diff options
Diffstat (limited to 'kernel/sched/fair.c')
| -rw-r--r-- | kernel/sched/fair.c | 5438 |
1 files changed, 3468 insertions, 1970 deletions
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 0f8736991427..da46c3164537 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -37,6 +37,7 @@ #include <linux/sched/cputime.h> #include <linux/sched/isolation.h> #include <linux/sched/nohz.h> +#include <linux/sched/prio.h> #include <linux/cpuidle.h> #include <linux/interrupt.h> @@ -47,38 +48,23 @@ #include <linux/psi.h> #include <linux/ratelimit.h> #include <linux/task_work.h> +#include <linux/rbtree_augmented.h> #include <asm/switch_to.h> -#include <linux/sched/cond_resched.h> +#include <uapi/linux/sched/types.h> #include "sched.h" #include "stats.h" #include "autogroup.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; - -/* * The initial- and re-scaling of tunables is configurable * * 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)) @@ -88,58 +74,20 @@ 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) - */ -unsigned int sysctl_sched_min_granularity = 750000ULL; -static unsigned int normalized_sysctl_sched_min_granularity = 750000ULL; - -/* - * Minimal preemption granularity for CPU-bound SCHED_IDLE tasks. - * Applies only when SCHED_IDLE tasks compete with normal tasks. - * - * (default: 0.75 msec) - */ -unsigned int sysctl_sched_idle_min_granularity = 750000ULL; - -/* - * This value is kept at sysctl_sched_latency/sysctl_sched_min_granularity + * (default: 0.70 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -static unsigned int sched_nr_latency = 8; +unsigned int sysctl_sched_base_slice = 700000ULL; +static unsigned int normalized_sysctl_sched_base_slice = 700000ULL; -/* - * 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; - -int sched_thermal_decay_shift; static int __init setup_sched_thermal_decay_shift(char *str) { - int _shift = 0; - - if (kstrtoint(str, 0, &_shift)) - pr_warn("Unable to set scheduler thermal pressure decay shift parameter\n"); - - sched_thermal_decay_shift = clamp(_shift, 0, 10); + 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. */ @@ -162,7 +110,6 @@ int __weak arch_asym_cpu_priority(int cpu) * (default: ~5%) */ #define capacity_greater(cap1, cap2) ((cap1) * 1024 > (cap2) * 1078) -#endif #ifdef CONFIG_CFS_BANDWIDTH /* @@ -184,14 +131,7 @@ static unsigned int sysctl_numa_balancing_promote_rate_limit = 65536; #endif #ifdef CONFIG_SYSCTL -static struct ctl_table sched_fair_sysctls[] = { - { - .procname = "sched_child_runs_first", - .data = &sysctl_sched_child_runs_first, - .maxlen = sizeof(unsigned int), - .mode = 0644, - .proc_handler = proc_dointvec, - }, +static const struct ctl_table sched_fair_sysctls[] = { #ifdef CONFIG_CFS_BANDWIDTH { .procname = "sched_cfs_bandwidth_slice_us", @@ -212,7 +152,6 @@ static struct ctl_table sched_fair_sysctls[] = { .extra1 = SYSCTL_ZERO, }, #endif /* CONFIG_NUMA_BALANCING */ - {} }; static int __init sched_fair_sysctl_init(void) @@ -221,7 +160,7 @@ static int __init sched_fair_sysctl_init(void) return 0; } late_initcall(sched_fair_sysctl_init); -#endif +#endif /* CONFIG_SYSCTL */ static inline void update_load_add(struct load_weight *lw, unsigned long inc) { @@ -277,9 +216,7 @@ 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 } @@ -347,6 +284,16 @@ static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight 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; @@ -435,8 +382,8 @@ static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) /* * With cfs_rq being unthrottled/throttled during an enqueue, - * it can happen the tmp_alone_branch points the a leaf that - * we finally want to del. In this case, tmp_alone_branch moves + * 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. */ @@ -450,10 +397,10 @@ static inline void list_del_leaf_cfs_rq(struct cfs_rq *cfs_rq) static inline void assert_list_leaf_cfs_rq(struct rq *rq) { - SCHED_WARN_ON(rq->tmp_alone_branch != &rq->leaf_cfs_rq_list); + WARN_ON_ONCE(rq->tmp_alone_branch != &rq->leaf_cfs_rq_list); } -/* Iterate thr' all leaf cfs_rq's on a runqueue */ +/* 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) @@ -468,7 +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; } @@ -522,7 +469,7 @@ static int se_is_idle(struct sched_entity *se) return cfs_rq_is_idle(group_cfs_rq(se)); } -#else /* !CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED: */ #define for_each_sched_entity(se) \ for (; se; se = NULL) @@ -565,10 +512,10 @@ static int cfs_rq_is_idle(struct cfs_rq *cfs_rq) static int se_is_idle(struct sched_entity *se) { - return 0; + 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); @@ -577,7 +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) @@ -586,7 +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) @@ -595,202 +542,511 @@ static inline u64 min_vruntime(u64 min_vruntime, u64 vruntime) return min_vruntime; } -static inline bool 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) +{ + /* + * Tiebreak on vruntime seems unnecessary since it can + * hardly happen. + */ + return (s64)(a->deadline - b->deadline) < 0; +} + +static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return (s64)(a->vruntime - b->vruntime) < 0; + return (s64)(se->vruntime - cfs_rq->zero_vruntime); } #define __node_2_se(node) \ rb_entry((node), struct sched_entity, run_node) -static void update_min_vruntime(struct cfs_rq *cfs_rq) +/* + * 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 = __node_2_se(leftmost); - - 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. */ - u64_u32_store(cfs_rq->min_vruntime, - max_vruntime(cfs_rq->min_vruntime, vruntime)); + return cfs_rq->zero_vruntime + avg; } -static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) +/* + * 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 update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return entity_before(__node_2_se(a), __node_2_se(b)); + s64 vlag, limit; + + 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); } /* - * Enqueue an entity into the rb-tree: + * 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 void __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) +static int vruntime_eligible(struct cfs_rq *cfs_rq, u64 vruntime) { - rb_add_cached(&se->run_node, &cfs_rq->tasks_timeline, __entity_less); + 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; + } + + 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 __node_2_se(left); + 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 __node_2_se(next); + 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 __node_2_se(last); +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: - */ +static inline void __min_slice_update(struct sched_entity *se, struct rb_node *node) +{ + if (node) { + struct sched_entity *rse = __node_2_se(node); + if (rse->min_slice < se->min_slice) + se->min_slice = rse->min_slice; + } +} -int sched_update_scaling(void) +/* + * se->min_vruntime = min(se->vruntime, {left,right}->min_vruntime) + */ +static inline bool min_vruntime_update(struct sched_entity *se, bool exit) { - unsigned int factor = get_update_sysctl_factor(); + 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 bool sched_idle_cfs_rq(struct cfs_rq *cfs_rq); +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. * - * s = p*P[w/rw] + * 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: + * + * 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) { - unsigned int nr_running = cfs_rq->nr_running; - struct sched_entity *init_se = se; - unsigned int min_gran; - u64 slice; + 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; - if (sched_feat(ALT_PERIOD)) - nr_running = rq_of(cfs_rq)->cfs.h_nr_running; + /* + * 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; + + /* + * 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; + } - slice = __sched_period(nr_running + !se->on_rq); + if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr))) + curr = NULL; - for_each_sched_entity(se) { - struct load_weight *load; - struct load_weight lw; - struct cfs_rq *qcfs_rq; + if (curr && protect && protect_slice(curr)) + return curr; - qcfs_rq = cfs_rq_of(se); - load = &qcfs_rq->load; + /* Pick the leftmost entity if it's eligible */ + if (se && entity_eligible(cfs_rq, se)) { + best = se; + goto found; + } - if (unlikely(!se->on_rq)) { - lw = qcfs_rq->load; + /* 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; } - slice = __calc_delta(slice, se->load.weight, load); - } - if (sched_feat(BASE_SLICE)) { - if (se_is_idle(init_se) && !sched_idle_cfs_rq(cfs_rq)) - min_gran = sysctl_sched_idle_min_granularity; - else - min_gran = sysctl_sched_min_granularity; + 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 = max_t(u64, slice, min_gran); + node = node->rb_right; } +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 slice; + 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; } #include "pelt.h" -#ifdef CONFIG_SMP static int select_idle_sibling(struct task_struct *p, int prev_cpu, int cpu); static unsigned long task_h_load(struct task_struct *p); @@ -812,14 +1068,15 @@ void init_entity_runnable_average(struct sched_entity *se) if (entity_is_task(se)) 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 */ } /* * 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 @@ -866,7 +1123,7 @@ void post_init_entity_util_avg(struct task_struct *p) 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) @@ -879,64 +1136,111 @@ void post_init_entity_util_avg(struct task_struct *p) sa->runnable_avg = sa->util_avg; } -#else /* !CONFIG_SMP */ -void init_entity_runnable_average(struct sched_entity *se) -{ -} -void post_init_entity_util_avg(struct task_struct *p) +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 *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; + } + + if (schedstat_enabled()) { + struct sched_statistics *stats; + + 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) + +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; - - if (schedstat_enabled()) { - struct sched_statistics *stats; - - stats = __schedstats_from_se(curr); - __schedstat_set(stats->exec_max, - max(delta_exec, stats->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 @@ -1064,6 +1368,23 @@ 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 @@ -1158,7 +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; @@ -1358,7 +1679,7 @@ static unsigned long score_nearby_nodes(struct task_struct *p, int nid, 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; @@ -1484,7 +1805,7 @@ static bool pgdat_free_space_enough(struct pglist_data *pgdat) continue; if (zone_watermark_ok(zone, 0, - wmark_pages(zone, WMARK_PROMO) + enough_wmark, + promo_wmark_pages(zone) + enough_wmark, ZONE_MOVABLE, 0)) return true; } @@ -1503,12 +1824,12 @@ static bool pgdat_free_space_enough(struct pglist_data *pgdat) * The smaller the hint page fault latency, the higher the possibility * for the page to be hot. */ -static int numa_hint_fault_latency(struct page *page) +static int numa_hint_fault_latency(struct folio *folio) { int last_time, time; time = jiffies_to_msecs(jiffies); - last_time = xchg_page_access_time(page, time); + last_time = folio_xchg_access_time(folio, time); return (time - last_time) & PAGE_ACCESS_TIME_MASK; } @@ -1565,7 +1886,7 @@ static void numa_promotion_adjust_threshold(struct pglist_data *pgdat, } } -bool should_numa_migrate_memory(struct task_struct *p, struct page * page, +bool should_numa_migrate_memory(struct task_struct *p, struct folio *folio, int src_nid, int dst_cpu) { struct numa_group *ng = deref_curr_numa_group(p); @@ -1573,38 +1894,43 @@ bool should_numa_migrate_memory(struct task_struct *p, struct page * page, 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 (sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING && - !node_is_toptier(src_nid)) { + 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 = sysctl_numa_balancing_promote_rate_limit << \ - (20 - PAGE_SHIFT); + 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(page); + latency = numa_hint_fault_latency(folio); if (latency >= th) return false; - return !numa_promotion_rate_limit(pgdat, rate_limit, - thp_nr_pages(page)); + 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)) @@ -1700,23 +2026,6 @@ struct numa_stats { int idle_cpu; }; -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; -} - struct task_numa_env { struct task_struct *p; @@ -1772,12 +2081,12 @@ static inline int numa_idle_core(int idle_core, int cpu) return idle_core; } -#else +#else /* !CONFIG_SCHED_SMT: */ static inline int numa_idle_core(int idle_core, int cpu) { return idle_core; } -#endif +#endif /* !CONFIG_SCHED_SMT */ /* * Gather all necessary information to make NUMA balancing placement @@ -1801,10 +2110,10 @@ static void update_numa_stats(struct task_numa_env *env, ns->load += cpu_load(rq); ns->runnable += cpu_runnable(rq); ns->util += cpu_util_cfs(cpu); - ns->nr_running += rq->cfs.h_nr_running; + ns->nr_running += rq->cfs.h_nr_runnable; ns->compute_capacity += capacity_of(cpu); - if (find_idle && !rq->nr_running && idle_cpu(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; @@ -1836,7 +2145,7 @@ static void task_numa_assign(struct task_numa_env *env, int start = env->dst_cpu; /* Find alternative idle CPU. */ - for_each_cpu_wrap(cpu, cpumask_of_node(env->dst_nid), start) { + 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; @@ -1931,7 +2240,8 @@ static bool task_numa_compare(struct task_numa_env *env, rcu_read_lock(); cur = rcu_dereference(dst_rq->curr); - if (cur && ((cur->flags & PF_EXITING) || is_idle_task(cur))) + if (cur && ((cur->flags & (PF_EXITING | PF_KTHREAD)) || + !cur->mm)) cur = NULL; /* @@ -2645,19 +2955,7 @@ static void task_numa_placement(struct task_struct *p) } /* Cannot migrate task to CPU-less node */ - if (max_nid != NUMA_NO_NODE && !node_state(max_nid, N_CPU)) { - int near_nid = max_nid; - int distance, near_distance = INT_MAX; - - for_each_node_state(nid, N_CPU) { - distance = node_distance(max_nid, nid); - if (distance < near_distance) { - near_nid = nid; - near_distance = distance; - } - } - max_nid = near_nid; - } + max_nid = numa_nearest_node(max_nid, N_CPU); if (ng) { numa_group_count_active_nodes(ng); @@ -2928,6 +3226,45 @@ 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(). @@ -2938,13 +3275,15 @@ static void task_numa_work(struct callback_head *work) struct task_struct *p = current; struct mm_struct *mm = p->mm; u64 runtime = p->se.sum_exec_runtime; - MA_STATE(mas, &mm->mm_mt, 0, 0); struct vm_area_struct *vma; 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; /* @@ -2958,6 +3297,15 @@ static 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); @@ -2985,7 +3333,6 @@ static 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 */ @@ -2995,37 +3342,118 @@ static void task_numa_work(struct callback_head *work) if (!mmap_read_trylock(mm)) return; - mas_set(&mas, start); - vma = mas_find(&mas, ULONG_MAX); + + /* + * 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; - mas_set(&mas, start); - vma = mas_find(&mas, ULONG_MAX); + vma_iter_set(&vmi, start); + vma = vma_next(&vmi); } - for (; vma; vma = mas_find(&mas, ULONG_MAX)) { + 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_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_is_accessible(vma)) + 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); @@ -3036,7 +3464,7 @@ static 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. @@ -3051,6 +3479,26 @@ static 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: @@ -3078,7 +3526,7 @@ out: } } -void init_numa_balancing(unsigned long clone_flags, struct task_struct *p) +void init_numa_balancing(u64 clone_flags, struct task_struct *p) { int mm_users = 0; struct mm_struct *mm = p->mm; @@ -3192,7 +3640,8 @@ static void update_scan_period(struct task_struct *p, int new_cpu) 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) { } @@ -3209,38 +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); -#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++; - if (se_is_idle(se)) - cfs_rq->idle_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); -#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--; - if (se_is_idle(se)) - cfs_rq->idle_nr_running--; + cfs_rq->nr_queued--; } /* @@ -3291,7 +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_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { @@ -3308,55 +3748,67 @@ dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) 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_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) { + bool curr = cfs_rq->curr == se; + if (se->on_rq) { /* commit outstanding execution time */ - if (cfs_rq->curr == se) - update_curr(cfs_rq); + 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); + /* + * 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 = get_pelt_divider(&se->avg); se->avg.load_avg = div_u64(se_weight(se) * se->avg.load_sum, divider); } while (0); -#endif enqueue_load_avg(cfs_rq, se); - if (se->on_rq) + if (se->on_rq) { + 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); - 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. @@ -3463,7 +3915,6 @@ static long calc_group_shares(struct cfs_rq *cfs_rq) */ return clamp_t(long, shares, MIN_SHARES, tg_shares); } -#endif /* CONFIG_SMP */ /* * Recomputes the group entity based on the current state of its group @@ -3474,29 +3925,23 @@ static void update_cfs_group(struct sched_entity *se) struct cfs_rq *gcfs_rq = group_cfs_rq(se); long shares; - if (!gcfs_rq) - return; - - if (throttled_hierarchy(gcfs_rq)) - return; - -#ifndef CONFIG_SMP - shares = READ_ONCE(gcfs_rq->tg->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); -#endif - reweight_entity(cfs_rq_of(se), se, shares); + 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) { @@ -3521,7 +3966,6 @@ static inline void cfs_rq_util_change(struct cfs_rq *cfs_rq, int flags) } } -#ifdef CONFIG_SMP static inline bool load_avg_is_decayed(struct sched_avg *sa) { if (sa->load_sum) @@ -3538,7 +3982,7 @@ static inline bool load_avg_is_decayed(struct sched_avg *sa) * Make sure that rounding and/or propagation of PELT values never * break this. */ - SCHED_WARN_ON(sa->load_avg || + WARN_ON_ONCE(sa->load_avg || sa->util_avg || sa->runnable_avg); @@ -3563,15 +4007,17 @@ 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 { - struct rq *rq = rq_of(cfs_rq); - 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); @@ -3588,6 +4034,9 @@ static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) if (child_cfs_rq_on_list(cfs_rq)) return false; + if (cfs_rq->tg_load_avg_contrib) + return false; + return true; } @@ -3607,7 +4056,8 @@ static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) */ 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. @@ -3615,10 +4065,67 @@ static inline void update_tg_load_avg(struct cfs_rq *cfs_rq) if (cfs_rq->tg == &root_task_group) return; + /* 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); } /* @@ -3913,10 +4420,12 @@ 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 clear_tg_offline_cfs_rqs(struct rq *rq) {} + static inline int propagate_entity_load_avg(struct sched_entity *se) { return 0; @@ -3924,7 +4433,7 @@ 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) @@ -4005,9 +4514,9 @@ static inline void migrate_se_pelt_lag(struct sched_entity *se) __update_load_avg_blocked_se(now, se); } -#else +#else /* !CONFIG_NO_HZ_COMMON: */ static void migrate_se_pelt_lag(struct sched_entity *se) {} -#endif +#endif /* !CONFIG_NO_HZ_COMMON */ /** * update_cfs_rq_load_avg - update the cfs_rq's load/util averages @@ -4189,7 +4698,7 @@ static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s /* * 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, cfs_rq, se); @@ -4272,40 +4781,27 @@ static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) return cfs_rq->avg.load_avg; } -static int newidle_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 max(ue.ewma, (ue.enqueued & ~UTIL_AVG_UNCHANGED)); + return READ_ONCE(p->se.avg.runnable_avg); } -static inline unsigned long task_util_est(struct task_struct *p) +static inline unsigned long _task_util_est(struct task_struct *p) { - return max(task_util(p), _task_util_est(p)); + return READ_ONCE(p->se.avg.util_est) & ~UTIL_AVG_UNCHANGED; } -#ifdef CONFIG_UCLAMP_TASK -static inline unsigned long uclamp_task_util(struct task_struct *p, - unsigned long uclamp_min, - unsigned long uclamp_max) -{ - return clamp(task_util_est(p), uclamp_min, uclamp_max); -} -#else -static inline unsigned long uclamp_task_util(struct task_struct *p, - unsigned long uclamp_min, - unsigned long uclamp_max) +static inline unsigned long task_util_est(struct task_struct *p) { - return task_util_est(p); + return max(task_util(p), _task_util_est(p)); } -#endif static inline void util_est_enqueue(struct cfs_rq *cfs_rq, struct task_struct *p) @@ -4316,9 +4812,9 @@ 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); trace_sched_util_est_cfs_tp(cfs_rq); } @@ -4332,34 +4828,20 @@ static inline void util_est_dequeue(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 -= min_t(unsigned int, enqueued, _task_util_est(p)); - WRITE_ONCE(cfs_rq->avg.util_est.enqueued, enqueued); + WRITE_ONCE(cfs_rq->avg.util_est, enqueued); trace_sched_util_est_cfs_tp(cfs_rq); } #define UTIL_EST_MARGIN (SCHED_CAPACITY_SCALE / 100) -/* - * 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 + margin < INT_MAX. - */ -static inline bool within_margin(int value, int margin) -{ - return ((unsigned int)(value + margin - 1) < (2 * margin - 1)); -} - static inline void util_est_update(struct cfs_rq *cfs_rq, struct task_struct *p, bool task_sleep) { - long last_ewma_diff, last_enqueued_diff; - struct util_est ue; + unsigned int ewma, dequeued, last_ewma_diff; if (!sched_feat(UTIL_EST)) return; @@ -4371,82 +4853,86 @@ static inline void util_est_update(struct cfs_rq *cfs_rq, 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; - last_enqueued_diff = ue.enqueued; + /* Get utilization at dequeue */ + dequeued = task_util(p); /* * Reset EWMA on utilization increases, the moving average is used only * to smooth utilization decreases. */ - ue.enqueued = task_util(p); - if (sched_feat(UTIL_EST_FASTUP)) { - if (ue.ewma < ue.enqueued) { - ue.ewma = ue.enqueued; - goto done; - } + 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. */ - last_ewma_diff = ue.enqueued - ue.ewma; - last_enqueued_diff -= ue.enqueued; - if (within_margin(last_ewma_diff, UTIL_EST_MARGIN)) { - if (!within_margin(last_enqueued_diff, UTIL_EST_MARGIN)) - goto done; - - return; - } + last_ewma_diff = ewma - dequeued; + if (last_ewma_diff < UTIL_EST_MARGIN) + goto done; /* - * To avoid overestimation of actual task utilization, skip updates if - * we cannot grant there is idle time in this CPU. + * To avoid underestimate of task utilization, skip updates of EWMA if + * we cannot grant that thread got all CPU time it wanted. */ - if (task_util(p) > capacity_orig_of(cpu_of(rq_of(cfs_rq)))) - return; + 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; + ewma <<= UTIL_EST_WEIGHT_SHIFT; + ewma -= last_ewma_diff; + ewma >>= UTIL_EST_WEIGHT_SHIFT; done: - ue.enqueued |= UTIL_AVG_UNCHANGED; - WRITE_ONCE(p->se.avg.util_est, ue); + ewma |= UTIL_AVG_UNCHANGED; + WRITE_ONCE(p->se.avg.util_est, ewma); trace_sched_util_est_se_tp(&p->se); } +static inline unsigned long get_actual_cpu_capacity(int cpu) +{ + unsigned long capacity = arch_scale_cpu_capacity(cpu); + + capacity -= max(hw_load_avg(cpu_rq(cpu)), cpufreq_get_pressure(cpu)); + + return capacity; +} + static inline int util_fits_cpu(unsigned long util, unsigned long uclamp_min, unsigned long uclamp_max, int cpu) { - unsigned long capacity_orig, capacity_orig_thermal; unsigned long capacity = capacity_of(cpu); + unsigned long capacity_orig; bool fits, uclamp_max_fits; /* @@ -4458,17 +4944,17 @@ static inline int util_fits_cpu(unsigned long util, return fits; /* - * We must use capacity_orig_of() for comparing against uclamp_min and + * 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 capacity_orig_of(little_cpu), it + * 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 thermal pressure since it has a direct impact + * 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 @@ -4476,17 +4962,8 @@ static inline int util_fits_cpu(unsigned long util, * * 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. - * - * In case of capacity inversion we should honour the inverted capacity - * for both uclamp_min and uclamp_max all the time. */ - capacity_orig = cpu_in_capacity_inversion(cpu); - if (capacity_orig) { - capacity_orig_thermal = capacity_orig; - } else { - capacity_orig = capacity_orig_of(cpu); - capacity_orig_thermal = capacity_orig - arch_scale_thermal_pressure(cpu); - } + capacity_orig = arch_scale_cpu_capacity(cpu); /* * We want to force a task to fit a cpu as implied by uclamp_max. @@ -4503,14 +4980,14 @@ static inline int util_fits_cpu(unsigned long util, * | | | | | | | * | | | | | | | * +---------------------------------------- - * cpu0 cpu1 cpu2 + * 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 + * * 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 @@ -4541,7 +5018,7 @@ static inline int util_fits_cpu(unsigned long util, * | | | | | | | * | | | | | | | (region c, boosted, util < uclamp_min) * +---------------------------------------- - * cpu0 cpu1 cpu2 + * 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 @@ -4561,8 +5038,9 @@ static inline int util_fits_cpu(unsigned long util, * handle the case uclamp_min > uclamp_max. */ uclamp_min = min(uclamp_min, uclamp_max); - if (util < uclamp_min && capacity_orig != SCHED_CAPACITY_SCALE) - fits = fits && (uclamp_min <= capacity_orig_thermal); + if (fits && (util < uclamp_min) && + (uclamp_min > get_actual_cpu_capacity(cpu))) + return -1; return fits; } @@ -4572,20 +5050,28 @@ static inline int task_fits_cpu(struct task_struct *p, int cpu) 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 util_fits_cpu(util, uclamp_min, uclamp_max, cpu); + /* + * 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 update_misfit_status(struct task_struct *p, struct rq *rq) { + int cpu = cpu_of(rq); + if (!sched_asym_cpucap_active()) return; - if (!p || p->nr_cpus_allowed == 1) { - rq->misfit_task_load = 0; - return; - } + /* + * 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)) { - if (task_fits_cpu(p, cpu_of(rq))) { rq->misfit_task_load = 0; return; } @@ -4597,195 +5083,199 @@ static inline void update_misfit_status(struct task_struct *p, struct rq *rq) rq->misfit_task_load = max_t(unsigned long, task_h_load(p), 1); } -#else /* CONFIG_SMP */ - -static inline bool cfs_rq_is_decayed(struct cfs_rq *cfs_rq) +void __setparam_fair(struct task_struct *p, const struct sched_attr *attr) { - return true; -} - -#define UPDATE_TG 0x0 -#define SKIP_AGE_LOAD 0x0 -#define DO_ATTACH 0x0 -#define DO_DETACH 0x0 - -static inline void update_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se, int not_used1) -{ - cfs_rq_util_change(cfs_rq, 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) {} -static inline void -detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) {} - -static inline int newidle_balance(struct rq *rq, struct rq_flags *rf) -{ - return 0; -} - -static inline void -util_est_enqueue(struct cfs_rq *cfs_rq, struct task_struct *p) {} - -static inline void -util_est_dequeue(struct cfs_rq *cfs_rq, struct task_struct *p) {} - -static inline void -util_est_update(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) {} - -#endif /* CONFIG_SMP */ - -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; - - if (d < 0) - d = -d; + 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); - - /* sleeps up to a single latency don't count. */ - if (!initial) { - unsigned long thresh; + if (sched_feat(PLACE_LAG) && cfs_rq->nr_queued && se->vlag) { + struct sched_entity *curr = cfs_rq->curr; + unsigned long load; - if (se_is_idle(se)) - thresh = sysctl_sched_min_granularity; - else - 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); + + lag *= load + scale_load_down(se->load.weight); + if (WARN_ON_ONCE(!load)) + load = 1; + lag = div_s64(lag, load); + } + + se->vruntime = vruntime - lag; - vruntime -= thresh; + if (se->rel_deadline) { + se->deadline += se->vruntime; + se->rel_deadline = 0; + return; } - /* ensure we never gain time by being placed backwards. */ - se->vruntime = max_vruntime(se->vruntime, vruntime); + /* + * 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); -static inline bool cfs_bandwidth_used(void); - -/* - * 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. * - For group_entity, update its runnable_weight to reflect the new - * h_nr_running of its group cfs_rq. + * h_nr_runnable of its group cfs_rq. * - For group_entity, update its weight to reflect the new share of * its group cfs_rq * - Add its new weight to cfs_rq->load.weight */ 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); + + /* + * 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_fair(cfs_rq, se, flags); - check_spread(cfs_rq, se); if (!curr) __enqueue_entity(cfs_rq, se); se->on_rq = 1; - if (cfs_rq->nr_running == 1) { + if (cfs_rq->nr_queued == 1) { check_enqueue_throttle(cfs_rq); - if (!throttled_hierarchy(cfs_rq)) - list_add_leaf_cfs_rq(cfs_rq); - } -} - -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 } } @@ -4800,49 +5290,98 @@ 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; - if (cfs_rq->skip == se) - __clear_buddies_skip(se); + for_each_sched_entity(se) { + struct cfs_rq *cfs_rq = cfs_rq_of(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) { + 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; /* - * Update run-time statistics of the 'current'. - */ - update_curr(cfs_rq); - - /* * When dequeuing a sched_entity, we must: * - Update loads to have both entity and cfs_rq synced with now. * - For group_entity, update its runnable_weight to reflect the new - * h_nr_running of its group cfs_rq. + * h_nr_runnable of its group cfs_rq. * - Subtract its previous weight from cfs_rq->load.weight. * - For group entity, update its weight to reflect the new share * of its group cfs_rq. @@ -4852,78 +5391,39 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) update_stats_dequeue_fair(cfs_rq, se, flags); - clear_buddies(cfs_rq, se); + update_entity_lag(cfs_rq, se); + if (sched_feat(PLACE_REL_DEADLINE) && !sleep) { + se->deadline -= se->vruntime; + 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); - if (cfs_rq->nr_running == 0) + if (cfs_rq->nr_queued == 0) { update_idle_cfs_rq_clock_pelt(cfs_rq); -} - -/* - * 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; +#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 @@ -4941,14 +5441,17 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *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() && @@ -4964,8 +5467,7 @@ set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) 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: @@ -4975,51 +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 && 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; - } - - if (cfs_rq->next && wakeup_preempt_entity(cfs_rq->next, left) < 1) { - /* - * Someone really wants this to run. If it's not unfair, run it. - */ - se = cfs_rq->next; - } else if (cfs_rq->last && wakeup_preempt_entity(cfs_rq->last, left) < 1) { + se = pick_eevdf(cfs_rq); + if (se->sched_delayed) { + dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); /* - * Prefer last buddy, try to return the CPU to a preempted task. + * Must not reference @se again, see __block_task(). */ - se = cfs_rq->last; + return NULL; } - return se; } @@ -5037,8 +5506,6 @@ 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_fair(cfs_rq, prev); /* Put 'current' back into the tree. */ @@ -5046,6 +5513,7 @@ static void put_prev_entity(struct cfs_rq *cfs_rq, struct sched_entity *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; } @@ -5069,19 +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); } @@ -5108,7 +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; @@ -5116,16 +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) { @@ -5235,59 +5685,253 @@ 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) +{ + WARN_ON_ONCE(p->se.on_rq); + list_del_init(&p->throttle_node); + + /* task blocked after throttled */ + if (flags & DEQUEUE_SLEEP) { + p->throttled = false; + return; + } + + /* + * 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 *src_cfs_rq, *dest_cfs_rq; + struct cfs_rq *cfs_rq = cfs_rq_of(&p->se); - src_cfs_rq = tg->cfs_rq[src_cpu]; - dest_cfs_rq = tg->cfs_rq[dest_cpu]; + /* @p should have gone through dequeue_throttled_task() first */ + WARN_ON_ONCE(!list_empty(&p->throttle_node)); - return throttled_hierarchy(src_cfs_rq) || - throttled_hierarchy(dest_cfs_rq); + /* + * 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; - cfs_rq->throttle_count--; - if (!cfs_rq->throttle_count) { + 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; + } - /* 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); + if (cfs_rq->throttled_clock_self) { + u64 delta = rq_clock(rq) - cfs_rq->throttled_clock_self; + + cfs_rq->throttled_clock_self = 0; + + 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_pelt = rq_clock_pelt(rq); + 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; } - cfs_rq->throttle_count++; + WARN_ON_ONCE(cfs_rq->throttled_clock_self); + WARN_ON_ONCE(!list_empty(&cfs_rq->throttled_limbo_list)); return 0; } @@ -5295,8 +5939,7 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); - struct sched_entity *se; - long task_delta, idle_task_delta, dequeue = 1; + int dequeue = 1; raw_spin_lock(&cfs_b->lock); /* This will start the period timer if necessary */ @@ -5319,62 +5962,17 @@ static bool throttle_cfs_rq(struct cfs_rq *cfs_rq) if (!dequeue) return false; /* Throttle no longer required. */ - se = cfs_rq->tg->se[cpu_of(rq_of(cfs_rq))]; - /* 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; - idle_task_delta = cfs_rq->idle_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) - goto done; - - dequeue_entity(qcfs_rq, se, DEQUEUE_SLEEP); - - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; - - qcfs_rq->h_nr_running -= task_delta; - qcfs_rq->idle_h_nr_running -= idle_task_delta; - - if (qcfs_rq->load.weight) { - /* Avoid re-evaluating load for this entity: */ - se = parent_entity(se); - break; - } - } - - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); - /* throttled entity or throttle-on-deactivate */ - if (!se->on_rq) - goto done; - - update_load_avg(qcfs_rq, se, 0); - se_update_runnable(se); - - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; - - qcfs_rq->h_nr_running -= task_delta; - qcfs_rq->idle_h_nr_running -= idle_task_delta; - } - - /* At this point se is NULL and we are at root level*/ - sub_nr_running(rq, task_delta); - -done: /* * Note: distribution will already see us throttled via the * throttled-list. rq->lock protects completion. */ cfs_rq->throttled = 1; - cfs_rq->throttled_clock = rq_clock(rq); + WARN_ON_ONCE(cfs_rq->throttled_clock); return true; } @@ -5382,17 +5980,29 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) { struct rq *rq = rq_of(cfs_rq); struct cfs_bandwidth *cfs_b = tg_cfs_bandwidth(cfs_rq->tg); - struct sched_entity *se; - long task_delta, idle_task_delta; + 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); @@ -5410,74 +6020,115 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) if (list_add_leaf_cfs_rq(cfs_rq_of(se))) break; } - goto unthrottle_throttle; } - task_delta = cfs_rq->h_nr_running; - idle_task_delta = cfs_rq->idle_h_nr_running; - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); + assert_list_leaf_cfs_rq(rq); - if (se->on_rq) - break; - enqueue_entity(qcfs_rq, se, ENQUEUE_WAKEUP); + /* 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_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; +static void __cfsb_csd_unthrottle(void *arg) +{ + struct cfs_rq *cursor, *tmp; + struct rq *rq = arg; + struct rq_flags rf; - qcfs_rq->h_nr_running += task_delta; - qcfs_rq->idle_h_nr_running += idle_task_delta; + rq_lock(rq, &rf); - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(qcfs_rq)) - goto unthrottle_throttle; - } + /* + * 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); - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); + /* + * 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(); - update_load_avg(qcfs_rq, se, UPDATE_TG); - se_update_runnable(se); + list_for_each_entry_safe(cursor, tmp, &rq->cfsb_csd_list, + throttled_csd_list) { + list_del_init(&cursor->throttled_csd_list); - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_task_delta = cfs_rq->h_nr_running; + if (cfs_rq_throttled(cursor)) + unthrottle_cfs_rq(cursor); + } - qcfs_rq->h_nr_running += task_delta; - qcfs_rq->idle_h_nr_running += idle_task_delta; + rcu_read_unlock(); + + rq_clock_stop_loop_update(rq); + rq_unlock(rq, &rf); +} + +static inline void __unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) +{ + struct rq *rq = rq_of(cfs_rq); + bool first; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(qcfs_rq)) - goto unthrottle_throttle; + if (rq == this_rq()) { + unthrottle_cfs_rq(cfs_rq); + return; } - /* At this point se is NULL and we are at root level*/ - add_nr_running(rq, task_delta); + /* Already enqueued */ + if (WARN_ON_ONCE(!list_empty(&cfs_rq->throttled_csd_list))) + return; -unthrottle_throttle: - assert_list_leaf_cfs_rq(rq); + 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); +} - /* Determine whether we need to wake up potentially idle CPU: */ - if (rq->curr == rq->idle && rq->cfs.nr_running) - resched_curr(rq); +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 void distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) +static bool distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) { - struct cfs_rq *cfs_rq; + 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_irqsave(rq, &rf); if (!cfs_rq_throttled(cfs_rq)) goto next; - /* By the above check, this should never be true */ - SCHED_WARN_ON(cfs_rq->runtime_remaining > 0); + /* 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; @@ -5490,16 +6141,44 @@ static void distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) cfs_rq->runtime_remaining += runtime; /* 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_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 throttled; } /* @@ -5544,10 +6223,8 @@ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun, u while (throttled && cfs_b->runtime > 0) { raw_spin_unlock_irqrestore(&cfs_b->lock, flags); /* we can't nest cfs_b->lock while distributing bandwidth */ - distribute_cfs_runtime(cfs_b); + throttled = distribute_cfs_runtime(cfs_b); raw_spin_lock_irqsave(&cfs_b->lock, flags); - - throttled = !list_empty(&cfs_b->throttled_cfs_rq); } /* @@ -5642,7 +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); @@ -5716,6 +6393,16 @@ static void sync_throttle(struct task_group *tg, int cpu) cfs_rq->throttle_count = pcfs_rq->throttle_count; 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() */ @@ -5747,8 +6434,6 @@ static enum hrtimer_restart sched_cfs_slack_timer(struct hrtimer *timer) return HRTIMER_NORESTART; } -extern const u64 max_cfs_quota_period; - static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) { struct cfs_bandwidth *cfs_b = @@ -5775,7 +6460,7 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) * to fail. */ new = old * 2; - if (new < max_cfs_quota_period) { + 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; @@ -5804,19 +6489,24 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) 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; + 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; } @@ -5824,6 +6514,8 @@ 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) @@ -5840,12 +6532,36 @@ void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) 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); + } } /* @@ -5881,6 +6597,17 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_rq_held(rq); + // Do not unthrottle for an active CPU + if (cpumask_test_cpu(cpu_of(rq), cpu_active_mask)) + return; + + /* + * The rq clock has already been updated in the + * set_rq_offline(), so we should skip updating + * 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)]; @@ -5889,54 +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(); -} -#else /* CONFIG_CFS_BANDWIDTH */ + rq_clock_stop_loop_update(rq); +} -static inline bool cfs_bandwidth_used(void) +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; } +#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) +{ + 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 @@ -5947,8 +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: @@ -5958,17 +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 (task_current(rq, p)) + if (task_current_donor(rq, p)) resched_curr(rq); return; } @@ -5983,15 +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_fair(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) { @@ -6000,52 +6781,92 @@ hrtick_start_fair(struct rq *rq, struct task_struct *p) static inline void hrtick_update(struct rq *rq) { } -#endif +#endif /* !CONFIG_SCHED_HRTICK */ -#ifdef CONFIG_SMP static inline bool cpu_overutilized(int cpu) { - unsigned long rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN); - unsigned long rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX); + 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); - trace_sched_overutilized_tp(rq->rd, 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); } -#else -static inline void update_overutilized_status(struct rq *rq) { } -#endif /* Runqueue only has SCHED_IDLE tasks enqueued */ static int sched_idle_rq(struct rq *rq) { - return unlikely(rq->nr_running == rq->cfs.idle_h_nr_running && + return unlikely(rq->nr_running == rq->cfs.h_nr_idle && rq->nr_running); } -/* - * Returns true if cfs_rq only has SCHED_IDLE entities enqueued. Note the use - * of idle_nr_running, which does not consider idle descendants of normal - * entities. - */ -static bool sched_idle_cfs_rq(struct cfs_rq *cfs_rq) +static int sched_idle_cpu(int cpu) { - return cfs_rq->nr_running && - cfs_rq->nr_running == cfs_rq->idle_nr_running; + return sched_idle_rq(cpu_rq(cpu)); } -#ifdef CONFIG_SMP -static int sched_idle_cpu(int cpu) +static void +requeue_delayed_entity(struct sched_entity *se) { - return sched_idle_rq(cpu_rq(cpu)); + 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); } -#endif /* * The enqueue_task method is called before nr_running is @@ -6057,8 +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 idle_h_nr_running = task_has_idle_policy(p); + int h_nr_idle = task_has_idle_policy(p); + int h_nr_runnable = 1; int task_new = !(flags & ENQUEUE_WAKEUP); + int rq_h_nr_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 @@ -6066,7 +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 @@ -6076,21 +6909,35 @@ 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); + + /* + * 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 (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_running++; - cfs_rq->idle_h_nr_running += idle_h_nr_running; + cfs_rq->h_nr_runnable += h_nr_runnable; + cfs_rq->h_nr_queued++; + cfs_rq->h_nr_idle += h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; - - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto enqueue_throttle; + h_nr_idle = 1; flags = ENQUEUE_WAKEUP; } @@ -6102,17 +6949,22 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) se_update_runnable(se); update_cfs_group(se); - cfs_rq->h_nr_running++; - cfs_rq->idle_h_nr_running += idle_h_nr_running; + se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); + slice = cfs_rq_min_slice(cfs_rq); - if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; + cfs_rq->h_nr_runnable += h_nr_runnable; + cfs_rq->h_nr_queued++; + cfs_rq->h_nr_idle += h_nr_idle; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto enqueue_throttle; + 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); @@ -6131,58 +6983,80 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) * and the following generally works well enough in practice. */ if (!task_new) - update_overutilized_status(rq); + check_update_overutilized_status(rq); -enqueue_throttle: 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) { - struct cfs_rq *cfs_rq; - struct sched_entity *se = &p->se; - int task_sleep = flags & DEQUEUE_SLEEP; - int idle_h_nr_running = task_has_idle_policy(p); 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; + u64 slice = 0; - util_est_dequeue(&rq->cfs, p); + 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); - cfs_rq->h_nr_running--; - cfs_rq->idle_h_nr_running -= idle_h_nr_running; + 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_runnable -= h_nr_runnable; + cfs_rq->h_nr_queued -= h_nr_queued; + cfs_rq->h_nr_idle -= h_nr_idle; if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; + h_nr_idle = h_nr_queued; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto dequeue_throttle; + 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) { @@ -6192,35 +7066,82 @@ static void dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) se_update_runnable(se); update_cfs_group(se); - cfs_rq->h_nr_running--; - cfs_rq->idle_h_nr_running -= idle_h_nr_running; + se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); + slice = cfs_rq_min_slice(cfs_rq); - if (cfs_rq_is_idle(cfs_rq)) - idle_h_nr_running = 1; + cfs_rq->h_nr_runnable -= h_nr_runnable; + cfs_rq->h_nr_queued -= h_nr_queued; + cfs_rq->h_nr_idle -= h_nr_idle; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto dequeue_throttle; + 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); } - /* At this point se is NULL and we are at root level*/ - sub_nr_running(rq, 1); + sub_nr_running(rq, h_nr_queued); /* balance early to pull high priority tasks */ if (unlikely(!was_sched_idle && sched_idle_rq(rq))) rq->next_balance = jiffies; -dequeue_throttle: - util_est_update(&rq->cfs, p, task_sleep); + 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; +} + +/* + * 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 bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) +{ + if (task_is_throttled(p)) { + dequeue_throttled_task(p, flags); + return true; + } + + if (!p->se.sched_delayed) + util_est_dequeue(&rq->cfs, p); + + util_est_update(&rq->cfs, p, flags & DEQUEUE_SLEEP); + if (dequeue_entities(rq, &p->se, flags) < 0) + return false; + + /* + * Must not reference @p after dequeue_entities(DEQUEUE_DELAYED). + */ + hrtick_update(rq); + return true; } -#ifdef CONFIG_SMP +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: load_balance, load_balance_newidle. */ +/* 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 @@ -6376,8 +7297,12 @@ 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; @@ -6440,7 +7365,7 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, target = wake_affine_weight(sd, p, this_cpu, prev_cpu, sync); schedstat_inc(p->stats.nr_wakeups_affine_attempts); - if (target == nr_cpumask_bits) + if (target != this_cpu) return prev_cpu; schedstat_inc(sd->ttwu_move_affine); @@ -6449,13 +7374,13 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, } static struct sched_group * -find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu); +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; @@ -6511,7 +7436,7 @@ 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; @@ -6536,13 +7461,13 @@ static inline int find_idlest_cpu(struct sched_domain *sd, struct task_struct *p continue; } - group = find_idlest_group(sd, p, cpu); + 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; @@ -6640,7 +7565,7 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu if (!available_idle_cpu(cpu)) { idle = false; if (*idle_cpu == -1) { - if (sched_idle_cpu(cpu) && cpumask_test_cpu(cpu, p->cpus_ptr)) { + if (sched_idle_cpu(cpu) && cpumask_test_cpu(cpu, cpus)) { *idle_cpu = cpu; break; } @@ -6648,7 +7573,7 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu } break; } - if (*idle_cpu == -1 && cpumask_test_cpu(cpu, p->cpus_ptr)) + if (*idle_cpu == -1 && cpumask_test_cpu(cpu, cpus)) *idle_cpu = cpu; } @@ -6662,13 +7587,19 @@ static int select_idle_core(struct task_struct *p, int core, struct cpumask *cpu /* * Scan the local SMT mask for idle CPUs. */ -static int select_idle_smt(struct task_struct *p, int target) +static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) { int cpu; 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) || sched_idle_cpu(cpu)) return cpu; } @@ -6676,7 +7607,7 @@ static int select_idle_smt(struct task_struct *p, int target) return -1; } -#else /* CONFIG_SCHED_SMT */ +#else /* !CONFIG_SCHED_SMT: */ static inline void set_idle_cores(int cpu, int val) { @@ -6692,12 +7623,12 @@ static inline int select_idle_core(struct task_struct *p, int core, struct cpuma return __select_idle_cpu(core, p); } -static inline int select_idle_smt(struct task_struct *p, int target) +static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int target) { return -1; } -#endif /* CONFIG_SCHED_SMT */ +#endif /* !CONFIG_SCHED_SMT */ /* * Scan the LLC domain for idle CPUs; this is dynamically regulated by @@ -6709,45 +7640,9 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool 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; - struct rq *this_rq = this_rq(); - int this = smp_processor_id(); - struct sched_domain *this_sd = NULL; - u64 time = 0; cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); - if (sched_feat(SIS_PROP) && !has_idle_core) { - u64 avg_cost, avg_idle, span_avg; - unsigned long now = jiffies; - - this_sd = rcu_dereference(*this_cpu_ptr(&sd_llc)); - if (!this_sd) - return -1; - - /* - * If we're busy, the assumption that the last idle period - * predicts the future is flawed; age away the remaining - * predicted idle time. - */ - if (unlikely(this_rq->wake_stamp < now)) { - while (this_rq->wake_stamp < now && this_rq->wake_avg_idle) { - this_rq->wake_stamp++; - this_rq->wake_avg_idle >>= 1; - } - } - - avg_idle = this_rq->wake_avg_idle; - avg_cost = this_sd->avg_scan_cost + 1; - - span_avg = sd->span_weight * avg_idle; - if (span_avg > 4*avg_cost) - nr = div_u64(span_avg, avg_cost); - else - nr = 4; - - time = cpu_clock(this); - } - if (sched_feat(SIS_UTIL)) { sd_share = rcu_dereference(per_cpu(sd_llc_shared, target)); if (sd_share) { @@ -6759,6 +7654,30 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool } } + 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)); + } + } + for_each_cpu_wrap(cpu, cpus, target + 1) { if (has_idle_core) { i = select_idle_core(p, cpu, cpus, &idle_cpu); @@ -6766,7 +7685,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool return i; } else { - if (!--nr) + if (--nr <= 0) return -1; idle_cpu = __select_idle_cpu(cpu, p); if ((unsigned int)idle_cpu < nr_cpumask_bits) @@ -6777,18 +7696,6 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, bool if (has_idle_core) set_idle_cores(target, false); - if (sched_feat(SIS_PROP) && this_sd && !has_idle_core) { - time = cpu_clock(this) - time; - - /* - * Account for the scan cost of wakeups against the average - * idle time. - */ - this_rq->wake_avg_idle -= min(this_rq->wake_avg_idle, time); - - update_avg(&this_sd->avg_scan_cost, time); - } - return idle_cpu; } @@ -6801,6 +7708,7 @@ 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; @@ -6816,12 +7724,28 @@ select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target) if (!available_idle_cpu(cpu) && !sched_idle_cpu(cpu)) continue; - if (util_fits_cpu(task_util, util_min, util_max, cpu)) + + 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); - if (cpu_cap > best_cap) { + /* + * 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; } } @@ -6834,7 +7758,11 @@ static inline bool asym_fits_cpu(unsigned long util, int cpu) { if (sched_asym_cpucap_active()) - return util_fits_cpu(util, util_min, util_max, cpu); + /* + * 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; } @@ -6847,11 +7775,11 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) bool has_idle_core = false; struct sched_domain *sd; unsigned long task_util, util_min, util_max; - int i, recent_used_cpu; + 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. + * that the task fits with CPU's capacity. */ if (sched_asym_cpucap_active()) { sync_entity_load_avg(&p->se); @@ -6874,8 +7802,14 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) */ 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)) - return 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 @@ -6900,9 +7834,15 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) recent_used_cpu != target && cpus_share_cache(recent_used_cpu, target) && (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) && - cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr) && + cpumask_test_cpu(recent_used_cpu, p->cpus_ptr) && asym_fits_cpu(task_util, util_min, util_max, recent_used_cpu)) { - return 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; } /* @@ -6933,7 +7873,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) has_idle_core = test_idle_cores(target); if (!has_idle_core && cpus_share_cache(prev, target)) { - i = select_idle_smt(p, prev); + i = select_idle_smt(p, sd, prev); if ((unsigned int)i < nr_cpumask_bits) return i; } @@ -6943,17 +7883,72 @@ static int select_idle_sibling(struct task_struct *p, int prev, int 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; } -/* - * Predicts what cpu_util(@cpu) would return if @p was removed from @cpu - * (@dst_cpu = -1) or migrated to @dst_cpu. - */ -static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_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 = &cpu_rq(cpu)->cfs; unsigned long util = READ_ONCE(cfs_rq->avg.util_avg); + unsigned long runnable; + + if (boost) { + runnable = READ_ONCE(cfs_rq->avg.runnable_avg); + util = max(util, runnable); + } /* * If @dst_cpu is -1 or @p migrates from @cpu to @dst_cpu remove its @@ -6961,24 +7956,24 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) * contribution. In all the other cases @cpu is not impacted by the * migration so its util_avg is already correct. */ - if (task_cpu(p) == cpu && dst_cpu != cpu) + if (p && task_cpu(p) == cpu && dst_cpu != cpu) lsub_positive(&util, task_util(p)); - else if (task_cpu(p) != cpu && dst_cpu == cpu) + else if (p && task_cpu(p) != cpu && dst_cpu == cpu) util += task_util(p); if (sched_feat(UTIL_EST)) { unsigned long util_est; - util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued); + 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.enqueued. + * 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.enqueued. + * 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 @@ -6999,13 +7994,23 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) */ if (dst_cpu == cpu) util_est += _task_util_est(p); - else if (unlikely(task_on_rq_queued(p) || current == 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, capacity_orig_of(cpu)); + 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); } /* @@ -7025,9 +8030,108 @@ static unsigned long cpu_util_without(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 cpu_util_cfs(cpu); + p = NULL; - return cpu_util_next(cpu, p, -1); + return cpu_util(cpu, p, -1, 0); +} + +/* + * 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); + + /* + * 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(). + */ + irq = cpu_util_irq(rq); + if (unlikely(irq >= scale)) { + if (min) + *min = scale; + if (max) + *max = scale; + return scale; + } + + if (min) { + /* + * 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. + */ + *min = max(irq + cpu_bw_dl(rq), uclamp_rq_get(rq, UCLAMP_MIN)); + + /* + * 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); + } + + /* + * 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. + */ + 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); } /* @@ -7074,7 +8178,7 @@ static inline void eenv_task_busy_time(struct energy_env *eenv, * cpu_capacity. * * The contribution of the task @p for which we want to estimate the - * energy cost is removed (by cpu_util_next()) and must be calculated + * 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 @@ -7095,9 +8199,9 @@ static inline void eenv_pd_busy_time(struct energy_env *eenv, int cpu; for_each_cpu(cpu, pd_cpus) { - unsigned long util = cpu_util_next(cpu, p, -1); + unsigned long util = cpu_util(cpu, p, -1, 0); - busy_time += effective_cpu_util(cpu, util, ENERGY_UTIL, NULL); + busy_time += effective_cpu_util(cpu, util, NULL, NULL); } eenv->pd_busy_time = min(eenv->pd_cap, busy_time); @@ -7119,18 +8223,34 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, for_each_cpu(cpu, pd_cpus) { struct task_struct *tsk = (cpu == dst_cpu) ? p : NULL; - unsigned long util = cpu_util_next(cpu, p, dst_cpu); - unsigned long cpu_util; + unsigned long util = cpu_util(cpu, p, dst_cpu, 1); + unsigned long eff_util, min, max; /* * 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 - * FREQUENCY_UTIL's utilization can be max OPP. + * NOTE: in case RT tasks are running, by default the min + * utilization can be max OPP. */ - cpu_util = effective_cpu_util(cpu, util, FREQUENCY_UTIL, tsk); - max_util = max(max_util, cpu_util); + eff_util = effective_cpu_util(cpu, util, &min, &max); + + /* 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(max_util, eenv->cpu_cap); @@ -7147,11 +8267,16 @@ compute_energy(struct energy_env *eenv, struct perf_domain *pd, { 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; if (dst_cpu >= 0) busy_time = min(eenv->pd_cap, busy_time + eenv->task_busy_time); - return em_cpu_energy(pd->em_pd, max_util, busy_time, eenv->cpu_cap); + 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; } /* @@ -7186,7 +8311,7 @@ compute_energy(struct energy_env *eenv, 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 @@ -7201,13 +8326,16 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) 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)) + if (!pd) goto unlock; /* @@ -7223,37 +8351,36 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) target = prev_cpu; sync_entity_load_avg(&p->se); - if (!uclamp_task_util(p, p_util_min, p_util_max)) + 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 util_min = p_util_min, util_max = p_util_max; - unsigned long cpu_cap, cpu_thermal_cap, util; - unsigned long cur_delta, max_spare_cap = 0; + 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 prev_spare_cap = 0; + unsigned long cur_delta, base_energy; int max_spare_cap_cpu = -1; - unsigned long base_energy; + int fits, max_fits = -1; cpumask_and(cpus, perf_domain_span(pd), cpu_online_mask); if (cpumask_empty(cpus)) continue; - /* Account thermal pressure for the energy estimation */ + /* Account external pressure for the energy estimation */ cpu = cpumask_first(cpus); - cpu_thermal_cap = arch_scale_cpu_capacity(cpu); - cpu_thermal_cap -= arch_scale_thermal_pressure(cpu); + cpu_actual_cap = get_actual_cpu_capacity(cpu); - eenv.cpu_cap = cpu_thermal_cap; + 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_thermal_cap; + eenv.pd_cap += cpu_actual_cap; if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) continue; @@ -7261,7 +8388,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) if (!cpumask_test_cpu(cpu, p->cpus_ptr)) continue; - util = cpu_util_next(cpu, p, cpu); + util = cpu_util(cpu, p, cpu, 0); cpu_cap = capacity_of(cpu); /* @@ -7274,7 +8401,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) if (uclamp_is_used() && !uclamp_rq_is_idle(rq)) { /* * Open code uclamp_rq_util_with() except for - * the clamp() part. Ie: apply max aggregation + * 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}. @@ -7285,7 +8412,9 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) util_min = max(rq_util_min, p_util_min); util_max = max(rq_util_max, p_util_max); } - if (!util_fits_cpu(util, util_min, util_max, cpu)) + + fits = util_fits_cpu(util, util_min, util_max, cpu); + if (!fits) continue; lsub_positive(&cpu_cap, util); @@ -7293,7 +8422,9 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) if (cpu == prev_cpu) { /* Always use prev_cpu as a candidate. */ prev_spare_cap = cpu_cap; - } else if (cpu_cap > max_spare_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 @@ -7301,10 +8432,11 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) */ max_spare_cap = cpu_cap; max_spare_cap_cpu = cpu; + max_fits = fits; } } - if (max_spare_cap_cpu < 0 && prev_spare_cap == 0) + if (max_spare_cap_cpu < 0 && prev_spare_cap < 0) continue; eenv_pd_busy_time(&eenv, cpus, p); @@ -7312,33 +8444,57 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) base_energy = compute_energy(&eenv, pd, cpus, p, -1); /* Evaluate the energy impact of using prev_cpu. */ - if (prev_spare_cap > 0) { + 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; - if (cur_delta < best_delta) { - best_delta = cur_delta; - best_energy_cpu = max_spare_cap_cpu; - } + + /* + * 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; } } rcu_read_unlock(); - if (best_delta < prev_delta) + 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 target; @@ -7377,7 +8533,11 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) if (wake_flags & WF_TTWU) { record_wakee(p); - if (sched_energy_enabled()) { + 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; @@ -7415,7 +8575,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) if (unlikely(sd)) { /* Slow path */ - new_cpu = find_idlest_cpu(sd, p, cpu, prev_cpu, sd_flag); + 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); @@ -7434,18 +8594,6 @@ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu) { struct sched_entity *se = &p->se; - /* - * 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 (READ_ONCE(p->__state) == TASK_WAKING) { - struct cfs_rq *cfs_rq = cfs_rq_of(se); - - se->vruntime -= u64_u32_load(cfs_rq->min_vruntime); - } - if (!task_on_rq_migrating(p)) { remove_entity_load_avg(se); @@ -7465,91 +8613,62 @@ static void migrate_task_rq_fair(struct task_struct *p, int new_cpu) /* Tell new CPU we are migrated */ se->avg.last_update_time = 0; - /* We have migrated, no longer consider this task hot */ - se->exec_start = 0; - update_scan_period(p, new_cpu); } static void task_dead_fair(struct task_struct *p) { - remove_entity_load_avg(&p->se); -} - -static int -balance_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) -{ - if (rq->nr_running) - return 1; + struct sched_entity *se = &p->se; - return newidle_balance(rq, rf) != 0; -} -#endif /* CONFIG_SMP */ + if (se->sched_delayed) { + struct rq_flags rf; + struct rq *rq; -static unsigned long wakeup_gran(struct sched_entity *se) -{ - unsigned long gran = sysctl_sched_wakeup_granularity; + 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); + } - /* - * 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); + 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) { - for_each_sched_entity(se) { - if (SCHED_WARN_ON(!se->on_rq)) - return; - if (se_is_idle(se)) - 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) { 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; @@ -7557,22 +8676,80 @@ static void set_next_buddy(struct sched_entity *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)) @@ -7580,18 +8757,13 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ /* * This is possible from callers such as attach_tasks(), in which we - * unconditionally check_preempt_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. @@ -7602,19 +8774,10 @@ 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; - /* Idle tasks are by definition preempted by non-idle tasks. */ - if (unlikely(task_has_idle_policy(curr)) && - likely(!task_has_idle_policy(p))) - goto preempt; - - /* - * Batch and idle tasks do not preempt non-idle tasks (their preemption - * is driven by the tick): - */ - if (unlikely(p->policy != SCHED_NORMAL) || !sched_feat(WAKEUP_PREEMPTION)) + if (!sched_feat(WAKEUP_PREEMPTION)) return; find_matching_se(&se, &pse); @@ -7624,146 +8787,169 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_ pse_is_idle = se_is_idle(pse); /* - * Preempt an idle group in favor of a non-idle group (and don't preempt + * 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) + 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; - update_curr(cfs_rq_of(se)); - if (wakeup_preempt_entity(se, pse) == 1) { + /* + * BATCH and IDLE tasks do not preempt others. + */ + if (unlikely(!normal_policy(p->policy))) + return; + + 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); } -#ifdef CONFIG_SMP -static struct task_struct *pick_task_fair(struct rq *rq) +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_running) + if (!cfs_rq->nr_queued) return NULL; - do { - struct sched_entity *curr = cfs_rq->curr; + throttled = false; - /* When we pick for a remote RQ, we'll not have done put_prev_entity() */ - if (curr) { - if (curr->on_rq) - update_curr(cfs_rq); - else - curr = NULL; + do { + /* Might not have done put_prev_entity() */ + if (cfs_rq->curr && cfs_rq->curr->on_rq) + update_curr(cfs_rq); - if (unlikely(check_cfs_rq_runtime(cfs_rq))) - goto again; - } + throttled |= check_cfs_rq_runtime(cfs_rq); - se = pick_next_entity(cfs_rq, curr); + se = pick_next_entity(rq, cfs_rq); + if (!se) + goto again; cfs_rq = group_cfs_rq(se); } while (cfs_rq); - return task_of(se); + p = task_of(se); + if (unlikely(throttled)) + task_throttle_setup_work(p); + return p; } -#endif + +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 (!sched_fair_runnable(rq)) + p = pick_task_fair(rq, rf); + if (!p) goto idle; + se = &p->se; #ifdef CONFIG_FAIR_GROUP_SCHED - if (!prev || prev->sched_class != &fair_sched_class) + 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; @@ -7781,55 +8967,33 @@ again: put_prev_entity(cfs_rq, pse); set_next_entity(cfs_rq, se); - } - - goto done; -simple: -#endif - if (prev) - put_prev_task(rq, prev); - - do { - se = pick_next_entity(cfs_rq, NULL); - set_next_entity(cfs_rq, se); - cfs_rq = group_cfs_rq(se); - } while (cfs_rq); - p = task_of(se); - -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. - */ - list_move(&p->se.group_node, &rq->cfs_tasks); -#endif - - if (hrtick_enabled_fair(rq)) - hrtick_start_fair(rq, p); + __set_next_task_fair(rq, p, true); + } - update_misfit_status(p, rq); + return p; +simple: +#endif /* CONFIG_FAIR_GROUP_SCHED */ + put_prev_set_next_task(rq, prev, p); return p; idle: - if (!rf) - return NULL; + if (rf) { + new_tasks = sched_balance_newidle(rq, rf); - new_tasks = newidle_balance(rq, rf); - - /* - * Because newidle_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; + /* + * 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; - if (new_tasks > 0) - goto again; + if (new_tasks > 0) + goto again; + } /* * rq is about to be idle, check if we need to update the @@ -7840,15 +9004,25 @@ idle: return NULL; } -static struct task_struct *__pick_next_task_fair(struct rq *rq) +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); +} + +void fair_server_init(struct rq *rq) { - return pick_next_task_fair(rq, NULL, NULL); + struct sched_dl_entity *dl_se = &rq->fair_server; + + init_dl_entity(dl_se); + + 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; @@ -7861,12 +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; @@ -7878,32 +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) { 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); @@ -7911,7 +9092,6 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p) return true; } -#ifdef CONFIG_SMP /************************************************** * Fair scheduling class load-balancing methods. * @@ -7961,7 +9141,7 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p) * 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: @@ -8055,6 +9235,11 @@ enum group_type { */ 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. @@ -8135,8 +9320,7 @@ static int task_hot(struct task_struct *p, struct lb_env *env) * 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) @@ -8159,43 +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) { @@ -8206,16 +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? @@ -8223,24 +9431,44 @@ 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_rq_held(env->src_rq); + if (p->sched_task_hot) + p->sched_task_hot = 0; /* * We do not migrate tasks that are: - * 1) throttled_lb_pair, or - * 2) cannot be migrated to this CPU due to cpus_ptr, or - * 3) running (obviously), or - * 4) are cache-hot on their current CPU. + * 1) delayed dequeued unless we migrate load, or + * 2) 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 pcpu kthreads; they are where they need to be. */ + /* 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_ptr)) { int cpu; @@ -8263,12 +9491,11 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) 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_ptr)) { - 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; @@ -8277,7 +9504,8 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) /* Record that we found at least one task that could run on dst_cpu */ env->flags &= ~LBF_ALL_PINNED; - if (task_on_cpu(env->src_rq, p)) { + 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; } @@ -8292,16 +9520,15 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) if (env->flags & LBF_ACTIVE_LB) return 1; - tsk_cache_hot = migrate_degrades_locality(p, env); - if (tsk_cache_hot == -1) - tsk_cache_hot = task_hot(p, env); + degrades = migrate_degrades_locality(p, env); + if (!degrades) + hot = task_hot(p, env); + else + hot = degrades > 0; - if (tsk_cache_hot <= 0 || - env->sd->nr_balance_failed > env->sd->cache_nice_tries) { - if (tsk_cache_hot == 1) { - schedstat_inc(env->sd->lb_hot_gained[env->idle]); - schedstat_inc(p->stats.nr_forced_migrations); - } + if (!hot || env->sd->nr_balance_failed > env->sd->cache_nice_tries) { + if (hot) + p->sched_task_hot = 1; return 1; } @@ -8316,6 +9543,15 @@ static void detach_task(struct task_struct *p, struct lb_env *env) { lockdep_assert_rq_held(env->src_rq); + if (p->sched_task_hot) { + p->sched_task_hot = 0; + schedstat_inc(env->sd->lb_hot_gained[env->idle]); + schedstat_inc(p->stats.nr_forced_migrations); + } + + WARN_ON(task_current(env->src_rq, p)); + WARN_ON(task_current_donor(env->src_rq, p)); + deactivate_task(env->src_rq, p, DEQUEUE_NOCLOCK); set_task_cpu(p, env->dst_cpu); } @@ -8383,16 +9619,12 @@ 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; env->loop++; - /* - * We've more or less seen every task there is, call it quits - * unless we haven't found any movable task yet. - */ - if (env->loop > env->loop_max && - !(env->flags & LBF_ALL_PINNED)) + /* We've more or less seen every task there is, call it quits */ + if (env->loop > env->loop_max) break; /* take a breather every nr_migrate tasks */ @@ -8437,7 +9669,7 @@ static int detach_tasks(struct lb_env *env) case migrate_util: util = task_util_est(p); - if (util > env->imbalance) + if (shr_bound(util, env->sd->nr_balance_failed) > env->imbalance) goto next; env->imbalance -= util; @@ -8480,6 +9712,9 @@ static int detach_tasks(struct lb_env *env) continue; next: + if (p->sched_task_hot) + schedstat_inc(p->stats.nr_failed_migrations_hot); + list_move(&p->se.group_node, tasks); } @@ -8502,7 +9737,7 @@ static void attach_task(struct rq *rq, struct task_struct *p) WARN_ON_ONCE(task_rq(p) != rq); activate_task(rq, p, ENQUEUE_NOCLOCK); - check_preempt_curr(rq, p, 0); + wakeup_preempt(rq, p, 0); } /* @@ -8556,19 +9791,17 @@ 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; - if (thermal_load_avg(rq)) + if (hw_load_avg(rq)) return true; -#ifdef CONFIG_HAVE_SCHED_AVG_IRQ - if (READ_ONCE(rq->avg_irq.util_avg)) + if (cpu_util_irq(rq)) return true; -#endif return false; } @@ -8583,37 +9816,27 @@ static inline void update_blocked_load_status(struct rq *rq, bool has_blocked) if (!has_blocked) rq->has_blocked_load = 0; } -#else +#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 +#endif /* !CONFIG_NO_HZ_COMMON */ static bool __update_blocked_others(struct rq *rq, bool *done) { - const struct sched_class *curr_class; - u64 now = rq_clock_pelt(rq); - unsigned long thermal_pressure; - bool decayed; + bool updated; /* * update_load_avg() can call cpufreq_update_util(). Make sure that RT, * DL and IRQ signals have been updated before updating CFS. */ - curr_class = rq->curr->sched_class; - - thermal_pressure = arch_scale_thermal_pressure(cpu_of(rq)); - - decayed = update_rt_rq_load_avg(now, rq, curr_class == &rt_sched_class) | - update_dl_rq_load_avg(now, rq, curr_class == &dl_sched_class) | - update_thermal_load_avg(rq_clock_thermal(rq), rq, thermal_pressure) | - update_irq_load_avg(rq, 0); + updated = update_other_load_avgs(rq); if (others_have_blocked(rq)) *done = false; - return decayed; + return updated; } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -8634,7 +9857,7 @@ static bool __update_blocked_fair(struct rq *rq, bool *done) if (update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq)) { update_tg_load_avg(cfs_rq); - if (cfs_rq->nr_running == 0) + if (cfs_rq->nr_queued == 0) update_idle_cfs_rq_clock_pelt(cfs_rq); if (cfs_rq == &rq->cfs) @@ -8707,7 +9930,7 @@ 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 +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static bool __update_blocked_fair(struct rq *rq, bool *done) { struct cfs_rq *cfs_rq = &rq->cfs; @@ -8724,9 +9947,9 @@ static unsigned long task_h_load(struct task_struct *p) { return p->se.avg.load_avg; } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ -static void update_blocked_averages(int cpu) +static void sched_balance_update_blocked_averages(int cpu) { bool decayed = false, done = true; struct rq *rq = cpu_rq(cpu); @@ -8745,24 +9968,25 @@ static void update_blocked_averages(int cpu) rq_unlock_irqrestore(rq, &rf); } -/********** Helpers for find_busiest_group ************************/ +/********** 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 group_capacity; - 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 tasks running in the group */ - unsigned int sum_h_nr_running; /* Nr of CFS 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; - unsigned int group_asym_packing; /* Tasks should be moved to preferred CPU */ - 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; @@ -8770,19 +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_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 */ + 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) @@ -8808,8 +10031,8 @@ static inline void init_sd_lb_stats(struct sd_lb_stats *sds) 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(cpu); unsigned long used, free; unsigned long irq; @@ -8821,12 +10044,9 @@ static unsigned long scale_rt_capacity(int cpu) /* * avg_rt.util_avg and avg_dl.util_avg track binary signals * (running and not running) with weights 0 and 1024 respectively. - * avg_thermal.load_avg tracks thermal pressure and the weighted - * average uses the actual delta max capacity(load). */ - used = READ_ONCE(rq->avg_rt.util_avg); - used += READ_ONCE(rq->avg_dl.util_avg); - used += thermal_load_avg(rq); + used = cpu_util_rt(rq); + used += cpu_util_dl(rq); if (unlikely(used >= max)) return 1; @@ -8838,82 +10058,14 @@ static unsigned long scale_rt_capacity(int cpu) static void update_cpu_capacity(struct sched_domain *sd, int cpu) { - unsigned long capacity_orig = arch_scale_cpu_capacity(cpu); unsigned long capacity = scale_rt_capacity(cpu); struct sched_group *sdg = sd->groups; - struct rq *rq = cpu_rq(cpu); - - rq->cpu_capacity_orig = capacity_orig; if (!capacity) capacity = 1; - rq->cpu_capacity = capacity; - - /* - * Detect if the performance domain is in capacity inversion state. - * - * Capacity inversion happens when another perf domain with equal or - * lower capacity_orig_of() ends up having higher capacity than this - * domain after subtracting thermal pressure. - * - * We only take into account thermal pressure in this detection as it's - * the only metric that actually results in *real* reduction of - * capacity due to performance points (OPPs) being dropped/become - * unreachable due to thermal throttling. - * - * We assume: - * * That all cpus in a perf domain have the same capacity_orig - * (same uArch). - * * Thermal pressure will impact all cpus in this perf domain - * equally. - */ - if (sched_energy_enabled()) { - unsigned long inv_cap = capacity_orig - thermal_load_avg(rq); - struct perf_domain *pd; - - rcu_read_lock(); - - pd = rcu_dereference(rq->rd->pd); - rq->cpu_capacity_inverted = 0; - - for (; pd; pd = pd->next) { - struct cpumask *pd_span = perf_domain_span(pd); - unsigned long pd_cap_orig, pd_cap; - - /* We can't be inverted against our own pd */ - if (cpumask_test_cpu(cpu_of(rq), pd_span)) - continue; - - cpu = cpumask_any(pd_span); - pd_cap_orig = arch_scale_cpu_capacity(cpu); - - if (capacity_orig < pd_cap_orig) - continue; - - /* - * handle the case of multiple perf domains have the - * same capacity_orig but one of them is under higher - * thermal pressure. We record it as capacity - * inversion. - */ - if (capacity_orig == pd_cap_orig) { - pd_cap = pd_cap_orig - thermal_load_avg(cpu_rq(cpu)); - - if (pd_cap > inv_cap) { - rq->cpu_capacity_inverted = inv_cap; - break; - } - } else if (pd_cap_orig > inv_cap) { - rq->cpu_capacity_inverted = inv_cap; - break; - } - } - - rcu_read_unlock(); - } - - trace_sched_cpu_capacity_tp(rq); + cpu_rq(cpu)->cpu_capacity = capacity; + trace_sched_cpu_capacity_tp(cpu_rq(cpu)); sdg->sgc->capacity = capacity; sdg->sgc->min_capacity = capacity; @@ -8940,9 +10092,9 @@ 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. */ @@ -8955,7 +10107,7 @@ void update_group_capacity(struct sched_domain *sd, int cpu) } } else { /* - * !SD_OVERLAP domains can assume that child groups + * !SD_NUMA domains can assume that child groups * span the current group. */ @@ -8984,19 +10136,13 @@ 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 whether a rq has a misfit task and if it looks like we can actually - * help that task: we can migrate the task to a CPU of higher capacity, or - * the task's current CPU is heavily pressured. - */ -static inline int check_misfit_status(struct rq *rq, struct sched_domain *sd) +/* Check if the rq has a misfit task */ +static inline bool check_misfit_status(struct rq *rq) { - return rq->misfit_task_load && - (rq->cpu_capacity_orig < rq->rd->max_cpu_capacity || - check_cpu_capacity(rq, sd)); + return rq->misfit_task_load; } /* @@ -9020,7 +10166,7 @@ static inline int check_misfit_status(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 @@ -9101,6 +10247,9 @@ group_type group_classify(unsigned int imbalance_pct, 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; @@ -9111,98 +10260,126 @@ group_type group_classify(unsigned int imbalance_pct, } /** - * asym_smt_can_pull_tasks - Check whether the load balancing CPU can pull tasks - * @dst_cpu: Destination CPU of the load balancing - * @sds: Load-balancing data with statistics of the local group - * @sgs: Load-balancing statistics of the candidate busiest group - * @sg: The candidate busiest group - * - * Check the state of the SMT siblings of both @sds::local and @sg and decide - * if @dst_cpu can pull tasks. + * sched_use_asym_prio - Check whether asym_packing priority must be used + * @sd: The scheduling domain of the load balancing + * @cpu: A CPU * - * If @dst_cpu does not have SMT siblings, it can pull tasks if two or more of - * the SMT siblings of @sg are busy. If only one CPU in @sg is busy, pull tasks - * only if @dst_cpu has higher priority. + * 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. * - * If both @dst_cpu and @sg have SMT siblings, and @sg has exactly one more - * busy CPU than @sds::local, let @dst_cpu pull tasks if it has higher priority. - * Bigger imbalances in the number of busy CPUs will be dealt with in - * update_sd_pick_busiest(). - * - * If @sg does not have SMT siblings, only pull tasks if all of the SMT siblings - * of @dst_cpu are idle and @sg has lower priority. - * - * Return: true if @dst_cpu can pull tasks, false otherwise. + * Returns: True if the priority of @cpu must be followed. False otherwise. */ -static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds, - struct sg_lb_stats *sgs, - struct sched_group *sg) +static bool sched_use_asym_prio(struct sched_domain *sd, int cpu) { -#ifdef CONFIG_SCHED_SMT - bool local_is_smt, sg_is_smt; - int sg_busy_cpus; + if (!(sd->flags & SD_ASYM_PACKING)) + return false; - local_is_smt = sds->local->flags & SD_SHARE_CPUCAPACITY; - sg_is_smt = sg->flags & SD_SHARE_CPUCAPACITY; + if (!sched_smt_active()) + return true; - sg_busy_cpus = sgs->group_weight - sgs->idle_cpus; + return sd->flags & SD_SHARE_CPUCAPACITY || is_core_idle(cpu); +} - if (!local_is_smt) { - /* - * If we are here, @dst_cpu is idle and does not have SMT - * siblings. Pull tasks if candidate group has two or more - * busy CPUs. - */ - if (sg_busy_cpus >= 2) /* implies sg_is_smt */ - return true; +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); +} - /* - * @dst_cpu does not have SMT siblings. @sg may have SMT - * siblings and only one is busy. In such case, @dst_cpu - * can help if it has higher priority and is idle (i.e., - * it has no running tasks). - */ - return sched_asym_prefer(dst_cpu, sg->asym_prefer_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; - /* @dst_cpu has SMT siblings. */ + return sched_asym(env->sd, env->dst_cpu, READ_ONCE(group->asym_prefer_cpu)); +} - if (sg_is_smt) { - int local_busy_cpus = sds->local->group_weight - - sds->local_stat.idle_cpus; - int busy_cpus_delta = sg_busy_cpus - local_busy_cpus; +/* 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 (busy_cpus_delta == 1) - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); + return (sg1->flags & SD_SHARE_CPUCAPACITY) != + (sg2->flags & SD_SHARE_CPUCAPACITY); +} +static inline bool smt_balance(struct lb_env *env, struct sg_lb_stats *sgs, + struct sched_group *group) +{ + if (!env->idle) return false; - } /* - * @sg does not have SMT siblings. Ensure that @sds::local does not end - * up with more than one busy SMT sibling and only pull tasks if there - * are not busy CPUs (i.e., no CPU has running tasks). + * 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 (!sds->local_stat.sum_nr_running) - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); + if (group->flags & SD_SHARE_CPUCAPACITY && + sgs->sum_h_nr_running > 1) + return true; return false; -#else - /* Always return false so that callers deal with non-SMT cases. */ - return false; -#endif } -static inline bool -sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs, - struct sched_group *group) +static inline long sibling_imbalance(struct lb_env *env, + struct sd_lb_stats *sds, + struct sg_lb_stats *busiest, + struct sg_lb_stats *local) { - /* Only do SMT checks if either local or candidate have SMT siblings */ - if ((sds->local->flags & SD_SHARE_CPUCAPACITY) || - (group->flags & SD_SHARE_CPUCAPACITY)) - return asym_smt_can_pull_tasks(env->dst_cpu, sds, sgs, group); + 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 sched_asym_prefer(env->dst_cpu, group->asym_prefer_cpu); + return imbalance; } static inline bool @@ -9212,7 +10389,7 @@ sched_reduced_capacity(struct rq *rq, struct sched_domain *sd) * When there is more than 1 task, the group_overloaded case already * takes care of cpu with reduced capacity */ - if (rq->cfs.h_nr_running != 1) + if (rq->cfs.h_nr_runnable != 1) return false; return check_cpu_capacity(rq, sd); @@ -9224,15 +10401,18 @@ sched_reduced_capacity(struct rq *rq, struct sched_domain *sd) * @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 i, nr_running, local_group; + int i, nr_running, local_group, sd_flags = env->sd->flags; + bool balancing_at_rd = !env->sd->parent; memset(sgs, 0, sizeof(*sgs)); @@ -9245,21 +10425,14 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->group_load += load; sgs->group_util += cpu_util_cfs(i); sgs->group_runnable += cpu_runnable(rq); - sgs->sum_h_nr_running += rq->cfs.h_nr_running; + sgs->sum_h_nr_running += rq->cfs.h_nr_runnable; nr_running = rq->nr_running; sgs->sum_nr_running += nr_running; - if (nr_running > 1) - *sg_status |= SG_OVERLOAD; - 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 /* * No need to call idle_cpu() if nr_running is not 0 */ @@ -9269,17 +10442,27 @@ static inline void update_sg_lb_stats(struct lb_env *env, continue; } + /* Overload indicator is only updated at root domain */ + if (balancing_at_rd && nr_running > 1) + *sg_overloaded = 1; + +#ifdef CONFIG_NUMA_BALANCING + /* Only fbq_classify_group() uses this to classify NUMA groups */ + if (sd_flags & SD_NUMA) { + sgs->nr_numa_running += rq->nr_numa_running; + sgs->nr_preferred_running += rq->nr_preferred_running; + } +#endif if (local_group) continue; - if (env->sd->flags & SD_ASYM_CPUCAPACITY) { + if (sd_flags & SD_ASYM_CPUCAPACITY) { /* Check for a misfit task on the cpu */ if (sgs->group_misfit_task_load < rq->misfit_task_load) { sgs->group_misfit_task_load = rq->misfit_task_load; - *sg_status |= SG_OVERLOAD; + *sg_overloaded = 1; } - } else if ((env->idle != CPU_NOT_IDLE) && - sched_reduced_capacity(rq, env->sd)) { + } 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; @@ -9291,11 +10474,13 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->group_weight = group->group_weight; /* Check if dst CPU is idle and preferred to this group */ - if (!local_group && env->sd->flags & SD_ASYM_PACKING && - env->idle != CPU_NOT_IDLE && sgs->sum_h_nr_running && - sched_asym(env, sds, sgs, 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); @@ -9355,9 +10540,7 @@ static bool update_sd_pick_busiest(struct lb_env *env, switch (sgs->group_type) { case group_overloaded: /* Select the overloaded group with highest avg_load. */ - if (sgs->avg_load <= busiest->avg_load) - return false; - break; + return sgs->avg_load > busiest->avg_load; case group_imbalanced: /* @@ -9368,18 +10551,25 @@ static bool update_sd_pick_busiest(struct lb_env *env, case group_asym_packing: /* Prefer to move from lowest priority CPU's work */ - if (sched_asym_prefer(sg->asym_prefer_cpu, sds->busiest->asym_prefer_cpu)) - return false; - break; + return sched_asym_prefer(READ_ONCE(sds->busiest->asym_prefer_cpu), + READ_ONCE(sg->asym_prefer_cpu)); case group_misfit_task: /* * If we have more than one misfit sg go with the biggest * misfit. */ - if (sgs->group_misfit_task_load < busiest->group_misfit_task_load) - return false; - break; + return sgs->group_misfit_task_load > busiest->group_misfit_task_load; + + 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: /* @@ -9390,15 +10580,40 @@ static bool update_sd_pick_busiest(struct lb_env *env, * contention when accessing shared HW resources. * * XXX for now avg_load is not computed and always 0 so we - * select the 1st one. + * select the 1st one, except if @sg is composed of SMT + * siblings. */ - if (sgs->avg_load <= busiest->avg_load) + + 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: /* - * Select not overloaded group with lowest number of idle cpus + * 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 @@ -9445,7 +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; @@ -9455,7 +10670,7 @@ 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; @@ -9496,10 +10711,8 @@ static int idle_cpu_without(int cpu, struct task_struct *p) * be computed and tested before calling idle_cpu_without(). */ -#ifdef CONFIG_SMP if (rq->ttwu_pending) return 0; -#endif return 1; } @@ -9524,7 +10737,7 @@ static inline void update_sg_wakeup_stats(struct sched_domain *sd, if (sd->flags & SD_ASYM_CPUCAPACITY) sgs->group_misfit_task_load = 1; - for_each_cpu(i, sched_group_span(group)) { + for_each_cpu_and(i, sched_group_span(group), p->cpus_ptr) { struct rq *rq = cpu_rq(i); unsigned int local; @@ -9532,7 +10745,7 @@ static inline void update_sg_wakeup_stats(struct sched_domain *sd, sgs->group_util += cpu_util_without(i, p); sgs->group_runnable += cpu_runnable_without(rq, p); local = task_running_on_cpu(i, p); - sgs->sum_h_nr_running += rq->cfs.h_nr_running - local; + sgs->sum_h_nr_running += rq->cfs.h_nr_runnable - local; nr_running = rq->nr_running - local; sgs->sum_nr_running += nr_running; @@ -9593,6 +10806,7 @@ static bool update_pick_idlest(struct sched_group *idlest, case group_imbalanced: case group_asym_packing: + case group_smt_balance: /* Those types are not used in the slow wakeup path */ return false; @@ -9619,13 +10833,13 @@ static bool update_pick_idlest(struct sched_group *idlest, } /* - * find_idlest_group() finds and returns the least busy CPU group within the + * 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 * -find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) +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; @@ -9724,6 +10938,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu) case group_imbalanced: case group_asym_packing: + case group_smt_balance: /* Those type are not used in the slow wakeup path */ return NULL; @@ -9868,12 +11083,11 @@ static void update_idle_cpu_scan(struct lb_env *env, static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sds) { - 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; unsigned long sum_util = 0; - int sg_status = 0; + bool sg_overloaded = 0, sg_overutilized = 0; do { struct sg_lb_stats *sgs = &tmp_sgs; @@ -9889,18 +11103,13 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd update_group_capacity(env->sd, env->dst_cpu); } - update_sg_lb_stats(env, sds, sg, sgs, &sg_status); - - if (local_group) - goto next_group; + update_sg_lb_stats(env, sds, sg, sgs, &sg_overloaded, &sg_overutilized); - - if (update_sd_pick_busiest(env, sds, sg, sgs)) { + if (!local_group && update_sd_pick_busiest(env, sds, sg, sgs)) { sds->busiest = sg; sds->busiest_stat = *sgs; } -next_group: /* Now, start updating sd_lb_stats */ sds->total_load += sgs->group_load; sds->total_capacity += sgs->group_capacity; @@ -9909,27 +11118,26 @@ next_group: sg = sg->next; } while (sg != env->sd->groups); - /* Tag domain that child domain prefers tasks go to siblings first */ - sds->prefer_sibling = child && child->flags & SD_PREFER_SIBLING; + /* + * 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); if (env->sd->flags & SD_NUMA) env->fbq_type = fbq_classify_group(&sds->busiest_stat); if (!env->sd->parent) { - struct root_domain *rd = env->dst_rq->rd; - /* update overload indicator if we are at root domain */ - WRITE_ONCE(rd->overload, sg_status & SG_OVERLOAD); + set_rd_overloaded(env->dst_rq->rd, sg_overloaded); /* Update over-utilization (tipping point, U >= 0) indicator */ - WRITE_ONCE(rd->overutilized, sg_status & SG_OVERUTILIZED); - trace_sched_overutilized_tp(rd, sg_status & SG_OVERUTILIZED); - } else if (sg_status & SG_OVERUTILIZED) { - struct root_domain *rd = env->dst_rq->rd; - - WRITE_ONCE(rd->overutilized, SG_OVERUTILIZED); - trace_sched_overutilized_tp(rd, SG_OVERUTILIZED); + set_rd_overutilized(env->dst_rq->rd, sg_overutilized); + } else if (sg_overutilized) { + set_rd_overutilized(env->dst_rq->rd, sg_overutilized); } update_idle_cpu_scan(env, sum_util); @@ -9974,6 +11182,13 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s 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 @@ -9992,7 +11207,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ if (local->group_type == group_has_spare) { if ((busiest->group_type > group_fully_busy) && - !(env->sd->flags & SD_SHARE_PKG_RESOURCES)) { + !(env->sd->flags & SD_SHARE_LLC)) { /* * If busiest is overloaded, try to fill spare * capacity. This might end up creating spare capacity @@ -10012,7 +11227,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * waiting task in this overloaded busiest group. Let's * try to pull it. */ - if (env->idle != CPU_NOT_IDLE && env->imbalance == 0) { + if (env->idle && env->imbalance == 0) { env->migration_type = migrate_task; env->imbalance = 1; } @@ -10021,19 +11236,17 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s } if (busiest->group_weight == 1 || sds->prefer_sibling) { - unsigned int nr_diff = busiest->sum_nr_running; /* * When prefer sibling, evenly spread running tasks on * groups. */ env->migration_type = migrate_task; - lsub_positive(&nr_diff, local->sum_nr_running); - env->imbalance = nr_diff; + env->imbalance = sibling_imbalance(env, sds, busiest, local); } else { /* * If there is no overload, we just want to even the number of - * idle cpus. + * idle CPUs. */ env->migration_type = migrate_task; env->imbalance = max_t(long, 0, @@ -10079,6 +11292,16 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s 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; + } + } /* @@ -10096,7 +11319,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s ) / 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: @@ -10119,7 +11342,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s */ /** - * 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. * @@ -10128,7 +11351,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s * * 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; @@ -10141,24 +11364,20 @@ static struct sched_group *find_busiest_group(struct lb_env *env) */ update_sd_lb_stats(env, &sds); - if (sched_energy_enabled()) { - struct root_domain *rd = env->dst_rq->rd; - - if (rcu_dereference(rd->pd) && !READ_ONCE(rd->overutilized)) - goto out_balanced; - } - - local = &sds.local_stat; - busiest = &sds.busiest_stat; - /* There is no busy sibling group to pull tasks from */ if (!sds.busiest) goto out_balanced; + busiest = &sds.busiest_stat; + /* Misfit tasks should be dealt with regardless of the avg load */ if (busiest->group_type == group_misfit_task) goto force_balance; + if (!is_rd_overutilized(env->dst_rq->rd) && + rcu_dereference(env->dst_rq->rd->pd)) + goto out_balanced; + /* ASYM feature bypasses nice load balance check */ if (busiest->group_type == group_asym_packing) goto force_balance; @@ -10171,6 +11390,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env) if (busiest->group_type == group_imbalanced) 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. @@ -10210,22 +11430,32 @@ static struct sched_group *find_busiest_group(struct lb_env *env) goto out_balanced; } - /* Try to move all excess tasks to child's sibling domain */ + /* + * 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 && - busiest->sum_nr_running > local->sum_nr_running + 1) + sibling_imbalance(env, &sds, busiest, local) > 1) goto force_balance; if (busiest->group_type != group_overloaded) { - if (env->idle == CPU_NOT_IDLE) + 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)) + local->idle_cpus <= (busiest->idle_cpus + 1)) { /* * If the busiest group is not overloaded * and there is no imbalance between this and busiest @@ -10236,12 +11466,14 @@ static struct sched_group *find_busiest_group(struct lb_env *env) * there is more than 1 CPU per group. */ goto out_balanced; + } - if (busiest->sum_h_nr_running == 1) + if (busiest->sum_h_nr_running == 1) { /* * busiest doesn't have any tasks waiting to run */ goto out_balanced; + } } force_balance: @@ -10255,9 +11487,9 @@ 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; @@ -10295,7 +11527,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, if (rt > env->fbq_type) continue; - nr_running = rq->cfs.h_nr_running; + nr_running = rq->cfs.h_nr_runnable; if (!nr_running) continue; @@ -10312,10 +11544,14 @@ static struct rq *find_busiest_queue(struct lb_env *env, nr_running == 1) continue; - /* Make sure we only pull tasks from a CPU of lower priority */ - if ((env->sd->flags & SD_ASYM_PACKING) && - sched_asym_prefer(i, env->dst_cpu) && - nr_running == 1) + /* + * 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 (sched_asym(env->sd, i, env->dst_cpu) && nr_running == 1) continue; switch (env->migration_type) { @@ -10351,7 +11587,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, break; case migrate_util: - util = cpu_util_cfs(i); + util = cpu_util_cfs_boost(i); /* * Don't try to pull utilization from a CPU with one @@ -10402,12 +11638,18 @@ 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. + * 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 != CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_PACKING) && - sched_asym_prefer(env->dst_cpu, env->src_cpu); + 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 @@ -10443,8 +11685,8 @@ 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; @@ -10460,8 +11702,9 @@ 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; + int cpu, idle_smt = -1; /* * Ensure the balancing environment is consistent; can happen @@ -10483,24 +11726,88 @@ static int should_we_balance(struct lb_env *env) 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; - /* Are we the first idle CPU? */ + /* + * 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; } + /* Are we the first idle CPU with busy siblings? */ + if (idle_smt != -1) + return idle_smt == 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) { @@ -10514,13 +11821,14 @@ static int load_balance(int this_cpu, struct rq *this_rq, .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, .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); @@ -10532,13 +11840,21 @@ 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; @@ -10546,7 +11862,7 @@ redo: WARN_ON_ONCE(busiest == env.dst_rq); - schedstat_add(sd->lb_imbalance[idle], env.imbalance); + update_lb_imbalance_stat(&env, sd, idle); env.src_cpu = busiest->cpu; env.src_rq = busiest; @@ -10556,7 +11872,7 @@ redo: 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. @@ -10592,9 +11908,7 @@ more_balance: if (env.flags & LBF_NEED_BREAK) { env.flags &= ~LBF_NEED_BREAK; - /* Stop if we tried all running tasks */ - if (env.loop < busiest->nr_running) - goto more_balance; + goto more_balance; } /* @@ -10671,8 +11985,12 @@ 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)) { @@ -10703,13 +12021,15 @@ more_balance: busiest->push_cpu = this_cpu; active_balance = 1; } - raw_spin_rq_unlock_irqrestore(busiest, 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); } + preempt_enable(); } } else { sd->nr_balance_failed = 0; @@ -10749,12 +12069,17 @@ out_one_pinned: ld_moved = 0; /* - * newidle_balance() disregards balance intervals, so we could + * 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 */ @@ -10763,6 +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; } @@ -10887,10 +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) @@ -10898,24 +12224,43 @@ void update_max_interval(void) max_load_balance_interval = HZ*num_online_cpus()/10; } -static inline bool update_newidle_cost(struct sched_domain *sd, u64 cost) +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 = jiffies; - } else if (time_after(jiffies, sd->last_decay_max_lb_cost + HZ)) { + 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 = jiffies; - + sd->last_decay_max_lb_cost = now; return true; } @@ -10928,7 +12273,7 @@ static inline bool update_newidle_cost(struct sched_domain *sd, u64 cost) * * 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; @@ -10938,7 +12283,7 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle) /* 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(); @@ -10947,7 +12292,7 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle) * Decay the newidle max times here because this is a regular * visit to all the domains. */ - need_decay = update_newidle_cost(sd, 0); + need_decay = update_newidle_cost(sd, 0, 0); max_cost += sd->max_newidle_lb_cost; /* @@ -10962,29 +12307,19 @@ static void rebalance_domains(struct rq *rq, enum cpu_idle_type idle) } interval = get_sd_balance_interval(sd, busy); - - need_serialize = sd->flags & SD_SERIALIZE; - if (need_serialize) { - if (!spin_trylock(&balancing)) - goto out; - } - 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; - busy = idle != CPU_IDLE && !sched_idle_cpu(cpu); + idle = idle_cpu(cpu); + busy = !idle && !sched_idle_cpu(cpu); } sd->last_balance = jiffies; 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; @@ -11017,36 +12352,37 @@ 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. - * - HK_TYPE_MISC CPUs are used for this task, because HK_TYPE_SCHED not set - * anywhere yet. */ - static inline int find_new_ilb(void) { - int ilb; const struct cpumask *hk_mask; + int ilb_cpu; - hk_mask = housekeeping_cpumask(HK_TYPE_MISC); + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); - for_each_cpu_and(ilb, nohz.idle_cpus_mask, hk_mask) { + for_each_cpu_and(ilb_cpu, nohz.idle_cpus_mask, hk_mask) { - if (ilb == smp_processor_id()) + if (ilb_cpu == smp_processor_id()) continue; - if (idle_cpu(ilb)) - return ilb; + if (idle_cpu(ilb_cpu)) + return ilb_cpu; } - return nr_cpu_ids; + return -1; } /* - * Kick a CPU to do the nohz balancing, if it is time for it. We pick any - * idle CPU in the HK_TYPE_MISC housekeeping set (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) { @@ -11060,8 +12396,14 @@ static void kick_ilb(unsigned int flags) 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; /* @@ -11074,7 +12416,7 @@ static void kick_ilb(unsigned int flags) /* * This way we generate an IPI on the target CPU which - * is idle. And the softirq performing nohz idle load balance + * is idle, and the softirq performing NOHZ idle load balancing * will be run before returning from the IPI. */ smp_call_function_single_async(ilb_cpu, &cpu_rq(ilb_cpu)->nohz_csd); @@ -11103,7 +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; @@ -11125,11 +12467,10 @@ static void nohz_balancer_kick(struct rq *rq) sd = rcu_dereference(rq->sd); if (sd) { /* - * If there's a CFS task and the current CPU has reduced - * capacity; kick the ILB to see if there's a better CPU to run - * on. + * If there's a runnable CFS task and the current CPU has reduced + * capacity, kick the ILB to see if there's a better CPU to run on: */ - if (rq->cfs.h_nr_running >= 1 && check_cpu_capacity(rq, sd)) { + if (rq->cfs.h_nr_runnable >= 1 && check_cpu_capacity(rq, sd)) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } @@ -11141,9 +12482,12 @@ static void nohz_balancer_kick(struct rq *rq) * 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_prefer(i, cpu)) { + if (sched_asym(sd, i, cpu)) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } @@ -11156,7 +12500,7 @@ static void nohz_balancer_kick(struct rq *rq) * When ASYM_CPUCAPACITY; see if there's a higher capacity CPU * to run the misfit task on. */ - if (check_misfit_status(rq, sd)) { + if (check_misfit_status(rq)) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } @@ -11175,11 +12519,11 @@ static void nohz_balancer_kick(struct rq *rq) if (sds) { /* * If there is an imbalance between LLC domains (IOW we could - * increase the overall cache use), we need some less-loaded LLC - * domain to pull some load. Likewise, we may need to spread + * 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 + * 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); @@ -11216,7 +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; @@ -11252,16 +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_TYPE_SCHED)) - return; - /* * Can be set safely without rq->lock held * If a clear happens, it will have evaluated last additions because @@ -11300,7 +12640,7 @@ void nohz_balance_enter_idle(int cpu) 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); } @@ -11318,13 +12658,13 @@ static bool update_nohz_stats(struct rq *rq) if (!time_after(jiffies, READ_ONCE(rq->last_blocked_load_update_tick))) return true; - update_blocked_averages(cpu); + 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. */ @@ -11339,7 +12679,7 @@ static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) int balance_cpu; 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 @@ -11375,7 +12715,7 @@ static void _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()) { + if (!idle_cpu(this_cpu) && need_resched()) { if (flags & NOHZ_STATS_KICK) has_blocked_load = true; if (flags & NOHZ_NEXT_KICK) @@ -11400,7 +12740,7 @@ static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) 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)) { @@ -11429,7 +12769,7 @@ abort: /* * 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) { @@ -11449,8 +12789,19 @@ static bool nohz_idle_balance(struct rq *this_rq, enum cpu_idle_type idle) } /* - * Check if we need to run the ILB for updating blocked load before entering - * idle state. + * 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) { @@ -11460,7 +12811,7 @@ void nohz_run_idle_balance(int cpu) /* * Update the blocked load only if no SCHED_SOFTIRQ is about to happen - * (ie NOHZ_STATS_KICK set) and will do the same. + * (i.e. NOHZ_STATS_KICK set) and will do the same. */ if ((flags == NOHZ_NEWILB_KICK) && !need_resched()) _nohz_idle_balance(cpu_rq(cpu), NOHZ_STATS_KICK); @@ -11470,13 +12821,6 @@ static void nohz_newidle_balance(struct rq *this_rq) { int this_cpu = this_rq->cpu; - /* - * This CPU doesn't want to be disturbed by scheduler - * housekeeping - */ - if (!housekeeping_cpu(this_cpu, HK_TYPE_SCHED)) - return; - /* Will wake up very soon. No time for doing anything else*/ if (this_rq->avg_idle < sysctl_sched_migration_cost) return; @@ -11493,7 +12837,7 @@ static void nohz_newidle_balance(struct rq *this_rq) 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) @@ -11502,10 +12846,10 @@ 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 */ /* - * newidle_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: @@ -11513,10 +12857,11 @@ static inline void nohz_newidle_balance(struct rq *this_rq) { } * 0 - failed, no new tasks * > 0 - success, new (fair) tasks present */ -static int newidle_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; @@ -11531,8 +12876,9 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf) 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); @@ -11552,26 +12898,28 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf) rcu_read_lock(); sd = rcu_dereference_check_sched_domain(this_rq->sd); + if (!sd) { + rcu_read_unlock(); + goto out; + } - if (!READ_ONCE(this_rq->rd->overload) || - (sd && this_rq->avg_idle < sd->max_newidle_lb_cost)) { + if (!get_rd_overloaded(this_rq->rd) || + this_rq->avg_idle < sd->max_newidle_lb_cost) { - if (sd) - update_next_balance(sd, &next_balance); + update_next_balance(sd, &next_balance); rcu_read_unlock(); - goto out; } rcu_read_unlock(); + rq_modified_clear(this_rq); raw_spin_rq_unlock(this_rq); t0 = sched_clock_cpu(this_cpu); - update_blocked_averages(this_cpu); + sched_balance_update_blocked_averages(this_cpu); rcu_read_lock(); for_each_domain(this_cpu, sd) { - int continue_balancing = 1; u64 domain_cost; update_next_balance(sd, &next_balance); @@ -11580,25 +12928,44 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf) break; if (sd->flags & SD_BALANCE_NEWIDLE) { + 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); t1 = sched_clock_cpu(this_cpu); domain_cost = t1 - t0; - update_newidle_cost(sd, domain_cost); - curr_cost += domain_cost; t0 = t1; + + /* + * 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 || - this_rq->ttwu_pending) + if (pulled_task || !continue_balancing) break; } rcu_read_unlock(); @@ -11613,11 +12980,11 @@ static int newidle_balance(struct rq *this_rq, struct rq_flags *rf) * have been enqueued in the meantime. Since we're not going idle, * pretend we pulled a task. */ - if (this_rq->cfs.h_nr_running && !pulled_task) + if (this_rq->cfs.h_nr_queued && !pulled_task) pulled_task = 1; - /* Is there a task of a high priority class? */ - if (this_rq->nr_running != this_rq->cfs.h_nr_running) + /* If a higher prio class was modified, restart the pick */ + if (rq_modified_above(this_rq, &fair_sched_class)) pulled_task = -1; out: @@ -11636,19 +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. @@ -11657,14 +13026,14 @@ 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 or @@ -11692,16 +13061,17 @@ static void rq_offline_fair(struct rq *rq) /* Ensure any throttled groups are reachable by pick_next_task */ unthrottle_offline_cfs_rqs(rq); -} -#endif /* CONFIG_SMP */ + /* 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 slice = sched_slice(cfs_rq_of(se), se); u64 rtime = se->sum_exec_runtime - se->prev_sum_exec_runtime; + u64 slice = se->slice; return (rtime * min_nr_tasks > slice); } @@ -11726,15 +13096,179 @@ static inline void task_tick_core(struct rq *rq, struct task_struct *curr) * MIN_NR_TASKS_DURING_FORCEIDLE - 1 tasks and use that to check * if we need to give up the CPU. */ - if (rq->core->core_forceidle_count && rq->cfs.nr_running == 1 && + if (rq->core->core_forceidle_count && rq->cfs.nr_queued == 1 && __entity_slice_used(&curr->se, MIN_NR_TASKS_DURING_FORCEIDLE)) resched_curr(rq); } /* - * se_fi_update - Update the cfs_rq->min_vruntime_fi in a CFS hierarchy if needed. + * 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(struct sched_entity *se, unsigned int fi_seq, bool forceidle) +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); @@ -11745,7 +13279,7 @@ static void se_fi_update(struct sched_entity *se, unsigned int fi_seq, bool forc cfs_rq->forceidle_seq = fi_seq; } - cfs_rq->min_vruntime_fi = cfs_rq->min_vruntime; + cfs_rq->zero_vruntime_fi = cfs_rq->zero_vruntime; } } @@ -11759,16 +13293,17 @@ void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi) se_fi_update(se, rq->core->core_forceidle_seq, in_fi); } -bool cfs_prio_less(struct task_struct *a, struct task_struct *b, bool 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); - struct sched_entity *sea = &a->se; - struct sched_entity *seb = &b->se; + 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; - SCHED_WARN_ON(task_rq(b)->core != rq->core); + WARN_ON_ONCE(task_rq(b)->core != rq->core); #ifdef CONFIG_FAIR_GROUP_SCHED /* @@ -11790,24 +13325,36 @@ bool cfs_prio_less(struct task_struct *a, struct task_struct *b, bool in_fi) cfs_rqa = sea->cfs_rq; cfs_rqb = seb->cfs_rq; -#else +#else /* !CONFIG_FAIR_GROUP_SCHED: */ cfs_rqa = &task_rq(a)->cfs; cfs_rqb = &task_rq(b)->cfs; -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ /* * Find delta after normalizing se's vruntime with its cfs_rq's - * min_vruntime_fi, which would have been updated in prior calls + * zero_vruntime_fi, which would have been updated in prior calls * to se_fi_update(). */ delta = (s64)(sea->vruntime - seb->vruntime) + - (s64)(cfs_rqb->min_vruntime_fi - cfs_rqa->min_vruntime_fi); + (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 -static inline void task_tick_core(struct rq *rq, struct task_struct *curr) {} + 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. @@ -11831,7 +13378,7 @@ 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); } @@ -11843,33 +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); } /* @@ -11877,12 +13398,15 @@ 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 (rq->cfs.nr_running == 1) + if (p->prio == oldprio) + return; + + if (rq->cfs.nr_queued == 1) return; /* @@ -11890,39 +13414,12 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) * our priority decreased, or if we are not currently running on * this runqueue and our priority is higher than the current's */ - if (task_current(rq, 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 || - (READ_ONCE(p->__state) == TASK_WAKING && p->sched_remote_wakeup)) - return true; - - return false; + } else { + wakeup_preempt(rq, p, 0); + } } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -11934,10 +13431,13 @@ static void propagate_entity_cfs_rq(struct sched_entity *se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - if (cfs_rq_throttled(cfs_rq)) - return; - - if (!throttled_hierarchy(cfs_rq)) + /* + * 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 */ @@ -11948,22 +13448,20 @@ static void propagate_entity_cfs_rq(struct sched_entity *se) update_load_avg(cfs_rq, se, UPDATE_TG); - if (cfs_rq_throttled(cfs_rq)) - break; - - if (!throttled_hierarchy(cfs_rq)) + 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); -#ifdef CONFIG_SMP /* * In case the task sched_avg hasn't been attached: * - A forked task which hasn't been woken up by wake_up_new_task(). @@ -11972,7 +13470,6 @@ static void detach_entity_cfs_rq(struct sched_entity *se) */ if (!se->avg.last_update_time) return; -#endif /* Catch up with the cfs_rq and remove our load when we leave */ update_load_avg(cfs_rq, se, 0); @@ -11995,16 +13492,6 @@ static void attach_entity_cfs_rq(struct sched_entity *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); } @@ -12012,12 +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) @@ -12027,31 +13516,29 @@ 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 (task_current(rq, 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. - * - * This routine is mostly called to set cfs_rq->curr field when a task - * migrates between groups/classes. - */ -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 sched_entity *se = &p->se; -#ifdef CONFIG_SMP if (task_on_rq_queued(p)) { /* * Move the next running task to the front of the list, so our @@ -12059,7 +13546,27 @@ static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) */ list_move(&se->group_node, &rq->cfs_tasks); } -#endif + 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_next_task_fair(struct rq *rq, struct task_struct *p, bool first) +{ + struct sched_entity *se = &p->se; for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); @@ -12068,15 +13575,15 @@ static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) /* 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; - u64_u32_store(cfs_rq->min_vruntime, (u64)(-(1LL << 20))); -#ifdef CONFIG_SMP + cfs_rq->zero_vruntime = (u64)(-(1LL << 20)); raw_spin_lock_init(&cfs_rq->removed.lock); -#endif } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -12091,10 +13598,8 @@ static void task_change_group_fair(struct task_struct *p) detach_task_cfs_rq(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); } @@ -12129,7 +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), @@ -12175,28 +13680,35 @@ void online_fair_sched_group(struct task_group *tg) 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_rq_lock_irqsave(rq, flags); - list_del_leaf_cfs_rq(tg->cfs_rq[cpu]); - raw_spin_rq_unlock_irqrestore(rq, flags); + if (cfs_rq->on_list) { + guard(rq_lock_irqsave)(rq); + list_del_leaf_cfs_rq(cfs_rq); + } } } @@ -12305,7 +13817,7 @@ int sched_group_set_idle(struct task_group *tg, long idle) for_each_possible_cpu(i) { struct rq *rq = cpu_rq(i); struct sched_entity *se = tg->se[i]; - struct cfs_rq *parent_cfs_rq, *grp_cfs_rq = tg->cfs_rq[i]; + struct cfs_rq *grp_cfs_rq = tg->cfs_rq[i]; bool was_idle = cfs_rq_is_idle(grp_cfs_rq); long idle_task_delta; struct rq_flags rf; @@ -12316,16 +13828,8 @@ int sched_group_set_idle(struct task_group *tg, long idle) if (WARN_ON_ONCE(was_idle == cfs_rq_is_idle(grp_cfs_rq))) goto next_cpu; - if (se->on_rq) { - parent_cfs_rq = cfs_rq_of(se); - if (cfs_rq_is_idle(grp_cfs_rq)) - parent_cfs_rq->idle_nr_running++; - else - parent_cfs_rq->idle_nr_running--; - } - - idle_task_delta = grp_cfs_rq->h_nr_running - - grp_cfs_rq->idle_h_nr_running; + idle_task_delta = grp_cfs_rq->h_nr_queued - + grp_cfs_rq->h_nr_idle; if (!cfs_rq_is_idle(grp_cfs_rq)) idle_task_delta *= -1; @@ -12335,7 +13839,7 @@ int sched_group_set_idle(struct task_group *tg, long idle) if (!se->on_rq) break; - cfs_rq->idle_h_nr_running += idle_task_delta; + cfs_rq->h_nr_idle += idle_task_delta; /* Already accounted at parent level and above. */ if (cfs_rq_is_idle(cfs_rq)) @@ -12356,19 +13860,6 @@ next_cpu: 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) -{ - return 1; -} - -void online_fair_sched_group(struct task_group *tg) { } - -void unregister_fair_sched_group(struct task_group *tg) { } - #endif /* CONFIG_FAIR_GROUP_SCHED */ @@ -12382,7 +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; } @@ -12392,20 +13883,20 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task */ 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_next_task = __pick_next_task_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 - .balance = balance_fair, - .pick_task = pick_task_fair, .select_task_rq = select_task_rq_fair, .migrate_task_rq = migrate_task_rq_fair, @@ -12413,13 +13904,14 @@ DEFINE_SCHED_CLASS(fair) = { .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, .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, @@ -12431,12 +13923,15 @@ DEFINE_SCHED_CLASS(fair) = { .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, *pos; @@ -12470,25 +13965,28 @@ void show_numa_stats(struct task_struct *p, struct seq_file *m) rcu_read_unlock(); } #endif /* CONFIG_NUMA_BALANCING */ -#endif /* CONFIG_SCHED_DEBUG */ __init void init_sched_fair_class(void) { -#ifdef CONFIG_SMP 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, run_rebalance_domains); + 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 */ - } |