abs(local_sgs.idle_cpus - idlest_sgs.idle_cpus); + if (!adjust_numa_imbalance(imbalance, + local_sgs.sum_nr_running + 1, + imb_numa_nr)) { + return NULL; + } + } +#endif /* CONFIG_NUMA */ + + /* + * Select group with highest number of idle CPUs. We could also + * compare the utilization which is more stable but it can end + * up that the group has less spare capacity but finally more + * idle CPUs which means more opportunity to run task. + */ + if (local_sgs.idle_cpus >= idlest_sgs.idle_cpus) + return NULL; + break; } + + return idlest; } -/** - * check_asym_packing - Check to see if the group is packed into the - * sched domain. - * - * This is primarily intended to used at the sibling level. Some - * cores like POWER7 prefer to use lower numbered SMT threads. In the - * case of POWER7, it can move to lower SMT modes only when higher - * threads are idle. When in lower SMT modes, the threads will - * perform better since they share less core resources. Hence when we - * have idle threads, we want them to be the higher ones. - * - * This packing function is run on idle threads. It checks to see if - * the busiest CPU in this domain (core in the P7 case) has a higher - * CPU number than the packing function is being run on. Here we are - * assuming lower CPU number will be equivalent to lower a SMT thread - * number. - * - * Return: 1 when packing is required and a task should be moved to - * this CPU. The amount of the imbalance is returned in env->imbalance. - * - * @env: The load balancing environment. - * @sds: Statistics of the sched_domain which is to be packed - */ -static int check_asym_packing(struct lb_env *env, struct sd_lb_stats *sds) +static void update_idle_cpu_scan(struct lb_env *env, + unsigned long sum_util) { - int busiest_cpu; - - if (!(env->sd->flags & SD_ASYM_PACKING)) - return 0; + struct sched_domain_shared *sd_share; + int llc_weight, pct; + u64 x, y, tmp; + /* + * Update the number of CPUs to scan in LLC domain, which could + * be used as a hint in select_idle_cpu(). The update of sd_share + * could be expensive because it is within a shared cache line. + * So the write of this hint only occurs during periodic load + * balancing, rather than CPU_NEWLY_IDLE, because the latter + * can fire way more frequently than the former. + */ + if (!sched_feat(SIS_UTIL) || env->idle == CPU_NEWLY_IDLE) + return; - if (env->idle == CPU_NOT_IDLE) - return 0; + llc_weight = per_cpu(sd_llc_size, env->dst_cpu); + if (env->sd->span_weight != llc_weight) + return; - if (!sds->busiest) - return 0; + sd_share = rcu_dereference(per_cpu(sd_llc_shared, env->dst_cpu)); + if (!sd_share) + return; - busiest_cpu = sds->busiest->asym_prefer_cpu; - if (sched_asym_prefer(busiest_cpu, env->dst_cpu)) - return 0; + /* + * The number of CPUs to search drops as sum_util increases, when + * sum_util hits 85% or above, the scan stops. + * The reason to choose 85% as the threshold is because this is the + * imbalance_pct(117) when a LLC sched group is overloaded. + * + * let y = SCHED_CAPACITY_SCALE - p * x^2 [1] + * and y'= y / SCHED_CAPACITY_SCALE + * + * x is the ratio of sum_util compared to the CPU capacity: + * x = sum_util / (llc_weight * SCHED_CAPACITY_SCALE) + * y' is the ratio of CPUs to be scanned in the LLC domain, + * and the number of CPUs to scan is calculated by: + * + * nr_scan = llc_weight * y' [2] + * + * When x hits the threshold of overloaded, AKA, when + * x = 100 / pct, y drops to 0. According to [1], + * p should be SCHED_CAPACITY_SCALE * pct^2 / 10000 + * + * Scale x by SCHED_CAPACITY_SCALE: + * x' = sum_util / llc_weight; [3] + * + * and finally [1] becomes: + * y = SCHED_CAPACITY_SCALE - + * x'^2 * pct^2 / (10000 * SCHED_CAPACITY_SCALE) [4] + * + */ + /* equation [3] */ + x = sum_util; + do_div(x, llc_weight); - env->imbalance = DIV_ROUND_CLOSEST( - sds->busiest_stat.avg_load * sds->busiest_stat.group_capacity, - SCHED_CAPACITY_SCALE); + /* equation [4] */ + pct = env->sd->imbalance_pct; + tmp = x * x * pct * pct; + do_div(tmp, 10000 * SCHED_CAPACITY_SCALE); + tmp = min_t(long, tmp, SCHED_CAPACITY_SCALE); + y = SCHED_CAPACITY_SCALE - tmp; - return 1; + /* equation [2] */ + y *= llc_weight; + do_div(y, SCHED_CAPACITY_SCALE); + if ((int)y != sd_share->nr_idle_scan) + WRITE_ONCE(sd_share->nr_idle_scan, (int)y); } /** - * fix_small_imbalance - Calculate the minor imbalance that exists - * amongst the groups of a sched_domain, during - * load balancing. + * update_sd_lb_stats - Update sched_domain's statistics for load balancing. * @env: The load balancing environment. - * @sds: Statistics of the sched_domain whose imbalance is to be calculated. + * @sds: variable to hold the statistics for this sched_domain. */ -static inline -void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds) + +static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sds) { - unsigned long tmp, capa_now = 0, capa_move = 0; - unsigned int imbn = 2; - unsigned long scaled_busy_load_per_task; - struct sg_lb_stats *local, *busiest; + struct sched_group *sg = env->sd->groups; + struct sg_lb_stats *local = &sds->local_stat; + struct sg_lb_stats tmp_sgs; + unsigned long sum_util = 0; + bool sg_overloaded = 0, sg_overutilized = 0; - local = &sds->local_stat; - busiest = &sds->busiest_stat; + do { + struct sg_lb_stats *sgs = &tmp_sgs; + int local_group; - if (!local->sum_nr_running) - local->load_per_task = cpu_avg_load_per_task(env->dst_cpu); - else if (busiest->load_per_task > local->load_per_task) - imbn = 1; + local_group = cpumask_test_cpu(env->dst_cpu, sched_group_span(sg)); + if (local_group) { + sds->local = sg; + sgs = local; - scaled_busy_load_per_task = - (busiest->load_per_task * SCHED_CAPACITY_SCALE) / - busiest->group_capacity; + if (env->idle != CPU_NEWLY_IDLE || + time_after_eq(jiffies, sg->sgc->next_update)) + update_group_capacity(env->sd, env->dst_cpu); + } - if (busiest->avg_load + scaled_busy_load_per_task >= - local->avg_load + (scaled_busy_load_per_task * imbn)) { - env->imbalance = busiest->load_per_task; - return; - } + update_sg_lb_stats(env, sds, sg, sgs, &sg_overloaded, &sg_overutilized); + + if (!local_group && update_sd_pick_busiest(env, sds, sg, sgs)) { + sds->busiest = sg; + sds->busiest_stat = *sgs; + } + + /* Now, start updating sd_lb_stats */ + sds->total_load += sgs->group_load; + sds->total_capacity += sgs->group_capacity; + + sum_util += sgs->group_util; + sg = sg->next; + } while (sg != env->sd->groups); /* - * OK, we don't have enough imbalance to justify moving tasks, - * however we may be able to increase total CPU capacity used by - * moving them. + * Indicate that the child domain of the busiest group prefers tasks + * go to a child's sibling domains first. NB the flags of a sched group + * are those of the child domain. */ + if (sds->busiest) + sds->prefer_sibling = !!(sds->busiest->flags & SD_PREFER_SIBLING); - capa_now += busiest->group_capacity * - min(busiest->load_per_task, busiest->avg_load); - capa_now += local->group_capacity * - min(local->load_per_task, local->avg_load); - capa_now /= SCHED_CAPACITY_SCALE; - /* Amount of load we'd subtract */ - if (busiest->avg_load > scaled_busy_load_per_task) { - capa_move += busiest->group_capacity * - min(busiest->load_per_task, - busiest->avg_load - scaled_busy_load_per_task); - } + if (env->sd->flags & SD_NUMA) + env->fbq_type = fbq_classify_group(&sds->busiest_stat); - /* Amount of load we'd add */ - if (busiest->avg_load * busiest->group_capacity < - busiest->load_per_task * SCHED_CAPACITY_SCALE) { - tmp = (busiest->avg_load * busiest->group_capacity) / - local->group_capacity; - } else { - tmp = (busiest->load_per_task * SCHED_CAPACITY_SCALE) / - local->group_capacity; + if (!env->sd->parent) { + /* update overload indicator if we are at root domain */ + set_rd_overloaded(env->dst_rq->rd, sg_overloaded); + + /* Update over-utilization (tipping point, U >= 0) indicator */ + set_rd_overutilized(env->dst_rq->rd, sg_overutilized); + } else if (sg_overutilized) { + set_rd_overutilized(env->dst_rq->rd, sg_overutilized); } - capa_move += local->group_capacity * - min(local->load_per_task, local->avg_load + tmp); - capa_move /= SCHED_CAPACITY_SCALE; - /* Move if we gain throughput */ - if (capa_move > capa_now) - env->imbalance = busiest->load_per_task; + update_idle_cpu_scan(env, sum_util); } /** @@ -8536,93 +11151,207 @@ void fix_small_imbalance(struct lb_env *env, struct sd_lb_stats *sds) */ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *sds) { - unsigned long max_pull, load_above_capacity = ~0UL; struct sg_lb_stats *local, *busiest; local = &sds->local_stat; busiest = &sds->busiest_stat; + if (busiest->group_type == group_misfit_task) { + if (env->sd->flags & SD_ASYM_CPUCAPACITY) { + /* Set imbalance to allow misfit tasks to be balanced. */ + env->migration_type = migrate_misfit; + env->imbalance = 1; + } else { + /* + * Set load imbalance to allow moving task from cpu + * with reduced capacity. + */ + env->migration_type = migrate_load; + env->imbalance = busiest->group_misfit_task_load; + } + return; + } + + if (busiest->group_type == group_asym_packing) { + /* + * In case of asym capacity, we will try to migrate all load to + * the preferred CPU. + */ + env->migration_type = migrate_task; + env->imbalance = busiest->sum_h_nr_running; + return; + } + + if (busiest->group_type == group_smt_balance) { + /* Reduce number of tasks sharing CPU capacity */ + env->migration_type = migrate_task; + env->imbalance = 1; + return; + } + if (busiest->group_type == group_imbalanced) { /* * In the group_imb case we cannot rely on group-wide averages - * to ensure CPU-load equilibrium, look at wider averages. XXX + * to ensure CPU-load equilibrium, try to move any task to fix + * the imbalance. The next load balance will take care of + * balancing back the system. */ - busiest->load_per_task = - min(busiest->load_per_task, sds->avg_load); + env->migration_type = migrate_task; + env->imbalance = 1; + return; } /* - * Avg load of busiest sg can be less and avg load of local sg can - * be greater than avg load across all sgs of sd because avg load - * factors in sg capacity and sgs with smaller group_type are - * skipped when updating the busiest sg: + * Try to use spare capacity of local group without overloading it or + * emptying busiest. */ - if (busiest->group_type != group_misfit_task && - (busiest->avg_load <= sds->avg_load || - local->avg_load >= sds->avg_load)) { - env->imbalance = 0; - return fix_small_imbalance(env, sds); - } + if (local->group_type == group_has_spare) { + if ((busiest->group_type > group_fully_busy) && + !(env->sd->flags & SD_SHARE_LLC)) { + /* + * If busiest is overloaded, try to fill spare + * capacity. This might end up creating spare capacity + * in busiest or busiest still being overloaded but + * there is no simple way to directly compute the + * amount of load to migrate in order to balance the + * system. + */ + env->migration_type = migrate_util; + env->imbalance = max(local->group_capacity, local->group_util) - + local->group_util; - /* - * If there aren't any idle CPUs, avoid creating some. - */ - if (busiest->group_type == group_overloaded && - local->group_type == group_overloaded) { - load_above_capacity = busiest->sum_nr_running * SCHED_CAPACITY_SCALE; - if (load_above_capacity > busiest->group_capacity) { - load_above_capacity -= busiest->group_capacity; - load_above_capacity *= scale_load_down(NICE_0_LOAD); - load_above_capacity /= busiest->group_capacity; - } else - load_above_capacity = ~0UL; + /* + * In some cases, the group's utilization is max or even + * higher than capacity because of migrations but the + * local CPU is (newly) idle. There is at least one + * waiting task in this overloaded busiest group. Let's + * try to pull it. + */ + if (env->idle && env->imbalance == 0) { + env->migration_type = migrate_task; + env->imbalance = 1; + } + + return; + } + + if (busiest->group_weight == 1 || sds->prefer_sibling) { + /* + * When prefer sibling, evenly spread running tasks on + * groups. + */ + env->migration_type = migrate_task; + env->imbalance = sibling_imbalance(env, sds, busiest, local); + } else { + + /* + * If there is no overload, we just want to even the number of + * idle CPUs. + */ + env->migration_type = migrate_task; + env->imbalance = max_t(long, 0, + (local->idle_cpus - busiest->idle_cpus)); + } + +#ifdef CONFIG_NUMA + /* Consider allowing a small imbalance between NUMA groups */ + if (env->sd->flags & SD_NUMA) { + env->imbalance = adjust_numa_imbalance(env->imbalance, + local->sum_nr_running + 1, + env->sd->imb_numa_nr); + } +#endif + + /* Number of tasks to move to restore balance */ + env->imbalance >>= 1; + + return; } /* - * We're trying to get all the CPUs to the average_load, so we don't - * want to push ourselves above the average load, nor do we wish to - * reduce the max loaded CPU below the average load. At the same time, - * we also don't want to reduce the group load below the group - * capacity. Thus we look for the minimum possible imbalance. + * Local is fully busy but has to take more load to relieve the + * busiest group */ - max_pull = min(busiest->avg_load - sds->avg_load, load_above_capacity); + if (local->group_type < group_overloaded) { + /* + * Local will become overloaded so the avg_load metrics are + * finally needed. + */ - /* How much load to actually move to equalise the imbalance */ - env->imbalance = min( - max_pull * busiest->group_capacity, - (sds->avg_load - local->avg_load) * local->group_capacity - ) / SCHED_CAPACITY_SCALE; + local->avg_load = (local->group_load * SCHED_CAPACITY_SCALE) / + local->group_capacity; + + /* + * If the local group is more loaded than the selected + * busiest group don't try to pull any tasks. + */ + if (local->avg_load >= busiest->avg_load) { + env->imbalance = 0; + return; + } + + sds->avg_load = (sds->total_load * SCHED_CAPACITY_SCALE) / + sds->total_capacity; + + /* + * If the local group is more loaded than the average system + * load, don't try to pull any tasks. + */ + if (local->avg_load >= sds->avg_load) { + env->imbalance = 0; + return; + } - /* Boost imbalance to allow misfit task to be balanced. */ - if (busiest->group_type == group_misfit_task) { - env->imbalance = max_t(long, env->imbalance, - busiest->group_misfit_task_load); } /* - * if *imbalance is less than the average load per runnable task - * there is no guarantee that any tasks will be moved so we'll have - * a think about bumping its value to force at least one task to be - * moved + * Both group are or will become overloaded and we're trying to get all + * the CPUs to the average_load, so we don't want to push ourselves + * above the average load, nor do we wish to reduce the max loaded CPU + * below the average load. At the same time, we also don't want to + * reduce the group load below the group capacity. Thus we look for + * the minimum possible imbalance. */ - if (env->imbalance < busiest->load_per_task) - return fix_small_imbalance(env, sds); + env->migration_type = migrate_load; + env->imbalance = min( + (busiest->avg_load - sds->avg_load) * busiest->group_capacity, + (sds->avg_load - local->avg_load) * local->group_capacity + ) / SCHED_CAPACITY_SCALE; } -/******* find_busiest_group() helpers end here *********************/ +/******* sched_balance_find_src_group() helpers end here *********************/ + +/* + * Decision matrix according to the local and busiest group type: + * + * busiest \ local has_spare fully_busy misfit asym imbalanced overloaded + * has_spare nr_idle balanced N/A N/A balanced balanced + * fully_busy nr_idle nr_idle N/A N/A balanced balanced + * misfit_task force N/A N/A N/A N/A N/A + * asym_packing force force N/A N/A force force + * imbalanced force force N/A N/A force force + * overloaded force force N/A N/A force avg_load + * + * N/A : Not Applicable because already filtered while updating + * statistics. + * balanced : The system is balanced for these 2 groups. + * force : Calculate the imbalance as load migration is probably needed. + * avg_load : Only if imbalance is significant enough. + * nr_idle : dst_cpu is not busy and the number of idle CPUs is quite + * different in groups. + */ /** - * find_busiest_group - Returns the busiest group within the sched_domain + * sched_balance_find_src_group - Returns the busiest group within the sched_domain * if there is an imbalance. + * @env: The load balancing environment. * - * Also calculates the amount of weighted load which should be moved + * Also calculates the amount of runnable load which should be moved * to restore balance. * - * @env: The load balancing environment. - * * Return: - The busiest group if imbalance exists. */ -static struct sched_group *find_busiest_group(struct lb_env *env) +static struct sched_group *sched_balance_find_src_group(struct lb_env *env) { struct sg_lb_stats *local, *busiest; struct sd_lb_stats sds; @@ -8630,91 +11359,125 @@ static struct sched_group *find_busiest_group(struct lb_env *env) init_sd_lb_stats(&sds); /* - * Compute the various statistics relavent for load balancing at + * Compute the various statistics relevant for load balancing at * this level. */ update_sd_lb_stats(env, &sds); - if (static_branch_unlikely(&sched_energy_present)) { - struct root_domain *rd = env->dst_rq->rd; - - if (rcu_dereference(rd->pd) && !READ_ONCE(rd->overutilized)) - goto out_balanced; - } + /* There is no busy sibling group to pull tasks from */ + if (!sds.busiest) + goto out_balanced; - local = &sds.local_stat; busiest = &sds.busiest_stat; - /* ASYM feature bypasses nice load balance check */ - if (check_asym_packing(env, &sds)) - return sds.busiest; + /* Misfit tasks should be dealt with regardless of the avg load */ + if (busiest->group_type == group_misfit_task) + goto force_balance; - /* There is no busy sibling group to pull tasks from */ - if (!sds.busiest || busiest->sum_nr_running == 0) + if (!is_rd_overutilized(env->dst_rq->rd) && + rcu_dereference(env->dst_rq->rd->pd)) goto out_balanced; - /* XXX broken for overlapping NUMA groups */ - sds.avg_load = (SCHED_CAPACITY_SCALE * sds.total_load) - / sds.total_capacity; + /* ASYM feature bypasses nice load balance check */ + if (busiest->group_type == group_asym_packing) + goto force_balance; /* * If the busiest group is imbalanced the below checks don't * work because they assume all things are equal, which typically - * isn't true due to cpus_allowed constraints and the like. + * isn't true due to cpus_ptr constraints and the like. */ if (busiest->group_type == group_imbalanced) goto force_balance; - /* - * When dst_cpu is idle, prevent SMP nice and/or asymmetric group - * capacities from resulting in underutilization due to avg_load. - */ - if (env->idle != CPU_NOT_IDLE && group_has_capacity(env, local) && - busiest->group_no_capacity) - goto force_balance; - - /* Misfit tasks should be dealt with regardless of the avg load */ - if (busiest->group_type == group_misfit_task) - goto force_balance; - + local = &sds.local_stat; /* * If the local group is busier than the selected busiest group * don't try and pull any tasks. */ - if (local->avg_load >= busiest->avg_load) + if (local->group_type > busiest->group_type) goto out_balanced; /* - * Don't pull any tasks if this group is already above the domain - * average load. + * When groups are overloaded, use the avg_load to ensure fairness + * between tasks. */ - if (local->avg_load >= sds.avg_load) - goto out_balanced; + if (local->group_type == group_overloaded) { + /* + * If the local group is more loaded than the selected + * busiest group don't try to pull any tasks. + */ + if (local->avg_load >= busiest->avg_load) + goto out_balanced; + + /* XXX broken for overlapping NUMA groups */ + sds.avg_load = (sds.total_load * SCHED_CAPACITY_SCALE) / + sds.total_capacity; - if (env->idle == CPU_IDLE) { /* - * This CPU is idle. If the busiest group is not overloaded - * and there is no imbalance between this and busiest group - * wrt idle CPUs, it is balanced. The imbalance becomes - * significant if the diff is greater than 1 otherwise we - * might end up to just move the imbalance on another group + * Don't pull any tasks if this group is already above the + * domain average load. */ - if ((busiest->group_type != group_overloaded) && - (local->idle_cpus <= (busiest->idle_cpus + 1))) + if (local->avg_load >= sds.avg_load) goto out_balanced; - } else { + /* - * In the CPU_NEWLY_IDLE, CPU_NOT_IDLE cases, use - * imbalance_pct to be conservative. + * If the busiest group is more loaded, use imbalance_pct to be + * conservative. */ if (100 * busiest->avg_load <= env->sd->imbalance_pct * local->avg_load) goto out_balanced; } + /* + * Try to move all excess tasks to a sibling domain of the busiest + * group's child domain. + */ + if (sds.prefer_sibling && local->group_type == group_has_spare && + sibling_imbalance(env, &sds, busiest, local) > 1) + goto force_balance; + + if (busiest->group_type != group_overloaded) { + if (!env->idle) { + /* + * If the busiest group is not overloaded (and as a + * result the local one too) but this CPU is already + * busy, let another idle CPU try to pull task. + */ + goto out_balanced; + } + + if (busiest->group_type == group_smt_balance && + smt_vs_nonsmt_groups(sds.local, sds.busiest)) { + /* Let non SMT CPU pull from SMT CPU sharing with sibling */ + goto force_balance; + } + + if (busiest->group_weight > 1 && + local->idle_cpus <= (busiest->idle_cpus + 1)) { + /* + * If the busiest group is not overloaded + * and there is no imbalance between this and busiest + * group wrt idle CPUs, it is balanced. The imbalance + * becomes significant if the diff is greater than 1 + * otherwise we might end up to just move the imbalance + * on another group. Of course this applies only if + * there is more than 1 CPU per group. + */ + goto out_balanced; + } + + if (busiest->sum_h_nr_running == 1) { + /* + * busiest doesn't have any tasks waiting to run + */ + goto out_balanced; + } + } + force_balance: /* Looks like there is an imbalance. Compute it */ - env->src_grp_type = busiest->group_type; calculate_imbalance(env, &sds); return env->imbalance ? sds.busiest : NULL; @@ -8724,17 +11487,19 @@ out_balanced: } /* - * find_busiest_queue - find the busiest runqueue among the CPUs in the group. + * sched_balance_find_src_rq - find the busiest runqueue among the CPUs in the group. */ -static struct rq *find_busiest_queue(struct lb_env *env, +static struct rq *sched_balance_find_src_rq(struct lb_env *env, struct sched_group *group) { struct rq *busiest = NULL, *rq; - unsigned long busiest_load = 0, busiest_capacity = 1; + unsigned long busiest_util = 0, busiest_load = 0, busiest_capacity = 1; + unsigned int busiest_nr = 0; int i; for_each_cpu_and(i, sched_group_span(group), env->cpus) { - unsigned long capacity, wl; + unsigned long capacity, load, util; + unsigned int nr_running; enum fbq_type rt; rq = cpu_rq(i); @@ -8762,18 +11527,9 @@ static struct rq *find_busiest_queue(struct lb_env *env, if (rt > env->fbq_type) continue; - /* - * For ASYM_CPUCAPACITY domains with misfit tasks we simply - * seek the "biggest" misfit task. - */ - if (env->src_grp_type == group_misfit_task) { - if (rq->misfit_task_load > busiest_load) { - busiest_load = rq->misfit_task_load; - busiest = rq; - } - + nr_running = rq->cfs.h_nr_runnable; + if (!nr_running) continue; - } capacity = capacity_of(i); @@ -8784,36 +11540,88 @@ static struct rq *find_busiest_queue(struct lb_env *env, * average load. */ if (env->sd->flags & SD_ASYM_CPUCAPACITY && - capacity_of(env->dst_cpu) < capacity && - rq->nr_running == 1) + !capacity_greater(capacity_of(env->dst_cpu), capacity) && + nr_running == 1) continue; - wl = weighted_cpuload(rq); - /* - * When comparing with imbalance, use weighted_cpuload() - * which is not scaled with the CPU capacity. + * Make sure we only pull tasks from a CPU of lower priority + * when balancing between SMT siblings. + * + * If balancing between cores, let lower priority CPUs help + * SMT cores with more than one busy sibling. */ - - if (rq->nr_running == 1 && wl > env->imbalance && - !check_cpu_capacity(rq, env->sd)) + if (sched_asym(env->sd, i, env->dst_cpu) && nr_running == 1) continue; - /* - * For the load comparisons with the other CPU's, consider - * the weighted_cpuload() scaled with the CPU capacity, so - * that the load can be moved away from the CPU that is - * potentially running at a lower capacity. - * - * Thus we're looking for max(wl_i / capacity_i), crosswise - * multiplication to rid ourselves of the division works out - * to: wl_i * capacity_j > wl_j * capacity_i; where j is - * our previous maximum. - */ - if (wl * busiest_capacity > busiest_load * capacity) { - busiest_load = wl; - busiest_capacity = capacity; - busiest = rq; + switch (env->migration_type) { + case migrate_load: + /* + * When comparing with load imbalance, use cpu_load() + * which is not scaled with the CPU capacity. + */ + load = cpu_load(rq); + + if (nr_running == 1 && load > env->imbalance && + !check_cpu_capacity(rq, env->sd)) + break; + + /* + * For the load comparisons with the other CPUs, + * consider the cpu_load() scaled with the CPU + * capacity, so that the load can be moved away + * from the CPU that is potentially running at a + * lower capacity. + * + * Thus we're looking for max(load_i / capacity_i), + * crosswise multiplication to rid ourselves of the + * division works out to: + * load_i * capacity_j > load_j * capacity_i; + * where j is our previous maximum. + */ + if (load * busiest_capacity > busiest_load * capacity) { + busiest_load = load; + busiest_capacity = capacity; + busiest = rq; + } + break; + + case migrate_util: + util = cpu_util_cfs_boost(i); + + /* + * Don't try to pull utilization from a CPU with one + * running task. Whatever its utilization, we will fail + * detach the task. + */ + if (nr_running <= 1) + continue; + + if (busiest_util < util) { + busiest_util = util; + busiest = rq; + } + break; + + case migrate_task: + if (busiest_nr < nr_running) { + busiest_nr = nr_running; + busiest = rq; + } + break; + + case migrate_misfit: + /* + * For ASYM_CPUCAPACITY domains with misfit tasks we + * simply seek the "biggest" misfit task. + */ + if (rq->misfit_task_load > busiest_load) { + busiest_load = rq->misfit_task_load; + busiest = rq; + } + + break; + } } @@ -8826,21 +11634,50 @@ static struct rq *find_busiest_queue(struct lb_env *env, */ #define MAX_PINNED_INTERVAL 512 +static inline bool +asym_active_balance(struct lb_env *env) +{ + /* + * ASYM_PACKING needs to force migrate tasks from busy but lower + * priority CPUs in order to pack all tasks in the highest priority + * CPUs. When done between cores, do it only if the whole core if the + * whole core is idle. + * + * If @env::src_cpu is an SMT core with busy siblings, let + * the lower priority @env::dst_cpu help it. Do not follow + * CPU priority. + */ + return env->idle && sched_use_asym_prio(env->sd, env->dst_cpu) && + (sched_asym_prefer(env->dst_cpu, env->src_cpu) || + !sched_use_asym_prio(env->sd, env->src_cpu)); +} + +static inline bool +imbalanced_active_balance(struct lb_env *env) +{ + struct sched_domain *sd = env->sd; + + /* + * The imbalanced case includes the case of pinned tasks preventing a fair + * distribution of the load on the system but also the even distribution of the + * threads on a system with spare capacity + */ + if ((env->migration_type == migrate_task) && + (sd->nr_balance_failed > sd->cache_nice_tries+2)) + return 1; + + return 0; +} + static int need_active_balance(struct lb_env *env) { struct sched_domain *sd = env->sd; - if (env->idle == CPU_NEWLY_IDLE) { + if (asym_active_balance(env)) + return 1; - /* - * ASYM_PACKING needs to force migrate tasks from busy but - * lower priority CPUs in order to pack all tasks in the - * highest priority CPUs. - */ - if ((sd->flags & SD_ASYM_PACKING) && - sched_asym_prefer(env->dst_cpu, env->src_cpu)) - return 1; - } + if (imbalanced_active_balance(env)) + return 1; /* * The dst_cpu is idle and the src_cpu CPU has only 1 CFS task. @@ -8848,25 +11685,26 @@ static int need_active_balance(struct lb_env *env) * because of other sched_class or IRQs if more capacity stays * available on dst_cpu. */ - if ((env->idle != CPU_NOT_IDLE) && - (env->src_rq->cfs.h_nr_running == 1)) { + if (env->idle && + (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; } - if (env->src_grp_type == group_misfit_task) + if (env->migration_type == migrate_misfit) return 1; - return unlikely(sd->nr_balance_failed > sd->cache_nice_tries+2); + return 0; } static int active_load_balance_cpu_stop(void *data); static int should_we_balance(struct lb_env *env) { + struct cpumask *swb_cpus = this_cpu_cpumask_var_ptr(should_we_balance_tmpmask); struct sched_group *sg = env->sd->groups; - int cpu, balance_cpu = -1; + int cpu, idle_smt = -1; /* * Ensure the balancing environment is consistent; can happen @@ -8878,34 +11716,98 @@ static int should_we_balance(struct lb_env *env) /* * In the newly idle case, we will allow all the CPUs * to do the newly idle load balance. + * + * However, we bail out if we already have tasks or a wakeup pending, + * to optimize wakeup latency. */ - if (env->idle == CPU_NEWLY_IDLE) + if (env->idle == CPU_NEWLY_IDLE) { + if (env->dst_rq->nr_running > 0 || env->dst_rq->ttwu_pending) + return 0; return 1; + } + cpumask_copy(swb_cpus, group_balance_mask(sg)); /* Try to find first idle CPU */ - for_each_cpu_and(cpu, group_balance_mask(sg), env->cpus) { + for_each_cpu_and(cpu, swb_cpus, env->cpus) { if (!idle_cpu(cpu)) continue; - balance_cpu = cpu; - break; + /* + * Don't balance to idle SMT in busy core right away when + * balancing cores, but remember the first idle SMT CPU for + * later consideration. Find CPU on an idle core first. + */ + if (!(env->sd->flags & SD_SHARE_CPUCAPACITY) && !is_core_idle(cpu)) { + if (idle_smt == -1) + idle_smt = cpu; + /* + * If the core is not idle, and first SMT sibling which is + * idle has been found, then its not needed to check other + * SMT siblings for idleness: + */ +#ifdef CONFIG_SCHED_SMT + cpumask_andnot(swb_cpus, swb_cpus, cpu_smt_mask(cpu)); +#endif + continue; + } + + /* + * Are we the first idle core in a non-SMT domain or higher, + * or the first idle CPU in a SMT domain? + */ + return cpu == env->dst_cpu; } - if (balance_cpu == -1) - balance_cpu = group_balance_cpu(sg); + /* Are we the first idle CPU with busy siblings? */ + if (idle_smt != -1) + return idle_smt == env->dst_cpu; - /* - * First idle CPU or the first CPU(busiest) in this sched group - * is eligible for doing load balancing at this and above domains. - */ - return balance_cpu == env->dst_cpu; + /* Are we the first CPU of this group ? */ + 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; + } +} + +/* + * This flag serializes load-balancing passes over large domains + * (above the NODE topology level) - only one load-balancing instance + * may run at a time, to reduce overhead on very large systems with + * lots of CPUs and large NUMA distances. + * + * - Note that load-balancing passes triggered while another one + * is executing are skipped and not re-tried. + * + * - Also note that this does not serialize rebalance_domains() + * execution, as non-SD_SERIALIZE domains will still be + * load-balanced in parallel. + */ +static atomic_t sched_balance_running = ATOMIC_INIT(0); + /* * Check this_cpu to ensure it is balanced within domain. Attempt to move * tasks if there is an imbalance. */ -static int load_balance(int this_cpu, struct rq *this_rq, +static int sched_balance_rq(int this_cpu, struct rq *this_rq, struct sched_domain *sd, enum cpu_idle_type idle, int *continue_balancing) { @@ -8915,18 +11817,18 @@ static int load_balance(int this_cpu, struct rq *this_rq, struct rq *busiest; struct rq_flags rf; struct cpumask *cpus = this_cpu_cpumask_var_ptr(load_balance_mask); - struct lb_env env = { .sd = sd, .dst_cpu = this_cpu, .dst_rq = this_rq, - .dst_grpmask = sched_group_span(sd->groups), + .dst_grpmask = group_balance_mask(sd->groups), .idle = idle, - .loop_break = sched_nr_migrate_break, + .loop_break = SCHED_NR_MIGRATE_BREAK, .cpus = cpus, .fbq_type = all, .tasks = LIST_HEAD_INIT(env.tasks), }; + bool need_unlock = false; cpumask_and(cpus, sched_domain_span(sd), cpu_active_mask); @@ -8938,34 +11840,43 @@ redo: goto out_balanced; } - group = find_busiest_group(&env); + if (!need_unlock && (sd->flags & SD_SERIALIZE)) { + int zero = 0; + if (!atomic_try_cmpxchg_acquire(&sched_balance_running, &zero, 1)) + goto out_balanced; + + need_unlock = true; + } + + group = sched_balance_find_src_group(&env); if (!group) { schedstat_inc(sd->lb_nobusyg[idle]); goto out_balanced; } - busiest = find_busiest_queue(&env, group); + busiest = sched_balance_find_src_rq(&env, group); if (!busiest) { schedstat_inc(sd->lb_nobusyq[idle]); goto out_balanced; } - BUG_ON(busiest == env.dst_rq); + 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; ld_moved = 0; + /* Clear this flag as soon as we find a pullable task */ + env.flags |= LBF_ALL_PINNED; if (busiest->nr_running > 1) { /* - * Attempt to move tasks. If find_busiest_group has found + * Attempt to move tasks. If sched_balance_find_src_group has found * an imbalance but busiest->nr_running <= 1, the group is * still unbalanced. ld_moved simply stays zero, so it is * correctly treated as an imbalance. */ - env.flags |= LBF_ALL_PINNED; env.loop_max = min(sysctl_sched_nr_migrate, busiest->nr_running); more_balance: @@ -9014,7 +11925,7 @@ more_balance: * load to given_cpu. In rare situations, this may cause * conflicts (balance_cpu and given_cpu/ilb_cpu deciding * _independently_ and at _same_ time to move some load to - * given_cpu) causing exceess load to be moved to given_cpu. + * given_cpu) causing excess load to be moved to given_cpu. * This however should not happen so much in practice and * moreover subsequent load balance cycles should correct the * excess load moved. @@ -9022,13 +11933,13 @@ more_balance: if ((env.flags & LBF_DST_PINNED) && env.imbalance > 0) { /* Prevent to re-select dst_cpu via env's CPUs */ - cpumask_clear_cpu(env.dst_cpu, env.cpus); + __cpumask_clear_cpu(env.dst_cpu, env.cpus); env.dst_rq = cpu_rq(env.new_dst_cpu); env.dst_cpu = env.new_dst_cpu; env.flags &= ~LBF_DST_PINNED; env.loop = 0; - env.loop_break = sched_nr_migrate_break; + env.loop_break = SCHED_NR_MIGRATE_BREAK; /* * Go back to "more_balance" rather than "redo" since we @@ -9049,7 +11960,7 @@ more_balance: /* All tasks on this runqueue were pinned by CPU affinity */ if (unlikely(env.flags & LBF_ALL_PINNED)) { - cpumask_clear_cpu(cpu_of(busiest), cpus); + __cpumask_clear_cpu(cpu_of(busiest), cpus); /* * Attempting to continue load balancing at the current * sched_domain level only makes sense if there are @@ -9060,7 +11971,7 @@ more_balance: */ if (!cpumask_subset(cpus, env.dst_grpmask)) { env.loop = 0; - env.loop_break = sched_nr_migrate_break; + env.loop_break = SCHED_NR_MIGRATE_BREAK; goto redo; } goto out_all_pinned; @@ -9074,27 +11985,32 @@ more_balance: * We do not want newidle balance, which can be very * frequent, pollute the failure counter causing * excessive cache_hot migrations and active balances. + * + * Similarly for migration_misfit which is not related to + * load/util migration, don't pollute nr_balance_failed. */ - if (idle != CPU_NEWLY_IDLE) + if (idle != CPU_NEWLY_IDLE && + env.migration_type != migrate_misfit) sd->nr_balance_failed++; if (need_active_balance(&env)) { unsigned long flags; - raw_spin_lock_irqsave(&busiest->lock, flags); + raw_spin_rq_lock_irqsave(busiest, flags); /* * Don't kick the active_load_balance_cpu_stop, * if the curr task on busiest CPU can't be * moved to this_cpu: */ - if (!cpumask_test_cpu(this_cpu, &busiest->curr->cpus_allowed)) { - raw_spin_unlock_irqrestore(&busiest->lock, - flags); - env.flags |= LBF_ALL_PINNED; + if (!cpumask_test_cpu(this_cpu, busiest->curr->cpus_ptr)) { + raw_spin_rq_unlock_irqrestore(busiest, flags); goto out_one_pinned; } + /* Record that we found at least one task that could run on this_cpu */ + env.flags &= ~LBF_ALL_PINNED; + /* * ->active_balance synchronizes accesses to * ->active_balance_work. Once set, it's cleared @@ -9105,32 +12021,23 @@ more_balance: busiest->push_cpu = this_cpu; active_balance = 1; } - raw_spin_unlock_irqrestore(&busiest->lock, flags); + preempt_disable(); + raw_spin_rq_unlock_irqrestore(busiest, flags); if (active_balance) { stop_one_cpu_nowait(cpu_of(busiest), active_load_balance_cpu_stop, busiest, &busiest->active_balance_work); } - - /* We've kicked active balancing, force task migration. */ - sd->nr_balance_failed = sd->cache_nice_tries+1; + preempt_enable(); } - } else + } else { sd->nr_balance_failed = 0; + } - if (likely(!active_balance)) { + if (likely(!active_balance) || need_active_balance(&env)) { /* We were unbalanced, so reset the balancing interval */ sd->balance_interval = sd->min_interval; - } else { - /* - * If we've begun active balancing, start to back off. This - * case may not be covered by the all_pinned logic if there - * is only 1 task on the busy runqueue (because we don't call - * detach_tasks). - */ - if (sd->balance_interval < sd->max_interval) - sd->balance_interval *= 2; } goto out; @@ -9138,9 +12045,10 @@ more_balance: out_balanced: /* * We reach balance although we may have faced some affinity - * constraints. Clear the imbalance flag if it was set. + * constraints. Clear the imbalance flag only if other tasks got + * a chance to move and fix the imbalance. */ - if (sd_parent) { + if (sd_parent && !(env.flags & LBF_ALL_PINNED)) { int *group_imbalance = &sd_parent->groups->sgc->imbalance; if (*group_imbalance) @@ -9161,12 +12069,17 @@ out_one_pinned: ld_moved = 0; /* - * idle_balance() disregards balance intervals, so we could repeatedly - * reach this code, which would lead to balance_interval skyrocketting - * in a short amount of time. Skip the balance_interval increase logic - * to avoid that. + * sched_balance_newidle() disregards balance intervals, so we could + * repeatedly reach this code, which would lead to balance_interval + * skyrocketing in a short amount of time. Skip the balance_interval + * increase logic to avoid that. + * + * Similarly misfit migration which is not necessarily an indication of + * the system being busy and requires lb to backoff to let it settle + * down. */ - if (env.idle == CPU_NEWLY_IDLE) + if (env.idle == CPU_NEWLY_IDLE || + env.migration_type == migrate_misfit) goto out; /* tune up the balancing interval */ @@ -9175,6 +12088,9 @@ out_one_pinned: sd->balance_interval < sd->max_interval) sd->balance_interval *= 2; out: + if (need_unlock) + atomic_set_release(&sched_balance_running, 0); + return ld_moved; } @@ -9188,6 +12104,15 @@ get_sd_balance_interval(struct sched_domain *sd, int cpu_busy) /* scale ms to jiffies */ interval = msecs_to_jiffies(interval); + + /* + * Reduce likelihood of busy balancing at higher domains racing with + * balancing at lower domains by preventing their balancing periods + * from being multiples of each other. + */ + if (cpu_busy) + interval -= 1; + interval = clamp(interval, 1UL, max_load_balance_interval); return interval; @@ -9245,14 +12170,13 @@ static int active_load_balance_cpu_stop(void *data) * we need to fix it. Originally reported by * Bjorn Helgaas on a 128-CPU setup. */ - BUG_ON(busiest_rq == target_rq); + WARN_ON_ONCE(busiest_rq == target_rq); /* Search for an sd spanning us and the target CPU. */ rcu_read_lock(); for_each_domain(target_cpu, sd) { - if ((sd->flags & SD_LOAD_BALANCE) && - cpumask_test_cpu(busiest_cpu, sched_domain_span(sd))) - break; + if (cpumask_test_cpu(busiest_cpu, sched_domain_span(sd))) + break; } if (likely(sd)) { @@ -9263,13 +12187,7 @@ static int active_load_balance_cpu_stop(void *data) .src_cpu = busiest_rq->cpu, .src_rq = busiest_rq, .idle = CPU_IDLE, - /* - * can_migrate_task() doesn't need to compute new_dst_cpu - * for active balancing. Since we have CPU_IDLE, but no - * @dst_grpmask we need to make that test go away with lying - * about DST_PINNED. - */ - .flags = LBF_DST_PINNED, + .flags = LBF_ACTIVE_LB, }; schedstat_inc(sd->alb_count); @@ -9297,10 +12215,8 @@ out_unlock: return 0; } -static DEFINE_SPINLOCK(balancing); - /* - * Scale the max load_balance interval with the number of CPUs in the system. + * Scale the max sched_balance_rq interval with the number of CPUs in the system. * This trades load-balance latency on larger machines for less cross talk. */ void update_max_interval(void) @@ -9308,41 +12224,77 @@ void update_max_interval(void) max_load_balance_interval = HZ*num_online_cpus()/10; } +static inline void update_newidle_stats(struct sched_domain *sd, unsigned int success) +{ + sd->newidle_call++; + sd->newidle_success += success; + + if (sd->newidle_call >= 1024) { + sd->newidle_ratio = sd->newidle_success; + sd->newidle_call /= 2; + sd->newidle_success /= 2; + } +} + +static inline bool +update_newidle_cost(struct sched_domain *sd, u64 cost, unsigned int success) +{ + unsigned long next_decay = sd->last_decay_max_lb_cost + HZ; + unsigned long now = jiffies; + + if (cost) + update_newidle_stats(sd, success); + + if (cost > sd->max_newidle_lb_cost) { + /* + * Track max cost of a domain to make sure to not delay the + * next wakeup on the CPU. + */ + sd->max_newidle_lb_cost = cost; + sd->last_decay_max_lb_cost = now; + + } else if (time_after(now, next_decay)) { + /* + * Decay the newidle max times by ~1% per second to ensure that + * it is not outdated and the current max cost is actually + * shorter. + */ + sd->max_newidle_lb_cost = (sd->max_newidle_lb_cost * 253) / 256; + sd->last_decay_max_lb_cost = now; + return true; + } + + return false; +} + /* * It checks each scheduling domain to see if it is due to be balanced, * and initiates a balancing operation if so. * * Balancing parameters are set up in init_sched_domains. */ -static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle) +static void sched_balance_domains(struct rq *rq, enum cpu_idle_type idle) { int continue_balancing = 1; int cpu = rq->cpu; + int busy = idle != CPU_IDLE && !sched_idle_cpu(cpu); unsigned long interval; struct sched_domain *sd; /* Earliest time when we have to do rebalance again */ unsigned long next_balance = jiffies + 60*HZ; int update_next_balance = 0; - int need_serialize, need_decay = 0; + int need_decay = 0; u64 max_cost = 0; rcu_read_lock(); for_each_domain(cpu, sd) { /* * Decay the newidle max times here because this is a regular - * visit to all the domains. Decay ~1% per second. + * visit to all the domains. */ - if (time_after(jiffies, sd->next_decay_max_lb_cost)) { - sd->max_newidle_lb_cost = - (sd->max_newidle_lb_cost * 253) / 256; - sd->next_decay_max_lb_cost = jiffies + HZ; - need_decay = 1; - } + need_decay = update_newidle_cost(sd, 0, 0); max_cost += sd->max_newidle_lb_cost; - if (!(sd->flags & SD_LOAD_BALANCE)) - continue; - /* * Stop the load balance at this level. There is another * CPU in our sched group which is doing load balancing more @@ -9354,29 +12306,20 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle) break; } - interval = get_sd_balance_interval(sd, idle != CPU_IDLE); - - need_serialize = sd->flags & SD_SERIALIZE; - if (need_serialize) { - if (!spin_trylock(&balancing)) - goto out; - } - + interval = get_sd_balance_interval(sd, busy); if (time_after_eq(jiffies, sd->last_balance + interval)) { - if (load_balance(cpu, rq, sd, idle, &continue_balancing)) { + if (sched_balance_rq(cpu, rq, sd, idle, &continue_balancing)) { /* * The LBF_DST_PINNED logic could have changed * env->dst_cpu, so we can't know our idle * state even if we migrated tasks. Update it. */ - idle = idle_cpu(cpu) ? CPU_IDLE : CPU_NOT_IDLE; + idle = idle_cpu(cpu); + busy = !idle && !sched_idle_cpu(cpu); } sd->last_balance = jiffies; - interval = get_sd_balance_interval(sd, idle != CPU_IDLE); + interval = get_sd_balance_interval(sd, busy); } - if (need_serialize) - spin_unlock(&balancing); -out: if (time_after(next_balance, sd->last_balance + interval)) { next_balance = sd->last_balance + interval; update_next_balance = 1; @@ -9397,22 +12340,9 @@ out: * When the cpu is attached to null domain for ex, it will not be * updated. */ - if (likely(update_next_balance)) { + if (likely(update_next_balance)) rq->next_balance = next_balance; -#ifdef CONFIG_NO_HZ_COMMON - /* - * If this CPU has been elected to perform the nohz idle - * balance. Other idle CPUs have already rebalanced with - * nohz_idle_balance() and nohz.next_balance has been - * updated accordingly. This CPU is now running the idle load - * balance for itself and we need to update the - * nohz.next_balance accordingly. - */ - if ((idle == CPU_IDLE) && time_after(nohz.next_balance, rq->next_balance)) - nohz.next_balance = rq->next_balance; -#endif - } } static inline int on_null_domain(struct rq *rq) @@ -9422,61 +12352,79 @@ static inline int on_null_domain(struct rq *rq) #ifdef CONFIG_NO_HZ_COMMON /* - * idle load balancing details - * - When one of the busy CPUs notice that there may be an idle rebalancing + * NOHZ idle load balancing (ILB) details: + * + * - 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. */ - static inline int find_new_ilb(void) { - int ilb = cpumask_first(nohz.idle_cpus_mask); + const struct cpumask *hk_mask; + int ilb_cpu; - if (ilb < nr_cpu_ids && idle_cpu(ilb)) - return ilb; + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); - return nr_cpu_ids; + for_each_cpu_and(ilb_cpu, nohz.idle_cpus_mask, hk_mask) { + + if (ilb_cpu == smp_processor_id()) + continue; + + if (idle_cpu(ilb_cpu)) + return ilb_cpu; + } + + return -1; } /* - * Kick a CPU to do the nohz balancing, if it is time for it. We pick the - * nohz_load_balancer CPU (if there is one) otherwise fallback to any idle - * CPU (if there is one). + * 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_KERNEL_NOISE housekeeping set + * (if there is one). */ static void kick_ilb(unsigned int flags) { int ilb_cpu; - nohz.next_balance++; + /* + * Increase nohz.next_balance only when if full ilb is triggered but + * not if we only update stats. + */ + if (flags & NOHZ_BALANCE_KICK) + nohz.next_balance = jiffies+1; ilb_cpu = find_new_ilb(); + if (ilb_cpu < 0) + return; - if (ilb_cpu >= nr_cpu_ids) + /* + * Don't bother if no new NOHZ balance work items for ilb_cpu, + * i.e. all bits in flags are already set in ilb_cpu. + */ + if ((atomic_read(nohz_flags(ilb_cpu)) & flags) == flags) return; + /* + * Access to rq::nohz_csd is serialized by NOHZ_KICK_MASK; he who sets + * the first flag owns it; cleared by nohz_csd_func(). + */ flags = atomic_fetch_or(flags, nohz_flags(ilb_cpu)); if (flags & NOHZ_KICK_MASK) return; /* - * Use smp_send_reschedule() instead of resched_cpu(). - * This way we generate a sched IPI on the target CPU which - * is idle. And the softirq performing nohz idle load balance + * This way we generate an IPI on the target CPU which + * is idle, and the softirq performing NOHZ idle load balancing * will be run before returning from the IPI. */ - smp_send_reschedule(ilb_cpu); + smp_call_function_single_async(ilb_cpu, &cpu_rq(ilb_cpu)->nohz_csd); } /* - * Current heuristic for kicking the idle load balancer in the presence - * of an idle cpu in the system. - * - This rq has more than one task. - * - This rq has at least one CFS task and the capacity of the CPU is - * significantly reduced because of RT tasks or IRQs. - * - At parent of LLC scheduler domain level, this cpu's scheduler group has - * multiple busy cpu. - * - For SD_ASYM_PACKING, if the lower numbered cpu's in the scheduler - * domain span are idle. + * Current decision point for kicking the idle load balancer in the presence + * of idle CPUs in the system. */ static void nohz_balancer_kick(struct rq *rq) { @@ -9497,7 +12445,7 @@ static void nohz_balancer_kick(struct rq *rq) /* * None are in tickless mode and hence no need for NOHZ idle load - * balancing. + * balancing: */ if (likely(!atomic_read(&nohz.nr_cpus))) return; @@ -9509,51 +12457,87 @@ static void nohz_balancer_kick(struct rq *rq) if (time_before(now, nohz.next_balance)) goto out; - if (rq->nr_running >= 2 || rq->misfit_task_load) { - flags = NOHZ_KICK_MASK; + if (rq->nr_running >= 2) { + flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto out; } rcu_read_lock(); - sds = rcu_dereference(per_cpu(sd_llc_shared, cpu)); - if (sds) { + + sd = rcu_dereference(rq->sd); + if (sd) { /* - * XXX: write a coherent comment on why we do this. - * See also: http://lkml.kernel.org/r/20111202010832.602203411@sbsiddha-desk.sc.intel.com + * 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: */ - nr_busy = atomic_read(&sds->nr_busy_cpus); - if (nr_busy > 1) { - flags = NOHZ_KICK_MASK; + if (rq->cfs.h_nr_runnable >= 1 && check_cpu_capacity(rq, sd)) { + flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } - } - sd = rcu_dereference(rq->sd); + sd = rcu_dereference(per_cpu(sd_asym_packing, cpu)); if (sd) { - if ((rq->cfs.h_nr_running >= 1) && - check_cpu_capacity(rq, sd)) { - flags = NOHZ_KICK_MASK; - goto unlock; + /* + * When ASYM_PACKING; see if there's a more preferred CPU + * currently idle; in which case, kick the ILB to move tasks + * around. + * + * When balancing between cores, all the SMT siblings of the + * preferred CPU must be idle. + */ + for_each_cpu_and(i, sched_domain_span(sd), nohz.idle_cpus_mask) { + if (sched_asym(sd, i, cpu)) { + flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; + goto unlock; + } } } - sd = rcu_dereference(per_cpu(sd_asym_packing, cpu)); + sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, cpu)); if (sd) { - for_each_cpu(i, sched_domain_span(sd)) { - if (i == cpu || - !cpumask_test_cpu(i, nohz.idle_cpus_mask)) - continue; + /* + * When ASYM_CPUCAPACITY; see if there's a higher capacity CPU + * to run the misfit task on. + */ + if (check_misfit_status(rq)) { + flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; + goto unlock; + } - if (sched_asym_prefer(i, cpu)) { - flags = NOHZ_KICK_MASK; - goto unlock; - } + /* + * For asymmetric systems, we do not want to nicely balance + * cache use, instead we want to embrace asymmetry and only + * ensure tasks have enough CPU capacity. + * + * Skip the LLC logic because it's not relevant in that case. + */ + goto unlock; + } + + sds = rcu_dereference(per_cpu(sd_llc_shared, cpu)); + if (sds) { + /* + * If there is an imbalance between LLC domains (IOW we could + * increase the overall cache utilization), we need a less-loaded LLC + * domain to pull some load from. Likewise, we may need to spread + * load within the current LLC domain (e.g. packed SMT cores but + * other CPUs are idle). We can't really know from here how busy + * the others are - so just get a NOHZ balance going if it looks + * like this LLC domain has tasks we could move. + */ + nr_busy = atomic_read(&sds->nr_busy_cpus); + if (nr_busy > 1) { + flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; + goto unlock; } } unlock: rcu_read_unlock(); out: + if (READ_ONCE(nohz.needs_update)) + flags |= NOHZ_NEXT_KICK; + if (flags) kick_ilb(flags); } @@ -9576,7 +12560,7 @@ unlock: void nohz_balance_exit_idle(struct rq *rq) { - SCHED_WARN_ON(rq != this_rq()); + WARN_ON_ONCE(rq != this_rq()); if (likely(!rq->nohz_tick_stopped)) return; @@ -9612,16 +12596,12 @@ void nohz_balance_enter_idle(int cpu) { struct rq *rq = cpu_rq(cpu); - SCHED_WARN_ON(cpu != smp_processor_id()); + WARN_ON_ONCE(cpu != smp_processor_id()); /* If this CPU is going down, then nothing needs to be done: */ if (!cpu_active(cpu)) return; - /* Spare idle load balancing on CPUs that don't want to be disturbed: */ - if (!housekeeping_cpu(cpu, HK_FLAG_SCHED)) - return; - /* * Can be set safely without rq->lock held * If a clear happens, it will have evaluated last additions because @@ -9650,29 +12630,45 @@ void nohz_balance_enter_idle(int cpu) /* * Ensures that if nohz_idle_balance() fails to observe our * @idle_cpus_mask store, it must observe the @has_blocked - * store. + * and @needs_update stores. */ smp_mb__after_atomic(); set_cpu_sd_state_idle(cpu); + WRITE_ONCE(nohz.needs_update, 1); out: /* * Each time a cpu enter idle, we assume that it has blocked load and - * enable the periodic update of the load of idle cpus + * enable the periodic update of the load of idle CPUs */ WRITE_ONCE(nohz.has_blocked, 1); } +static bool update_nohz_stats(struct rq *rq) +{ + unsigned int cpu = rq->cpu; + + if (!rq->has_blocked_load) + return false; + + if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask)) + return false; + + if (!time_after(jiffies, READ_ONCE(rq->last_blocked_load_update_tick))) + return true; + + sched_balance_update_blocked_averages(cpu); + + return rq->has_blocked_load; +} + /* - * Internal function that runs load balance for all idle cpus. The load balance + * Internal function that runs load balance for all idle CPUs. The load balance * can be a simple update of blocked load or a complete load balance with * tasks movement depending of flags. - * The function returns false if the loop has stopped before running - * through all idle CPUs. */ -static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, - enum cpu_idle_type idle) +static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) { /* Earliest time when we have to do rebalance again */ unsigned long now = jiffies; @@ -9681,20 +12677,24 @@ static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, int update_next_balance = 0; int this_cpu = this_rq->cpu; int balance_cpu; - int ret = false; struct rq *rq; - SCHED_WARN_ON((flags & NOHZ_KICK_MASK) == NOHZ_BALANCE_KICK); + WARN_ON_ONCE((flags & NOHZ_KICK_MASK) == NOHZ_BALANCE_KICK); /* * We assume there will be no idle load after this update and clear * the has_blocked flag. If a cpu enters idle in the mean time, it will - * set the has_blocked flag and trig another update of idle load. + * set the has_blocked flag and trigger another update of idle load. * Because a cpu that becomes idle, is added to idle_cpus_mask before * setting the flag, we are sure to not clear the state and not * check the load of an idle cpu. + * + * Same applies to idle_cpus_mask vs needs_update. */ - WRITE_ONCE(nohz.has_blocked, 0); + if (flags & NOHZ_STATS_KICK) + WRITE_ONCE(nohz.has_blocked, 0); + if (flags & NOHZ_NEXT_KICK) + WRITE_ONCE(nohz.needs_update, 0); /* * Ensures that if we miss the CPU, we must see the has_blocked @@ -9702,8 +12702,12 @@ static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, */ smp_mb(); - for_each_cpu(balance_cpu, nohz.idle_cpus_mask) { - if (balance_cpu == this_cpu || !idle_cpu(balance_cpu)) + /* + * Start with the next CPU after this_cpu so we will end with this_cpu and let a + * chance for other idle cpu to pull load. + */ + for_each_cpu_wrap(balance_cpu, nohz.idle_cpus_mask, this_cpu+1) { + if (!idle_cpu(balance_cpu)) continue; /* @@ -9711,14 +12715,18 @@ static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, * work being done for other CPUs. Next load * balancing owner will pick it up. */ - if (need_resched()) { - has_blocked_load = true; + if (!idle_cpu(this_cpu) && need_resched()) { + if (flags & NOHZ_STATS_KICK) + has_blocked_load = true; + if (flags & NOHZ_NEXT_KICK) + WRITE_ONCE(nohz.needs_update, 1); goto abort; } rq = cpu_rq(balance_cpu); - has_blocked_load |= update_nohz_stats(rq, true); + if (flags & NOHZ_STATS_KICK) + has_blocked_load |= update_nohz_stats(rq); /* * If time for next balance is due, @@ -9729,11 +12737,10 @@ static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, rq_lock_irqsave(rq, &rf); update_rq_clock(rq); - cpu_load_update_idle(rq); rq_unlock_irqrestore(rq, &rf); if (flags & NOHZ_BALANCE_KICK) - rebalance_domains(rq, CPU_IDLE); + sched_balance_domains(rq, CPU_IDLE); } if (time_after(next_balance, rq->next_balance)) { @@ -9742,26 +12749,6 @@ static bool _nohz_idle_balance(struct rq *this_rq, unsigned int flags, } } - /* Newly idle CPU doesn't need an update */ - if (idle != CPU_NEWLY_IDLE) { - update_blocked_averages(this_cpu); - has_blocked_load |= this_rq->has_blocked_load; - } - - if (flags & NOHZ_BALANCE_KICK) - rebalance_domains(this_rq, CPU_IDLE); - - WRITE_ONCE(nohz.next_blocked, - now + msecs_to_jiffies(LOAD_AVG_PERIOD)); - - /* The full idle balance loop has been done */ - ret = true; - -abort: - /* There is still blocked load, enable periodic update */ - if (has_blocked_load) - WRITE_ONCE(nohz.has_blocked, 1); - /* * next_balance will be updated only when there is a need. * When the CPU is attached to null domain for ex, it will not be @@ -9770,46 +12757,69 @@ abort: if (likely(update_next_balance)) nohz.next_balance = next_balance; - return ret; + if (flags & NOHZ_STATS_KICK) + WRITE_ONCE(nohz.next_blocked, + now + msecs_to_jiffies(LOAD_AVG_PERIOD)); + +abort: + /* There is still blocked load, enable periodic update */ + if (has_blocked_load) + WRITE_ONCE(nohz.has_blocked, 1); } /* * In CONFIG_NO_HZ_COMMON case, the idle balance kickee will do the - * rebalancing for all the cpus for whom scheduler ticks are stopped. + * rebalancing for all the CPUs for whom scheduler ticks are stopped. */ static bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) { - int this_cpu = this_rq->cpu; - unsigned int flags; + unsigned int flags = this_rq->nohz_idle_balance; - if (!(atomic_read(nohz_flags(this_cpu)) & NOHZ_KICK_MASK)) + if (!flags) return false; - if (idle != CPU_IDLE) { - atomic_andnot(NOHZ_KICK_MASK, nohz_flags(this_cpu)); - return false; - } + this_rq->nohz_idle_balance = 0; - /* could be _relaxed() */ - flags = atomic_fetch_andnot(NOHZ_KICK_MASK, nohz_flags(this_cpu)); - if (!(flags & NOHZ_KICK_MASK)) + if (idle != CPU_IDLE) return false; - _nohz_idle_balance(this_rq, flags, idle); + _nohz_idle_balance(this_rq, flags); return true; } -static void nohz_newidle_balance(struct rq *this_rq) +/* + * Check if we need to directly run the ILB for updating blocked load before + * entering idle state. Here we run ILB directly without issuing IPIs. + * + * Note that when this function is called, the tick may not yet be stopped on + * this CPU yet. nohz.idle_cpus_mask is updated only when tick is stopped and + * cleared on the next busy tick. In other words, nohz.idle_cpus_mask updates + * don't align with CPUs enter/exit idle to avoid bottlenecks due to high idle + * entry/exit rate (usec). So it is possible that _nohz_idle_balance() is + * called from this function on (this) CPU that's not yet in the mask. That's + * OK because the goal of nohz_run_idle_balance() is to run ILB only for + * updating the blocked load of already idle CPUs without waking up one of + * those idle CPUs and outside the preempt disable / IRQ off phase of the local + * cpu about to enter idle, because it can take a long time. + */ +void nohz_run_idle_balance(int cpu) { - int this_cpu = this_rq->cpu; + unsigned int flags; + + flags = atomic_fetch_andnot(NOHZ_NEWILB_KICK, nohz_flags(cpu)); /* - * This CPU doesn't want to be disturbed by scheduler - * housekeeping + * Update the blocked load only if no SCHED_SOFTIRQ is about to happen + * (i.e. NOHZ_STATS_KICK set) and will do the same. */ - if (!housekeeping_cpu(this_cpu, HK_FLAG_SCHED)) - return; + if ((flags == NOHZ_NEWILB_KICK) && !need_resched()) + _nohz_idle_balance(cpu_rq(cpu), NOHZ_STATS_KICK); +} + +static void nohz_newidle_balance(struct rq *this_rq) +{ + int this_cpu = this_rq->cpu; /* Will wake up very soon. No time for doing anything else*/ if (this_rq->avg_idle < sysctl_sched_migration_cost) @@ -9820,19 +12830,14 @@ static void nohz_newidle_balance(struct rq *this_rq) time_before(jiffies, READ_ONCE(nohz.next_blocked))) return; - raw_spin_unlock(&this_rq->lock); /* - * This CPU is going to be idle and blocked load of idle CPUs - * need to be updated. Run the ilb locally as it is a good - * candidate for ilb instead of waking up another idle CPU. - * Kick an normal ilb if we failed to do the update. + * Set the need to trigger ILB in order to update blocked load + * before entering idle state. */ - if (!_nohz_idle_balance(this_rq, NOHZ_STATS_KICK, CPU_NEWLY_IDLE)) - kick_ilb(NOHZ_STATS_KICK); - raw_spin_lock(&this_rq->lock); + atomic_or(NOHZ_NEWILB_KICK, nohz_flags(this_cpu)); } -#else /* !CONFIG_NO_HZ_COMMON */ +#else /* !CONFIG_NO_HZ_COMMON: */ static inline void nohz_balancer_kick(struct rq *rq) { } static inline bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) @@ -9841,23 +12846,39 @@ static inline bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle } static inline void nohz_newidle_balance(struct rq *this_rq) { } -#endif /* CONFIG_NO_HZ_COMMON */ +#endif /* !CONFIG_NO_HZ_COMMON */ /* - * idle_balance is called by schedule() if this_cpu is about to become + * sched_balance_newidle is called by schedule() if this_cpu is about to become * idle. Attempts to pull tasks from other CPUs. + * + * Returns: + * < 0 - we released the lock and there are !fair tasks present + * 0 - failed, no new tasks + * > 0 - success, new (fair) tasks present */ -static int idle_balance(struct rq *this_rq, struct rq_flags *rf) +static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) { unsigned long next_balance = jiffies + HZ; int this_cpu = this_rq->cpu; + int continue_balancing = 1; + u64 t0, t1, curr_cost = 0; struct sched_domain *sd; int pulled_task = 0; - u64 curr_cost = 0; + + update_misfit_status(NULL, this_rq); + + /* + * There is a task waiting to run. No need to search for one. + * Return 0; the task will be enqueued when switching to idle. + */ + if (this_rq->ttwu_pending) + return 0; /* - * We must set idle_stamp _before_ calling idle_balance(), such that we - * measure the duration of idle_balance() as idle time. + * We must set idle_stamp _before_ calling sched_balance_rq() + * for CPU_NEWLY_IDLE, such that we measure the this duration + * as idle time. */ this_rq->idle_stamp = rq_clock(this_rq); @@ -9875,85 +12896,106 @@ static int idle_balance(struct rq *this_rq, struct rq_flags *rf) */ rq_unpin_lock(this_rq, rf); - if (this_rq->avg_idle < sysctl_sched_migration_cost || - !READ_ONCE(this_rq->rd->overload)) { - - rcu_read_lock(); - sd = rcu_dereference_check_sched_domain(this_rq->sd); - if (sd) - update_next_balance(sd, &next_balance); + rcu_read_lock(); + sd = rcu_dereference_check_sched_domain(this_rq->sd); + if (!sd) { rcu_read_unlock(); + goto out; + } - nohz_newidle_balance(this_rq); + if (!get_rd_overloaded(this_rq->rd) || + this_rq->avg_idle < sd->max_newidle_lb_cost) { + update_next_balance(sd, &next_balance); + rcu_read_unlock(); goto out; } + rcu_read_unlock(); - raw_spin_unlock(&this_rq->lock); + rq_modified_clear(this_rq); + raw_spin_rq_unlock(this_rq); + + t0 = sched_clock_cpu(this_cpu); + sched_balance_update_blocked_averages(this_cpu); - update_blocked_averages(this_cpu); rcu_read_lock(); for_each_domain(this_cpu, sd) { - int continue_balancing = 1; - u64 t0, domain_cost; + u64 domain_cost; - if (!(sd->flags & SD_LOAD_BALANCE)) - continue; + update_next_balance(sd, &next_balance); - if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) { - update_next_balance(sd, &next_balance); + if (this_rq->avg_idle < curr_cost + sd->max_newidle_lb_cost) break; - } if (sd->flags & SD_BALANCE_NEWIDLE) { - t0 = sched_clock_cpu(this_cpu); + unsigned int weight = 1; + + if (sched_feat(NI_RANDOM)) { + /* + * Throw a 1k sided dice; and only run + * newidle_balance according to the success + * rate. + */ + u32 d1k = sched_rng() % 1024; + weight = 1 + sd->newidle_ratio; + if (d1k > weight) { + update_newidle_stats(sd, 0); + continue; + } + weight = (1024 + weight/2) / weight; + } - pulled_task = load_balance(this_cpu, this_rq, + pulled_task = sched_balance_rq(this_cpu, this_rq, sd, CPU_NEWLY_IDLE, &continue_balancing); - domain_cost = sched_clock_cpu(this_cpu) - t0; - if (domain_cost > sd->max_newidle_lb_cost) - sd->max_newidle_lb_cost = domain_cost; - + t1 = sched_clock_cpu(this_cpu); + domain_cost = t1 - t0; curr_cost += domain_cost; - } + t0 = t1; - update_next_balance(sd, &next_balance); + /* + * Track max cost of a domain to make sure to not delay the + * next wakeup on the CPU. + */ + update_newidle_cost(sd, domain_cost, weight * !!pulled_task); + } /* * Stop searching for tasks to pull if there are * now runnable tasks on this rq. */ - if (pulled_task || this_rq->nr_running > 0) + if (pulled_task || !continue_balancing) break; } rcu_read_unlock(); - raw_spin_lock(&this_rq->lock); + raw_spin_rq_lock(this_rq); if (curr_cost > this_rq->max_idle_balance_cost) this_rq->max_idle_balance_cost = curr_cost; -out: /* * While browsing the domains, we released the rq lock, a task could * 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; + /* If a higher prio class was modified, restart the pick */ + if (rq_modified_above(this_rq, &fair_sched_class)) + pulled_task = -1; + +out: /* Move the next balance forward */ if (time_after(this_rq->next_balance, next_balance)) this_rq->next_balance = next_balance; - /* Is there a task of a high priority class? */ - if (this_rq->nr_running != this_rq->cfs.h_nr_running) - pulled_task = -1; - if (pulled_task) this_rq->idle_stamp = 0; + else + nohz_newidle_balance(this_rq); rq_repin_lock(this_rq, rf); @@ -9961,19 +13003,21 @@ out: } /* - * run_rebalance_domains is triggered when needed from the scheduler tick. - * Also triggered for nohz idle balancing (with nohz_balancing_kick set). + * This softirq handler is triggered via SCHED_SOFTIRQ from two places: + * + * - directly from the local sched_tick() for periodic load balancing + * + * - indirectly from a remote sched_tick() for NOHZ idle balancing + * through the SMP cross-call nohz_csd_func() */ -static __latent_entropy void run_rebalance_domains(struct softirq_action *h) +static __latent_entropy void sched_balance_softirq(void) { struct rq *this_rq = this_rq(); - enum cpu_idle_type idle = this_rq->idle_balance ? - CPU_IDLE : CPU_NOT_IDLE; - + enum cpu_idle_type idle = this_rq->idle_balance; /* - * If this CPU has a pending nohz_balance_kick, then do the + * If this CPU has a pending NOHZ_BALANCE_KICK, then do the * balancing on behalf of the other idle CPUs whose ticks are - * stopped. Do nohz_idle_balance *before* rebalance_domains to + * stopped. Do nohz_idle_balance *before* sched_balance_domains to * give the idle CPUs a chance to load balance. Else we may * load balance only within the local sched_domain hierarchy * and abort nohz_idle_balance altogether if we pull some load. @@ -9982,17 +13026,20 @@ static __latent_entropy void run_rebalance_domains(struct softirq_action *h) return; /* normal load balance */ - update_blocked_averages(this_rq->cpu); - rebalance_domains(this_rq, idle); + sched_balance_update_blocked_averages(this_rq->cpu); + sched_balance_domains(this_rq, idle); } /* * Trigger the SCHED_SOFTIRQ if it is time to do periodic load balancing. */ -void trigger_load_balance(struct rq *rq) +void sched_balance_trigger(struct rq *rq) { - /* Don't need to rebalance while attached to NULL domain */ - if (unlikely(on_null_domain(rq))) + /* + * Don't need to rebalance while attached to NULL domain or + * runqueue CPU is not active + */ + if (unlikely(on_null_domain(rq) || !cpu_active(cpu_of(rq)))) return; if (time_after_eq(jiffies, rq->next_balance)) @@ -10014,9 +13061,300 @@ static void rq_offline_fair(struct rq *rq) /* Ensure any throttled groups are reachable by pick_next_task */ unthrottle_offline_cfs_rqs(rq); + + /* Ensure that we remove rq contribution to group share: */ + clear_tg_offline_cfs_rqs(rq); +} + +#ifdef CONFIG_SCHED_CORE +static inline bool +__entity_slice_used(struct sched_entity *se, int min_nr_tasks) +{ + u64 rtime = se->sum_exec_runtime - se->prev_sum_exec_runtime; + u64 slice = se->slice; + + return (rtime * min_nr_tasks > slice); +} + +#define MIN_NR_TASKS_DURING_FORCEIDLE 2 +static inline void task_tick_core(struct rq *rq, struct task_struct *curr) +{ + if (!sched_core_enabled(rq)) + return; + + /* + * If runqueue has only one task which used up its slice and + * if the sibling is forced idle, then trigger schedule to + * give forced idle task a chance. + * + * sched_slice() considers only this active rq and it gets the + * whole slice. But during force idle, we have siblings acting + * like a single runqueue and hence we need to consider runnable + * tasks on this CPU and the forced idle CPU. Ideally, we should + * go through the forced idle rq, but that would be a perf hit. + * We can assume that the forced idle CPU has at least + * 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_queued == 1 && + __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE)) + resched_curr(rq); +} + +/* + * Consider any infeasible weight scenario. Take for instance two tasks, + * each bound to their respective sibling, one with weight 1 and one with + * weight 2. Then the lower weight task will run ahead of the higher weight + * task without bound. + * + * This utterly destroys the concept of a shared time base. + * + * Remember; all this is about a proportionally fair scheduling, where each + * tasks receives: + * + * w_i + * dt_i = ---------- dt (1) + * \Sum_j w_j + * + * which we do by tracking a virtual time, s_i: + * + * 1 + * s_i = --- d[t]_i (2) + * w_i + * + * Where d[t] is a delta of discrete time, while dt is an infinitesimal. + * The immediate corollary is that the ideal schedule S, where (2) to use + * an infinitesimal delta, is: + * + * 1 + * S = ---------- dt (3) + * \Sum_i w_i + * + * From which we can define the lag, or deviation from the ideal, as: + * + * lag(i) = S - s_i (4) + * + * And since the one and only purpose is to approximate S, we get that: + * + * \Sum_i w_i lag(i) := 0 (5) + * + * If this were not so, we no longer converge to S, and we can no longer + * claim our scheduler has any of the properties we derive from S. This is + * exactly what you did above, you broke it! + * + * + * Let's continue for a while though; to see if there is anything useful to + * be learned. We can combine (1)-(3) or (4)-(5) and express S in s_i: + * + * \Sum_i w_i s_i + * S = -------------- (6) + * \Sum_i w_i + * + * Which gives us a way to compute S, given our s_i. Now, if you've read + * our code, you know that we do not in fact do this, the reason for this + * is two-fold. Firstly, computing S in that way requires a 64bit division + * for every time we'd use it (see 12), and secondly, this only describes + * the steady-state, it doesn't handle dynamics. + * + * Anyway, in (6): s_i -> x + (s_i - x), to get: + * + * \Sum_i w_i (s_i - x) + * S - x = -------------------- (7) + * \Sum_i w_i + * + * Which shows that S and s_i transform alike (which makes perfect sense + * given that S is basically the (weighted) average of s_i). + * + * So the thing to remember is that the above is strictly UP. It is + * possible to generalize to multiple runqueues -- however it gets really + * yuck when you have to add affinity support, as illustrated by our very + * first counter-example. + * + * Luckily I think we can avoid needing a full multi-queue variant for + * core-scheduling (or load-balancing). The crucial observation is that we + * only actually need this comparison in the presence of forced-idle; only + * then do we need to tell if the stalled rq has higher priority over the + * other. + * + * [XXX assumes SMT2; better consider the more general case, I suspect + * it'll work out because our comparison is always between 2 rqs and the + * answer is only interesting if one of them is forced-idle] + * + * And (under assumption of SMT2) when there is forced-idle, there is only + * a single queue, so everything works like normal. + * + * Let, for our runqueue 'k': + * + * T_k = \Sum_i w_i s_i + * W_k = \Sum_i w_i ; for all i of k (8) + * + * Then we can write (6) like: + * + * T_k + * S_k = --- (9) + * W_k + * + * From which immediately follows that: + * + * T_k + T_l + * S_k+l = --------- (10) + * W_k + W_l + * + * On which we can define a combined lag: + * + * lag_k+l(i) := S_k+l - s_i (11) + * + * And that gives us the tools to compare tasks across a combined runqueue. + * + * + * Combined this gives the following: + * + * a) when a runqueue enters force-idle, sync it against it's sibling rq(s) + * using (7); this only requires storing single 'time'-stamps. + * + * b) when comparing tasks between 2 runqueues of which one is forced-idle, + * compare the combined lag, per (11). + * + * Now, of course cgroups (I so hate them) make this more interesting in + * that a) seems to suggest we need to iterate all cgroup on a CPU at such + * boundaries, but I think we can avoid that. The force-idle is for the + * whole CPU, all it's rqs. So we can mark it in the root and lazily + * propagate downward on demand. + */ + +/* + * So this sync is basically a relative reset of S to 0. + * + * So with 2 queues, when one goes idle, we drop them both to 0 and one + * then increases due to not being idle, and the idle one builds up lag to + * get re-elected. So far so simple, right? + * + * When there's 3, we can have the situation where 2 run and one is idle, + * we sync to 0 and let the idle one build up lag to get re-election. Now + * suppose another one also drops idle. At this point dropping all to 0 + * again would destroy the built-up lag from the queue that was already + * idle, not good. + * + * So instead of syncing everything, we can: + * + * less := !((s64)(s_a - s_b) <= 0) + * + * (v_a - S_a) - (v_b - S_b) == v_a - v_b - S_a + S_b + * == v_a - (v_b - S_a + S_b) + * + * IOW, we can recast the (lag) comparison to a one-sided difference. + * So if then, instead of syncing the whole queue, sync the idle queue + * against the active queue with S_a + S_b at the point where we sync. + * + * (XXX consider the implication of living in a cyclic group: N / 2^n N) + * + * This gives us means of syncing single queues against the active queue, + * and for already idle queues to preserve their build-up lag. + * + * Of course, then we get the situation where there's 2 active and one + * going idle, who do we pick to sync against? Theory would have us sync + * against the combined S, but as we've already demonstrated, there is no + * such thing in infeasible weight scenarios. + * + * One thing I've considered; and this is where that core_active rudiment + * came from, is having active queues sync up between themselves after + * every tick. This limits the observed divergence due to the work + * conservancy. + * + * On top of that, we can improve upon things by employing (10) here. + */ + +/* + * se_fi_update - Update the cfs_rq->zero_vruntime_fi in a CFS hierarchy if needed. + */ +static void se_fi_update(const struct sched_entity *se, unsigned int fi_seq, + bool forceidle) +{ + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + if (forceidle) { + if (cfs_rq->forceidle_seq == fi_seq) + break; + cfs_rq->forceidle_seq = fi_seq; + } + + cfs_rq->zero_vruntime_fi = cfs_rq->zero_vruntime; + } } -#endif /* CONFIG_SMP */ +void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi) +{ + struct sched_entity *se = &p->se; + + if (p->sched_class != &fair_sched_class) + return; + + se_fi_update(se, rq->core->core_forceidle_seq, in_fi); +} + +bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, + bool in_fi) +{ + struct rq *rq = task_rq(a); + const struct sched_entity *sea = &a->se; + const struct sched_entity *seb = &b->se; + struct cfs_rq *cfs_rqa; + struct cfs_rq *cfs_rqb; + s64 delta; + + WARN_ON_ONCE(task_rq(b)->core != rq->core); + +#ifdef CONFIG_FAIR_GROUP_SCHED + /* + * Find an se in the hierarchy for tasks a and b, such that the se's + * are immediate siblings. + */ + while (sea->cfs_rq->tg != seb->cfs_rq->tg) { + int sea_depth = sea->depth; + int seb_depth = seb->depth; + + if (sea_depth >= seb_depth) + sea = parent_entity(sea); + if (sea_depth <= seb_depth) + seb = parent_entity(seb); + } + + se_fi_update(sea, rq->core->core_forceidle_seq, in_fi); + se_fi_update(seb, rq->core->core_forceidle_seq, in_fi); + + cfs_rqa = sea->cfs_rq; + cfs_rqb = seb->cfs_rq; +#else /* !CONFIG_FAIR_GROUP_SCHED: */ + cfs_rqa = &task_rq(a)->cfs; + cfs_rqb = &task_rq(b)->cfs; +#endif /* !CONFIG_FAIR_GROUP_SCHED */ + + /* + * Find delta after normalizing se's vruntime with its cfs_rq's + * zero_vruntime_fi, which would have been updated in prior calls + * to se_fi_update(). + */ + delta = (s64)(sea->vruntime - seb->vruntime) + + (s64)(cfs_rqb->zero_vruntime_fi - cfs_rqa->zero_vruntime_fi); + + return delta > 0; +} + +static int task_is_throttled_fair(struct task_struct *p, int cpu) +{ + struct cfs_rq *cfs_rq; + +#ifdef CONFIG_FAIR_GROUP_SCHED + cfs_rq = task_group(p)->cfs_rq[cpu]; +#else + cfs_rq = &cpu_rq(cpu)->cfs; +#endif + return throttled_hierarchy(cfs_rq); +} +#else /* !CONFIG_SCHED_CORE: */ +static inline void task_tick_core(struct rq *rq, struct task_struct *curr) {} +#endif /* !CONFIG_SCHED_CORE */ /* * scheduler tick hitting a task of our scheduling class. @@ -10040,7 +13378,9 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) task_tick_numa(rq, curr); update_misfit_status(curr, rq); - update_overutilized_status(task_rq(curr)); + check_update_overutilized_status(task_rq(curr)); + + task_tick_core(rq, curr); } /* @@ -10050,33 +13390,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) */ static void task_fork_fair(struct task_struct *p) { - struct cfs_rq *cfs_rq; - struct sched_entity *se = &p->se, *curr; - struct rq *rq = this_rq(); - struct rq_flags rf; - - rq_lock(rq, &rf); - update_rq_clock(rq); - - cfs_rq = task_cfs_rq(current); - curr = cfs_rq->curr; - if (curr) { - update_curr(cfs_rq); - se->vruntime = curr->vruntime; - } - place_entity(cfs_rq, se, 1); - - if (sysctl_sched_child_runs_first && curr && entity_before(curr, se)) { - /* - * Upon rescheduling, sched_class::put_prev_task() will place - * 'current' within the tree based on its new key value. - */ - swap(curr->vruntime, se->vruntime); - resched_curr(rq); - } - - se->vruntime -= cfs_rq->min_vruntime; - rq_unlock(rq, &rf); + set_task_max_allowed_capacity(p); } /* @@ -10084,49 +13398,28 @@ static void task_fork_fair(struct task_struct *p) * the current task. */ static void -prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) +prio_changed_fair(struct rq *rq, struct task_struct *p, u64 oldprio) { if (!task_on_rq_queued(p)) return; + if (p->prio == oldprio) + return; + + if (rq->cfs.nr_queued == 1) + return; + /* * Reschedule if we are currently running on this runqueue and * our priority decreased, or if we are not currently running on * this runqueue and our priority is higher than the current's */ - if (rq->curr == p) { + if (task_current_donor(rq, p)) { if (p->prio > oldprio) resched_curr(rq); - } else - check_preempt_curr(rq, p, 0); -} - -static inline bool vruntime_normalized(struct task_struct *p) -{ - struct sched_entity *se = &p->se; - - /* - * In both the TASK_ON_RQ_QUEUED and TASK_ON_RQ_MIGRATING cases, - * the dequeue_entity(.flags=0) will already have normalized the - * vruntime. - */ - if (p->on_rq) - return true; - - /* - * When !on_rq, vruntime of the task has usually NOT been normalized. - * But there are some cases where it has already been normalized: - * - * - A forked child which is waiting for being woken up by - * wake_up_new_task(). - * - A task which has been woken up by try_to_wake_up() and - * waiting for actually being woken up by sched_ttwu_pending(). - */ - if (!se->sum_exec_runtime || - (p->state == TASK_WAKING && p->sched_remote_wakeup)) - return true; - - return false; + } else { + wakeup_preempt(rq, p, 0); + } } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -10136,7 +13429,16 @@ static inline bool vruntime_normalized(struct task_struct *p) */ static void propagate_entity_cfs_rq(struct sched_entity *se) { - struct cfs_rq *cfs_rq; + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + /* + * If a task gets attached to this cfs_rq and before being queued, + * it gets migrated to another CPU due to reasons like affinity + * change, make sure this cfs_rq stays on leaf cfs_rq list to have + * that removed load decayed or it can cause faireness problem. + */ + if (!cfs_rq_pelt_clock_throttled(cfs_rq)) + list_add_leaf_cfs_rq(cfs_rq); /* Start to propagate at parent */ se = se->parent; @@ -10144,24 +13446,35 @@ static void propagate_entity_cfs_rq(struct sched_entity *se) for_each_sched_entity(se) { cfs_rq = cfs_rq_of(se); - if (cfs_rq_throttled(cfs_rq)) - break; - update_load_avg(cfs_rq, se, UPDATE_TG); + + if (!cfs_rq_pelt_clock_throttled(cfs_rq)) + list_add_leaf_cfs_rq(cfs_rq); } + + assert_list_leaf_cfs_rq(rq_of(cfs_rq)); } -#else +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static void propagate_entity_cfs_rq(struct sched_entity *se) { } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ static void detach_entity_cfs_rq(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); + /* + * In case the task sched_avg hasn't been attached: + * - A forked task which hasn't been woken up by wake_up_new_task(). + * - A task which has been woken up by try_to_wake_up() but is + * waiting for actually being woken up by sched_ttwu_pending(). + */ + if (!se->avg.last_update_time) + return; + /* Catch up with the cfs_rq and remove our load when we leave */ update_load_avg(cfs_rq, se, 0); detach_entity_load_avg(cfs_rq, se); - update_tg_load_avg(cfs_rq, false); + update_tg_load_avg(cfs_rq); propagate_entity_cfs_rq(se); } @@ -10169,34 +13482,16 @@ static void attach_entity_cfs_rq(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); -#ifdef CONFIG_FAIR_GROUP_SCHED - /* - * Since the real-depth could have been changed (only FAIR - * class maintain depth value), reset depth properly. - */ - se->depth = se->parent ? se->parent->depth + 1 : 0; -#endif - /* Synchronize entity with its cfs_rq */ update_load_avg(cfs_rq, se, sched_feat(ATTACH_AGE_LOAD) ? 0 : SKIP_AGE_LOAD); - attach_entity_load_avg(cfs_rq, se, 0); - update_tg_load_avg(cfs_rq, false); + attach_entity_load_avg(cfs_rq, se); + update_tg_load_avg(cfs_rq); propagate_entity_cfs_rq(se); } static void detach_task_cfs_rq(struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); - - if (!vruntime_normalized(p)) { - /* - * Fix up our vruntime so that the current sleep doesn't - * cause 'unlimited' sleep bonus. - */ - place_entity(cfs_rq, se, 0); - se->vruntime -= cfs_rq->min_vruntime; - } detach_entity_cfs_rq(se); } @@ -10204,12 +13499,14 @@ static void detach_task_cfs_rq(struct task_struct *p) static void attach_task_cfs_rq(struct task_struct *p) { struct sched_entity *se = &p->se; - struct cfs_rq *cfs_rq = cfs_rq_of(se); attach_entity_cfs_rq(se); +} - if (!vruntime_normalized(p)) - se->vruntime += cfs_rq->min_vruntime; +static void switching_from_fair(struct rq *rq, struct task_struct *p) +{ + if (p->se.sched_delayed) + dequeue_task(rq, p, DEQUEUE_SLEEP | DEQUEUE_DELAYED | DEQUEUE_NOCLOCK); } static void switched_from_fair(struct rq *rq, struct task_struct *p) @@ -10219,29 +13516,57 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p) static void switched_to_fair(struct rq *rq, struct task_struct *p) { + WARN_ON_ONCE(p->se.sched_delayed); + attach_task_cfs_rq(p); + set_task_max_allowed_capacity(p); + if (task_on_rq_queued(p)) { /* * We were most likely switched from sched_rt, so * kick off the schedule if running, otherwise just see * if we can still preempt the current task. */ - if (rq->curr == p) + if (task_current_donor(rq, p)) resched_curr(rq); else - check_preempt_curr(rq, p, 0); + wakeup_preempt(rq, p, 0); } } -/* Account for a task changing its policy or group. +static void __set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) +{ + struct sched_entity *se = &p->se; + + if (task_on_rq_queued(p)) { + /* + * Move the next running task to the front of the list, so our + * cfs_tasks list becomes MRU one. + */ + list_move(&se->group_node, &rq->cfs_tasks); + } + if (!first) + return; + + WARN_ON_ONCE(se->sched_delayed); + + if (hrtick_enabled_fair(rq)) + hrtick_start_fair(rq, p); + + update_misfit_status(p, rq); + sched_fair_update_stop_tick(rq, p); +} + +/* + * Account for a task changing its policy or group. * * This routine is mostly called to set cfs_rq->curr field when a task * migrates between groups/classes. */ -static void set_curr_task_fair(struct rq *rq) +static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) { - struct sched_entity *se = &rq->curr->se; + struct sched_entity *se = &p->se; for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); @@ -10250,60 +13575,39 @@ static void set_curr_task_fair(struct rq *rq) /* ensure bandwidth has been allocated on our new cfs_rq */ account_cfs_rq_runtime(cfs_rq, 0); } + + __set_next_task_fair(rq, p, first); } void init_cfs_rq(struct cfs_rq *cfs_rq) { cfs_rq->tasks_timeline = RB_ROOT_CACHED; - cfs_rq->min_vruntime = (u64)(-(1LL << 20)); -#ifndef CONFIG_64BIT - cfs_rq->min_vruntime_copy = cfs_rq->min_vruntime; -#endif -#ifdef CONFIG_SMP + cfs_rq->zero_vruntime = (u64)(-(1LL << 20)); raw_spin_lock_init(&cfs_rq->removed.lock); -#endif } #ifdef CONFIG_FAIR_GROUP_SCHED -static void task_set_group_fair(struct task_struct *p) +static void task_change_group_fair(struct task_struct *p) { - struct sched_entity *se = &p->se; - - set_task_rq(p, task_cpu(p)); - se->depth = se->parent ? se->parent->depth + 1 : 0; -} + /* + * We couldn't detach or attach a forked task which + * hasn't been woken up by wake_up_new_task(). + */ + if (READ_ONCE(p->__state) == TASK_NEW) + return; -static void task_move_group_fair(struct task_struct *p) -{ detach_task_cfs_rq(p); - set_task_rq(p, task_cpu(p)); -#ifdef CONFIG_SMP /* Tell se's cfs_rq has been changed -- migrated */ p->se.avg.last_update_time = 0; -#endif + set_task_rq(p, task_cpu(p)); attach_task_cfs_rq(p); } -static void task_change_group_fair(struct task_struct *p, int type) -{ - switch (type) { - case TASK_SET_GROUP: - task_set_group_fair(p); - break; - - case TASK_MOVE_GROUP: - task_move_group_fair(p); - break; - } -} - void free_fair_sched_group(struct task_group *tg) { int i; - destroy_cfs_bandwidth(tg_cfs_bandwidth(tg)); - for_each_possible_cpu(i) { if (tg->cfs_rq) kfree(tg->cfs_rq[i]); @@ -10330,7 +13634,7 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) tg->shares = NICE_0_LOAD; - init_cfs_bandwidth(tg_cfs_bandwidth(tg)); + init_cfs_bandwidth(tg_cfs_bandwidth(tg), tg_cfs_bandwidth(parent)); for_each_possible_cpu(i) { cfs_rq = kzalloc_node(sizeof(struct cfs_rq), @@ -10338,7 +13642,7 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) if (!cfs_rq) goto err; - se = kzalloc_node(sizeof(struct sched_entity), + se = kzalloc_node(sizeof(struct sched_entity_stats), GFP_KERNEL, cpu_to_node(i)); if (!se) goto err_free_rq; @@ -10359,43 +13663,52 @@ err: void online_fair_sched_group(struct task_group *tg) { struct sched_entity *se; + struct rq_flags rf; struct rq *rq; int i; for_each_possible_cpu(i) { rq = cpu_rq(i); se = tg->se[i]; - - raw_spin_lock_irq(&rq->lock); + rq_lock_irq(rq, &rf); update_rq_clock(rq); attach_entity_cfs_rq(se); sync_throttle(tg, i); - raw_spin_unlock_irq(&rq->lock); + rq_unlock_irq(rq, &rf); } } void unregister_fair_sched_group(struct task_group *tg) { - unsigned long flags; - struct rq *rq; int cpu; + destroy_cfs_bandwidth(tg_cfs_bandwidth(tg)); + for_each_possible_cpu(cpu) { - if (tg->se[cpu]) - remove_entity_load_avg(tg->se[cpu]); + struct cfs_rq *cfs_rq = tg->cfs_rq[cpu]; + struct sched_entity *se = tg->se[cpu]; + struct rq *rq = cpu_rq(cpu); + + if (se) { + if (se->sched_delayed) { + guard(rq_lock_irqsave)(rq); + if (se->sched_delayed) { + update_rq_clock(rq); + dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); + } + list_del_leaf_cfs_rq(cfs_rq); + } + remove_entity_load_avg(se); + } /* * Only empty task groups can be destroyed; so we can speculatively * check on_list without danger of it being re-added. */ - if (!tg->cfs_rq[cpu]->on_list) - continue; - - rq = cpu_rq(cpu); - - raw_spin_lock_irqsave(&rq->lock, flags); - list_del_leaf_cfs_rq(tg->cfs_rq[cpu]); - raw_spin_unlock_irqrestore(&rq->lock, flags); + if (cfs_rq->on_list) { + guard(rq_lock_irqsave)(rq); + list_del_leaf_cfs_rq(cfs_rq); + } } } @@ -10432,10 +13745,12 @@ void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, static DEFINE_MUTEX(shares_mutex); -int sched_group_set_shares(struct task_group *tg, unsigned long shares) +static int __sched_group_set_shares(struct task_group *tg, unsigned long shares) { int i; + lockdep_assert_held(&shares_mutex); + /* * We can't change the weight of the root cgroup. */ @@ -10444,9 +13759,8 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) shares = clamp(shares, scale_load(MIN_SHARES), scale_load(MAX_SHARES)); - mutex_lock(&shares_mutex); if (tg->shares == shares) - goto done; + return 0; tg->shares = shares; for_each_possible_cpu(i) { @@ -10464,22 +13778,87 @@ int sched_group_set_shares(struct task_group *tg, unsigned long shares) rq_unlock_irqrestore(rq, &rf); } -done: - mutex_unlock(&shares_mutex); return 0; } -#else /* CONFIG_FAIR_GROUP_SCHED */ -void free_fair_sched_group(struct task_group *tg) { } - -int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) +int sched_group_set_shares(struct task_group *tg, unsigned long shares) { - return 1; + int ret; + + mutex_lock(&shares_mutex); + if (tg_is_idle(tg)) + ret = -EINVAL; + else + ret = __sched_group_set_shares(tg, shares); + mutex_unlock(&shares_mutex); + + return ret; } -void online_fair_sched_group(struct task_group *tg) { } +int sched_group_set_idle(struct task_group *tg, long idle) +{ + int i; + + if (tg == &root_task_group) + return -EINVAL; + + if (idle < 0 || idle > 1) + return -EINVAL; + + mutex_lock(&shares_mutex); + + if (tg->idle == idle) { + mutex_unlock(&shares_mutex); + return 0; + } + + tg->idle = idle; + + for_each_possible_cpu(i) { + struct rq *rq = cpu_rq(i); + struct sched_entity *se = tg->se[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; + + rq_lock_irqsave(rq, &rf); + + grp_cfs_rq->idle = idle; + if (WARN_ON_ONCE(was_idle == cfs_rq_is_idle(grp_cfs_rq))) + goto next_cpu; + + 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; + + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(se); + + if (!se->on_rq) + break; + + cfs_rq->h_nr_idle += idle_task_delta; + + /* Already accounted at parent level and above. */ + if (cfs_rq_is_idle(cfs_rq)) + break; + } + +next_cpu: + rq_unlock_irqrestore(rq, &rf); + } -void unregister_fair_sched_group(struct task_group *tg) { } + /* Idle groups have minimum weight. */ + if (tg_is_idle(tg)) + __sched_group_set_shares(tg, scale_load(WEIGHT_IDLEPRIO)); + else + __sched_group_set_shares(tg, NICE_0_LOAD); + + mutex_unlock(&shares_mutex); + return 0; +} #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -10494,7 +13873,7 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task * idle runqueue: */ if (rq->cfs.load.weight) - rr_interval = NS_TO_JIFFIES(sched_slice(cfs_rq_of(se), se)); + rr_interval = NS_TO_JIFFIES(se->slice); return rr_interval; } @@ -10502,19 +13881,22 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task /* * All the scheduling class methods: */ -const struct sched_class fair_sched_class = { - .next = &idle_sched_class, +DEFINE_SCHED_CLASS(fair) = { + + .queue_mask = 2, + .enqueue_task = enqueue_task_fair, .dequeue_task = dequeue_task_fair, .yield_task = yield_task_fair, .yield_to_task = yield_to_task_fair, - .check_preempt_curr = check_preempt_wakeup, + .wakeup_preempt = check_preempt_wakeup_fair, + .pick_task = pick_task_fair, .pick_next_task = pick_next_task_fair, .put_prev_task = put_prev_task_fair, + .set_next_task = set_next_task_fair, -#ifdef CONFIG_SMP .select_task_rq = select_task_rq_fair, .migrate_task_rq = migrate_task_rq_fair, @@ -10522,14 +13904,14 @@ const struct sched_class fair_sched_class = { .rq_offline = rq_offline_fair, .task_dead = task_dead_fair, - .set_cpus_allowed = set_cpus_allowed_common, -#endif + .set_cpus_allowed = set_cpus_allowed_fair, - .set_curr_task = set_curr_task_fair, .task_tick = task_tick_fair, .task_fork = task_fork_fair, + .reweight_task = reweight_task_fair, .prio_changed = prio_changed_fair, + .switching_from = switching_from_fair, .switched_from = switched_from_fair, .switched_to = switched_to_fair, @@ -10540,15 +13922,22 @@ const struct sched_class fair_sched_class = { #ifdef CONFIG_FAIR_GROUP_SCHED .task_change_group = task_change_group_fair, #endif + +#ifdef CONFIG_SCHED_CORE + .task_is_throttled = task_is_throttled_fair, +#endif + +#ifdef CONFIG_UCLAMP_TASK + .uclamp_enabled = 1, +#endif }; -#ifdef CONFIG_SCHED_DEBUG void print_cfs_stats(struct seq_file *m, int cpu) { - struct cfs_rq *cfs_rq; + struct cfs_rq *cfs_rq, *pos; rcu_read_lock(); - for_each_leaf_cfs_rq(cpu_rq(cpu), cfs_rq) + for_each_leaf_cfs_rq_safe(cpu_rq(cpu), cfs_rq, pos) print_cfs_rq(m, cpu, cfs_rq); rcu_read_unlock(); } @@ -10558,32 +13947,46 @@ void show_numa_stats(struct task_struct *p, struct seq_file *m) { int node; unsigned long tsf = 0, tpf = 0, gsf = 0, gpf = 0; + struct numa_group *ng; + rcu_read_lock(); + ng = rcu_dereference(p->numa_group); for_each_online_node(node) { if (p->numa_faults) { tsf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 0)]; tpf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 1)]; } - if (p->numa_group) { - gsf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 0)], - gpf = p->numa_group->faults[task_faults_idx(NUMA_MEM, node, 1)]; + if (ng) { + gsf = ng->faults[task_faults_idx(NUMA_MEM, node, 0)], + gpf = ng->faults[task_faults_idx(NUMA_MEM, node, 1)]; } print_numa_stats(m, node, tsf, tpf, gsf, gpf); } + rcu_read_unlock(); } #endif /* CONFIG_NUMA_BALANCING */ -#endif /* CONFIG_SCHED_DEBUG */ __init void init_sched_fair_class(void) { -#ifdef CONFIG_SMP - open_softirq(SCHED_SOFTIRQ, run_rebalance_domains); + int i; + + for_each_possible_cpu(i) { + zalloc_cpumask_var_node(&per_cpu(load_balance_mask, i), GFP_KERNEL, cpu_to_node(i)); + zalloc_cpumask_var_node(&per_cpu(select_rq_mask, i), GFP_KERNEL, cpu_to_node(i)); + zalloc_cpumask_var_node(&per_cpu(should_we_balance_tmpmask, i), + GFP_KERNEL, cpu_to_node(i)); + +#ifdef CONFIG_CFS_BANDWIDTH + INIT_CSD(&cpu_rq(i)->cfsb_csd, __cfsb_csd_unthrottle, cpu_rq(i)); + INIT_LIST_HEAD(&cpu_rq(i)->cfsb_csd_list); +#endif + } + + open_softirq(SCHED_SOFTIRQ, sched_balance_softirq); #ifdef CONFIG_NO_HZ_COMMON nohz.next_balance = jiffies; nohz.next_blocked = jiffies; zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT); #endif -#endif /* SMP */ - } |