diff options
Diffstat (limited to 'kernel/sched')
46 files changed, 14802 insertions, 7548 deletions
diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile index 976092b7bd45..b1f1a367034f 100644 --- a/kernel/sched/Makefile +++ b/kernel/sched/Makefile @@ -1,5 +1,8 @@ # SPDX-License-Identifier: GPL-2.0 +CONTEXT_ANALYSIS_core.o := y +CONTEXT_ANALYSIS_fair.o := y + # The compilers are complaining about unused variables inside an if(0) scope # block. This is daft, shut them up. ccflags-y += $(call cc-disable-warning, unused-but-set-variable) @@ -22,6 +25,11 @@ ifneq ($(CONFIG_SCHED_OMIT_FRAME_POINTER),y) CFLAGS_core.o := $(PROFILING) -fno-omit-frame-pointer endif +# Branch profiling isn't noinstr-safe +ifdef CONFIG_TRACE_BRANCH_PROFILING +CFLAGS_build_policy.o += -DDISABLE_BRANCH_PROFILING +CFLAGS_build_utility.o += -DDISABLE_BRANCH_PROFILING +endif # # Build efficiency: # diff --git a/kernel/sched/autogroup.c b/kernel/sched/autogroup.c index 2b331822c7e7..e380cf9372bb 100644 --- a/kernel/sched/autogroup.c +++ b/kernel/sched/autogroup.c @@ -4,6 +4,9 @@ * Auto-group scheduling implementation: */ +#include "autogroup.h" +#include "sched.h" + unsigned int __read_mostly sysctl_sched_autogroup_enabled = 1; static struct autogroup autogroup_default; static atomic_t autogroup_seq_nr; @@ -25,9 +28,9 @@ static void __init sched_autogroup_sysctl_init(void) { register_sysctl_init("kernel", sched_autogroup_sysctls); } -#else +#else /* !CONFIG_SYSCTL: */ #define sched_autogroup_sysctl_init() do { } while (0) -#endif +#endif /* !CONFIG_SYSCTL */ void __init autogroup_init(struct task_struct *init_task) { @@ -83,7 +86,7 @@ static inline struct autogroup *autogroup_task_get(struct task_struct *p) static inline struct autogroup *autogroup_create(void) { - struct autogroup *ag = kzalloc(sizeof(*ag), GFP_KERNEL); + struct autogroup *ag = kzalloc_obj(*ag); struct task_group *tg; if (!ag) @@ -108,7 +111,7 @@ static inline struct autogroup *autogroup_create(void) free_rt_sched_group(tg); tg->rt_se = root_task_group.rt_se; tg->rt_rq = root_task_group.rt_rq; -#endif +#endif /* CONFIG_RT_GROUP_SCHED */ tg->autogroup = ag; sched_online_group(tg, &root_task_group); @@ -175,8 +178,8 @@ autogroup_move_group(struct task_struct *p, struct autogroup *ag) * this process can already run with task_group() == prev->tg or we can * race with cgroup code which can read autogroup = prev under rq->lock. * In the latter case for_each_thread() can not miss a migrating thread, - * cpu_cgroup_attach() must not be possible after cgroup_exit() and it - * can't be removed from thread list, we hold ->siglock. + * cpu_cgroup_attach() must not be possible after cgroup_task_exit() + * and it can't be removed from thread list, we hold ->siglock. * * If an exiting thread was already removed from thread list we rely on * sched_autogroup_exit_task(). diff --git a/kernel/sched/autogroup.h b/kernel/sched/autogroup.h index 90d69f2c5eaf..06c82b2bdfb5 100644 --- a/kernel/sched/autogroup.h +++ b/kernel/sched/autogroup.h @@ -2,6 +2,8 @@ #ifndef _KERNEL_SCHED_AUTOGROUP_H #define _KERNEL_SCHED_AUTOGROUP_H +#include "sched.h" + #ifdef CONFIG_SCHED_AUTOGROUP struct autogroup { @@ -41,7 +43,7 @@ autogroup_task_group(struct task_struct *p, struct task_group *tg) extern int autogroup_path(struct task_group *tg, char *buf, int buflen); -#else /* !CONFIG_SCHED_AUTOGROUP */ +#else /* !CONFIG_SCHED_AUTOGROUP: */ static inline void autogroup_init(struct task_struct *init_task) { } static inline void autogroup_free(struct task_group *tg) { } @@ -61,6 +63,6 @@ static inline int autogroup_path(struct task_group *tg, char *buf, int buflen) return 0; } -#endif /* CONFIG_SCHED_AUTOGROUP */ +#endif /* !CONFIG_SCHED_AUTOGROUP */ #endif /* _KERNEL_SCHED_AUTOGROUP_H */ diff --git a/kernel/sched/build_policy.c b/kernel/sched/build_policy.c index fae1f5c921eb..755883faf751 100644 --- a/kernel/sched/build_policy.c +++ b/kernel/sched/build_policy.c @@ -50,17 +50,17 @@ #include "idle.c" #include "rt.c" +#include "cpudeadline.c" -#ifdef CONFIG_SMP -# include "cpudeadline.c" -# include "pelt.c" -#endif +#include "pelt.c" #include "cputime.c" #include "deadline.c" #ifdef CONFIG_SCHED_CLASS_EXT +# include "ext_internal.h" # include "ext.c" +# include "ext_idle.c" #endif #include "syscalls.c" diff --git a/kernel/sched/build_utility.c b/kernel/sched/build_utility.c index 80a3df49ab47..e2cf3b08d4e9 100644 --- a/kernel/sched/build_utility.c +++ b/kernel/sched/build_utility.c @@ -68,9 +68,7 @@ # include "cpufreq_schedutil.c" #endif -#ifdef CONFIG_SCHED_DEBUG -# include "debug.c" -#endif +#include "debug.c" #ifdef CONFIG_SCHEDSTATS # include "stats.c" @@ -82,11 +80,10 @@ #include "wait_bit.c" #include "wait.c" -#ifdef CONFIG_SMP -# include "cpupri.c" -# include "stop_task.c" -# include "topology.c" -#endif +#include "cpupri.c" +#include "stop_task.c" + +#include "topology.c" #ifdef CONFIG_SCHED_CORE # include "core_sched.c" diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index a09655b48140..2ae4fbf13431 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -54,6 +54,9 @@ * */ +#include <linux/sched/clock.h> +#include "sched.h" + /* * Scheduler clock - returns current time in nanosec units. * This is default implementation. @@ -170,6 +173,7 @@ notrace static void __sched_clock_work(struct work_struct *work) scd->tick_gtod, __gtod_offset, scd->tick_raw, __sched_clock_offset); + disable_sched_clock_irqtime(); static_branch_disable(&__sched_clock_stable); } @@ -235,6 +239,8 @@ static int __init sched_clock_init_late(void) if (__sched_clock_stable_early) __set_sched_clock_stable(); + else + disable_sched_clock_irqtime(); /* disable if clock unstable. */ return 0; } @@ -471,7 +477,7 @@ notrace void sched_clock_idle_wakeup_event(void) } EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); -#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ +#else /* !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK: */ void __init sched_clock_init(void) { @@ -489,7 +495,7 @@ notrace u64 sched_clock_cpu(int cpu) return sched_clock(); } -#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ +#endif /* !CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ /* * Running clock - returns the time that has elapsed while a guest has been diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c index 3561ab533dd4..19ee702273c0 100644 --- a/kernel/sched/completion.c +++ b/kernel/sched/completion.c @@ -13,6 +13,11 @@ * Waiting for completion is a typically sync point, but not an exclusion point. */ +#include <linux/linkage.h> +#include <linux/sched/debug.h> +#include <linux/completion.h> +#include "sched.h" + static void complete_with_flags(struct completion *x, int wake_flags) { unsigned long flags; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 042351c7afce..b8871449d3c6 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7,6 +7,8 @@ * Copyright (C) 1991-2002 Linus Torvalds * Copyright (C) 1998-2024 Ingo Molnar, Red Hat */ +#define INSTANTIATE_EXPORTED_MIGRATE_DISABLE +#include <linux/sched.h> #include <linux/highmem.h> #include <linux/hrtimer_api.h> #include <linux/ktime_api.h> @@ -66,10 +68,11 @@ #include <linux/vtime.h> #include <linux/wait_api.h> #include <linux/workqueue_api.h> +#include <linux/livepatch_sched.h> #ifdef CONFIG_PREEMPT_DYNAMIC -# ifdef CONFIG_GENERIC_ENTRY -# include <linux/entry-common.h> +# ifdef CONFIG_GENERIC_IRQ_ENTRY +# include <linux/irq-entry-common.h> # endif #endif @@ -91,11 +94,11 @@ #include "autogroup.h" #include "pelt.h" #include "smp.h" -#include "stats.h" #include "../workqueue_internal.h" #include "../../io_uring/io-wq.h" #include "../smpboot.h" +#include "../locking/mutex.h" EXPORT_TRACEPOINT_SYMBOL_GPL(ipi_send_cpu); EXPORT_TRACEPOINT_SYMBOL_GPL(ipi_send_cpumask); @@ -116,10 +119,47 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(sched_util_est_cfs_tp); EXPORT_TRACEPOINT_SYMBOL_GPL(sched_util_est_se_tp); EXPORT_TRACEPOINT_SYMBOL_GPL(sched_update_nr_running_tp); EXPORT_TRACEPOINT_SYMBOL_GPL(sched_compute_energy_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_entry_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_exit_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_set_need_resched_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_dl_throttle_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_dl_replenish_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_dl_update_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_dl_server_start_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_dl_server_stop_tp); DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); +DEFINE_PER_CPU(struct rnd_state, sched_rnd_state); + +#ifdef CONFIG_SCHED_PROXY_EXEC +DEFINE_STATIC_KEY_TRUE(__sched_proxy_exec); +static int __init setup_proxy_exec(char *str) +{ + bool proxy_enable = true; + + if (*str && kstrtobool(str + 1, &proxy_enable)) { + pr_warn("Unable to parse sched_proxy_exec=\n"); + return 0; + } + + if (proxy_enable) { + pr_info("sched_proxy_exec enabled via boot arg\n"); + static_branch_enable(&__sched_proxy_exec); + } else { + pr_info("sched_proxy_exec disabled via boot arg\n"); + static_branch_disable(&__sched_proxy_exec); + } + return 1; +} +#else +static int __init setup_proxy_exec(char *str) +{ + pr_warn("CONFIG_SCHED_PROXY_EXEC=n, so it cannot be enabled or disabled at boot time\n"); + return 0; +} +#endif +__setup("sched_proxy_exec", setup_proxy_exec); -#ifdef CONFIG_SCHED_DEBUG /* * Debugging: various feature bits * @@ -129,7 +169,7 @@ DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); */ #define SCHED_FEAT(name, enabled) \ (1UL << __SCHED_FEAT_##name) * enabled | -const_debug unsigned int sysctl_sched_features = +__read_mostly unsigned int sysctl_sched_features = #include "features.h" 0; #undef SCHED_FEAT @@ -143,13 +183,12 @@ const_debug unsigned int sysctl_sched_features = */ __read_mostly int sysctl_resched_latency_warn_ms = 100; __read_mostly int sysctl_resched_latency_warn_once = 1; -#endif /* CONFIG_SCHED_DEBUG */ /* * Number of tasks to iterate in a single balance run. * Limited because this is done with IRQs disabled. */ -const_debug unsigned int sysctl_sched_nr_migrate = SCHED_NR_MIGRATE_BREAK; +__read_mostly unsigned int sysctl_sched_nr_migrate = SCHED_NR_MIGRATE_BREAK; __read_mostly int scheduler_running; @@ -365,6 +404,8 @@ static atomic_t sched_core_count; static struct cpumask sched_core_mask; static void sched_core_lock(int cpu, unsigned long *flags) + __context_unsafe(/* acquires multiple */) + __acquires(&runqueues.__lock) /* overapproximation */ { const struct cpumask *smt_mask = cpu_smt_mask(cpu); int t, i = 0; @@ -375,6 +416,8 @@ static void sched_core_lock(int cpu, unsigned long *flags) } static void sched_core_unlock(int cpu, unsigned long *flags) + __context_unsafe(/* releases multiple */) + __releases(&runqueues.__lock) /* overapproximation */ { const struct cpumask *smt_mask = cpu_smt_mask(cpu); int t; @@ -483,13 +526,23 @@ void sched_core_put(void) schedule_work(&_work); } -#else /* !CONFIG_SCHED_CORE */ +#else /* !CONFIG_SCHED_CORE: */ static inline void sched_core_enqueue(struct rq *rq, struct task_struct *p) { } static inline void sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) { } -#endif /* CONFIG_SCHED_CORE */ +#endif /* !CONFIG_SCHED_CORE */ + +/* need a wrapper since we may need to trace from modules */ +EXPORT_TRACEPOINT_SYMBOL(sched_set_state_tp); + +/* Call via the helper macro trace_set_current_state. */ +void __trace_set_current_state(int state_value) +{ + trace_sched_set_state_tp(current, state_value); +} +EXPORT_SYMBOL(__trace_set_current_state); /* * Serialization rules: @@ -543,8 +596,8 @@ sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) { } * * p->on_rq <- { 0, 1 = TASK_ON_RQ_QUEUED, 2 = TASK_ON_RQ_MIGRATING }: * - * is set by activate_task() and cleared by deactivate_task(), under - * rq->lock. Non-zero indicates the task is runnable, the special + * is set by activate_task() and cleared by deactivate_task()/block_task(), + * under rq->lock. Non-zero indicates the task is runnable, the special * ON_RQ_MIGRATING state is used for migration without holding both * rq->locks. It indicates task_cpu() is not stable, see task_rq_lock(). * @@ -589,6 +642,7 @@ sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) { } */ void raw_spin_rq_lock_nested(struct rq *rq, int subclass) + __context_unsafe() { raw_spinlock_t *lock; @@ -614,6 +668,7 @@ void raw_spin_rq_lock_nested(struct rq *rq, int subclass) } bool raw_spin_rq_trylock(struct rq *rq) + __context_unsafe() { raw_spinlock_t *lock; bool ret; @@ -637,12 +692,6 @@ bool raw_spin_rq_trylock(struct rq *rq) } } -void raw_spin_rq_unlock(struct rq *rq) -{ - raw_spin_unlock(rq_lockp(rq)); -} - -#ifdef CONFIG_SMP /* * double_rq_lock - safely lock two runqueues */ @@ -656,16 +705,16 @@ void double_rq_lock(struct rq *rq1, struct rq *rq2) raw_spin_rq_lock(rq1); if (__rq_lockp(rq1) != __rq_lockp(rq2)) raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING); + else + __acquire_ctx_lock(__rq_lockp(rq2)); /* fake acquire */ double_rq_clock_clear_update(rq1, rq2); } -#endif /* - * __task_rq_lock - lock the rq @p resides on. + * ___task_rq_lock - lock the rq @p resides on. */ -struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) - __acquires(rq->lock) +struct rq *___task_rq_lock(struct task_struct *p, struct rq_flags *rf) { struct rq *rq; @@ -688,9 +737,7 @@ struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) /* * task_rq_lock - lock p->pi_lock and lock the rq @p resides on. */ -struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) - __acquires(p->pi_lock) - __acquires(rq->lock) +struct rq *_task_rq_lock(struct task_struct *p, struct rq_flags *rf) { struct rq *rq; @@ -731,6 +778,11 @@ struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) * RQ-clock updating methods: */ +/* Use CONFIG_PARAVIRT as this will avoid more #ifdef in arch code. */ +#ifdef CONFIG_PARAVIRT +struct static_key paravirt_steal_rq_enabled; +#endif + static void update_rq_clock_task(struct rq *rq, s64 delta) { /* @@ -800,11 +852,10 @@ void update_rq_clock(struct rq *rq) if (rq->clock_update_flags & RQCF_ACT_SKIP) return; -#ifdef CONFIG_SCHED_DEBUG if (sched_feat(WARN_DOUBLE_CLOCK)) - SCHED_WARN_ON(rq->clock_update_flags & RQCF_UPDATED); + WARN_ON_ONCE(rq->clock_update_flags & RQCF_UPDATED); rq->clock_update_flags |= RQCF_UPDATED; -#endif + clock = sched_clock_cpu(cpu_of(rq)); scx_rq_clock_update(rq, clock); @@ -821,7 +872,14 @@ void update_rq_clock(struct rq *rq) * Use HR-timers to deliver accurate preemption points. */ -static void hrtick_clear(struct rq *rq) +enum { + HRTICK_SCHED_NONE = 0, + HRTICK_SCHED_DEFER = BIT(1), + HRTICK_SCHED_START = BIT(2), + HRTICK_SCHED_REARM_HRTIMER = BIT(3) +}; + +static void __used hrtick_clear(struct rq *rq) { if (hrtimer_active(&rq->hrtick_timer)) hrtimer_cancel(&rq->hrtick_timer); @@ -840,20 +898,30 @@ static enum hrtimer_restart hrtick(struct hrtimer *timer) rq_lock(rq, &rf); update_rq_clock(rq); - rq->donor->sched_class->task_tick(rq, rq->curr, 1); + rq->donor->sched_class->task_tick(rq, rq->donor, 1); rq_unlock(rq, &rf); return HRTIMER_NORESTART; } -#ifdef CONFIG_SMP +static inline bool hrtick_needs_rearm(struct hrtimer *timer, ktime_t expires) +{ + /* + * Queued is false when the timer is not started or currently + * running the callback. In both cases, restart. If queued check + * whether the expiry time actually changes substantially. + */ + return !hrtimer_is_queued(timer) || + abs(expires - hrtimer_get_expires(timer)) > 5000; +} -static void __hrtick_restart(struct rq *rq) +static void hrtick_cond_restart(struct rq *rq) { struct hrtimer *timer = &rq->hrtick_timer; ktime_t time = rq->hrtick_time; - hrtimer_start(timer, time, HRTIMER_MODE_ABS_PINNED_HARD); + if (hrtick_needs_rearm(timer, time)) + hrtimer_start(timer, time, HRTIMER_MODE_ABS_PINNED_HARD); } /* @@ -865,7 +933,7 @@ static void __hrtick_start(void *arg) struct rq_flags rf; rq_lock(rq, &rf); - __hrtick_restart(rq); + hrtick_cond_restart(rq); rq_unlock(rq, &rf); } @@ -876,7 +944,6 @@ static void __hrtick_start(void *arg) */ void hrtick_start(struct rq *rq, u64 delay) { - struct hrtimer *timer = &rq->hrtick_timer; s64 delta; /* @@ -884,50 +951,68 @@ void hrtick_start(struct rq *rq, u64 delay) * doesn't make sense and can cause timer DoS. */ delta = max_t(s64, delay, 10000LL); - rq->hrtick_time = ktime_add_ns(timer->base->get_time(), delta); + + /* + * If this is in the middle of schedule() only note the delay + * and let hrtick_schedule_exit() deal with it. + */ + if (rq->hrtick_sched) { + rq->hrtick_sched |= HRTICK_SCHED_START; + rq->hrtick_delay = delta; + return; + } + + rq->hrtick_time = ktime_add_ns(ktime_get(), delta); + if (!hrtick_needs_rearm(&rq->hrtick_timer, rq->hrtick_time)) + return; if (rq == this_rq()) - __hrtick_restart(rq); + hrtimer_start(&rq->hrtick_timer, rq->hrtick_time, HRTIMER_MODE_ABS_PINNED_HARD); else smp_call_function_single_async(cpu_of(rq), &rq->hrtick_csd); } -#else -/* - * Called to set the hrtick timer state. - * - * called with rq->lock held and IRQs disabled - */ -void hrtick_start(struct rq *rq, u64 delay) +static inline void hrtick_schedule_enter(struct rq *rq) { - /* - * Don't schedule slices shorter than 10000ns, that just - * doesn't make sense. Rely on vruntime for fairness. - */ - delay = max_t(u64, delay, 10000LL); - hrtimer_start(&rq->hrtick_timer, ns_to_ktime(delay), - HRTIMER_MODE_REL_PINNED_HARD); + rq->hrtick_sched = HRTICK_SCHED_DEFER; + if (hrtimer_test_and_clear_rearm_deferred()) + rq->hrtick_sched |= HRTICK_SCHED_REARM_HRTIMER; } -#endif /* CONFIG_SMP */ - -static void hrtick_rq_init(struct rq *rq) -{ -#ifdef CONFIG_SMP - INIT_CSD(&rq->hrtick_csd, __hrtick_start, rq); -#endif - hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); - rq->hrtick_timer.function = hrtick; -} -#else /* CONFIG_SCHED_HRTICK */ -static inline void hrtick_clear(struct rq *rq) +static inline void hrtick_schedule_exit(struct rq *rq) { + if (rq->hrtick_sched & HRTICK_SCHED_START) { + rq->hrtick_time = ktime_add_ns(ktime_get(), rq->hrtick_delay); + hrtick_cond_restart(rq); + } else if (idle_rq(rq)) { + /* + * No need for using hrtimer_is_active(). The timer is CPU local + * and interrupts are disabled, so the callback cannot be + * running and the queued state is valid. + */ + if (hrtimer_is_queued(&rq->hrtick_timer)) + hrtimer_cancel(&rq->hrtick_timer); + } + + if (rq->hrtick_sched & HRTICK_SCHED_REARM_HRTIMER) + __hrtimer_rearm_deferred(); + + rq->hrtick_sched = HRTICK_SCHED_NONE; } -static inline void hrtick_rq_init(struct rq *rq) +static void hrtick_rq_init(struct rq *rq) { + INIT_CSD(&rq->hrtick_csd, __hrtick_start, rq); + rq->hrtick_sched = HRTICK_SCHED_NONE; + hrtimer_setup(&rq->hrtick_timer, hrtick, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_HARD | HRTIMER_MODE_LAZY_REARM); } -#endif /* CONFIG_SCHED_HRTICK */ +#else /* !CONFIG_SCHED_HRTICK: */ +static inline void hrtick_clear(struct rq *rq) { } +static inline void hrtick_rq_init(struct rq *rq) { } +static inline void hrtick_schedule_enter(struct rq *rq) { } +static inline void hrtick_schedule_exit(struct rq *rq) { } +#endif /* !CONFIG_SCHED_HRTICK */ /* * try_cmpxchg based fetch_or() macro so it works for different integer types: @@ -943,7 +1028,7 @@ static inline void hrtick_rq_init(struct rq *rq) _val; \ }) -#if defined(CONFIG_SMP) && defined(TIF_POLLING_NRFLAG) +#ifdef TIF_POLLING_NRFLAG /* * Atomically set TIF_NEED_RESCHED and test for TIF_POLLING_NRFLAG, * this avoids any races wrt polling state changes and thereby avoids @@ -982,13 +1067,11 @@ static inline bool set_nr_and_not_polling(struct thread_info *ti, int tif) return true; } -#ifdef CONFIG_SMP static inline bool set_nr_if_polling(struct task_struct *p) { return false; } #endif -#endif static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task) { @@ -1104,6 +1187,7 @@ static void __resched_curr(struct rq *rq, int tif) cpu = cpu_of(rq); + trace_sched_set_need_resched_tp(curr, cpu, tif); if (cpu == smp_processor_id()) { set_ti_thread_flag(cti, tif); if (tif == TIF_NEED_RESCHED) @@ -1119,6 +1203,12 @@ static void __resched_curr(struct rq *rq, int tif) } } +void __trace_set_need_resched(struct task_struct *curr, int tif) +{ + trace_sched_set_need_resched_tp(curr, smp_processor_id(), tif); +} +EXPORT_SYMBOL_GPL(__trace_set_need_resched); + void resched_curr(struct rq *rq) { __resched_curr(rq, TIF_NEED_RESCHED); @@ -1161,7 +1251,6 @@ void resched_cpu(int cpu) raw_spin_rq_unlock_irqrestore(rq, flags); } -#ifdef CONFIG_SMP #ifdef CONFIG_NO_HZ_COMMON /* * In the semi idle case, use the nearest busy CPU for migrating timers @@ -1368,10 +1457,8 @@ bool sched_can_stop_tick(struct rq *rq) return true; } #endif /* CONFIG_NO_HZ_FULL */ -#endif /* CONFIG_SMP */ -#if defined(CONFIG_RT_GROUP_SCHED) || (defined(CONFIG_FAIR_GROUP_SCHED) && \ - (defined(CONFIG_SMP) || defined(CONFIG_CFS_BANDWIDTH))) +#if defined(CONFIG_RT_GROUP_SCHED) || defined(CONFIG_FAIR_GROUP_SCHED) /* * Iterate task_group tree rooted at *from, calling @down when first entering a * node and @up when leaving it for the final time. @@ -1720,7 +1807,7 @@ static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p, bucket = &uc_rq->bucket[uc_se->bucket_id]; - SCHED_WARN_ON(!bucket->tasks); + WARN_ON_ONCE(!bucket->tasks); if (likely(bucket->tasks)) bucket->tasks--; @@ -1740,14 +1827,14 @@ static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p, * Defensive programming: this should never happen. If it happens, * e.g. due to future modification, warn and fix up the expected value. */ - SCHED_WARN_ON(bucket->value > rq_clamp); + WARN_ON_ONCE(bucket->value > rq_clamp); if (bucket->value >= rq_clamp) { bkt_clamp = uclamp_rq_max_value(rq, clamp_id, uc_se->value); uclamp_rq_set(rq, clamp_id, bkt_clamp); } } -static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p) +static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p, int flags) { enum uclamp_id clamp_id; @@ -1757,13 +1844,14 @@ static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p) * The condition is constructed such that a NOP is generated when * sched_uclamp_used is disabled. */ - if (!static_branch_unlikely(&sched_uclamp_used)) + if (!uclamp_is_used()) return; if (unlikely(!p->sched_class->uclamp_enabled)) return; - if (p->se.sched_delayed) + /* Only inc the delayed task which being woken up. */ + if (p->se.sched_delayed && !(flags & ENQUEUE_DELAYED)) return; for_each_clamp_id(clamp_id) @@ -1784,7 +1872,7 @@ static inline void uclamp_rq_dec(struct rq *rq, struct task_struct *p) * The condition is constructed such that a NOP is generated when * sched_uclamp_used is disabled. */ - if (!static_branch_unlikely(&sched_uclamp_used)) + if (!uclamp_is_used()) return; if (unlikely(!p->sched_class->uclamp_enabled)) @@ -1942,12 +2030,12 @@ static int sysctl_sched_uclamp_handler(const struct ctl_table *table, int write, } if (update_root_tg) { - static_branch_enable(&sched_uclamp_used); + sched_uclamp_enable(); uclamp_update_root_tg(); } if (old_min_rt != sysctl_sched_uclamp_util_min_rt_default) { - static_branch_enable(&sched_uclamp_used); + sched_uclamp_enable(); uclamp_sync_util_min_rt_default(); } @@ -1964,7 +2052,7 @@ undo: sysctl_sched_uclamp_util_min_rt_default = old_min_rt; return result; } -#endif +#endif /* CONFIG_SYSCTL */ static void uclamp_fork(struct task_struct *p) { @@ -2030,13 +2118,13 @@ static void __init init_uclamp(void) } } -#else /* !CONFIG_UCLAMP_TASK */ -static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p) { } +#else /* !CONFIG_UCLAMP_TASK: */ +static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p, int flags) { } static inline void uclamp_rq_dec(struct rq *rq, struct task_struct *p) { } static inline void uclamp_fork(struct task_struct *p) { } static inline void uclamp_post_fork(struct task_struct *p) { } static inline void init_uclamp(void) { } -#endif /* CONFIG_UCLAMP_TASK */ +#endif /* !CONFIG_UCLAMP_TASK */ bool sched_task_on_rq(struct task_struct *p) { @@ -2067,12 +2155,14 @@ void enqueue_task(struct rq *rq, struct task_struct *p, int flags) if (!(flags & ENQUEUE_NOCLOCK)) update_rq_clock(rq); - p->sched_class->enqueue_task(rq, p, flags); /* - * Must be after ->enqueue_task() because ENQUEUE_DELAYED can clear - * ->sched_delayed. + * Can be before ->enqueue_task() because uclamp considers the + * ENQUEUE_DELAYED task before its ->sched_delayed gets cleared + * in ->enqueue_task(). */ - uclamp_rq_inc(rq, p); + uclamp_rq_inc(rq, p, flags); + + p->sched_class->enqueue_task(rq, p, flags); psi_enqueue(p, flags); @@ -2111,8 +2201,6 @@ void activate_task(struct rq *rq, struct task_struct *p, int flags) { if (task_on_rq_migrating(p)) flags |= ENQUEUE_MIGRATED; - if (flags & ENQUEUE_MIGRATED) - sched_mm_cid_migrate_to(rq, p); enqueue_task(rq, p, flags); @@ -2122,7 +2210,7 @@ void activate_task(struct rq *rq, struct task_struct *p, int flags) void deactivate_task(struct rq *rq, struct task_struct *p, int flags) { - SCHED_WARN_ON(flags & DEQUEUE_SLEEP); + WARN_ON_ONCE(flags & DEQUEUE_SLEEP); WRITE_ONCE(p->on_rq, TASK_ON_RQ_MIGRATING); ASSERT_EXCLUSIVE_WRITER(p->on_rq); @@ -2152,45 +2240,18 @@ inline int task_curr(const struct task_struct *p) return cpu_curr(task_cpu(p)) == p; } -/* - * ->switching_to() is called with the pi_lock and rq_lock held and must not - * mess with locking. - */ -void check_class_changing(struct rq *rq, struct task_struct *p, - const struct sched_class *prev_class) -{ - if (prev_class != p->sched_class && p->sched_class->switching_to) - p->sched_class->switching_to(rq, p); -} - -/* - * switched_from, switched_to and prio_changed must _NOT_ drop rq->lock, - * use the balance_callback list if you want balancing. - * - * this means any call to check_class_changed() must be followed by a call to - * balance_callback(). - */ -void check_class_changed(struct rq *rq, struct task_struct *p, - const struct sched_class *prev_class, - int oldprio) -{ - if (prev_class != p->sched_class) { - if (prev_class->switched_from) - prev_class->switched_from(rq, p); - - p->sched_class->switched_to(rq, p); - } else if (oldprio != p->prio || dl_task(p)) - p->sched_class->prio_changed(rq, p, oldprio); -} - void wakeup_preempt(struct rq *rq, struct task_struct *p, int flags) { struct task_struct *donor = rq->donor; - if (p->sched_class == donor->sched_class) - donor->sched_class->wakeup_preempt(rq, p, flags); - else if (sched_class_above(p->sched_class, donor->sched_class)) + if (p->sched_class == rq->next_class) { + rq->next_class->wakeup_preempt(rq, p, flags); + + } else if (sched_class_above(p->sched_class, rq->next_class)) { + rq->next_class->wakeup_preempt(rq, p, flags); resched_curr(rq); + rq->next_class = p->sched_class; + } /* * A queue event has occurred, and we're going to schedule. In @@ -2278,6 +2339,12 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state * just go back and repeat. */ rq = task_rq_lock(p, &rf); + /* + * If task is sched_delayed, force dequeue it, to avoid always + * hitting the tick timeout in the queued case + */ + if (p->se.sched_delayed) + dequeue_task(rq, p, DEQUEUE_SLEEP | DEQUEUE_DELAYED); trace_sched_wait_task(p); running = task_on_cpu(rq, p); queued = task_on_rq_queued(p); @@ -2338,10 +2405,8 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state return ncsw; } -#ifdef CONFIG_SMP - static void -__do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx); +do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx); static void migrate_disable_switch(struct rq *rq, struct task_struct *p) { @@ -2356,34 +2421,11 @@ static void migrate_disable_switch(struct rq *rq, struct task_struct *p) if (p->cpus_ptr != &p->cpus_mask) return; - /* - * Violates locking rules! See comment in __do_set_cpus_allowed(). - */ - __do_set_cpus_allowed(p, &ac); + scoped_guard (task_rq_lock, p) + do_set_cpus_allowed(p, &ac); } -void migrate_disable(void) -{ - struct task_struct *p = current; - - if (p->migration_disabled) { -#ifdef CONFIG_DEBUG_PREEMPT - /* - *Warn about overflow half-way through the range. - */ - WARN_ON_ONCE((s16)p->migration_disabled < 0); -#endif - p->migration_disabled++; - return; - } - - guard(preempt)(); - this_rq()->nr_pinned++; - p->migration_disabled = 1; -} -EXPORT_SYMBOL_GPL(migrate_disable); - -void migrate_enable(void) +void ___migrate_enable(void) { struct task_struct *p = current; struct affinity_context ac = { @@ -2391,35 +2433,19 @@ void migrate_enable(void) .flags = SCA_MIGRATE_ENABLE, }; -#ifdef CONFIG_DEBUG_PREEMPT - /* - * Check both overflow from migrate_disable() and superfluous - * migrate_enable(). - */ - if (WARN_ON_ONCE((s16)p->migration_disabled <= 0)) - return; -#endif + __set_cpus_allowed_ptr(p, &ac); +} +EXPORT_SYMBOL_GPL(___migrate_enable); - if (p->migration_disabled > 1) { - p->migration_disabled--; - return; - } +void migrate_disable(void) +{ + __migrate_disable(); +} +EXPORT_SYMBOL_GPL(migrate_disable); - /* - * Ensure stop_task runs either before or after this, and that - * __set_cpus_allowed_ptr(SCA_MIGRATE_ENABLE) doesn't schedule(). - */ - guard(preempt)(); - if (p->cpus_ptr != &p->cpus_mask) - __set_cpus_allowed_ptr(p, &ac); - /* - * Mustn't clear migration_disabled() until cpus_ptr points back at the - * regular cpus_mask, otherwise things that race (eg. - * select_fallback_rq) get confused. - */ - barrier(); - p->migration_disabled = 0; - this_rq()->nr_pinned--; +void migrate_enable(void) +{ + __migrate_enable(); } EXPORT_SYMBOL_GPL(migrate_enable); @@ -2479,6 +2505,7 @@ static inline bool is_cpu_allowed(struct task_struct *p, int cpu) */ static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, struct task_struct *p, int new_cpu) + __must_hold(__rq_lockp(rq)) { lockdep_assert_rq_held(rq); @@ -2525,6 +2552,7 @@ struct set_affinity_pending { */ static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf, struct task_struct *p, int dest_cpu) + __must_hold(__rq_lockp(rq)) { /* Affinity changed (again). */ if (!is_cpu_allowed(p, dest_cpu)) @@ -2561,6 +2589,12 @@ static int migration_cpu_stop(void *data) */ flush_smp_call_function_queue(); + /* + * We may change the underlying rq, but the locks held will + * appropriately be "transferred" when switching. + */ + context_unsafe_alias(rq); + raw_spin_lock(&p->pi_lock); rq_lock(rq, &rf); @@ -2629,7 +2663,8 @@ static int migration_cpu_stop(void *data) */ WARN_ON_ONCE(!pending->stop_pending); preempt_disable(); - task_rq_unlock(rq, p, &rf); + rq_unlock(rq, &rf); + raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); stop_one_cpu_nowait(task_cpu(p), migration_cpu_stop, &pending->arg, &pending->stop_work); preempt_enable(); @@ -2638,7 +2673,8 @@ static int migration_cpu_stop(void *data) out: if (pending) pending->stop_pending = false; - task_rq_unlock(rq, p, &rf); + rq_unlock(rq, &rf); + raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); if (complete) complete_all(&pending->done); @@ -2670,6 +2706,8 @@ int push_cpu_stop(void *arg) if (!lowest_rq) goto out_unlock; + lockdep_assert_rq_held(lowest_rq); + // XXX validate p is still the highest prio task if (task_rq(p) == rq) { move_queued_task_locked(rq, lowest_rq, p); @@ -2687,6 +2725,8 @@ out_unlock: return 0; } +static inline void mm_update_cpus_allowed(struct mm_struct *mm, const cpumask_t *affmask); + /* * sched_class::set_cpus_allowed must do the below, but is not required to * actually call this function. @@ -2700,6 +2740,7 @@ void set_cpus_allowed_common(struct task_struct *p, struct affinity_context *ctx cpumask_copy(&p->cpus_mask, ctx->new_mask); p->nr_cpus_allowed = cpumask_weight(ctx->new_mask); + mm_update_cpus_allowed(p->mm, ctx->new_mask); /* * Swap in a new user_cpus_ptr if SCA_USER flag set @@ -2709,56 +2750,17 @@ void set_cpus_allowed_common(struct task_struct *p, struct affinity_context *ctx } static void -__do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx) +do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx) { - struct rq *rq = task_rq(p); - bool queued, running; - - /* - * This here violates the locking rules for affinity, since we're only - * supposed to change these variables while holding both rq->lock and - * p->pi_lock. - * - * HOWEVER, it magically works, because ttwu() is the only code that - * accesses these variables under p->pi_lock and only does so after - * smp_cond_load_acquire(&p->on_cpu, !VAL), and we're in __schedule() - * before finish_task(). - * - * XXX do further audits, this smells like something putrid. - */ - if (ctx->flags & SCA_MIGRATE_DISABLE) - SCHED_WARN_ON(!p->on_cpu); - else - lockdep_assert_held(&p->pi_lock); - - queued = task_on_rq_queued(p); - running = task_current_donor(rq, p); - - if (queued) { - /* - * Because __kthread_bind() calls this on blocked tasks without - * holding rq->lock. - */ - lockdep_assert_rq_held(rq); - dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK); - } - if (running) - put_prev_task(rq, p); - - p->sched_class->set_cpus_allowed(p, ctx); - mm_set_cpus_allowed(p->mm, ctx->new_mask); - - if (queued) - enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); - if (running) - set_next_task(rq, p); + scoped_guard (sched_change, p, DEQUEUE_SAVE) + p->sched_class->set_cpus_allowed(p, ctx); } /* * Used for kthread_bind() and select_fallback_rq(), in both cases the user * affinity (if any) should be destroyed too. */ -void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +void set_cpus_allowed_force(struct task_struct *p, const struct cpumask *new_mask) { struct affinity_context ac = { .new_mask = new_mask, @@ -2770,7 +2772,8 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) struct rcu_head rcu; }; - __do_set_cpus_allowed(p, &ac); + scoped_guard (__task_rq_lock, p) + do_set_cpus_allowed(p, &ac); /* * Because this is called with p->pi_lock held, it is not possible @@ -2808,7 +2811,7 @@ int dup_user_cpus_ptr(struct task_struct *dst, struct task_struct *src, * Use pi_lock to protect content of user_cpus_ptr * * Though unlikely, user_cpus_ptr can be reset to NULL by a concurrent - * do_set_cpus_allowed(). + * set_cpus_allowed_force(). */ raw_spin_lock_irqsave(&src->pi_lock, flags); if (src->user_cpus_ptr) { @@ -2915,14 +2918,20 @@ void release_user_cpus_ptr(struct task_struct *p) */ static int affine_move_task(struct rq *rq, struct task_struct *p, struct rq_flags *rf, int dest_cpu, unsigned int flags) - __releases(rq->lock) - __releases(p->pi_lock) + __releases(__rq_lockp(rq), &p->pi_lock) { struct set_affinity_pending my_pending = { }, *pending = NULL; bool stop_pending, complete = false; - /* Can the task run on the task's current CPU? If so, we're done */ - if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask)) { + /* + * Can the task run on the task's current CPU? If so, we're done + * + * We are also done if the task is the current donor, boosting a lock- + * holding proxy, (and potentially has been migrated outside its + * current or previous affinity mask) + */ + if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask) || + (task_current_donor(rq, p) && !task_current(rq, p))) { struct task_struct *push_task = NULL; if ((flags & SCA_MIGRATE_ENABLE) && @@ -3064,8 +3073,7 @@ static int __set_cpus_allowed_ptr_locked(struct task_struct *p, struct affinity_context *ctx, struct rq *rq, struct rq_flags *rf) - __releases(rq->lock) - __releases(p->pi_lock) + __releases(__rq_lockp(rq), &p->pi_lock) { const struct cpumask *cpu_allowed_mask = task_cpu_possible_mask(p); const struct cpumask *cpu_valid_mask = cpu_active_mask; @@ -3073,8 +3081,6 @@ static int __set_cpus_allowed_ptr_locked(struct task_struct *p, unsigned int dest_cpu; int ret = 0; - update_rq_clock(rq); - if (kthread || is_migration_disabled(p)) { /* * Kernel threads are allowed on online && !active CPUs, @@ -3129,7 +3135,7 @@ static int __set_cpus_allowed_ptr_locked(struct task_struct *p, goto out; } - __do_set_cpus_allowed(p, ctx); + do_set_cpus_allowed(p, ctx); return affine_move_task(rq, p, rf, dest_cpu, ctx->flags); @@ -3290,9 +3296,10 @@ void relax_compatible_cpus_allowed_ptr(struct task_struct *p) WARN_ON_ONCE(ret); } +#ifdef CONFIG_SMP + void set_task_cpu(struct task_struct *p, unsigned int new_cpu) { -#ifdef CONFIG_SCHED_DEBUG unsigned int state = READ_ONCE(p->__state); /* @@ -3330,7 +3337,6 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) WARN_ON_ONCE(!cpu_online(new_cpu)); WARN_ON_ONCE(is_migration_disabled(p)); -#endif trace_sched_migrate_task(p, new_cpu); @@ -3338,13 +3344,12 @@ void set_task_cpu(struct task_struct *p, unsigned int new_cpu) if (p->sched_class->migrate_task_rq) p->sched_class->migrate_task_rq(p, new_cpu); p->se.nr_migrations++; - rseq_migrate(p); - sched_mm_cid_migrate_from(p); perf_event_task_migrate(p); } __set_task_cpu(p, new_cpu); } +#endif /* CONFIG_SMP */ #ifdef CONFIG_NUMA_BALANCING static void __migrate_swap_task(struct task_struct *p, int cpu) @@ -3537,13 +3542,7 @@ static int select_fallback_rq(int cpu, struct task_struct *p) } fallthrough; case possible: - /* - * XXX When called from select_task_rq() we only - * hold p->pi_lock and again violate locking order. - * - * More yuck to audit. - */ - do_set_cpus_allowed(p, task_cpu_fallback_mask(p)); + set_cpus_allowed_force(p, task_cpu_fallback_mask(p)); state = fail; break; case fail: @@ -3644,17 +3643,6 @@ void sched_set_stop_task(int cpu, struct task_struct *stop) } } -#else /* CONFIG_SMP */ - -static inline void migrate_disable_switch(struct rq *rq, struct task_struct *p) { } - -static inline bool rq_has_pinned_tasks(struct rq *rq) -{ - return false; -} - -#endif /* !CONFIG_SMP */ - static void ttwu_stat(struct task_struct *p, int cpu, int wake_flags) { @@ -3665,7 +3653,6 @@ ttwu_stat(struct task_struct *p, int cpu, int wake_flags) rq = this_rq(); -#ifdef CONFIG_SMP if (cpu == rq->cpu) { __schedstat_inc(rq->ttwu_local); __schedstat_inc(p->stats.nr_wakeups_local); @@ -3685,7 +3672,6 @@ ttwu_stat(struct task_struct *p, int cpu, int wake_flags) if (wake_flags & WF_MIGRATED) __schedstat_inc(p->stats.nr_wakeups_migrate); -#endif /* CONFIG_SMP */ __schedstat_inc(rq->ttwu_count); __schedstat_inc(p->stats.nr_wakeups); @@ -3703,6 +3689,18 @@ static inline void ttwu_do_wakeup(struct task_struct *p) trace_sched_wakeup(p); } +void update_rq_avg_idle(struct rq *rq) +{ + u64 delta = rq_clock(rq) - rq->idle_stamp; + u64 max = 2*rq->max_idle_balance_cost; + + update_avg(&rq->avg_idle, delta); + + if (rq->avg_idle > max) + rq->avg_idle = max; + rq->idle_stamp = 0; +} + static void ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, struct rq_flags *rf) @@ -3714,13 +3712,11 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, if (p->sched_contributes_to_load) rq->nr_uninterruptible--; -#ifdef CONFIG_SMP if (wake_flags & WF_RQ_SELECTED) en_flags |= ENQUEUE_RQ_SELECTED; if (wake_flags & WF_MIGRATED) en_flags |= ENQUEUE_MIGRATED; else -#endif if (p->in_iowait) { delayacct_blkio_end(p); atomic_dec(&task_rq(p)->nr_iowait); @@ -3731,7 +3727,6 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, ttwu_do_wakeup(p); -#ifdef CONFIG_SMP if (p->sched_class->task_woken) { /* * Our task @p is fully woken up and running; so it's safe to @@ -3741,19 +3736,6 @@ ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, p->sched_class->task_woken(rq, p); rq_repin_lock(rq, rf); } - - if (rq->idle_stamp) { - u64 delta = rq_clock(rq) - rq->idle_stamp; - u64 max = 2*rq->max_idle_balance_cost; - - update_avg(&rq->avg_idle, delta); - - if (rq->avg_idle > max) - rq->avg_idle = max; - - rq->idle_stamp = 0; - } -#endif } /* @@ -3802,12 +3784,11 @@ static int ttwu_runnable(struct task_struct *p, int wake_flags) ttwu_do_wakeup(p); ret = 1; } - __task_rq_unlock(rq, &rf); + __task_rq_unlock(rq, p, &rf); return ret; } -#ifdef CONFIG_SMP void sched_ttwu_pending(void *arg) { struct llist_node *llist = arg; @@ -3874,7 +3855,9 @@ static void __ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags p->sched_remote_wakeup = !!(wake_flags & WF_MIGRATED); WRITE_ONCE(rq->ttwu_pending, 1); +#ifdef CONFIG_SMP __smp_call_single_queue(cpu, &p->wake_entry.llist); +#endif } void wake_up_if_idle(int cpu) @@ -3922,14 +3905,16 @@ bool cpus_share_resources(int this_cpu, int that_cpu) static inline bool ttwu_queue_cond(struct task_struct *p, int cpu) { - /* - * The BPF scheduler may depend on select_task_rq() being invoked during - * wakeups. In addition, @p may end up executing on a different CPU - * regardless of what happens in the wakeup path making the ttwu_queue - * optimization less meaningful. Skip if on SCX. - */ - if (task_on_scx(p)) + int this_cpu = smp_processor_id(); + + /* See SCX_OPS_ALLOW_QUEUED_WAKEUP. */ + if (!scx_allow_ttwu_queue(p)) + return false; + +#ifdef CONFIG_SMP + if (p->sched_class == &stop_sched_class) return false; +#endif /* * Do not complicate things with the async wake_list while the CPU is @@ -3946,10 +3931,10 @@ static inline bool ttwu_queue_cond(struct task_struct *p, int cpu) * If the CPU does not share cache, then queue the task on the * remote rqs wakelist to avoid accessing remote data. */ - if (!cpus_share_cache(smp_processor_id(), cpu)) + if (!cpus_share_cache(this_cpu, cpu)) return true; - if (cpu == smp_processor_id()) + if (cpu == this_cpu) return false; /* @@ -3980,15 +3965,6 @@ static bool ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags) return false; } -#else /* !CONFIG_SMP */ - -static inline bool ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags) -{ - return false; -} - -#endif /* CONFIG_SMP */ - static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) { struct rq *rq = cpu_rq(cpu); @@ -4196,7 +4172,7 @@ int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) * - we're serialized against set_special_state() by virtue of * it disabling IRQs (this allows not taking ->pi_lock). */ - SCHED_WARN_ON(p->se.sched_delayed); + WARN_ON_ONCE(p->se.sched_delayed); if (!ttwu_state_match(p, state, &success)) goto out; @@ -4244,7 +4220,6 @@ int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags)) break; -#ifdef CONFIG_SMP /* * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be * possible to, falsely, observe p->on_cpu == 0. @@ -4265,7 +4240,7 @@ int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) * __schedule(). See the comment for smp_mb__after_spinlock(). * * Form a control-dep-acquire with p->on_rq == 0 above, to ensure - * schedule()'s deactivate_task() has 'happened' and p will no longer + * schedule()'s block_task() has 'happened' and p will no longer * care about it's own p->state. See the comment in __schedule(). */ smp_acquire__after_ctrl_dep(); @@ -4323,9 +4298,6 @@ int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) psi_ttwu_dequeue(p); set_task_cpu(p, cpu); } -#else - cpu = task_cpu(p); -#endif /* CONFIG_SMP */ ttwu_queue(p, cpu, wake_flags); } @@ -4358,14 +4330,12 @@ static bool __task_needs_rq_lock(struct task_struct *p) if (p->on_rq) return true; -#ifdef CONFIG_SMP /* * Ensure the task has finished __schedule() and will not be referenced * anymore. Again, see try_to_wake_up() for a longer comment. */ smp_rmb(); smp_cond_load_acquire(&p->on_cpu, !VAL); -#endif return false; } @@ -4387,29 +4357,30 @@ static bool __task_needs_rq_lock(struct task_struct *p) */ int task_call_func(struct task_struct *p, task_call_f func, void *arg) { - struct rq *rq = NULL; struct rq_flags rf; int ret; raw_spin_lock_irqsave(&p->pi_lock, rf.flags); - if (__task_needs_rq_lock(p)) - rq = __task_rq_lock(p, &rf); + if (__task_needs_rq_lock(p)) { + struct rq *rq = __task_rq_lock(p, &rf); - /* - * At this point the task is pinned; either: - * - blocked and we're holding off wakeups (pi->lock) - * - woken, and we're holding off enqueue (rq->lock) - * - queued, and we're holding off schedule (rq->lock) - * - running, and we're holding off de-schedule (rq->lock) - * - * The called function (@func) can use: task_curr(), p->on_rq and - * p->__state to differentiate between these states. - */ - ret = func(p, arg); + /* + * At this point the task is pinned; either: + * - blocked and we're holding off wakeups (pi->lock) + * - woken, and we're holding off enqueue (rq->lock) + * - queued, and we're holding off schedule (rq->lock) + * - running, and we're holding off de-schedule (rq->lock) + * + * The called function (@func) can use: task_curr(), p->on_rq and + * p->__state to differentiate between these states. + */ + ret = func(p, arg); - if (rq) - rq_unlock(rq, &rf); + __task_rq_unlock(rq, p, &rf); + } else { + ret = func(p, arg); + } raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); return ret; @@ -4476,7 +4447,7 @@ int wake_up_state(struct task_struct *p, unsigned int state) * __sched_fork() is basic setup which is also used by sched_init() to * initialize the boot CPU's idle task. */ -static void __sched_fork(unsigned long clone_flags, struct task_struct *p) +static void __sched_fork(u64 clone_flags, struct task_struct *p) { p->on_rq = 0; @@ -4487,13 +4458,17 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p) p->se.nr_migrations = 0; p->se.vruntime = 0; p->se.vlag = 0; + p->se.rel_deadline = 0; INIT_LIST_HEAD(&p->se.group_node); /* A delayed task cannot be in clone(). */ - SCHED_WARN_ON(p->se.sched_delayed); + WARN_ON_ONCE(p->se.sched_delayed); #ifdef CONFIG_FAIR_GROUP_SCHED p->se.cfs_rq = NULL; +#ifdef CONFIG_CFS_BANDWIDTH + init_cfs_throttle_work(p); +#endif #endif #ifdef CONFIG_SCHEDSTATS @@ -4521,11 +4496,8 @@ static void __sched_fork(unsigned long clone_flags, struct task_struct *p) p->capture_control = NULL; #endif init_numa_balancing(clone_flags, p); -#ifdef CONFIG_SMP p->wake_entry.u_flags = CSD_TYPE_TTWU; p->migration_pending = NULL; -#endif - init_sched_mm_cid(p); } DEFINE_STATIC_KEY_FALSE(sched_numa_balancing); @@ -4587,8 +4559,8 @@ static int sysctl_numa_balancing(const struct ctl_table *table, int write, } return err; } -#endif -#endif +#endif /* CONFIG_PROC_SYSCTL */ +#endif /* CONFIG_NUMA_BALANCING */ #ifdef CONFIG_SCHEDSTATS @@ -4713,7 +4685,7 @@ late_initcall(sched_core_sysctl_init); /* * fork()/clone()-time setup: */ -int sched_fork(unsigned long clone_flags, struct task_struct *p) +int sched_fork(u64 clone_flags, struct task_struct *p) { __sched_fork(clone_flags, p); /* @@ -4775,14 +4747,11 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p) if (likely(sched_info_on())) memset(&p->sched_info, 0, sizeof(p->sched_info)); #endif -#if defined(CONFIG_SMP) p->on_cpu = 0; -#endif init_task_preempt_count(p); -#ifdef CONFIG_SMP plist_node_init(&p->pushable_tasks, MAX_PRIO); RB_CLEAR_NODE(&p->pushable_dl_tasks); -#endif + return 0; } @@ -4804,7 +4773,6 @@ int sched_cgroup_fork(struct task_struct *p, struct kernel_clone_args *kargs) p->sched_task_group = tg; } #endif - rseq_migrate(p); /* * We're setting the CPU for the first time, we don't migrate, * so use __set_task_cpu(). @@ -4814,7 +4782,7 @@ int sched_cgroup_fork(struct task_struct *p, struct kernel_clone_args *kargs) p->sched_class->task_fork(p); raw_spin_unlock_irqrestore(&p->pi_lock, flags); - return scx_fork(p); + return scx_fork(p, kargs); } void sched_cancel_fork(struct task_struct *p) @@ -4822,13 +4790,16 @@ void sched_cancel_fork(struct task_struct *p) scx_cancel_fork(p); } +static void sched_mm_cid_fork(struct task_struct *t); + void sched_post_fork(struct task_struct *p) { + sched_mm_cid_fork(p); uclamp_post_fork(p); scx_post_fork(p); } -unsigned long to_ratio(u64 period, u64 runtime) +u64 to_ratio(u64 period, u64 runtime) { if (runtime == RUNTIME_INF) return BW_UNIT; @@ -4859,7 +4830,6 @@ void wake_up_new_task(struct task_struct *p) raw_spin_lock_irqsave(&p->pi_lock, rf.flags); WRITE_ONCE(p->__state, TASK_RUNNING); -#ifdef CONFIG_SMP /* * Fork balancing, do it here and not earlier because: * - cpus_ptr can change in the fork path @@ -4869,9 +4839,7 @@ void wake_up_new_task(struct task_struct *p) * as we're not fully set-up yet. */ p->recent_used_cpu = task_cpu(p); - rseq_migrate(p); __set_task_cpu(p, select_task_rq(p, task_cpu(p), &wake_flags)); -#endif rq = __task_rq_lock(p, &rf); update_rq_clock(rq); post_init_entity_util_avg(p); @@ -4879,7 +4847,6 @@ void wake_up_new_task(struct task_struct *p) activate_task(rq, p, ENQUEUE_NOCLOCK | ENQUEUE_INITIAL); trace_sched_wakeup_new(p); wakeup_preempt(rq, p, wake_flags); -#ifdef CONFIG_SMP if (p->sched_class->task_woken) { /* * Nothing relies on rq->lock after this, so it's fine to @@ -4889,7 +4856,6 @@ void wake_up_new_task(struct task_struct *p) p->sched_class->task_woken(rq, p); rq_repin_lock(rq, &rf); } -#endif task_rq_unlock(rq, p, &rf); } @@ -4966,7 +4932,7 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr, __fire_sched_out_preempt_notifiers(curr, next); } -#else /* !CONFIG_PREEMPT_NOTIFIERS */ +#else /* !CONFIG_PREEMPT_NOTIFIERS: */ static inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) { @@ -4978,11 +4944,10 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr, { } -#endif /* CONFIG_PREEMPT_NOTIFIERS */ +#endif /* !CONFIG_PREEMPT_NOTIFIERS */ static inline void prepare_task(struct task_struct *next) { -#ifdef CONFIG_SMP /* * Claim the task as running, we do this before switching to it * such that any running task will have this set. @@ -4991,12 +4956,10 @@ static inline void prepare_task(struct task_struct *next) * its ordering comment. */ WRITE_ONCE(next->on_cpu, 1); -#endif } static inline void finish_task(struct task_struct *prev) { -#ifdef CONFIG_SMP /* * This must be the very last reference to @prev from this CPU. After * p->on_cpu is cleared, the task can be moved to a different CPU. We @@ -5009,10 +4972,35 @@ static inline void finish_task(struct task_struct *prev) * Pairs with the smp_cond_load_acquire() in try_to_wake_up(). */ smp_store_release(&prev->on_cpu, 0); -#endif } -#ifdef CONFIG_SMP +/* + * Only called from __schedule context + * + * There are some cases where we are going to re-do the action + * that added the balance callbacks. We may not be in a state + * where we can run them, so just zap them so they can be + * properly re-added on the next time around. This is similar + * handling to running the callbacks, except we just don't call + * them. + */ +static void zap_balance_callbacks(struct rq *rq) +{ + struct balance_callback *next, *head; + bool found = false; + + lockdep_assert_rq_held(rq); + + head = rq->balance_callback; + while (head) { + if (head == &balance_push_callback) + found = true; + next = head->next; + head->next = NULL; + head = next; + } + rq->balance_callback = found ? &balance_push_callback : NULL; +} static void do_balance_callbacks(struct rq *rq, struct balance_callback *head) { @@ -5079,9 +5067,13 @@ struct balance_callback *splice_balance_callbacks(struct rq *rq) return __splice_balance_callbacks(rq, true); } -static void __balance_callbacks(struct rq *rq) +void __balance_callbacks(struct rq *rq, struct rq_flags *rf) { + if (rf) + rq_unpin_lock(rq, rf); do_balance_callbacks(rq, __splice_balance_callbacks(rq, false)); + if (rf) + rq_repin_lock(rq, rf); } void balance_callbacks(struct rq *rq, struct balance_callback *head) @@ -5095,16 +5087,10 @@ void balance_callbacks(struct rq *rq, struct balance_callback *head) } } -#else - -static inline void __balance_callbacks(struct rq *rq) -{ -} - -#endif - static inline void prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf) + __releases(__rq_lockp(rq)) + __acquires(__rq_lockp(this_rq())) { /* * Since the runqueue lock will be released by the next @@ -5118,9 +5104,15 @@ prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf /* this is a valid case when another task releases the spinlock */ rq_lockp(rq)->owner = next; #endif + /* + * Model the rq reference switcheroo. + */ + __release(__rq_lockp(rq)); + __acquire(__rq_lockp(this_rq())); } static inline void finish_lock_switch(struct rq *rq) + __releases(__rq_lockp(rq)) { /* * If we are tracking spinlock dependencies then we have to @@ -5128,7 +5120,8 @@ static inline void finish_lock_switch(struct rq *rq) * prev into current: */ spin_acquire(&__rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_); - __balance_callbacks(rq); + __balance_callbacks(rq, NULL); + hrtick_schedule_exit(rq); raw_spin_rq_unlock_irq(rq); } @@ -5176,11 +5169,11 @@ static inline void kmap_local_sched_in(void) static inline void prepare_task_switch(struct rq *rq, struct task_struct *prev, struct task_struct *next) + __must_hold(__rq_lockp(rq)) { kcov_prepare_switch(prev); sched_info_switch(rq, prev, next); perf_event_task_sched_out(prev, next); - rseq_preempt(prev); fire_sched_out_preempt_notifiers(prev, next); kmap_local_sched_out(); prepare_task(next); @@ -5207,7 +5200,7 @@ prepare_task_switch(struct rq *rq, struct task_struct *prev, * because prev may have moved to another CPU. */ static struct rq *finish_task_switch(struct task_struct *prev) - __releases(rq->lock) + __releases(__rq_lockp(this_rq())) { struct rq *rq = this_rq(); struct mm_struct *mm = rq->prev_mm; @@ -5281,6 +5274,14 @@ static struct rq *finish_task_switch(struct task_struct *prev) if (prev->sched_class->task_dead) prev->sched_class->task_dead(prev); + /* + * sched_ext_dead() must come before cgroup_task_dead() to + * prevent cgroups from being removed while its member tasks are + * visible to SCX schedulers. + */ + sched_ext_dead(prev); + cgroup_task_dead(prev); + /* Task is done with its stack. */ put_task_stack(prev); @@ -5295,7 +5296,7 @@ static struct rq *finish_task_switch(struct task_struct *prev) * @prev: the thread we just switched away from. */ asmlinkage __visible void schedule_tail(struct task_struct *prev) - __releases(rq->lock) + __releases(__rq_lockp(this_rq())) { /* * New tasks start with FORK_PREEMPT_COUNT, see there and @@ -5307,6 +5308,12 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev) */ finish_task_switch(prev); + /* + * This is a special case: the newly created task has just + * switched the context for the first time. It is returning from + * schedule for the first time in this path. + */ + trace_sched_exit_tp(true); preempt_enable(); if (current->set_child_tid) @@ -5321,6 +5328,7 @@ asmlinkage __visible void schedule_tail(struct task_struct *prev) static __always_inline struct rq * context_switch(struct rq *rq, struct task_struct *prev, struct task_struct *next, struct rq_flags *rf) + __releases(__rq_lockp(rq)) { prepare_task_switch(rq, prev, next); @@ -5337,19 +5345,16 @@ context_switch(struct rq *rq, struct task_struct *prev, * * kernel -> user switch + mmdrop_lazy_tlb() active * user -> user switch - * - * switch_mm_cid() needs to be updated if the barriers provided - * by context_switch() are modified. */ - if (!next->mm) { // to kernel + if (!next->mm) { // to kernel enter_lazy_tlb(prev->active_mm, next); next->active_mm = prev->active_mm; - if (prev->mm) // from user + if (prev->mm) // from user mmgrab_lazy_tlb(prev->active_mm); else prev->active_mm = NULL; - } else { // to user + } else { // to user membarrier_switch_mm(rq, prev->active_mm, next->mm); /* * sys_membarrier() requires an smp_mb() between setting @@ -5362,15 +5367,20 @@ context_switch(struct rq *rq, struct task_struct *prev, switch_mm_irqs_off(prev->active_mm, next->mm, next); lru_gen_use_mm(next->mm); - if (!prev->mm) { // from kernel + if (!prev->mm) { // from kernel /* will mmdrop_lazy_tlb() in finish_task_switch(). */ rq->prev_mm = prev->active_mm; prev->active_mm = NULL; } } - /* switch_mm_cid() requires the memory barriers above. */ - switch_mm_cid(rq, prev, next); + mm_cid_switch_to(prev, next); + + /* + * Tell rseq that the task was scheduled in. Must be after + * switch_mm_cid() to get the TIF flag set. + */ + rseq_sched_switch_event(next); prepare_lock_switch(rq, next, rf); @@ -5484,8 +5494,6 @@ unsigned int nr_iowait(void) return sum; } -#ifdef CONFIG_SMP - /* * sched_exec - execve() is a valuable balancing opportunity, because at * this point the task has the smallest effective memory and cache footprint. @@ -5509,8 +5517,6 @@ void sched_exec(void) stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg); } -#endif - DEFINE_PER_CPU(struct kernel_stat, kstat); DEFINE_PER_CPU(struct kernel_cpustat, kernel_cpustat); @@ -5545,7 +5551,7 @@ unsigned long long task_sched_runtime(struct task_struct *p) struct rq *rq; u64 ns; -#if defined(CONFIG_64BIT) && defined(CONFIG_SMP) +#ifdef CONFIG_64BIT /* * 64-bit doesn't need locks to atomically read a 64-bit value. * So we have a optimization chance when the task's delta_exec is 0. @@ -5578,7 +5584,6 @@ unsigned long long task_sched_runtime(struct task_struct *p) return ns; } -#ifdef CONFIG_SCHED_DEBUG static u64 cpu_resched_latency(struct rq *rq) { int latency_warn_ms = READ_ONCE(sysctl_resched_latency_warn_ms); @@ -5623,9 +5628,6 @@ static int __init setup_resched_latency_warn_ms(char *str) return 1; } __setup("resched_latency_warn_ms=", setup_resched_latency_warn_ms); -#else -static inline u64 cpu_resched_latency(struct rq *rq) { return 0; } -#endif /* CONFIG_SCHED_DEBUG */ /* * This function gets called by the timer code, with HZ frequency. @@ -5663,7 +5665,6 @@ void sched_tick(void) resched_latency = cpu_resched_latency(rq); calc_global_load_tick(rq); sched_core_tick(rq); - task_tick_mm_cid(rq, donor); scx_tick(rq); rq_unlock(rq, &rf); @@ -5676,12 +5677,10 @@ void sched_tick(void) if (donor->flags & PF_WQ_WORKER) wq_worker_tick(donor); -#ifdef CONFIG_SMP if (!scx_switched_all()) { rq->idle_balance = idle_cpu(cpu); sched_balance_trigger(rq); } -#endif } #ifdef CONFIG_NO_HZ_FULL @@ -5746,7 +5745,7 @@ static void sched_tick_remote(struct work_struct *work) * we are always sure that there is no proxy (only a * single task is running). */ - SCHED_WARN_ON(rq->curr != rq->donor); + WARN_ON_ONCE(rq->curr != rq->donor); update_rq_clock(rq); if (!is_idle_task(curr)) { @@ -5755,7 +5754,7 @@ static void sched_tick_remote(struct work_struct *work) * reasonable amount of time. */ u64 delta = rq_clock_task(rq) - curr->se.exec_start; - WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 3); + WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 30); } curr->sched_class->task_tick(rq, curr, 0); @@ -5772,7 +5771,7 @@ static void sched_tick_remote(struct work_struct *work) os = atomic_fetch_add_unless(&twork->state, -1, TICK_SCHED_REMOTE_RUNNING); WARN_ON_ONCE(os == TICK_SCHED_REMOTE_OFFLINE); if (os == TICK_SCHED_REMOTE_RUNNING) - queue_delayed_work(system_unbound_wq, dwork, HZ); + queue_delayed_work(system_dfl_wq, dwork, HZ); } static void sched_tick_start(int cpu) @@ -5791,7 +5790,7 @@ static void sched_tick_start(int cpu) if (os == TICK_SCHED_REMOTE_OFFLINE) { twork->cpu = cpu; INIT_DELAYED_WORK(&twork->work, sched_tick_remote); - queue_delayed_work(system_unbound_wq, &twork->work, HZ); + queue_delayed_work(system_dfl_wq, &twork->work, HZ); } } @@ -5821,10 +5820,10 @@ int __init sched_tick_offload_init(void) return 0; } -#else /* !CONFIG_NO_HZ_FULL */ +#else /* !CONFIG_NO_HZ_FULL: */ static inline void sched_tick_start(int cpu) { } static inline void sched_tick_stop(int cpu) { } -#endif +#endif /* !CONFIG_NO_HZ_FULL */ #if defined(CONFIG_PREEMPTION) && (defined(CONFIG_DEBUG_PREEMPT) || \ defined(CONFIG_TRACE_PREEMPT_TOGGLE)) @@ -5966,7 +5965,7 @@ static inline void schedule_debug(struct task_struct *prev, bool preempt) preempt_count_set(PREEMPT_DISABLED); } rcu_sleep_check(); - SCHED_WARN_ON(ct_state() == CT_STATE_USER); + WARN_ON_ONCE(ct_state() == CT_STATE_USER); profile_hit(SCHED_PROFILING, __builtin_return_address(0)); @@ -5979,19 +5978,6 @@ static void prev_balance(struct rq *rq, struct task_struct *prev, const struct sched_class *start_class = prev->sched_class; const struct sched_class *class; -#ifdef CONFIG_SCHED_CLASS_EXT - /* - * SCX requires a balance() call before every pick_task() including when - * waking up from SCHED_IDLE. If @start_class is below SCX, start from - * SCX instead. Also, set a flag to detect missing balance() call. - */ - if (scx_enabled()) { - rq->scx.flags |= SCX_RQ_BAL_PENDING; - if (sched_class_above(&ext_sched_class, start_class)) - start_class = &ext_sched_class; - } -#endif - /* * We must do the balancing pass before put_prev_task(), such * that when we release the rq->lock the task is in the same @@ -6011,6 +5997,7 @@ static void prev_balance(struct rq *rq, struct task_struct *prev, */ static inline struct task_struct * __pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) + __must_hold(__rq_lockp(rq)) { const struct sched_class *class; struct task_struct *p; @@ -6035,7 +6022,7 @@ __pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) /* Assume the next prioritized class is idle_sched_class */ if (!p) { - p = pick_task_idle(rq); + p = pick_task_idle(rq, rf); put_prev_set_next_task(rq, prev, p); } @@ -6047,11 +6034,15 @@ restart: for_each_active_class(class) { if (class->pick_next_task) { - p = class->pick_next_task(rq, prev); + p = class->pick_next_task(rq, prev, rf); + if (unlikely(p == RETRY_TASK)) + goto restart; if (p) return p; } else { - p = class->pick_task(rq); + p = class->pick_task(rq, rf); + if (unlikely(p == RETRY_TASK)) + goto restart; if (p) { put_prev_set_next_task(rq, prev, p); return p; @@ -6081,7 +6072,11 @@ static inline bool cookie_match(struct task_struct *a, struct task_struct *b) return a->core_cookie == b->core_cookie; } -static inline struct task_struct *pick_task(struct rq *rq) +/* + * Careful; this can return RETRY_TASK, it does not include the retry-loop + * itself due to the whole SMT pick retry thing below. + */ +static inline struct task_struct *pick_task(struct rq *rq, struct rq_flags *rf) { const struct sched_class *class; struct task_struct *p; @@ -6089,7 +6084,7 @@ static inline struct task_struct *pick_task(struct rq *rq) rq->dl_server = NULL; for_each_active_class(class) { - p = class->pick_task(rq); + p = class->pick_task(rq, rf); if (p) return p; } @@ -6103,8 +6098,9 @@ static void queue_core_balance(struct rq *rq); static struct task_struct * pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) + __must_hold(__rq_lockp(rq)) { - struct task_struct *next, *p, *max = NULL; + struct task_struct *next, *p, *max; const struct cpumask *smt_mask; bool fi_before = false; bool core_clock_updated = (rq == rq->core); @@ -6189,7 +6185,10 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) * and there are no cookied tasks running on siblings. */ if (!need_sync) { - next = pick_task(rq); +restart_single: + next = pick_task(rq, rf); + if (unlikely(next == RETRY_TASK)) + goto restart_single; if (!next->core_cookie) { rq->core_pick = NULL; rq->core_dl_server = NULL; @@ -6209,6 +6208,8 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) * * Tie-break prio towards the current CPU */ +restart_multi: + max = NULL; for_each_cpu_wrap(i, smt_mask, cpu) { rq_i = cpu_rq(i); @@ -6220,7 +6221,11 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) if (i != cpu && (rq_i != rq->core || !core_clock_updated)) update_rq_clock(rq_i); - rq_i->core_pick = p = pick_task(rq_i); + p = pick_task(rq_i, rf); + if (unlikely(p == RETRY_TASK)) + goto restart_multi; + + rq_i->core_pick = p; rq_i->core_dl_server = rq_i->dl_server; if (!max || prio_less(max, p, fi_before)) @@ -6242,7 +6247,7 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) if (cookie) p = sched_core_find(rq_i, cookie); if (!p) - p = idle_sched_class.pick_task(rq_i); + p = idle_sched_class.pick_task(rq_i, rf); } rq_i->core_pick = p; @@ -6402,6 +6407,7 @@ static bool steal_cookie_task(int cpu, struct sched_domain *sd) } static void sched_core_balance(struct rq *rq) + __must_hold(__rq_lockp(rq)) { struct sched_domain *sd; int cpu = cpu_of(rq); @@ -6539,7 +6545,7 @@ static inline void sched_core_cpu_dying(unsigned int cpu) rq->core = rq; } -#else /* !CONFIG_SCHED_CORE */ +#else /* !CONFIG_SCHED_CORE: */ static inline void sched_core_cpu_starting(unsigned int cpu) {} static inline void sched_core_cpu_deactivate(unsigned int cpu) {} @@ -6547,11 +6553,12 @@ static inline void sched_core_cpu_dying(unsigned int cpu) {} static struct task_struct * pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) + __must_hold(__rq_lockp(rq)) { return __pick_next_task(rq, prev, rf); } -#endif /* CONFIG_SCHED_CORE */ +#endif /* !CONFIG_SCHED_CORE */ /* * Constants for the sched_mode argument of __schedule(). @@ -6567,19 +6574,35 @@ pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) /* * Helper function for __schedule() * - * If a task does not have signals pending, deactivate it - * Otherwise marks the task's __state as RUNNING + * Tries to deactivate the task, unless the should_block arg + * is false or if a signal is pending. In the case a signal + * is pending, marks the task's __state as RUNNING (and clear + * blocked_on). */ static bool try_to_block_task(struct rq *rq, struct task_struct *p, - unsigned long task_state) + unsigned long *task_state_p, bool should_block) { + unsigned long task_state = *task_state_p; int flags = DEQUEUE_NOCLOCK; if (signal_pending_state(task_state, p)) { WRITE_ONCE(p->__state, TASK_RUNNING); + *task_state_p = TASK_RUNNING; + set_task_blocked_on_waking(p, NULL); + return false; } + /* + * We check should_block after signal_pending because we + * will want to wake the task in that case. But if + * should_block is false, its likely due to the task being + * blocked on a mutex, and we want to keep it on the runqueue + * to be selectable for proxy-execution. + */ + if (!should_block) + return false; + p->sched_contributes_to_load = (task_state & TASK_UNINTERRUPTIBLE) && !(task_state & TASK_NOLOAD) && @@ -6603,6 +6626,355 @@ static bool try_to_block_task(struct rq *rq, struct task_struct *p, return true; } +#ifdef CONFIG_SCHED_PROXY_EXEC +static inline void proxy_set_task_cpu(struct task_struct *p, int cpu) +{ + unsigned int wake_cpu; + + /* + * Since we are enqueuing a blocked task on a cpu it may + * not be able to run on, preserve wake_cpu when we + * __set_task_cpu so we can return the task to where it + * was previously runnable. + */ + wake_cpu = p->wake_cpu; + __set_task_cpu(p, cpu); + p->wake_cpu = wake_cpu; +} + +static inline struct task_struct *proxy_resched_idle(struct rq *rq) +{ + put_prev_set_next_task(rq, rq->donor, rq->idle); + rq_set_donor(rq, rq->idle); + set_tsk_need_resched(rq->idle); + return rq->idle; +} + +static bool proxy_deactivate(struct rq *rq, struct task_struct *donor) +{ + unsigned long state = READ_ONCE(donor->__state); + + /* Don't deactivate if the state has been changed to TASK_RUNNING */ + if (state == TASK_RUNNING) + return false; + /* + * Because we got donor from pick_next_task(), it is *crucial* + * that we call proxy_resched_idle() before we deactivate it. + * As once we deactivate donor, donor->on_rq is set to zero, + * which allows ttwu() to immediately try to wake the task on + * another rq. So we cannot use *any* references to donor + * after that point. So things like cfs_rq->curr or rq->donor + * need to be changed from next *before* we deactivate. + */ + proxy_resched_idle(rq); + return try_to_block_task(rq, donor, &state, true); +} + +static inline void proxy_release_rq_lock(struct rq *rq, struct rq_flags *rf) + __releases(__rq_lockp(rq)) +{ + /* + * The class scheduler may have queued a balance callback + * from pick_next_task() called earlier. + * + * So here we have to zap callbacks before unlocking the rq + * as another CPU may jump in and call sched_balance_rq + * which can trip the warning in rq_pin_lock() if we + * leave callbacks set. + * + * After we later reaquire the rq lock, we will force __schedule() + * to pick_again, so the callbacks will get re-established. + */ + zap_balance_callbacks(rq); + rq_unpin_lock(rq, rf); + raw_spin_rq_unlock(rq); +} + +static inline void proxy_reacquire_rq_lock(struct rq *rq, struct rq_flags *rf) + __acquires(__rq_lockp(rq)) +{ + raw_spin_rq_lock(rq); + rq_repin_lock(rq, rf); + update_rq_clock(rq); +} + +/* + * If the blocked-on relationship crosses CPUs, migrate @p to the + * owner's CPU. + * + * This is because we must respect the CPU affinity of execution + * contexts (owner) but we can ignore affinity for scheduling + * contexts (@p). So we have to move scheduling contexts towards + * potential execution contexts. + * + * Note: The owner can disappear, but simply migrate to @target_cpu + * and leave that CPU to sort things out. + */ +static void proxy_migrate_task(struct rq *rq, struct rq_flags *rf, + struct task_struct *p, int target_cpu) + __must_hold(__rq_lockp(rq)) +{ + struct rq *target_rq = cpu_rq(target_cpu); + + lockdep_assert_rq_held(rq); + WARN_ON(p == rq->curr); + /* + * Since we are migrating a blocked donor, it could be rq->donor, + * and we want to make sure there aren't any references from this + * rq to it before we drop the lock. This avoids another cpu + * jumping in and grabbing the rq lock and referencing rq->donor + * or cfs_rq->curr, etc after we have migrated it to another cpu, + * and before we pick_again in __schedule. + * + * So call proxy_resched_idle() to drop the rq->donor references + * before we release the lock. + */ + proxy_resched_idle(rq); + + deactivate_task(rq, p, DEQUEUE_NOCLOCK); + proxy_set_task_cpu(p, target_cpu); + + proxy_release_rq_lock(rq, rf); + + attach_one_task(target_rq, p); + + proxy_reacquire_rq_lock(rq, rf); +} + +static void proxy_force_return(struct rq *rq, struct rq_flags *rf, + struct task_struct *p) + __must_hold(__rq_lockp(rq)) +{ + struct rq *task_rq, *target_rq = NULL; + int cpu, wake_flag = WF_TTWU; + + lockdep_assert_rq_held(rq); + WARN_ON(p == rq->curr); + + if (p == rq->donor) + proxy_resched_idle(rq); + + proxy_release_rq_lock(rq, rf); + /* + * We drop the rq lock, and re-grab task_rq_lock to get + * the pi_lock (needed for select_task_rq) as well. + */ + scoped_guard (task_rq_lock, p) { + task_rq = scope.rq; + + /* + * Since we let go of the rq lock, the task may have been + * woken or migrated to another rq before we got the + * task_rq_lock. So re-check we're on the same RQ. If + * not, the task has already been migrated and that CPU + * will handle any futher migrations. + */ + if (task_rq != rq) + break; + + /* + * Similarly, if we've been dequeued, someone else will + * wake us + */ + if (!task_on_rq_queued(p)) + break; + + /* + * Since we should only be calling here from __schedule() + * -> find_proxy_task(), no one else should have + * assigned current out from under us. But check and warn + * if we see this, then bail. + */ + if (task_current(task_rq, p) || task_on_cpu(task_rq, p)) { + WARN_ONCE(1, "%s rq: %i current/on_cpu task %s %d on_cpu: %i\n", + __func__, cpu_of(task_rq), + p->comm, p->pid, p->on_cpu); + break; + } + + update_rq_clock(task_rq); + deactivate_task(task_rq, p, DEQUEUE_NOCLOCK); + cpu = select_task_rq(p, p->wake_cpu, &wake_flag); + set_task_cpu(p, cpu); + target_rq = cpu_rq(cpu); + clear_task_blocked_on(p, NULL); + } + + if (target_rq) + attach_one_task(target_rq, p); + + proxy_reacquire_rq_lock(rq, rf); +} + +/* + * Find runnable lock owner to proxy for mutex blocked donor + * + * Follow the blocked-on relation: + * task->blocked_on -> mutex->owner -> task... + * + * Lock order: + * + * p->pi_lock + * rq->lock + * mutex->wait_lock + * p->blocked_lock + * + * Returns the task that is going to be used as execution context (the one + * that is actually going to be run on cpu_of(rq)). + */ +static struct task_struct * +find_proxy_task(struct rq *rq, struct task_struct *donor, struct rq_flags *rf) + __must_hold(__rq_lockp(rq)) +{ + struct task_struct *owner = NULL; + bool curr_in_chain = false; + int this_cpu = cpu_of(rq); + struct task_struct *p; + struct mutex *mutex; + int owner_cpu; + + /* Follow blocked_on chain. */ + for (p = donor; (mutex = p->blocked_on); p = owner) { + /* if its PROXY_WAKING, do return migration or run if current */ + if (mutex == PROXY_WAKING) { + if (task_current(rq, p)) { + clear_task_blocked_on(p, PROXY_WAKING); + return p; + } + goto force_return; + } + + /* + * By taking mutex->wait_lock we hold off concurrent mutex_unlock() + * and ensure @owner sticks around. + */ + guard(raw_spinlock)(&mutex->wait_lock); + guard(raw_spinlock)(&p->blocked_lock); + + /* Check again that p is blocked with blocked_lock held */ + if (mutex != __get_task_blocked_on(p)) { + /* + * Something changed in the blocked_on chain and + * we don't know if only at this level. So, let's + * just bail out completely and let __schedule() + * figure things out (pick_again loop). + */ + return NULL; + } + + if (task_current(rq, p)) + curr_in_chain = true; + + owner = __mutex_owner(mutex); + if (!owner) { + /* + * If there is no owner, either clear blocked_on + * and return p (if it is current and safe to + * just run on this rq), or return-migrate the task. + */ + if (task_current(rq, p)) { + __clear_task_blocked_on(p, NULL); + return p; + } + goto force_return; + } + + if (!READ_ONCE(owner->on_rq) || owner->se.sched_delayed) { + /* XXX Don't handle blocked owners/delayed dequeue yet */ + if (curr_in_chain) + return proxy_resched_idle(rq); + goto deactivate; + } + + owner_cpu = task_cpu(owner); + if (owner_cpu != this_cpu) { + /* + * @owner can disappear, simply migrate to @owner_cpu + * and leave that CPU to sort things out. + */ + if (curr_in_chain) + return proxy_resched_idle(rq); + goto migrate_task; + } + + if (task_on_rq_migrating(owner)) { + /* + * One of the chain of mutex owners is currently migrating to this + * CPU, but has not yet been enqueued because we are holding the + * rq lock. As a simple solution, just schedule rq->idle to give + * the migration a chance to complete. Much like the migrate_task + * case we should end up back in find_proxy_task(), this time + * hopefully with all relevant tasks already enqueued. + */ + return proxy_resched_idle(rq); + } + + /* + * Its possible to race where after we check owner->on_rq + * but before we check (owner_cpu != this_cpu) that the + * task on another cpu was migrated back to this cpu. In + * that case it could slip by our checks. So double check + * we are still on this cpu and not migrating. If we get + * inconsistent results, try again. + */ + if (!task_on_rq_queued(owner) || task_cpu(owner) != this_cpu) + return NULL; + + if (owner == p) { + /* + * It's possible we interleave with mutex_unlock like: + * + * lock(&rq->lock); + * find_proxy_task() + * mutex_unlock() + * lock(&wait_lock); + * donor(owner) = current->blocked_donor; + * unlock(&wait_lock); + * + * wake_up_q(); + * ... + * ttwu_runnable() + * __task_rq_lock() + * lock(&wait_lock); + * owner == p + * + * Which leaves us to finish the ttwu_runnable() and make it go. + * + * So schedule rq->idle so that ttwu_runnable() can get the rq + * lock and mark owner as running. + */ + return proxy_resched_idle(rq); + } + /* + * OK, now we're absolutely sure @owner is on this + * rq, therefore holding @rq->lock is sufficient to + * guarantee its existence, as per ttwu_remote(). + */ + } + WARN_ON_ONCE(owner && !owner->on_rq); + return owner; + +deactivate: + if (proxy_deactivate(rq, donor)) + return NULL; + /* If deactivate fails, force return */ + p = donor; +force_return: + proxy_force_return(rq, rf, p); + return NULL; +migrate_task: + proxy_migrate_task(rq, rf, p, owner_cpu); + return NULL; +} +#else /* SCHED_PROXY_EXEC */ +static struct task_struct * +find_proxy_task(struct rq *rq, struct task_struct *donor, struct rq_flags *rf) +{ + WARN_ONCE(1, "This should never be called in the !SCHED_PROXY_EXEC case\n"); + return donor; +} +#endif /* SCHED_PROXY_EXEC */ + /* * __schedule() is the main scheduler function. * @@ -6650,23 +7022,27 @@ static void __sched notrace __schedule(int sched_mode) * as a preemption by schedule_debug() and RCU. */ bool preempt = sched_mode > SM_NONE; + bool is_switch = false; unsigned long *switch_count; unsigned long prev_state; struct rq_flags rf; struct rq *rq; int cpu; + /* Trace preemptions consistently with task switches */ + trace_sched_entry_tp(sched_mode == SM_PREEMPT); + cpu = smp_processor_id(); rq = cpu_rq(cpu); prev = rq->curr; schedule_debug(prev, preempt); - if (sched_feat(HRTICK) || sched_feat(HRTICK_DL)) - hrtick_clear(rq); + klp_sched_try_switch(prev); local_irq_disable(); rcu_note_context_switch(preempt); + migrate_disable_switch(rq, prev); /* * Make sure that signal_pending_state()->signal_pending() below @@ -6688,6 +7064,8 @@ static void __sched notrace __schedule(int sched_mode) rq_lock(rq, &rf); smp_mb__after_spinlock(); + hrtick_schedule_enter(rq); + /* Promote REQ to ACT */ rq->clock_update_flags <<= 1; update_rq_clock(rq); @@ -6707,29 +7085,76 @@ static void __sched notrace __schedule(int sched_mode) /* SCX must consult the BPF scheduler to tell if rq is empty */ if (!rq->nr_running && !scx_enabled()) { next = prev; + rq->next_class = &idle_sched_class; goto picked; } } else if (!preempt && prev_state) { - try_to_block_task(rq, prev, prev_state); + /* + * We pass task_is_blocked() as the should_block arg + * in order to keep mutex-blocked tasks on the runqueue + * for slection with proxy-exec (without proxy-exec + * task_is_blocked() will always be false). + */ + try_to_block_task(rq, prev, &prev_state, + !task_is_blocked(prev)); switch_count = &prev->nvcsw; } - next = pick_next_task(rq, prev, &rf); - rq_set_donor(rq, next); +pick_again: + assert_balance_callbacks_empty(rq); + next = pick_next_task(rq, rq->donor, &rf); + rq->next_class = next->sched_class; + if (sched_proxy_exec()) { + struct task_struct *prev_donor = rq->donor; + + rq_set_donor(rq, next); + if (unlikely(next->blocked_on)) { + next = find_proxy_task(rq, next, &rf); + if (!next) { + zap_balance_callbacks(rq); + goto pick_again; + } + if (next == rq->idle) { + zap_balance_callbacks(rq); + goto keep_resched; + } + } + if (rq->donor == prev_donor && prev != next) { + struct task_struct *donor = rq->donor; + /* + * When transitioning like: + * + * prev next + * donor: B B + * curr: A B or C + * + * then put_prev_set_next_task() will not have done + * anything, since B == B. However, A might have + * missed a RT/DL balance opportunity due to being + * on_cpu. + */ + donor->sched_class->put_prev_task(rq, donor, donor); + donor->sched_class->set_next_task(rq, donor, true); + } + } else { + rq_set_donor(rq, next); + } + picked: clear_tsk_need_resched(prev); clear_preempt_need_resched(); -#ifdef CONFIG_SCHED_DEBUG +keep_resched: rq->last_seen_need_resched_ns = 0; -#endif - if (likely(prev != next)) { + is_switch = prev != next; + if (likely(is_switch)) { rq->nr_switches++; /* * RCU users of rcu_dereference(rq->curr) may not see * changes to task_struct made by pick_next_task(). */ RCU_INIT_POINTER(rq->curr, next); + /* * The membarrier system call requires each architecture * to have a full memory barrier after updating @@ -6754,7 +7179,6 @@ picked: */ ++*switch_count; - migrate_disable_switch(rq, prev); psi_account_irqtime(rq, prev, next); psi_sched_switch(prev, next, !task_on_rq_queued(prev) || prev->se.sched_delayed); @@ -6765,9 +7189,11 @@ picked: rq = context_switch(rq, prev, next, &rf); } else { rq_unpin_lock(rq, &rf); - __balance_callbacks(rq); + __balance_callbacks(rq, NULL); + hrtick_schedule_exit(rq); raw_spin_rq_unlock_irq(rq); } + trace_sched_exit_tp(is_switch); } void __noreturn do_task_dead(void) @@ -6812,7 +7238,7 @@ static inline void sched_submit_work(struct task_struct *tsk) * deadlock if the callback attempts to acquire a lock which is * already acquired. */ - SCHED_WARN_ON(current->__state & TASK_RTLOCK_WAIT); + WARN_ON_ONCE(current->__state & TASK_RTLOCK_WAIT); /* * If we are going to sleep and we have plugged IO queued, @@ -6971,14 +7397,14 @@ NOKPROBE_SYMBOL(preempt_schedule); EXPORT_SYMBOL(preempt_schedule); #ifdef CONFIG_PREEMPT_DYNAMIC -#if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) -#ifndef preempt_schedule_dynamic_enabled -#define preempt_schedule_dynamic_enabled preempt_schedule -#define preempt_schedule_dynamic_disabled NULL -#endif +# ifdef CONFIG_HAVE_PREEMPT_DYNAMIC_CALL +# ifndef preempt_schedule_dynamic_enabled +# define preempt_schedule_dynamic_enabled preempt_schedule +# define preempt_schedule_dynamic_disabled NULL +# endif DEFINE_STATIC_CALL(preempt_schedule, preempt_schedule_dynamic_enabled); EXPORT_STATIC_CALL_TRAMP(preempt_schedule); -#elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) +# elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) static DEFINE_STATIC_KEY_TRUE(sk_dynamic_preempt_schedule); void __sched notrace dynamic_preempt_schedule(void) { @@ -6988,8 +7414,8 @@ void __sched notrace dynamic_preempt_schedule(void) } NOKPROBE_SYMBOL(dynamic_preempt_schedule); EXPORT_SYMBOL(dynamic_preempt_schedule); -#endif -#endif +# endif +#endif /* CONFIG_PREEMPT_DYNAMIC */ /** * preempt_schedule_notrace - preempt_schedule called by tracing @@ -7044,14 +7470,14 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) EXPORT_SYMBOL_GPL(preempt_schedule_notrace); #ifdef CONFIG_PREEMPT_DYNAMIC -#if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) -#ifndef preempt_schedule_notrace_dynamic_enabled -#define preempt_schedule_notrace_dynamic_enabled preempt_schedule_notrace -#define preempt_schedule_notrace_dynamic_disabled NULL -#endif +# if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) +# ifndef preempt_schedule_notrace_dynamic_enabled +# define preempt_schedule_notrace_dynamic_enabled preempt_schedule_notrace +# define preempt_schedule_notrace_dynamic_disabled NULL +# endif DEFINE_STATIC_CALL(preempt_schedule_notrace, preempt_schedule_notrace_dynamic_enabled); EXPORT_STATIC_CALL_TRAMP(preempt_schedule_notrace); -#elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) +# elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) static DEFINE_STATIC_KEY_TRUE(sk_dynamic_preempt_schedule_notrace); void __sched notrace dynamic_preempt_schedule_notrace(void) { @@ -7061,7 +7487,7 @@ void __sched notrace dynamic_preempt_schedule_notrace(void) } NOKPROBE_SYMBOL(dynamic_preempt_schedule_notrace); EXPORT_SYMBOL(dynamic_preempt_schedule_notrace); -#endif +# endif #endif #endif /* CONFIG_PREEMPTION */ @@ -7095,7 +7521,7 @@ asmlinkage __visible void __sched preempt_schedule_irq(void) int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags, void *key) { - WARN_ON_ONCE(IS_ENABLED(CONFIG_SCHED_DEBUG) && wake_flags & ~(WF_SYNC|WF_CURRENT_CPU)); + WARN_ON_ONCE(wake_flags & ~(WF_SYNC|WF_CURRENT_CPU)); return try_to_wake_up(curr->private, mode, wake_flags); } EXPORT_SYMBOL(default_wake_function); @@ -7157,7 +7583,7 @@ void rt_mutex_post_schedule(void) */ void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) { - int prio, oldprio, queued, running, queue_flag = + int prio, oldprio, queue_flag = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; const struct sched_class *prev_class, *next_class; struct rq_flags rf; @@ -7213,74 +7639,56 @@ void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) trace_sched_pi_setprio(p, pi_task); oldprio = p->prio; - if (oldprio == prio) + if (oldprio == prio && !dl_prio(prio)) queue_flag &= ~DEQUEUE_MOVE; prev_class = p->sched_class; next_class = __setscheduler_class(p->policy, prio); - if (prev_class != next_class && p->se.sched_delayed) - dequeue_task(rq, p, DEQUEUE_SLEEP | DEQUEUE_DELAYED | DEQUEUE_NOCLOCK); - - queued = task_on_rq_queued(p); - running = task_current_donor(rq, p); - if (queued) - dequeue_task(rq, p, queue_flag); - if (running) - put_prev_task(rq, p); + if (prev_class != next_class) + queue_flag |= DEQUEUE_CLASS; - /* - * Boosting condition are: - * 1. -rt task is running and holds mutex A - * --> -dl task blocks on mutex A - * - * 2. -dl task is running and holds mutex A - * --> -dl task blocks on mutex A and could preempt the - * running task - */ - if (dl_prio(prio)) { - if (!dl_prio(p->normal_prio) || - (pi_task && dl_prio(pi_task->prio) && - dl_entity_preempt(&pi_task->dl, &p->dl))) { - p->dl.pi_se = pi_task->dl.pi_se; - queue_flag |= ENQUEUE_REPLENISH; + scoped_guard (sched_change, p, queue_flag) { + /* + * Boosting condition are: + * 1. -rt task is running and holds mutex A + * --> -dl task blocks on mutex A + * + * 2. -dl task is running and holds mutex A + * --> -dl task blocks on mutex A and could preempt the + * running task + */ + if (dl_prio(prio)) { + if (!dl_prio(p->normal_prio) || + (pi_task && dl_prio(pi_task->prio) && + dl_entity_preempt(&pi_task->dl, &p->dl))) { + p->dl.pi_se = pi_task->dl.pi_se; + scope->flags |= ENQUEUE_REPLENISH; + } else { + p->dl.pi_se = &p->dl; + } + } else if (rt_prio(prio)) { + if (dl_prio(oldprio)) + p->dl.pi_se = &p->dl; + if (oldprio < prio) + scope->flags |= ENQUEUE_HEAD; } else { - p->dl.pi_se = &p->dl; + if (dl_prio(oldprio)) + p->dl.pi_se = &p->dl; + if (rt_prio(oldprio)) + p->rt.timeout = 0; } - } else if (rt_prio(prio)) { - if (dl_prio(oldprio)) - p->dl.pi_se = &p->dl; - if (oldprio < prio) - queue_flag |= ENQUEUE_HEAD; - } else { - if (dl_prio(oldprio)) - p->dl.pi_se = &p->dl; - if (rt_prio(oldprio)) - p->rt.timeout = 0; - } - - p->sched_class = next_class; - p->prio = prio; - check_class_changing(rq, p, prev_class); - - if (queued) - enqueue_task(rq, p, queue_flag); - if (running) - set_next_task(rq, p); - - check_class_changed(rq, p, prev_class, oldprio); + p->sched_class = next_class; + p->prio = prio; + } out_unlock: - /* Avoid rq from going away on us: */ - preempt_disable(); + /* Caller holds task_struct::pi_lock, IRQs are still disabled */ - rq_unpin_lock(rq, &rf); - __balance_callbacks(rq); - raw_spin_rq_unlock(rq); - - preempt_enable(); + __balance_callbacks(rq, &rf); + __task_rq_unlock(rq, p, &rf); } -#endif +#endif /* CONFIG_RT_MUTEXES */ #if !defined(CONFIG_PREEMPTION) || defined(CONFIG_PREEMPT_DYNAMIC) int __sched __cond_resched(void) @@ -7290,7 +7698,7 @@ int __sched __cond_resched(void) return 1; } /* - * In preemptible kernels, ->rcu_read_lock_nesting tells the tick + * In PREEMPT_RCU kernels, ->rcu_read_lock_nesting tells the tick * whether the current CPU is in an RCU read-side critical section, * so the tick can report quiescent states even for CPUs looping * in kernel context. In contrast, in non-preemptible kernels, @@ -7299,6 +7707,8 @@ int __sched __cond_resched(void) * RCU quiescent state. Therefore, the following code causes * cond_resched() to report a quiescent state, but only when RCU * is in urgent need of one. + * A third case, preemptible, but non-PREEMPT_RCU provides for + * urgently needed quiescent states via rcu_flavor_sched_clock_irq(). */ #ifndef CONFIG_PREEMPT_RCU rcu_all_qs(); @@ -7309,21 +7719,20 @@ EXPORT_SYMBOL(__cond_resched); #endif #ifdef CONFIG_PREEMPT_DYNAMIC -#if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) -#define cond_resched_dynamic_enabled __cond_resched -#define cond_resched_dynamic_disabled ((void *)&__static_call_return0) +# ifdef CONFIG_HAVE_PREEMPT_DYNAMIC_CALL +# define cond_resched_dynamic_enabled __cond_resched +# define cond_resched_dynamic_disabled ((void *)&__static_call_return0) DEFINE_STATIC_CALL_RET0(cond_resched, __cond_resched); EXPORT_STATIC_CALL_TRAMP(cond_resched); -#define might_resched_dynamic_enabled __cond_resched -#define might_resched_dynamic_disabled ((void *)&__static_call_return0) +# define might_resched_dynamic_enabled __cond_resched +# define might_resched_dynamic_disabled ((void *)&__static_call_return0) DEFINE_STATIC_CALL_RET0(might_resched, __cond_resched); EXPORT_STATIC_CALL_TRAMP(might_resched); -#elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) +# elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) static DEFINE_STATIC_KEY_FALSE(sk_dynamic_cond_resched); int __sched dynamic_cond_resched(void) { - klp_sched_try_switch(); if (!static_branch_unlikely(&sk_dynamic_cond_resched)) return 0; return __cond_resched(); @@ -7338,8 +7747,8 @@ int __sched dynamic_might_resched(void) return __cond_resched(); } EXPORT_SYMBOL(dynamic_might_resched); -#endif -#endif +# endif +#endif /* CONFIG_PREEMPT_DYNAMIC */ /* * __cond_resched_lock() - if a reschedule is pending, drop the given lock, @@ -7405,9 +7814,9 @@ EXPORT_SYMBOL(__cond_resched_rwlock_write); #ifdef CONFIG_PREEMPT_DYNAMIC -#ifdef CONFIG_GENERIC_ENTRY -#include <linux/entry-common.h> -#endif +# ifdef CONFIG_GENERIC_IRQ_ENTRY +# include <linux/irq-entry-common.h> +# endif /* * SC:cond_resched @@ -7462,40 +7871,39 @@ int preempt_dynamic_mode = preempt_dynamic_undefined; int sched_dynamic_mode(const char *str) { -#ifndef CONFIG_PREEMPT_RT +# if !(defined(CONFIG_PREEMPT_RT) || defined(CONFIG_ARCH_HAS_PREEMPT_LAZY)) if (!strcmp(str, "none")) return preempt_dynamic_none; if (!strcmp(str, "voluntary")) return preempt_dynamic_voluntary; -#endif +# endif if (!strcmp(str, "full")) return preempt_dynamic_full; -#ifdef CONFIG_ARCH_HAS_PREEMPT_LAZY +# ifdef CONFIG_ARCH_HAS_PREEMPT_LAZY if (!strcmp(str, "lazy")) return preempt_dynamic_lazy; -#endif +# endif return -EINVAL; } -#define preempt_dynamic_key_enable(f) static_key_enable(&sk_dynamic_##f.key) -#define preempt_dynamic_key_disable(f) static_key_disable(&sk_dynamic_##f.key) +# define preempt_dynamic_key_enable(f) static_key_enable(&sk_dynamic_##f.key) +# define preempt_dynamic_key_disable(f) static_key_disable(&sk_dynamic_##f.key) -#if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) -#define preempt_dynamic_enable(f) static_call_update(f, f##_dynamic_enabled) -#define preempt_dynamic_disable(f) static_call_update(f, f##_dynamic_disabled) -#elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) -#define preempt_dynamic_enable(f) preempt_dynamic_key_enable(f) -#define preempt_dynamic_disable(f) preempt_dynamic_key_disable(f) -#else -#error "Unsupported PREEMPT_DYNAMIC mechanism" -#endif +# if defined(CONFIG_HAVE_PREEMPT_DYNAMIC_CALL) +# define preempt_dynamic_enable(f) static_call_update(f, f##_dynamic_enabled) +# define preempt_dynamic_disable(f) static_call_update(f, f##_dynamic_disabled) +# elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) +# define preempt_dynamic_enable(f) preempt_dynamic_key_enable(f) +# define preempt_dynamic_disable(f) preempt_dynamic_key_disable(f) +# else +# error "Unsupported PREEMPT_DYNAMIC mechanism" +# endif static DEFINE_MUTEX(sched_dynamic_mutex); -static bool klp_override; static void __sched_dynamic_update(int mode) { @@ -7503,8 +7911,7 @@ static void __sched_dynamic_update(int mode) * Avoid {NONE,VOLUNTARY} -> FULL transitions from ever ending up in * the ZERO state, which is invalid. */ - if (!klp_override) - preempt_dynamic_enable(cond_resched); + preempt_dynamic_enable(cond_resched); preempt_dynamic_enable(might_resched); preempt_dynamic_enable(preempt_schedule); preempt_dynamic_enable(preempt_schedule_notrace); @@ -7513,8 +7920,7 @@ static void __sched_dynamic_update(int mode) switch (mode) { case preempt_dynamic_none: - if (!klp_override) - preempt_dynamic_enable(cond_resched); + preempt_dynamic_enable(cond_resched); preempt_dynamic_disable(might_resched); preempt_dynamic_disable(preempt_schedule); preempt_dynamic_disable(preempt_schedule_notrace); @@ -7525,8 +7931,7 @@ static void __sched_dynamic_update(int mode) break; case preempt_dynamic_voluntary: - if (!klp_override) - preempt_dynamic_enable(cond_resched); + preempt_dynamic_enable(cond_resched); preempt_dynamic_enable(might_resched); preempt_dynamic_disable(preempt_schedule); preempt_dynamic_disable(preempt_schedule_notrace); @@ -7537,8 +7942,7 @@ static void __sched_dynamic_update(int mode) break; case preempt_dynamic_full: - if (!klp_override) - preempt_dynamic_disable(cond_resched); + preempt_dynamic_disable(cond_resched); preempt_dynamic_disable(might_resched); preempt_dynamic_enable(preempt_schedule); preempt_dynamic_enable(preempt_schedule_notrace); @@ -7549,8 +7953,7 @@ static void __sched_dynamic_update(int mode) break; case preempt_dynamic_lazy: - if (!klp_override) - preempt_dynamic_disable(cond_resched); + preempt_dynamic_disable(cond_resched); preempt_dynamic_disable(might_resched); preempt_dynamic_enable(preempt_schedule); preempt_dynamic_enable(preempt_schedule_notrace); @@ -7571,36 +7974,6 @@ void sched_dynamic_update(int mode) mutex_unlock(&sched_dynamic_mutex); } -#ifdef CONFIG_HAVE_PREEMPT_DYNAMIC_CALL - -static int klp_cond_resched(void) -{ - __klp_sched_try_switch(); - return __cond_resched(); -} - -void sched_dynamic_klp_enable(void) -{ - mutex_lock(&sched_dynamic_mutex); - - klp_override = true; - static_call_update(cond_resched, klp_cond_resched); - - mutex_unlock(&sched_dynamic_mutex); -} - -void sched_dynamic_klp_disable(void) -{ - mutex_lock(&sched_dynamic_mutex); - - klp_override = false; - __sched_dynamic_update(preempt_dynamic_mode); - - mutex_unlock(&sched_dynamic_mutex); -} - -#endif /* CONFIG_HAVE_PREEMPT_DYNAMIC_CALL */ - static int __init setup_preempt_mode(char *str) { int mode = sched_dynamic_mode(str); @@ -7632,7 +8005,7 @@ static void __init preempt_dynamic_init(void) } } -#define PREEMPT_MODEL_ACCESSOR(mode) \ +# define PREEMPT_MODEL_ACCESSOR(mode) \ bool preempt_model_##mode(void) \ { \ WARN_ON_ONCE(preempt_dynamic_mode == preempt_dynamic_undefined); \ @@ -7647,10 +8020,57 @@ PREEMPT_MODEL_ACCESSOR(lazy); #else /* !CONFIG_PREEMPT_DYNAMIC: */ +#define preempt_dynamic_mode -1 + static inline void preempt_dynamic_init(void) { } #endif /* CONFIG_PREEMPT_DYNAMIC */ +const char *preempt_modes[] = { + "none", "voluntary", "full", "lazy", NULL, +}; + +const char *preempt_model_str(void) +{ + bool brace = IS_ENABLED(CONFIG_PREEMPT_RT) && + (IS_ENABLED(CONFIG_PREEMPT_DYNAMIC) || + IS_ENABLED(CONFIG_PREEMPT_LAZY)); + static char buf[128]; + + if (IS_ENABLED(CONFIG_PREEMPT_BUILD)) { + struct seq_buf s; + + seq_buf_init(&s, buf, sizeof(buf)); + seq_buf_puts(&s, "PREEMPT"); + + if (IS_ENABLED(CONFIG_PREEMPT_RT)) + seq_buf_printf(&s, "%sRT%s", + brace ? "_{" : "_", + brace ? "," : ""); + + if (IS_ENABLED(CONFIG_PREEMPT_DYNAMIC)) { + seq_buf_printf(&s, "(%s)%s", + preempt_dynamic_mode >= 0 ? + preempt_modes[preempt_dynamic_mode] : "undef", + brace ? "}" : ""); + return seq_buf_str(&s); + } + + if (IS_ENABLED(CONFIG_PREEMPT_LAZY)) { + seq_buf_printf(&s, "LAZY%s", + brace ? "}" : ""); + return seq_buf_str(&s); + } + + return seq_buf_str(&s); + } + + if (IS_ENABLED(CONFIG_PREEMPT_VOLUNTARY_BUILD)) + return "VOLUNTARY"; + + return "NONE"; +} + int io_schedule_prepare(void) { int old_iowait = current->in_iowait; @@ -7765,10 +8185,9 @@ void show_state_filter(unsigned int state_filter) sched_show_task(p); } -#ifdef CONFIG_SCHED_DEBUG if (!state_filter) sysrq_sched_debug_show(); -#endif + rcu_read_unlock(); /* * Only show locks if all tasks are dumped: @@ -7787,12 +8206,10 @@ void show_state_filter(unsigned int state_filter) */ void __init init_idle(struct task_struct *idle, int cpu) { -#ifdef CONFIG_SMP struct affinity_context ac = (struct affinity_context) { .new_mask = cpumask_of(cpu), .flags = 0, }; -#endif struct rq *rq = cpu_rq(cpu); unsigned long flags; @@ -7808,13 +8225,11 @@ void __init init_idle(struct task_struct *idle, int cpu) idle->flags |= PF_KTHREAD | PF_NO_SETAFFINITY; kthread_set_per_cpu(idle, cpu); -#ifdef CONFIG_SMP /* * No validation and serialization required at boot time and for * setting up the idle tasks of not yet online CPUs. */ set_cpus_allowed_common(idle, &ac); -#endif /* * We're having a chicken and egg problem, even though we are * holding rq->lock, the CPU isn't yet set to this CPU so the @@ -7833,9 +8248,7 @@ void __init init_idle(struct task_struct *idle, int cpu) rq_set_donor(rq, idle); rcu_assign_pointer(rq->curr, idle); idle->on_rq = TASK_ON_RQ_QUEUED; -#ifdef CONFIG_SMP idle->on_cpu = 1; -#endif raw_spin_rq_unlock(rq); raw_spin_unlock_irqrestore(&idle->pi_lock, flags); @@ -7848,13 +8261,9 @@ void __init init_idle(struct task_struct *idle, int cpu) idle->sched_class = &idle_sched_class; ftrace_graph_init_idle_task(idle, cpu); vtime_init_idle(idle, cpu); -#ifdef CONFIG_SMP sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); -#endif } -#ifdef CONFIG_SMP - int cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial) { @@ -7914,26 +8323,9 @@ int migrate_task_to(struct task_struct *p, int target_cpu) */ void sched_setnuma(struct task_struct *p, int nid) { - bool queued, running; - struct rq_flags rf; - struct rq *rq; - - rq = task_rq_lock(p, &rf); - queued = task_on_rq_queued(p); - running = task_current_donor(rq, p); - - if (queued) - dequeue_task(rq, p, DEQUEUE_SAVE); - if (running) - put_prev_task(rq, p); - - p->numa_preferred_nid = nid; - - if (queued) - enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); - if (running) - set_next_task(rq, p); - task_rq_unlock(rq, p, &rf); + guard(task_rq_lock)(p); + scoped_guard (sched_change, p, DEQUEUE_SAVE) + p->numa_preferred_nid = nid; } #endif /* CONFIG_NUMA_BALANCING */ @@ -7971,18 +8363,21 @@ static int __balance_push_cpu_stop(void *arg) struct rq_flags rf; int cpu; - raw_spin_lock_irq(&p->pi_lock); - rq_lock(rq, &rf); - - update_rq_clock(rq); + scoped_guard (raw_spinlock_irq, &p->pi_lock) { + /* + * We may change the underlying rq, but the locks held will + * appropriately be "transferred" when switching. + */ + context_unsafe_alias(rq); - if (task_rq(p) == rq && task_on_rq_queued(p)) { cpu = select_fallback_rq(rq->cpu, p); - rq = __migrate_task(rq, &rf, p, cpu); - } - rq_unlock(rq, &rf); - raw_spin_unlock_irq(&p->pi_lock); + rq_lock(rq, &rf); + update_rq_clock(rq); + if (task_rq(p) == rq && task_on_rq_queued(p)) + rq = __migrate_task(rq, &rf, p, cpu); + rq_unlock(rq, &rf); + } put_task_struct(p); @@ -7998,6 +8393,7 @@ static DEFINE_PER_CPU(struct cpu_stop_work, push_work); * effective when the hotplug motion is down. */ static void balance_push(struct rq *rq) + __must_hold(__rq_lockp(rq)) { struct task_struct *push_task = rq->curr; @@ -8090,7 +8486,7 @@ static void balance_hotplug_wait(void) TASK_UNINTERRUPTIBLE); } -#else +#else /* !CONFIG_HOTPLUG_CPU: */ static inline void balance_push(struct rq *rq) { @@ -8104,7 +8500,7 @@ static inline void balance_hotplug_wait(void) { } -#endif /* CONFIG_HOTPLUG_CPU */ +#endif /* !CONFIG_HOTPLUG_CPU */ void set_rq_online(struct rq *rq) { @@ -8183,7 +8579,7 @@ static void cpuset_cpu_active(void) * operation in the resume sequence, just build a single sched * domain, ignoring cpusets. */ - partition_sched_domains(1, NULL, NULL); + cpuset_reset_sched_domains(); if (--num_cpus_frozen) return; /* @@ -8202,7 +8598,7 @@ static void cpuset_cpu_inactive(unsigned int cpu) cpuset_update_active_cpus(); } else { num_cpus_frozen++; - partition_sched_domains(1, NULL, NULL); + cpuset_reset_sched_domains(); } } @@ -8401,10 +8797,15 @@ int sched_cpu_dying(unsigned int cpu) sched_tick_stop(cpu); rq_lock_irqsave(rq, &rf); + update_rq_clock(rq); if (rq->nr_running != 1 || rq_has_pinned_tasks(rq)) { WARN(true, "Dying CPU not properly vacated!"); dump_rq_tasks(rq, KERN_WARNING); } + dl_server_stop(&rq->fair_server); +#ifdef CONFIG_SCHED_CLASS_EXT + dl_server_stop(&rq->ext_server); +#endif rq_unlock_irqrestore(rq, &rf); calc_load_migrate(rq); @@ -8413,20 +8814,22 @@ int sched_cpu_dying(unsigned int cpu) sched_core_cpu_dying(cpu); return 0; } -#endif +#endif /* CONFIG_HOTPLUG_CPU */ void __init sched_init_smp(void) { sched_init_numa(NUMA_NO_NODE); + prandom_init_once(&sched_rnd_state); + /* * There's no userspace yet to cause hotplug operations; hence all the * CPU masks are stable and all blatant races in the below code cannot * happen. */ - mutex_lock(&sched_domains_mutex); + sched_domains_mutex_lock(); sched_init_domains(cpu_active_mask); - mutex_unlock(&sched_domains_mutex); + sched_domains_mutex_unlock(); /* Move init over to a non-isolated CPU */ if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_TYPE_DOMAIN)) < 0) @@ -8437,6 +8840,8 @@ void __init sched_init_smp(void) init_sched_rt_class(); init_sched_dl_class(); + sched_init_dl_servers(); + sched_smp_initialized = true; } @@ -8447,13 +8852,6 @@ static int __init migration_init(void) } early_initcall(migration_init); -#else -void __init sched_init_smp(void) -{ - sched_init_granularity(); -} -#endif /* CONFIG_SMP */ - int in_sched_functions(unsigned long addr) { return in_lock_functions(addr) || @@ -8479,9 +8877,7 @@ void __init sched_init(void) int i; /* Make sure the linker didn't screw up */ -#ifdef CONFIG_SMP BUG_ON(!sched_class_above(&stop_sched_class, &dl_sched_class)); -#endif BUG_ON(!sched_class_above(&dl_sched_class, &rt_sched_class)); BUG_ON(!sched_class_above(&rt_sched_class, &fair_sched_class)); BUG_ON(!sched_class_above(&fair_sched_class, &idle_sched_class)); @@ -8512,7 +8908,7 @@ void __init sched_init(void) init_cfs_bandwidth(&root_task_group.cfs_bandwidth, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_EXT_GROUP_SCHED - root_task_group.scx_weight = CGROUP_WEIGHT_DFL; + scx_tg_init(&root_task_group); #endif /* CONFIG_EXT_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED root_task_group.rt_se = (struct sched_rt_entity **)ptr; @@ -8524,9 +8920,7 @@ void __init sched_init(void) #endif /* CONFIG_RT_GROUP_SCHED */ } -#ifdef CONFIG_SMP init_defrootdomain(); -#endif #ifdef CONFIG_RT_GROUP_SCHED init_rt_bandwidth(&root_task_group.rt_bandwidth, @@ -8587,7 +8981,8 @@ void __init sched_init(void) rq->rt.rt_runtime = global_rt_runtime(); init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, NULL); #endif -#ifdef CONFIG_SMP + rq->next_class = &idle_sched_class; + rq->sd = NULL; rq->rd = NULL; rq->cpu_capacity = SCHED_CAPACITY_SCALE; @@ -8613,10 +9008,12 @@ void __init sched_init(void) #ifdef CONFIG_HOTPLUG_CPU rcuwait_init(&rq->hotplug_wait); #endif -#endif /* CONFIG_SMP */ hrtick_rq_init(rq); atomic_set(&rq->nr_iowait, 0); fair_server_init(rq); +#ifdef CONFIG_SCHED_CLASS_EXT + ext_server_init(rq); +#endif #ifdef CONFIG_SCHED_CORE rq->core = rq; @@ -8661,10 +9058,9 @@ void __init sched_init(void) calc_load_update = jiffies + LOAD_FREQ; -#ifdef CONFIG_SMP idle_thread_set_boot_cpu(); + balance_push_set(smp_processor_id(), false); -#endif init_sched_fair_class(); init_sched_ext_class(); @@ -8797,7 +9193,7 @@ void __cant_sleep(const char *file, int line, int preempt_offset) } EXPORT_SYMBOL_GPL(__cant_sleep); -#ifdef CONFIG_SMP +# ifdef CONFIG_SMP void __cant_migrate(const char *file, int line) { static unsigned long prev_jiffy; @@ -8828,8 +9224,8 @@ void __cant_migrate(const char *file, int line) add_taint(TAINT_WARN, LOCKDEP_STILL_OK); } EXPORT_SYMBOL_GPL(__cant_migrate); -#endif -#endif +# endif /* CONFIG_SMP */ +#endif /* CONFIG_DEBUG_ATOMIC_SLEEP */ #ifdef CONFIG_MAGIC_SYSRQ void normalize_rt_tasks(void) @@ -8869,7 +9265,7 @@ void normalize_rt_tasks(void) #endif /* CONFIG_MAGIC_SYSRQ */ -#if defined(CONFIG_KGDB_KDB) +#ifdef CONFIG_KGDB_KDB /* * These functions are only useful for KDB. * @@ -8893,7 +9289,7 @@ struct task_struct *curr_task(int cpu) return cpu_curr(cpu); } -#endif /* defined(CONFIG_KGDB_KDB) */ +#endif /* CONFIG_KGDB_KDB */ #ifdef CONFIG_CGROUP_SCHED /* task_group_lock serializes the addition/removal of task groups */ @@ -8952,7 +9348,7 @@ struct task_group *sched_create_group(struct task_group *parent) if (!alloc_rt_sched_group(tg, parent)) goto err; - scx_group_set_weight(tg, CGROUP_WEIGHT_DFL); + scx_tg_init(tg); alloc_uclamp_sched_group(tg, parent); return tg; @@ -8967,7 +9363,7 @@ void sched_online_group(struct task_group *tg, struct task_group *parent) unsigned long flags; spin_lock_irqsave(&task_group_lock, flags); - list_add_rcu(&tg->list, &task_groups); + list_add_tail_rcu(&tg->list, &task_groups); /* Root should already exist: */ WARN_ON(!parent); @@ -9047,38 +9443,29 @@ static void sched_change_group(struct task_struct *tsk) */ void sched_move_task(struct task_struct *tsk, bool for_autogroup) { - int queued, running, queue_flags = - DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; + unsigned int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE; + bool resched = false; + bool queued = false; struct rq *rq; CLASS(task_rq_lock, rq_guard)(tsk); rq = rq_guard.rq; - update_rq_clock(rq); - - running = task_current_donor(rq, tsk); - queued = task_on_rq_queued(tsk); - - if (queued) - dequeue_task(rq, tsk, queue_flags); - if (running) - put_prev_task(rq, tsk); - - sched_change_group(tsk); - if (!for_autogroup) - scx_cgroup_move_task(tsk); + scoped_guard (sched_change, tsk, queue_flags) { + sched_change_group(tsk); + if (!for_autogroup) + scx_cgroup_move_task(tsk); + if (scope->running) + resched = true; + queued = scope->queued; + } - if (queued) - enqueue_task(rq, tsk, queue_flags); - if (running) { - set_next_task(rq, tsk); - /* - * After changing group, the running task may have joined a - * throttled one but it's still the running task. Trigger a - * resched to make sure that task can still run. - */ + if (resched) resched_curr(rq); - } + else if (queued) + wakeup_preempt(rq, tsk, 0); + + __balance_callbacks(rq, &rq_guard.rf); } static struct cgroup_subsys_state * @@ -9153,11 +9540,15 @@ static int cpu_cgroup_can_attach(struct cgroup_taskset *tset) struct task_struct *task; struct cgroup_subsys_state *css; + if (!rt_group_sched_enabled()) + goto scx_check; + cgroup_taskset_for_each(task, css, tset) { if (!sched_rt_can_attach(css_tg(css), task)) return -EINVAL; } -#endif +scx_check: +#endif /* CONFIG_RT_GROUP_SCHED */ return scx_cgroup_can_attach(tset); } @@ -9168,8 +9559,6 @@ static void cpu_cgroup_attach(struct cgroup_taskset *tset) cgroup_taskset_for_each(task, css, tset) sched_move_task(task, false); - - scx_cgroup_finish_attach(); } static void cpu_cgroup_cancel_attach(struct cgroup_taskset *tset) @@ -9188,7 +9577,7 @@ static void cpu_util_update_eff(struct cgroup_subsys_state *css) unsigned int clamps; lockdep_assert_held(&uclamp_mutex); - SCHED_WARN_ON(!rcu_read_lock_held()); + WARN_ON_ONCE(!rcu_read_lock_held()); css_for_each_descendant_pre(css, top_css) { uc_parent = css_tg(css)->parent @@ -9280,7 +9669,7 @@ static ssize_t cpu_uclamp_write(struct kernfs_open_file *of, char *buf, if (req.ret) return req.ret; - static_branch_enable(&sched_uclamp_used); + sched_uclamp_enable(); guard(mutex)(&uclamp_mutex); guard(rcu)(); @@ -9357,7 +9746,7 @@ static unsigned long tg_weight(struct task_group *tg) #ifdef CONFIG_FAIR_GROUP_SCHED return scale_load_down(tg->shares); #else - return sched_weight_from_cgroup(tg->scx_weight); + return sched_weight_from_cgroup(tg->scx.weight); #endif } @@ -9385,47 +9774,23 @@ static u64 cpu_shares_read_u64(struct cgroup_subsys_state *css, #ifdef CONFIG_CFS_BANDWIDTH static DEFINE_MUTEX(cfs_constraints_mutex); -const u64 max_cfs_quota_period = 1 * NSEC_PER_SEC; /* 1s */ -static const u64 min_cfs_quota_period = 1 * NSEC_PER_MSEC; /* 1ms */ -/* More than 203 days if BW_SHIFT equals 20. */ -static const u64 max_cfs_runtime = MAX_BW * NSEC_PER_USEC; - static int __cfs_schedulable(struct task_group *tg, u64 period, u64 runtime); -static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota, - u64 burst) +static int tg_set_cfs_bandwidth(struct task_group *tg, + u64 period_us, u64 quota_us, u64 burst_us) { int i, ret = 0, runtime_enabled, runtime_was_enabled; struct cfs_bandwidth *cfs_b = &tg->cfs_bandwidth; + u64 period, quota, burst; - if (tg == &root_task_group) - return -EINVAL; - - /* - * Ensure we have at some amount of bandwidth every period. This is - * to prevent reaching a state of large arrears when throttled via - * entity_tick() resulting in prolonged exit starvation. - */ - if (quota < min_cfs_quota_period || period < min_cfs_quota_period) - return -EINVAL; + period = (u64)period_us * NSEC_PER_USEC; - /* - * Likewise, bound things on the other side by preventing insane quota - * periods. This also allows us to normalize in computing quota - * feasibility. - */ - if (period > max_cfs_quota_period) - return -EINVAL; - - /* - * Bound quota to defend quota against overflow during bandwidth shift. - */ - if (quota != RUNTIME_INF && quota > max_cfs_runtime) - return -EINVAL; + if (quota_us == RUNTIME_INF) + quota = RUNTIME_INF; + else + quota = (u64)quota_us * NSEC_PER_USEC; - if (quota != RUNTIME_INF && (burst > quota || - burst + quota > max_cfs_runtime)) - return -EINVAL; + burst = (u64)burst_us * NSEC_PER_USEC; /* * Prevent race between setting of cfs_rq->runtime_enabled and @@ -9468,7 +9833,7 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota, guard(rq_lock_irq)(rq); cfs_rq->runtime_enabled = runtime_enabled; - cfs_rq->runtime_remaining = 0; + cfs_rq->runtime_remaining = 1; if (cfs_rq->throttled) unthrottle_cfs_rq(cfs_rq); @@ -9480,28 +9845,22 @@ static int tg_set_cfs_bandwidth(struct task_group *tg, u64 period, u64 quota, return 0; } -static int tg_set_cfs_quota(struct task_group *tg, long cfs_quota_us) +static u64 tg_get_cfs_period(struct task_group *tg) { - u64 quota, period, burst; + u64 cfs_period_us; - period = ktime_to_ns(tg->cfs_bandwidth.period); - burst = tg->cfs_bandwidth.burst; - if (cfs_quota_us < 0) - quota = RUNTIME_INF; - else if ((u64)cfs_quota_us <= U64_MAX / NSEC_PER_USEC) - quota = (u64)cfs_quota_us * NSEC_PER_USEC; - else - return -EINVAL; + cfs_period_us = ktime_to_ns(tg->cfs_bandwidth.period); + do_div(cfs_period_us, NSEC_PER_USEC); - return tg_set_cfs_bandwidth(tg, period, quota, burst); + return cfs_period_us; } -static long tg_get_cfs_quota(struct task_group *tg) +static u64 tg_get_cfs_quota(struct task_group *tg) { u64 quota_us; if (tg->cfs_bandwidth.quota == RUNTIME_INF) - return -1; + return RUNTIME_INF; quota_us = tg->cfs_bandwidth.quota; do_div(quota_us, NSEC_PER_USEC); @@ -9509,45 +9868,7 @@ static long tg_get_cfs_quota(struct task_group *tg) return quota_us; } -static int tg_set_cfs_period(struct task_group *tg, long cfs_period_us) -{ - u64 quota, period, burst; - - if ((u64)cfs_period_us > U64_MAX / NSEC_PER_USEC) - return -EINVAL; - - period = (u64)cfs_period_us * NSEC_PER_USEC; - quota = tg->cfs_bandwidth.quota; - burst = tg->cfs_bandwidth.burst; - - return tg_set_cfs_bandwidth(tg, period, quota, burst); -} - -static long tg_get_cfs_period(struct task_group *tg) -{ - u64 cfs_period_us; - - cfs_period_us = ktime_to_ns(tg->cfs_bandwidth.period); - do_div(cfs_period_us, NSEC_PER_USEC); - - return cfs_period_us; -} - -static int tg_set_cfs_burst(struct task_group *tg, long cfs_burst_us) -{ - u64 quota, period, burst; - - if ((u64)cfs_burst_us > U64_MAX / NSEC_PER_USEC) - return -EINVAL; - - burst = (u64)cfs_burst_us * NSEC_PER_USEC; - period = ktime_to_ns(tg->cfs_bandwidth.period); - quota = tg->cfs_bandwidth.quota; - - return tg_set_cfs_bandwidth(tg, period, quota, burst); -} - -static long tg_get_cfs_burst(struct task_group *tg) +static u64 tg_get_cfs_burst(struct task_group *tg) { u64 burst_us; @@ -9557,42 +9878,6 @@ static long tg_get_cfs_burst(struct task_group *tg) return burst_us; } -static s64 cpu_cfs_quota_read_s64(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return tg_get_cfs_quota(css_tg(css)); -} - -static int cpu_cfs_quota_write_s64(struct cgroup_subsys_state *css, - struct cftype *cftype, s64 cfs_quota_us) -{ - return tg_set_cfs_quota(css_tg(css), cfs_quota_us); -} - -static u64 cpu_cfs_period_read_u64(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return tg_get_cfs_period(css_tg(css)); -} - -static int cpu_cfs_period_write_u64(struct cgroup_subsys_state *css, - struct cftype *cftype, u64 cfs_period_us) -{ - return tg_set_cfs_period(css_tg(css), cfs_period_us); -} - -static u64 cpu_cfs_burst_read_u64(struct cgroup_subsys_state *css, - struct cftype *cft) -{ - return tg_get_cfs_burst(css_tg(css)); -} - -static int cpu_cfs_burst_write_u64(struct cgroup_subsys_state *css, - struct cftype *cftype, u64 cfs_burst_us) -{ - return tg_set_cfs_burst(css_tg(css), cfs_burst_us); -} - struct cfs_schedulable_data { struct task_group *tg; u64 period, quota; @@ -9727,6 +10012,143 @@ static int cpu_cfs_local_stat_show(struct seq_file *sf, void *v) } #endif /* CONFIG_CFS_BANDWIDTH */ +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH +const u64 max_bw_quota_period_us = 1 * USEC_PER_SEC; /* 1s */ +static const u64 min_bw_quota_period_us = 1 * USEC_PER_MSEC; /* 1ms */ +/* More than 203 days if BW_SHIFT equals 20. */ +static const u64 max_bw_runtime_us = MAX_BW; + +static void tg_bandwidth(struct task_group *tg, + u64 *period_us_p, u64 *quota_us_p, u64 *burst_us_p) +{ +#ifdef CONFIG_CFS_BANDWIDTH + if (period_us_p) + *period_us_p = tg_get_cfs_period(tg); + if (quota_us_p) + *quota_us_p = tg_get_cfs_quota(tg); + if (burst_us_p) + *burst_us_p = tg_get_cfs_burst(tg); +#else /* !CONFIG_CFS_BANDWIDTH */ + if (period_us_p) + *period_us_p = tg->scx.bw_period_us; + if (quota_us_p) + *quota_us_p = tg->scx.bw_quota_us; + if (burst_us_p) + *burst_us_p = tg->scx.bw_burst_us; +#endif /* CONFIG_CFS_BANDWIDTH */ +} + +static u64 cpu_period_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + u64 period_us; + + tg_bandwidth(css_tg(css), &period_us, NULL, NULL); + return period_us; +} + +static int tg_set_bandwidth(struct task_group *tg, + u64 period_us, u64 quota_us, u64 burst_us) +{ + const u64 max_usec = U64_MAX / NSEC_PER_USEC; + int ret = 0; + + if (tg == &root_task_group) + return -EINVAL; + + /* Values should survive translation to nsec */ + if (period_us > max_usec || + (quota_us != RUNTIME_INF && quota_us > max_usec) || + burst_us > max_usec) + return -EINVAL; + + /* + * Ensure we have some amount of bandwidth every period. This is to + * prevent reaching a state of large arrears when throttled via + * entity_tick() resulting in prolonged exit starvation. + */ + if (quota_us < min_bw_quota_period_us || + period_us < min_bw_quota_period_us) + return -EINVAL; + + /* + * Likewise, bound things on the other side by preventing insane quota + * periods. This also allows us to normalize in computing quota + * feasibility. + */ + if (period_us > max_bw_quota_period_us) + return -EINVAL; + + /* + * Bound quota to defend quota against overflow during bandwidth shift. + */ + if (quota_us != RUNTIME_INF && quota_us > max_bw_runtime_us) + return -EINVAL; + + if (quota_us != RUNTIME_INF && (burst_us > quota_us || + burst_us + quota_us > max_bw_runtime_us)) + return -EINVAL; + +#ifdef CONFIG_CFS_BANDWIDTH + ret = tg_set_cfs_bandwidth(tg, period_us, quota_us, burst_us); +#endif /* CONFIG_CFS_BANDWIDTH */ + if (!ret) + scx_group_set_bandwidth(tg, period_us, quota_us, burst_us); + return ret; +} + +static s64 cpu_quota_read_s64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + u64 quota_us; + + tg_bandwidth(css_tg(css), NULL, "a_us, NULL); + return quota_us; /* (s64)RUNTIME_INF becomes -1 */ +} + +static u64 cpu_burst_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + u64 burst_us; + + tg_bandwidth(css_tg(css), NULL, NULL, &burst_us); + return burst_us; +} + +static int cpu_period_write_u64(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 period_us) +{ + struct task_group *tg = css_tg(css); + u64 quota_us, burst_us; + + tg_bandwidth(tg, NULL, "a_us, &burst_us); + return tg_set_bandwidth(tg, period_us, quota_us, burst_us); +} + +static int cpu_quota_write_s64(struct cgroup_subsys_state *css, + struct cftype *cftype, s64 quota_us) +{ + struct task_group *tg = css_tg(css); + u64 period_us, burst_us; + + if (quota_us < 0) + quota_us = RUNTIME_INF; + + tg_bandwidth(tg, &period_us, NULL, &burst_us); + return tg_set_bandwidth(tg, period_us, quota_us, burst_us); +} + +static int cpu_burst_write_u64(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 burst_us) +{ + struct task_group *tg = css_tg(css); + u64 period_us, quota_us; + + tg_bandwidth(tg, &period_us, "a_us, NULL); + return tg_set_bandwidth(tg, period_us, quota_us, burst_us); +} +#endif /* CONFIG_GROUP_SCHED_BANDWIDTH */ + #ifdef CONFIG_RT_GROUP_SCHED static int cpu_rt_runtime_write(struct cgroup_subsys_state *css, struct cftype *cft, s64 val) @@ -9770,7 +10192,7 @@ static int cpu_idle_write_s64(struct cgroup_subsys_state *css, scx_group_set_idle(css_tg(css), idle); return ret; } -#endif +#endif /* CONFIG_GROUP_SCHED_WEIGHT */ static struct cftype cpu_legacy_files[] = { #ifdef CONFIG_GROUP_SCHED_WEIGHT @@ -9785,22 +10207,24 @@ static struct cftype cpu_legacy_files[] = { .write_s64 = cpu_idle_write_s64, }, #endif -#ifdef CONFIG_CFS_BANDWIDTH +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH { - .name = "cfs_quota_us", - .read_s64 = cpu_cfs_quota_read_s64, - .write_s64 = cpu_cfs_quota_write_s64, + .name = "cfs_period_us", + .read_u64 = cpu_period_read_u64, + .write_u64 = cpu_period_write_u64, }, { - .name = "cfs_period_us", - .read_u64 = cpu_cfs_period_read_u64, - .write_u64 = cpu_cfs_period_write_u64, + .name = "cfs_quota_us", + .read_s64 = cpu_quota_read_s64, + .write_s64 = cpu_quota_write_s64, }, { .name = "cfs_burst_us", - .read_u64 = cpu_cfs_burst_read_u64, - .write_u64 = cpu_cfs_burst_write_u64, + .read_u64 = cpu_burst_read_u64, + .write_u64 = cpu_burst_write_u64, }, +#endif +#ifdef CONFIG_CFS_BANDWIDTH { .name = "stat", .seq_show = cpu_cfs_stat_show, @@ -9810,18 +10234,6 @@ static struct cftype cpu_legacy_files[] = { .seq_show = cpu_cfs_local_stat_show, }, #endif -#ifdef CONFIG_RT_GROUP_SCHED - { - .name = "rt_runtime_us", - .read_s64 = cpu_rt_runtime_read, - .write_s64 = cpu_rt_runtime_write, - }, - { - .name = "rt_period_us", - .read_u64 = cpu_rt_period_read_uint, - .write_u64 = cpu_rt_period_write_uint, - }, -#endif #ifdef CONFIG_UCLAMP_TASK_GROUP { .name = "uclamp.min", @@ -9839,6 +10251,55 @@ static struct cftype cpu_legacy_files[] = { { } /* Terminate */ }; +#ifdef CONFIG_RT_GROUP_SCHED +static struct cftype rt_group_files[] = { + { + .name = "rt_runtime_us", + .read_s64 = cpu_rt_runtime_read, + .write_s64 = cpu_rt_runtime_write, + }, + { + .name = "rt_period_us", + .read_u64 = cpu_rt_period_read_uint, + .write_u64 = cpu_rt_period_write_uint, + }, + { } /* Terminate */ +}; + +# ifdef CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED +DEFINE_STATIC_KEY_FALSE(rt_group_sched); +# else +DEFINE_STATIC_KEY_TRUE(rt_group_sched); +# endif + +static int __init setup_rt_group_sched(char *str) +{ + long val; + + if (kstrtol(str, 0, &val) || val < 0 || val > 1) { + pr_warn("Unable to set rt_group_sched\n"); + return 1; + } + if (val) + static_branch_enable(&rt_group_sched); + else + static_branch_disable(&rt_group_sched); + + return 1; +} +__setup("rt_group_sched=", setup_rt_group_sched); + +static int __init cpu_rt_group_init(void) +{ + if (!rt_group_sched_enabled()) + return 0; + + WARN_ON(cgroup_add_legacy_cftypes(&cpu_cgrp_subsys, rt_group_files)); + return 0; +} +subsys_initcall(cpu_rt_group_init); +#endif /* CONFIG_RT_GROUP_SCHED */ + static int cpu_extra_stat_show(struct seq_file *sf, struct cgroup_subsys_state *css) { @@ -9861,7 +10322,7 @@ static int cpu_extra_stat_show(struct seq_file *sf, cfs_b->nr_periods, cfs_b->nr_throttled, throttled_usec, cfs_b->nr_burst, burst_usec); } -#endif +#endif /* CONFIG_CFS_BANDWIDTH */ return 0; } @@ -9959,32 +10420,32 @@ static void __maybe_unused cpu_period_quota_print(struct seq_file *sf, } /* caller should put the current value in *@periodp before calling */ -static int __maybe_unused cpu_period_quota_parse(char *buf, - u64 *periodp, u64 *quotap) +static int __maybe_unused cpu_period_quota_parse(char *buf, u64 *period_us_p, + u64 *quota_us_p) { char tok[21]; /* U64_MAX */ - if (sscanf(buf, "%20s %llu", tok, periodp) < 1) + if (sscanf(buf, "%20s %llu", tok, period_us_p) < 1) return -EINVAL; - *periodp *= NSEC_PER_USEC; - - if (sscanf(tok, "%llu", quotap)) - *quotap *= NSEC_PER_USEC; - else if (!strcmp(tok, "max")) - *quotap = RUNTIME_INF; - else - return -EINVAL; + if (sscanf(tok, "%llu", quota_us_p) < 1) { + if (!strcmp(tok, "max")) + *quota_us_p = RUNTIME_INF; + else + return -EINVAL; + } return 0; } -#ifdef CONFIG_CFS_BANDWIDTH +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH static int cpu_max_show(struct seq_file *sf, void *v) { struct task_group *tg = css_tg(seq_css(sf)); + u64 period_us, quota_us; - cpu_period_quota_print(sf, tg_get_cfs_period(tg), tg_get_cfs_quota(tg)); + tg_bandwidth(tg, &period_us, "a_us, NULL); + cpu_period_quota_print(sf, period_us, quota_us); return 0; } @@ -9992,17 +10453,16 @@ static ssize_t cpu_max_write(struct kernfs_open_file *of, char *buf, size_t nbytes, loff_t off) { struct task_group *tg = css_tg(of_css(of)); - u64 period = tg_get_cfs_period(tg); - u64 burst = tg->cfs_bandwidth.burst; - u64 quota; + u64 period_us, quota_us, burst_us; int ret; - ret = cpu_period_quota_parse(buf, &period, "a); + tg_bandwidth(tg, &period_us, NULL, &burst_us); + ret = cpu_period_quota_parse(buf, &period_us, "a_us); if (!ret) - ret = tg_set_cfs_bandwidth(tg, period, quota, burst); + ret = tg_set_bandwidth(tg, period_us, quota_us, burst_us); return ret ?: nbytes; } -#endif +#endif /* CONFIG_CFS_BANDWIDTH */ static struct cftype cpu_files[] = { #ifdef CONFIG_GROUP_SCHED_WEIGHT @@ -10025,7 +10485,7 @@ static struct cftype cpu_files[] = { .write_s64 = cpu_idle_write_s64, }, #endif -#ifdef CONFIG_CFS_BANDWIDTH +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH { .name = "max", .flags = CFTYPE_NOT_ON_ROOT, @@ -10035,10 +10495,10 @@ static struct cftype cpu_files[] = { { .name = "max.burst", .flags = CFTYPE_NOT_ON_ROOT, - .read_u64 = cpu_cfs_burst_read_u64, - .write_u64 = cpu_cfs_burst_write_u64, + .read_u64 = cpu_burst_read_u64, + .write_u64 = cpu_burst_write_u64, }, -#endif +#endif /* CONFIG_CFS_BANDWIDTH */ #ifdef CONFIG_UCLAMP_TASK_GROUP { .name = "uclamp.min", @@ -10052,7 +10512,7 @@ static struct cftype cpu_files[] = { .seq_show = cpu_uclamp_max_show, .write = cpu_uclamp_max_write, }, -#endif +#endif /* CONFIG_UCLAMP_TASK_GROUP */ { } /* terminate */ }; @@ -10073,7 +10533,7 @@ struct cgroup_subsys cpu_cgrp_subsys = { .threaded = true, }; -#endif /* CONFIG_CGROUP_SCHED */ +#endif /* CONFIG_CGROUP_SCHED */ void dump_cpu_task(int cpu) { @@ -10141,558 +10601,636 @@ void call_trace_sched_update_nr_running(struct rq *rq, int count) } #ifdef CONFIG_SCHED_MM_CID - /* - * @cid_lock: Guarantee forward-progress of cid allocation. + * Concurrency IDentifier management * - * Concurrency ID allocation within a bitmap is mostly lock-free. The cid_lock - * is only used when contention is detected by the lock-free allocation so - * forward progress can be guaranteed. - */ -DEFINE_RAW_SPINLOCK(cid_lock); - -/* - * @use_cid_lock: Select cid allocation behavior: lock-free vs spinlock. + * Serialization rules: * - * When @use_cid_lock is 0, the cid allocation is lock-free. When contention is - * detected, it is set to 1 to ensure that all newly coming allocations are - * serialized by @cid_lock until the allocation which detected contention - * completes and sets @use_cid_lock back to 0. This guarantees forward progress - * of a cid allocation. - */ -int use_cid_lock; - -/* - * mm_cid remote-clear implements a lock-free algorithm to clear per-mm/cpu cid - * concurrently with respect to the execution of the source runqueue context - * switch. + * mm::mm_cid::mutex: Serializes fork() and exit() and therefore + * protects mm::mm_cid::users and mode switch + * transitions + * + * mm::mm_cid::lock: Serializes mm_update_max_cids() and + * mm_update_cpus_allowed(). Nests in mm_cid::mutex + * and runqueue lock. + * + * The mm_cidmask bitmap is not protected by any of the mm::mm_cid locks + * and can only be modified with atomic operations. + * + * The mm::mm_cid:pcpu per CPU storage is protected by the CPUs runqueue + * lock. * - * There is one basic properties we want to guarantee here: + * CID ownership: * - * (1) Remote-clear should _never_ mark a per-cpu cid UNSET when it is actively - * used by a task. That would lead to concurrent allocation of the cid and - * userspace corruption. + * A CID is either owned by a task (stored in task_struct::mm_cid.cid) or + * by a CPU (stored in mm::mm_cid.pcpu::cid). CIDs owned by CPUs have the + * MM_CID_ONCPU bit set. * - * Provide this guarantee by introducing a Dekker memory ordering to guarantee - * that a pair of loads observe at least one of a pair of stores, which can be - * shown as: + * During the transition of ownership mode, the MM_CID_TRANSIT bit is set + * on the CIDs. When this bit is set the tasks drop the CID back into the + * pool when scheduling out. * - * X = Y = 0 + * Both bits (ONCPU and TRANSIT) are filtered out by task_cid() when the + * CID is actually handed over to user space in the RSEQ memory. * - * w[X]=1 w[Y]=1 - * MB MB - * r[Y]=y r[X]=x + * Mode switching: * - * Which guarantees that x==0 && y==0 is impossible. But rather than using - * values 0 and 1, this algorithm cares about specific state transitions of the - * runqueue current task (as updated by the scheduler context switch), and the - * per-mm/cpu cid value. + * The ownership mode is per process and stored in mm:mm_cid::mode with the + * following possible states: * - * Let's introduce task (Y) which has task->mm == mm and task (N) which has - * task->mm != mm for the rest of the discussion. There are two scheduler state - * transitions on context switch we care about: + * 0: Per task ownership + * 0 | MM_CID_TRANSIT: Transition from per CPU to per task + * MM_CID_ONCPU: Per CPU ownership + * MM_CID_ONCPU | MM_CID_TRANSIT: Transition from per task to per CPU * - * (TSA) Store to rq->curr with transition from (N) to (Y) + * All transitions of ownership mode happen in two phases: * - * (TSB) Store to rq->curr with transition from (Y) to (N) + * 1) mm:mm_cid::mode has the MM_CID_TRANSIT bit set. This is OR'ed on the + * CIDs and denotes that the CID is only temporarily owned by a + * task. When the task schedules out it drops the CID back into the + * pool if this bit is set. * - * On the remote-clear side, there is one transition we care about: + * 2) The initiating context walks the per CPU space or the tasks to fixup + * or drop the CIDs and after completion it clears MM_CID_TRANSIT in + * mm:mm_cid::mode. After that point the CIDs are strictly task or CPU + * owned again. * - * (TMA) cmpxchg to *pcpu_cid to set the LAZY flag + * This two phase transition is required to prevent CID space exhaustion + * during the transition as a direct transfer of ownership would fail: * - * There is also a transition to UNSET state which can be performed from all - * sides (scheduler, remote-clear). It is always performed with a cmpxchg which - * guarantees that only a single thread will succeed: + * - On task to CPU mode switch if a task is scheduled in on one CPU and + * then migrated to another CPU before the fixup freed enough per task + * CIDs. * - * (TMB) cmpxchg to *pcpu_cid to mark UNSET + * - On CPU to task mode switch if two tasks are scheduled in on the same + * CPU before the fixup freed per CPU CIDs. * - * Just to be clear, what we do _not_ want to happen is a transition to UNSET - * when a thread is actively using the cid (property (1)). + * Both scenarios can result in a live lock because sched_in() is invoked + * with runqueue lock held and loops in search of a CID and the fixup + * thread can't make progress freeing them up because it is stuck on the + * same runqueue lock. * - * Let's looks at the relevant combinations of TSA/TSB, and TMA transitions. + * While MM_CID_TRANSIT is active during the transition phase the MM_CID + * bitmap can be contended, but that's a temporary contention bound to the + * transition period. After that everything goes back into steady state and + * nothing except fork() and exit() will touch the bitmap. This is an + * acceptable tradeoff as it completely avoids complex serialization, + * memory barriers and atomic operations for the common case. * - * Scenario A) (TSA)+(TMA) (from next task perspective) + * Aside of that this mechanism also ensures RT compability: * - * CPU0 CPU1 + * - The task which runs the fixup is fully preemptible except for the + * short runqueue lock held sections. * - * Context switch CS-1 Remote-clear - * - store to rq->curr: (N)->(Y) (TSA) - cmpxchg to *pcpu_id to LAZY (TMA) - * (implied barrier after cmpxchg) - * - switch_mm_cid() - * - memory barrier (see switch_mm_cid() - * comment explaining how this barrier - * is combined with other scheduler - * barriers) - * - mm_cid_get (next) - * - READ_ONCE(*pcpu_cid) - rcu_dereference(src_rq->curr) + * - The transient impact of the bitmap contention is only problematic + * when there is a thundering herd scenario of tasks scheduling in and + * out concurrently. There is not much which can be done about that + * except for avoiding mode switching by a proper overall system + * configuration. * - * This Dekker ensures that either task (Y) is observed by the - * rcu_dereference() or the LAZY flag is observed by READ_ONCE(), or both are - * observed. + * Switching to per CPU mode happens when the user count becomes greater + * than the maximum number of CIDs, which is calculated by: * - * If task (Y) store is observed by rcu_dereference(), it means that there is - * still an active task on the cpu. Remote-clear will therefore not transition - * to UNSET, which fulfills property (1). + * opt_cids = min(mm_cid::nr_cpus_allowed, mm_cid::users); + * max_cids = min(1.25 * opt_cids, num_possible_cpus()); * - * If task (Y) is not observed, but the lazy flag is observed by READ_ONCE(), - * it will move its state to UNSET, which clears the percpu cid perhaps - * uselessly (which is not an issue for correctness). Because task (Y) is not - * observed, CPU1 can move ahead to set the state to UNSET. Because moving - * state to UNSET is done with a cmpxchg expecting that the old state has the - * LAZY flag set, only one thread will successfully UNSET. + * The +25% allowance is useful for tight CPU masks in scenarios where only + * a few threads are created and destroyed to avoid frequent mode + * switches. Though this allowance shrinks, the closer opt_cids becomes to + * num_possible_cpus(), which is the (unfortunate) hard ABI limit. * - * If both states (LAZY flag and task (Y)) are observed, the thread on CPU0 - * will observe the LAZY flag and transition to UNSET (perhaps uselessly), and - * CPU1 will observe task (Y) and do nothing more, which is fine. + * At the point of switching to per CPU mode the new user is not yet + * visible in the system, so the task which initiated the fork() runs the + * fixup function. mm_cid_fixup_tasks_to_cpu() walks the thread list and + * either marks each task owned CID with MM_CID_TRANSIT if the task is + * running on a CPU or drops it into the CID pool if a task is not on a + * CPU. Tasks which schedule in before the task walk reaches them do the + * handover in mm_cid_schedin(). When mm_cid_fixup_tasks_to_cpus() + * completes it is guaranteed that no task related to that MM owns a CID + * anymore. * - * What we are effectively preventing with this Dekker is a scenario where - * neither LAZY flag nor store (Y) are observed, which would fail property (1) - * because this would UNSET a cid which is actively used. + * Switching back to task mode happens when the user count goes below the + * threshold which was recorded on the per CPU mode switch: + * + * pcpu_thrs = min(opt_cids - (opt_cids / 4), num_possible_cpus() / 2); + * + * This threshold is updated when a affinity change increases the number of + * allowed CPUs for the MM, which might cause a switch back to per task + * mode. + * + * If the switch back was initiated by a exiting task, then that task runs + * the fixup function. If it was initiated by a affinity change, then it's + * run either in the deferred update function in context of a workqueue or + * by a task which forks a new one or by a task which exits. Whatever + * happens first. mm_cid_fixup_cpus_to_task() walks through the possible + * CPUs and either marks the CPU owned CIDs with MM_CID_TRANSIT if a + * related task is running on the CPU or drops it into the pool. Tasks + * which are scheduled in before the fixup covered them do the handover + * themself. When mm_cid_fixup_cpus_to_tasks() completes it is guaranteed + * that no CID related to that MM is owned by a CPU anymore. */ -void sched_mm_cid_migrate_from(struct task_struct *t) +/* + * Update the CID range properties when the constraints change. Invoked via + * fork(), exit() and affinity changes + */ +static void __mm_update_max_cids(struct mm_mm_cid *mc) { - t->migrate_from_cpu = task_cpu(t); -} + unsigned int opt_cids, max_cids; -static -int __sched_mm_cid_migrate_from_fetch_cid(struct rq *src_rq, - struct task_struct *t, - struct mm_cid *src_pcpu_cid) -{ - struct mm_struct *mm = t->mm; - struct task_struct *src_task; - int src_cid, last_mm_cid; + /* Calculate the new optimal constraint */ + opt_cids = min(mc->nr_cpus_allowed, mc->users); - if (!mm) - return -1; - - last_mm_cid = t->last_mm_cid; - /* - * If the migrated task has no last cid, or if the current - * task on src rq uses the cid, it means the source cid does not need - * to be moved to the destination cpu. - */ - if (last_mm_cid == -1) - return -1; - src_cid = READ_ONCE(src_pcpu_cid->cid); - if (!mm_cid_is_valid(src_cid) || last_mm_cid != src_cid) - return -1; + /* Adjust the maximum CIDs to +25% limited by the number of possible CPUs */ + max_cids = min(opt_cids + (opt_cids / 4), num_possible_cpus()); + WRITE_ONCE(mc->max_cids, max_cids); +} - /* - * If we observe an active task using the mm on this rq, it means we - * are not the last task to be migrated from this cpu for this mm, so - * there is no need to move src_cid to the destination cpu. - */ - guard(rcu)(); - src_task = rcu_dereference(src_rq->curr); - if (READ_ONCE(src_task->mm_cid_active) && src_task->mm == mm) { - t->last_mm_cid = -1; - return -1; - } +static inline unsigned int mm_cid_calc_pcpu_thrs(struct mm_mm_cid *mc) +{ + unsigned int opt_cids; - return src_cid; + opt_cids = min(mc->nr_cpus_allowed, mc->users); + /* Has to be at least 1 because 0 indicates PCPU mode off */ + return max(min(opt_cids - opt_cids / 4, num_possible_cpus() / 2), 1); } -static -int __sched_mm_cid_migrate_from_try_steal_cid(struct rq *src_rq, - struct task_struct *t, - struct mm_cid *src_pcpu_cid, - int src_cid) +static bool mm_update_max_cids(struct mm_struct *mm) { - struct task_struct *src_task; - struct mm_struct *mm = t->mm; - int lazy_cid; + struct mm_mm_cid *mc = &mm->mm_cid; + bool percpu = cid_on_cpu(mc->mode); - if (src_cid == -1) - return -1; + lockdep_assert_held(&mm->mm_cid.lock); - /* - * Attempt to clear the source cpu cid to move it to the destination - * cpu. - */ - lazy_cid = mm_cid_set_lazy_put(src_cid); - if (!try_cmpxchg(&src_pcpu_cid->cid, &src_cid, lazy_cid)) - return -1; - - /* - * The implicit barrier after cmpxchg per-mm/cpu cid before loading - * rq->curr->mm matches the scheduler barrier in context_switch() - * between store to rq->curr and load of prev and next task's - * per-mm/cpu cid. - * - * The implicit barrier after cmpxchg per-mm/cpu cid before loading - * rq->curr->mm_cid_active matches the barrier in - * sched_mm_cid_exit_signals(), sched_mm_cid_before_execve(), and - * sched_mm_cid_after_execve() between store to t->mm_cid_active and - * load of per-mm/cpu cid. - */ + /* Clear deferred mode switch flag. A change is handled by the caller */ + mc->update_deferred = false; + __mm_update_max_cids(mc); - /* - * If we observe an active task using the mm on this rq after setting - * the lazy-put flag, this task will be responsible for transitioning - * from lazy-put flag set to MM_CID_UNSET. - */ - scoped_guard (rcu) { - src_task = rcu_dereference(src_rq->curr); - if (READ_ONCE(src_task->mm_cid_active) && src_task->mm == mm) { - /* - * We observed an active task for this mm, there is therefore - * no point in moving this cid to the destination cpu. - */ - t->last_mm_cid = -1; - return -1; - } + /* Check whether owner mode must be changed */ + if (!percpu) { + /* Enable per CPU mode when the number of users is above max_cids */ + if (mc->users > mc->max_cids) + mc->pcpu_thrs = mm_cid_calc_pcpu_thrs(mc); + } else { + /* Switch back to per task if user count under threshold */ + if (mc->users < mc->pcpu_thrs) + mc->pcpu_thrs = 0; } + /* Mode change required? */ + if (percpu == !!mc->pcpu_thrs) + return false; + + /* Flip the mode and set the transition flag to bridge the transfer */ + WRITE_ONCE(mc->mode, mc->mode ^ (MM_CID_TRANSIT | MM_CID_ONCPU)); /* - * The src_cid is unused, so it can be unset. + * Order the store against the subsequent fixups so that + * acquire(rq::lock) cannot be reordered by the CPU before the + * store. */ - if (!try_cmpxchg(&src_pcpu_cid->cid, &lazy_cid, MM_CID_UNSET)) - return -1; - WRITE_ONCE(src_pcpu_cid->recent_cid, MM_CID_UNSET); - return src_cid; + smp_mb(); + return true; } -/* - * Migration to dst cpu. Called with dst_rq lock held. - * Interrupts are disabled, which keeps the window of cid ownership without the - * source rq lock held small. - */ -void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) +static inline void mm_update_cpus_allowed(struct mm_struct *mm, const struct cpumask *affmsk) { - struct mm_cid *src_pcpu_cid, *dst_pcpu_cid; - struct mm_struct *mm = t->mm; - int src_cid, src_cpu; - bool dst_cid_is_set; - struct rq *src_rq; - - lockdep_assert_rq_held(dst_rq); + struct cpumask *mm_allowed; + struct mm_mm_cid *mc; + unsigned int weight; - if (!mm) - return; - src_cpu = t->migrate_from_cpu; - if (src_cpu == -1) { - t->last_mm_cid = -1; + if (!mm || !READ_ONCE(mm->mm_cid.users)) return; - } /* - * Move the src cid if the dst cid is unset. This keeps id - * allocation closest to 0 in cases where few threads migrate around - * many CPUs. - * - * If destination cid or recent cid is already set, we may have - * to just clear the src cid to ensure compactness in frequent - * migrations scenarios. - * - * It is not useful to clear the src cid when the number of threads is - * greater or equal to the number of allowed CPUs, because user-space - * can expect that the number of allowed cids can reach the number of - * allowed CPUs. - */ - dst_pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu_of(dst_rq)); - dst_cid_is_set = !mm_cid_is_unset(READ_ONCE(dst_pcpu_cid->cid)) || - !mm_cid_is_unset(READ_ONCE(dst_pcpu_cid->recent_cid)); - if (dst_cid_is_set && atomic_read(&mm->mm_users) >= READ_ONCE(mm->nr_cpus_allowed)) - return; - src_pcpu_cid = per_cpu_ptr(mm->pcpu_cid, src_cpu); - src_rq = cpu_rq(src_cpu); - src_cid = __sched_mm_cid_migrate_from_fetch_cid(src_rq, t, src_pcpu_cid); - if (src_cid == -1) - return; - src_cid = __sched_mm_cid_migrate_from_try_steal_cid(src_rq, t, src_pcpu_cid, - src_cid); - if (src_cid == -1) - return; - if (dst_cid_is_set) { - __mm_cid_put(mm, src_cid); + * mm::mm_cid::mm_cpus_allowed is the superset of each threads + * allowed CPUs mask which means it can only grow. + */ + mc = &mm->mm_cid; + guard(raw_spinlock)(&mc->lock); + mm_allowed = mm_cpus_allowed(mm); + weight = cpumask_weighted_or(mm_allowed, mm_allowed, affmsk); + if (weight == mc->nr_cpus_allowed) return; - } - /* Move src_cid to dst cpu. */ - mm_cid_snapshot_time(dst_rq, mm); - WRITE_ONCE(dst_pcpu_cid->cid, src_cid); - WRITE_ONCE(dst_pcpu_cid->recent_cid, src_cid); -} -static void sched_mm_cid_remote_clear(struct mm_struct *mm, struct mm_cid *pcpu_cid, - int cpu) -{ - struct rq *rq = cpu_rq(cpu); - struct task_struct *t; - int cid, lazy_cid; + WRITE_ONCE(mc->nr_cpus_allowed, weight); + __mm_update_max_cids(mc); + if (!cid_on_cpu(mc->mode)) + return; - cid = READ_ONCE(pcpu_cid->cid); - if (!mm_cid_is_valid(cid)) + /* Adjust the threshold to the wider set */ + mc->pcpu_thrs = mm_cid_calc_pcpu_thrs(mc); + /* Switch back to per task mode? */ + if (mc->users >= mc->pcpu_thrs) return; - /* - * Clear the cpu cid if it is set to keep cid allocation compact. If - * there happens to be other tasks left on the source cpu using this - * mm, the next task using this mm will reallocate its cid on context - * switch. - */ - lazy_cid = mm_cid_set_lazy_put(cid); - if (!try_cmpxchg(&pcpu_cid->cid, &cid, lazy_cid)) + /* Don't queue twice */ + if (mc->update_deferred) return; - /* - * The implicit barrier after cmpxchg per-mm/cpu cid before loading - * rq->curr->mm matches the scheduler barrier in context_switch() - * between store to rq->curr and load of prev and next task's - * per-mm/cpu cid. - * - * The implicit barrier after cmpxchg per-mm/cpu cid before loading - * rq->curr->mm_cid_active matches the barrier in - * sched_mm_cid_exit_signals(), sched_mm_cid_before_execve(), and - * sched_mm_cid_after_execve() between store to t->mm_cid_active and - * load of per-mm/cpu cid. - */ + /* Queue the irq work, which schedules the real work */ + mc->update_deferred = true; + irq_work_queue(&mc->irq_work); +} +static inline void mm_cid_complete_transit(struct mm_struct *mm, unsigned int mode) +{ /* - * If we observe an active task using the mm on this rq after setting - * the lazy-put flag, that task will be responsible for transitioning - * from lazy-put flag set to MM_CID_UNSET. + * Ensure that the store removing the TRANSIT bit cannot be + * reordered by the CPU before the fixups have been completed. */ - scoped_guard (rcu) { - t = rcu_dereference(rq->curr); - if (READ_ONCE(t->mm_cid_active) && t->mm == mm) - return; - } + smp_mb(); + WRITE_ONCE(mm->mm_cid.mode, mode); +} - /* - * The cid is unused, so it can be unset. - * Disable interrupts to keep the window of cid ownership without rq - * lock small. - */ - scoped_guard (irqsave) { - if (try_cmpxchg(&pcpu_cid->cid, &lazy_cid, MM_CID_UNSET)) - __mm_cid_put(mm, cid); +static inline void mm_cid_transit_to_task(struct task_struct *t, struct mm_cid_pcpu *pcp) +{ + if (cid_on_cpu(t->mm_cid.cid)) { + unsigned int cid = cpu_cid_to_cid(t->mm_cid.cid); + + t->mm_cid.cid = cid_to_transit_cid(cid); + pcp->cid = t->mm_cid.cid; } } -static void sched_mm_cid_remote_clear_old(struct mm_struct *mm, int cpu) +static void mm_cid_fixup_cpus_to_tasks(struct mm_struct *mm) { - struct rq *rq = cpu_rq(cpu); - struct mm_cid *pcpu_cid; - struct task_struct *curr; - u64 rq_clock; + unsigned int cpu; - /* - * rq->clock load is racy on 32-bit but one spurious clear once in a - * while is irrelevant. - */ - rq_clock = READ_ONCE(rq->clock); - pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu); + /* Walk the CPUs and fixup all stale CIDs */ + for_each_possible_cpu(cpu) { + struct mm_cid_pcpu *pcp = per_cpu_ptr(mm->mm_cid.pcpu, cpu); + struct rq *rq = cpu_rq(cpu); - /* - * In order to take care of infrequently scheduled tasks, bump the time - * snapshot associated with this cid if an active task using the mm is - * observed on this rq. - */ - scoped_guard (rcu) { - curr = rcu_dereference(rq->curr); - if (READ_ONCE(curr->mm_cid_active) && curr->mm == mm) { - WRITE_ONCE(pcpu_cid->time, rq_clock); - return; + /* Remote access to mm::mm_cid::pcpu requires rq_lock */ + guard(rq_lock_irq)(rq); + /* Is the CID still owned by the CPU? */ + if (cid_on_cpu(pcp->cid)) { + /* + * If rq->curr has @mm, transfer it with the + * transition bit set. Otherwise drop it. + */ + if (rq->curr->mm == mm && rq->curr->mm_cid.active) + mm_cid_transit_to_task(rq->curr, pcp); + else + mm_drop_cid_on_cpu(mm, pcp); + + } else if (rq->curr->mm == mm && rq->curr->mm_cid.active) { + unsigned int cid = rq->curr->mm_cid.cid; + + /* Ensure it has the transition bit set */ + if (!cid_in_transit(cid)) { + cid = cid_to_transit_cid(cid); + rq->curr->mm_cid.cid = cid; + pcp->cid = cid; + } } } - - if (rq_clock < pcpu_cid->time + SCHED_MM_CID_PERIOD_NS) - return; - sched_mm_cid_remote_clear(mm, pcpu_cid, cpu); + mm_cid_complete_transit(mm, 0); } -static void sched_mm_cid_remote_clear_weight(struct mm_struct *mm, int cpu, - int weight) +static inline void mm_cid_transit_to_cpu(struct task_struct *t, struct mm_cid_pcpu *pcp) { - struct mm_cid *pcpu_cid; - int cid; - - pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu); - cid = READ_ONCE(pcpu_cid->cid); - if (!mm_cid_is_valid(cid) || cid < weight) - return; - sched_mm_cid_remote_clear(mm, pcpu_cid, cpu); + if (cid_on_task(t->mm_cid.cid)) { + t->mm_cid.cid = cid_to_transit_cid(t->mm_cid.cid); + pcp->cid = t->mm_cid.cid; + } } -static void task_mm_cid_work(struct callback_head *work) +static void mm_cid_fixup_task_to_cpu(struct task_struct *t, struct mm_struct *mm) { - unsigned long now = jiffies, old_scan, next_scan; - struct task_struct *t = current; - struct cpumask *cidmask; - struct mm_struct *mm; - int weight, cpu; - - SCHED_WARN_ON(t != container_of(work, struct task_struct, cid_work)); - - work->next = work; /* Prevent double-add */ - if (t->flags & PF_EXITING) - return; - mm = t->mm; - if (!mm) - return; - old_scan = READ_ONCE(mm->mm_cid_next_scan); - next_scan = now + msecs_to_jiffies(MM_CID_SCAN_DELAY); - if (!old_scan) { - unsigned long res; - - res = cmpxchg(&mm->mm_cid_next_scan, old_scan, next_scan); - if (res != old_scan) - old_scan = res; + /* Remote access to mm::mm_cid::pcpu requires rq_lock */ + guard(task_rq_lock)(t); + if (cid_on_task(t->mm_cid.cid)) { + /* If running on the CPU, put the CID in transit mode, otherwise drop it */ + if (task_rq(t)->curr == t) + mm_cid_transit_to_cpu(t, per_cpu_ptr(mm->mm_cid.pcpu, task_cpu(t))); else - old_scan = next_scan; + mm_unset_cid_on_task(t); } - if (time_before(now, old_scan)) - return; - if (!try_cmpxchg(&mm->mm_cid_next_scan, &old_scan, next_scan)) - return; - cidmask = mm_cidmask(mm); - /* Clear cids that were not recently used. */ - for_each_possible_cpu(cpu) - sched_mm_cid_remote_clear_old(mm, cpu); - weight = cpumask_weight(cidmask); - /* - * Clear cids that are greater or equal to the cidmask weight to - * recompact it. - */ - for_each_possible_cpu(cpu) - sched_mm_cid_remote_clear_weight(mm, cpu, weight); } -void init_sched_mm_cid(struct task_struct *t) +static void mm_cid_fixup_tasks_to_cpus(void) { - struct mm_struct *mm = t->mm; - int mm_users = 0; + struct mm_struct *mm = current->mm; + struct task_struct *t; - if (mm) { - mm_users = atomic_read(&mm->mm_users); - if (mm_users == 1) - mm->mm_cid_next_scan = jiffies + msecs_to_jiffies(MM_CID_SCAN_DELAY); + lockdep_assert_held(&mm->mm_cid.mutex); + + hlist_for_each_entry(t, &mm->mm_cid.user_list, mm_cid.node) { + /* Current has already transferred before invoking the fixup. */ + if (t != current) + mm_cid_fixup_task_to_cpu(t, mm); } - t->cid_work.next = &t->cid_work; /* Protect against double add */ - init_task_work(&t->cid_work, task_mm_cid_work); + + mm_cid_complete_transit(mm, MM_CID_ONCPU); } -void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) +static bool sched_mm_cid_add_user(struct task_struct *t, struct mm_struct *mm) { - struct callback_head *work = &curr->cid_work; - unsigned long now = jiffies; + lockdep_assert_held(&mm->mm_cid.lock); - if (!curr->mm || (curr->flags & (PF_EXITING | PF_KTHREAD)) || - work->next != work) - return; - if (time_before(now, READ_ONCE(curr->mm->mm_cid_next_scan))) - return; - - /* No page allocation under rq lock */ - task_work_add(curr, work, TWA_RESUME); + t->mm_cid.active = 1; + hlist_add_head(&t->mm_cid.node, &mm->mm_cid.user_list); + mm->mm_cid.users++; + return mm_update_max_cids(mm); } -void sched_mm_cid_exit_signals(struct task_struct *t) +static void sched_mm_cid_fork(struct task_struct *t) { struct mm_struct *mm = t->mm; - struct rq *rq; + bool percpu; if (!mm) return; - preempt_disable(); - rq = this_rq(); - guard(rq_lock_irqsave)(rq); - preempt_enable_no_resched(); /* holding spinlock */ - WRITE_ONCE(t->mm_cid_active, 0); + WARN_ON_ONCE(t->mm_cid.cid != MM_CID_UNSET); + + guard(mutex)(&mm->mm_cid.mutex); + scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) { + struct mm_cid_pcpu *pcp = this_cpu_ptr(mm->mm_cid.pcpu); + + /* First user ? */ + if (!mm->mm_cid.users) { + sched_mm_cid_add_user(t, mm); + t->mm_cid.cid = mm_get_cid(mm); + /* Required for execve() */ + pcp->cid = t->mm_cid.cid; + return; + } + + if (!sched_mm_cid_add_user(t, mm)) { + if (!cid_on_cpu(mm->mm_cid.mode)) + t->mm_cid.cid = mm_get_cid(mm); + return; + } + + /* Handle the mode change and transfer current's CID */ + percpu = cid_on_cpu(mm->mm_cid.mode); + if (!percpu) + mm_cid_transit_to_task(current, pcp); + else + mm_cid_transit_to_cpu(current, pcp); + } + + if (percpu) { + mm_cid_fixup_tasks_to_cpus(); + } else { + mm_cid_fixup_cpus_to_tasks(mm); + t->mm_cid.cid = mm_get_cid(mm); + } +} + +static bool sched_mm_cid_remove_user(struct task_struct *t) +{ + lockdep_assert_held(&t->mm->mm_cid.lock); + + t->mm_cid.active = 0; + /* Clear the transition bit */ + t->mm_cid.cid = cid_from_transit_cid(t->mm_cid.cid); + mm_unset_cid_on_task(t); + hlist_del_init(&t->mm_cid.node); + t->mm->mm_cid.users--; + return mm_update_max_cids(t->mm); +} + +static bool __sched_mm_cid_exit(struct task_struct *t) +{ + struct mm_struct *mm = t->mm; + + if (!sched_mm_cid_remove_user(t)) + return false; /* - * Store t->mm_cid_active before loading per-mm/cpu cid. - * Matches barrier in sched_mm_cid_remote_clear_old(). + * Contrary to fork() this only deals with a switch back to per + * task mode either because the above decreased users or an + * affinity change increased the number of allowed CPUs and the + * deferred fixup did not run yet. */ - smp_mb(); - mm_cid_put(mm); - t->last_mm_cid = t->mm_cid = -1; + if (WARN_ON_ONCE(cid_on_cpu(mm->mm_cid.mode))) + return false; + /* + * A failed fork(2) cleanup never gets here, so @current must have + * the same MM as @t. That's true for exit() and the failed + * pthread_create() cleanup case. + */ + if (WARN_ON_ONCE(current->mm != mm)) + return false; + return true; } -void sched_mm_cid_before_execve(struct task_struct *t) +/* + * When a task exits, the MM CID held by the task is not longer required as + * the task cannot return to user space. + */ +void sched_mm_cid_exit(struct task_struct *t) { struct mm_struct *mm = t->mm; - struct rq *rq; - if (!mm) + if (!mm || !t->mm_cid.active) return; + /* + * Ensure that only one instance is doing MM CID operations within + * a MM. The common case is uncontended. The rare fixup case adds + * some overhead. + */ + scoped_guard(mutex, &mm->mm_cid.mutex) { + /* mm_cid::mutex is sufficient to protect mm_cid::users */ + if (likely(mm->mm_cid.users > 1)) { + scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) { + if (!__sched_mm_cid_exit(t)) + return; + /* + * Mode change. The task has the CID unset + * already and dealt with an eventually set + * TRANSIT bit. If the CID is owned by the CPU + * then drop it. + */ + mm_drop_cid_on_cpu(mm, this_cpu_ptr(mm->mm_cid.pcpu)); + } + mm_cid_fixup_cpus_to_tasks(mm); + return; + } + /* Last user */ + scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) { + /* Required across execve() */ + if (t == current) + mm_cid_transit_to_task(t, this_cpu_ptr(mm->mm_cid.pcpu)); + /* Ignore mode change. There is nothing to do. */ + sched_mm_cid_remove_user(t); + } + } - preempt_disable(); - rq = this_rq(); - guard(rq_lock_irqsave)(rq); - preempt_enable_no_resched(); /* holding spinlock */ - WRITE_ONCE(t->mm_cid_active, 0); /* - * Store t->mm_cid_active before loading per-mm/cpu cid. - * Matches barrier in sched_mm_cid_remote_clear_old(). + * As this is the last user (execve(), process exit or failed + * fork(2)) there is no concurrency anymore. + * + * Synchronize eventually pending work to ensure that there are no + * dangling references left. @t->mm_cid.users is zero so nothing + * can queue this work anymore. */ - smp_mb(); - mm_cid_put(mm); - t->last_mm_cid = t->mm_cid = -1; + irq_work_sync(&mm->mm_cid.irq_work); + cancel_work_sync(&mm->mm_cid.work); +} + +/* Deactivate MM CID allocation across execve() */ +void sched_mm_cid_before_execve(struct task_struct *t) +{ + sched_mm_cid_exit(t); } +/* Reactivate MM CID after execve() */ void sched_mm_cid_after_execve(struct task_struct *t) { - struct mm_struct *mm = t->mm; - struct rq *rq; + if (t->mm) + sched_mm_cid_fork(t); +} - if (!mm) +static void mm_cid_work_fn(struct work_struct *work) +{ + struct mm_struct *mm = container_of(work, struct mm_struct, mm_cid.work); + + guard(mutex)(&mm->mm_cid.mutex); + /* Did the last user task exit already? */ + if (!mm->mm_cid.users) return; - preempt_disable(); - rq = this_rq(); - scoped_guard (rq_lock_irqsave, rq) { - preempt_enable_no_resched(); /* holding spinlock */ - WRITE_ONCE(t->mm_cid_active, 1); - /* - * Store t->mm_cid_active before loading per-mm/cpu cid. - * Matches barrier in sched_mm_cid_remote_clear_old(). - */ - smp_mb(); - t->last_mm_cid = t->mm_cid = mm_cid_get(rq, t, mm); + scoped_guard(raw_spinlock_irq, &mm->mm_cid.lock) { + /* Have fork() or exit() handled it already? */ + if (!mm->mm_cid.update_deferred) + return; + /* This clears mm_cid::update_deferred */ + if (!mm_update_max_cids(mm)) + return; + /* Affinity changes can only switch back to task mode */ + if (WARN_ON_ONCE(cid_on_cpu(mm->mm_cid.mode))) + return; } - rseq_set_notify_resume(t); + mm_cid_fixup_cpus_to_tasks(mm); } -void sched_mm_cid_fork(struct task_struct *t) +static void mm_cid_irq_work(struct irq_work *work) { - WARN_ON_ONCE(!t->mm || t->mm_cid != -1); - t->mm_cid_active = 1; + struct mm_struct *mm = container_of(work, struct mm_struct, mm_cid.irq_work); + + /* + * Needs to be unconditional because mm_cid::lock cannot be held + * when scheduling work as mm_update_cpus_allowed() nests inside + * rq::lock and schedule_work() might end up in wakeup... + */ + schedule_work(&mm->mm_cid.work); } -#endif -#ifdef CONFIG_SCHED_CLASS_EXT -void sched_deq_and_put_task(struct task_struct *p, int queue_flags, - struct sched_enq_and_set_ctx *ctx) +void mm_init_cid(struct mm_struct *mm, struct task_struct *p) { + mm->mm_cid.max_cids = 0; + mm->mm_cid.mode = 0; + mm->mm_cid.nr_cpus_allowed = p->nr_cpus_allowed; + mm->mm_cid.users = 0; + mm->mm_cid.pcpu_thrs = 0; + mm->mm_cid.update_deferred = 0; + raw_spin_lock_init(&mm->mm_cid.lock); + mutex_init(&mm->mm_cid.mutex); + mm->mm_cid.irq_work = IRQ_WORK_INIT_HARD(mm_cid_irq_work); + INIT_WORK(&mm->mm_cid.work, mm_cid_work_fn); + INIT_HLIST_HEAD(&mm->mm_cid.user_list); + cpumask_copy(mm_cpus_allowed(mm), &p->cpus_mask); + bitmap_zero(mm_cidmask(mm), num_possible_cpus()); +} +#else /* CONFIG_SCHED_MM_CID */ +static inline void mm_update_cpus_allowed(struct mm_struct *mm, const struct cpumask *affmsk) { } +static inline void sched_mm_cid_fork(struct task_struct *t) { } +#endif /* !CONFIG_SCHED_MM_CID */ + +static DEFINE_PER_CPU(struct sched_change_ctx, sched_change_ctx); + +struct sched_change_ctx *sched_change_begin(struct task_struct *p, unsigned int flags) +{ + struct sched_change_ctx *ctx = this_cpu_ptr(&sched_change_ctx); struct rq *rq = task_rq(p); + /* + * Must exclusively use matched flags since this is both dequeue and + * enqueue. + */ + WARN_ON_ONCE(flags & 0xFFFF0000); + lockdep_assert_rq_held(rq); - *ctx = (struct sched_enq_and_set_ctx){ + if (!(flags & DEQUEUE_NOCLOCK)) { + update_rq_clock(rq); + flags |= DEQUEUE_NOCLOCK; + } + + if ((flags & DEQUEUE_CLASS) && p->sched_class->switching_from) + p->sched_class->switching_from(rq, p); + + *ctx = (struct sched_change_ctx){ .p = p, - .queue_flags = queue_flags, + .class = p->sched_class, + .flags = flags, .queued = task_on_rq_queued(p), - .running = task_current(rq, p), + .running = task_current_donor(rq, p), }; - update_rq_clock(rq); + if (!(flags & DEQUEUE_CLASS)) { + if (p->sched_class->get_prio) + ctx->prio = p->sched_class->get_prio(rq, p); + else + ctx->prio = p->prio; + } + if (ctx->queued) - dequeue_task(rq, p, queue_flags | DEQUEUE_NOCLOCK); + dequeue_task(rq, p, flags); if (ctx->running) put_prev_task(rq, p); + + if ((flags & DEQUEUE_CLASS) && p->sched_class->switched_from) + p->sched_class->switched_from(rq, p); + + return ctx; } -void sched_enq_and_set_task(struct sched_enq_and_set_ctx *ctx) +void sched_change_end(struct sched_change_ctx *ctx) { - struct rq *rq = task_rq(ctx->p); + struct task_struct *p = ctx->p; + struct rq *rq = task_rq(p); lockdep_assert_rq_held(rq); + /* + * Changing class without *QUEUE_CLASS is bad. + */ + WARN_ON_ONCE(p->sched_class != ctx->class && !(ctx->flags & ENQUEUE_CLASS)); + + if ((ctx->flags & ENQUEUE_CLASS) && p->sched_class->switching_to) + p->sched_class->switching_to(rq, p); + if (ctx->queued) - enqueue_task(rq, ctx->p, ctx->queue_flags | ENQUEUE_NOCLOCK); + enqueue_task(rq, p, ctx->flags); if (ctx->running) - set_next_task(rq, ctx->p); + set_next_task(rq, p); + + if (ctx->flags & ENQUEUE_CLASS) { + if (p->sched_class->switched_to) + p->sched_class->switched_to(rq, p); + + if (ctx->running) { + /* + * If this was a class promotion; let the old class + * know it got preempted. Note that none of the + * switch*_from() methods know the new class and none + * of the switch*_to() methods know the old class. + */ + if (sched_class_above(p->sched_class, ctx->class)) { + rq->next_class->wakeup_preempt(rq, p, 0); + rq->next_class = p->sched_class; + } + /* + * If this was a degradation in class; make sure to + * reschedule. + */ + if (sched_class_above(ctx->class, p->sched_class)) + resched_curr(rq); + } + } else { + p->sched_class->prio_changed(rq, p, ctx->prio); + } } -#endif /* CONFIG_SCHED_CLASS_EXT */ diff --git a/kernel/sched/core_sched.c b/kernel/sched/core_sched.c index 1ef98a93eb1d..73b6b2426911 100644 --- a/kernel/sched/core_sched.c +++ b/kernel/sched/core_sched.c @@ -4,13 +4,15 @@ * A simple wrapper around refcount. An allocated sched_core_cookie's * address is used to compute the cookie of the task. */ +#include "sched.h" + struct sched_core_cookie { refcount_t refcnt; }; static unsigned long sched_core_alloc_cookie(void) { - struct sched_core_cookie *ck = kmalloc(sizeof(*ck), GFP_KERNEL); + struct sched_core_cookie *ck = kmalloc_obj(*ck); if (!ck) return 0; @@ -65,7 +67,7 @@ static unsigned long sched_core_update_cookie(struct task_struct *p, * a cookie until after we've removed it, we must have core scheduling * enabled here. */ - SCHED_WARN_ON((p->core_cookie || cookie) && !sched_core_enabled(rq)); + WARN_ON_ONCE((p->core_cookie || cookie) && !sched_core_enabled(rq)); if (sched_core_enqueued(p)) sched_core_dequeue(rq, p, DEQUEUE_SAVE); diff --git a/kernel/sched/cpuacct.c b/kernel/sched/cpuacct.c index 0de9dda09949..ca9d52cb1ebb 100644 --- a/kernel/sched/cpuacct.c +++ b/kernel/sched/cpuacct.c @@ -6,6 +6,8 @@ * Based on the work by Paul Menage (menage@google.com) and Balbir Singh * (balbir@in.ibm.com). */ +#include <linux/sched/cputime.h> +#include "sched.h" /* Time spent by the tasks of the CPU accounting group executing in ... */ enum cpuacct_stat_index { @@ -59,7 +61,7 @@ cpuacct_css_alloc(struct cgroup_subsys_state *parent_css) if (!parent_css) return &root_cpuacct.css; - ca = kzalloc(sizeof(*ca), GFP_KERNEL); + ca = kzalloc_obj(*ca); if (!ca) goto out; diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c index 95baa12a1029..0a2b7e30fd10 100644 --- a/kernel/sched/cpudeadline.c +++ b/kernel/sched/cpudeadline.c @@ -6,6 +6,7 @@ * * Author: Juri Lelli <j.lelli@sssup.it> */ +#include "sched.h" static inline int parent(int i) { @@ -165,12 +166,13 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p, * cpudl_clear - remove a CPU from the cpudl max-heap * @cp: the cpudl max-heap context * @cpu: the target CPU + * @online: the online state of the deadline runqueue * * Notes: assumes cpu_rq(cpu)->lock is locked * * Returns: (void) */ -void cpudl_clear(struct cpudl *cp, int cpu) +void cpudl_clear(struct cpudl *cp, int cpu, bool online) { int old_idx, new_cpu; unsigned long flags; @@ -183,7 +185,7 @@ void cpudl_clear(struct cpudl *cp, int cpu) if (old_idx == IDX_INVALID) { /* * Nothing to remove if old_idx was invalid. - * This could happen if a rq_offline_dl is + * This could happen if rq_online_dl or rq_offline_dl is * called for a CPU without -dl tasks running. */ } else { @@ -194,9 +196,12 @@ void cpudl_clear(struct cpudl *cp, int cpu) cp->elements[new_cpu].idx = old_idx; cp->elements[cpu].idx = IDX_INVALID; cpudl_heapify(cp, old_idx); - - cpumask_set_cpu(cpu, cp->free_cpus); } + if (likely(online)) + __cpumask_set_cpu(cpu, cp->free_cpus); + else + __cpumask_clear_cpu(cpu, cp->free_cpus); + raw_spin_unlock_irqrestore(&cp->lock, flags); } @@ -227,7 +232,7 @@ void cpudl_set(struct cpudl *cp, int cpu, u64 dl) cp->elements[new_idx].cpu = cpu; cp->elements[cpu].idx = new_idx; cpudl_heapify_up(cp, new_idx); - cpumask_clear_cpu(cpu, cp->free_cpus); + __cpumask_clear_cpu(cpu, cp->free_cpus); } else { cp->elements[old_idx].dl = dl; cpudl_heapify(cp, old_idx); @@ -237,26 +242,6 @@ void cpudl_set(struct cpudl *cp, int cpu, u64 dl) } /* - * cpudl_set_freecpu - Set the cpudl.free_cpus - * @cp: the cpudl max-heap context - * @cpu: rd attached CPU - */ -void cpudl_set_freecpu(struct cpudl *cp, int cpu) -{ - cpumask_set_cpu(cpu, cp->free_cpus); -} - -/* - * cpudl_clear_freecpu - Clear the cpudl.free_cpus - * @cp: the cpudl max-heap context - * @cpu: rd attached CPU - */ -void cpudl_clear_freecpu(struct cpudl *cp, int cpu) -{ - cpumask_clear_cpu(cpu, cp->free_cpus); -} - -/* * cpudl_init - initialize the cpudl structure * @cp: the cpudl max-heap context */ @@ -267,9 +252,7 @@ int cpudl_init(struct cpudl *cp) raw_spin_lock_init(&cp->lock); cp->size = 0; - cp->elements = kcalloc(nr_cpu_ids, - sizeof(struct cpudl_item), - GFP_KERNEL); + cp->elements = kzalloc_objs(struct cpudl_item, nr_cpu_ids); if (!cp->elements) return -ENOMEM; diff --git a/kernel/sched/cpudeadline.h b/kernel/sched/cpudeadline.h index 0adeda93b5fb..d7699468eedd 100644 --- a/kernel/sched/cpudeadline.h +++ b/kernel/sched/cpudeadline.h @@ -1,4 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/types.h> +#include <linux/spinlock.h> #define IDX_INVALID -1 @@ -15,12 +17,8 @@ struct cpudl { struct cpudl_item *elements; }; -#ifdef CONFIG_SMP int cpudl_find(struct cpudl *cp, struct task_struct *p, struct cpumask *later_mask); void cpudl_set(struct cpudl *cp, int cpu, u64 dl); -void cpudl_clear(struct cpudl *cp, int cpu); +void cpudl_clear(struct cpudl *cp, int cpu, bool online); int cpudl_init(struct cpudl *cp); -void cpudl_set_freecpu(struct cpudl *cp, int cpu); -void cpudl_clear_freecpu(struct cpudl *cp, int cpu); void cpudl_cleanup(struct cpudl *cp); -#endif /* CONFIG_SMP */ diff --git a/kernel/sched/cpufreq.c b/kernel/sched/cpufreq.c index 5252fb191fae..742fb9e62e1a 100644 --- a/kernel/sched/cpufreq.c +++ b/kernel/sched/cpufreq.c @@ -5,6 +5,7 @@ * Copyright (C) 2016, Intel Corporation * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com> */ +#include "sched.h" DEFINE_PER_CPU(struct update_util_data __rcu *, cpufreq_update_util_data); diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 1a19d69b91ed..ae9fd211cec1 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -5,6 +5,8 @@ * Copyright (C) 2016, Intel Corporation * Author: Rafael J. Wysocki <rafael.j.wysocki@intel.com> */ +#include <uapi/linux/sched/types.h> +#include "sched.h" #define IOWAIT_BOOST_MIN (SCHED_CAPACITY_SCALE / 8) @@ -81,9 +83,23 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time) if (!cpufreq_this_cpu_can_update(sg_policy->policy)) return false; - if (unlikely(sg_policy->limits_changed)) { - sg_policy->limits_changed = false; - sg_policy->need_freq_update = cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS); + if (unlikely(READ_ONCE(sg_policy->limits_changed))) { + WRITE_ONCE(sg_policy->limits_changed, false); + sg_policy->need_freq_update = true; + + /* + * The above limits_changed update must occur before the reads + * of policy limits in cpufreq_driver_resolve_freq() or a policy + * limits update might be missed, so use a memory barrier to + * ensure it. + * + * This pairs with the write memory barrier in sugov_limits(). + */ + smp_mb(); + + return true; + } else if (sg_policy->need_freq_update) { + /* ignore_dl_rate_limit() wants a new frequency to be found. */ return true; } @@ -95,10 +111,22 @@ static bool sugov_should_update_freq(struct sugov_policy *sg_policy, u64 time) static bool sugov_update_next_freq(struct sugov_policy *sg_policy, u64 time, unsigned int next_freq) { - if (sg_policy->need_freq_update) + if (sg_policy->need_freq_update) { sg_policy->need_freq_update = false; - else if (sg_policy->next_freq == next_freq) + /* + * The policy limits have changed, but if the return value of + * cpufreq_driver_resolve_freq() after applying the new limits + * is still equal to the previously selected frequency, the + * driver callback need not be invoked unless the driver + * specifically wants that to happen on every update of the + * policy limits. + */ + if (sg_policy->next_freq == next_freq && + !cpufreq_driver_test_flags(CPUFREQ_NEED_UPDATE_LIMITS)) + return false; + } else if (sg_policy->next_freq == next_freq) { return false; + } sg_policy->next_freq = next_freq; sg_policy->last_freq_update_time = time; @@ -354,9 +382,9 @@ static bool sugov_hold_freq(struct sugov_cpu *sg_cpu) sg_cpu->saved_idle_calls = idle_calls; return ret; } -#else +#else /* !CONFIG_NO_HZ_COMMON: */ static inline bool sugov_hold_freq(struct sugov_cpu *sg_cpu) { return false; } -#endif /* CONFIG_NO_HZ_COMMON */ +#endif /* !CONFIG_NO_HZ_COMMON */ /* * Make sugov_should_update_freq() ignore the rate limit when DL @@ -365,7 +393,7 @@ static inline bool sugov_hold_freq(struct sugov_cpu *sg_cpu) { return false; } static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu) { if (cpu_bw_dl(cpu_rq(sg_cpu->cpu)) > sg_cpu->bw_min) - sg_cpu->sg_policy->limits_changed = true; + sg_cpu->sg_policy->need_freq_update = true; } static inline bool sugov_update_single_common(struct sugov_cpu *sg_cpu, @@ -433,6 +461,7 @@ static void sugov_update_single_perf(struct update_util_data *hook, u64 time, unsigned int flags) { struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util); + struct sugov_policy *sg_policy = sg_cpu->sg_policy; unsigned long prev_util = sg_cpu->util; unsigned long max_cap; @@ -454,10 +483,10 @@ static void sugov_update_single_perf(struct update_util_data *hook, u64 time, if (sugov_hold_freq(sg_cpu) && sg_cpu->util < prev_util) sg_cpu->util = prev_util; - cpufreq_driver_adjust_perf(sg_cpu->cpu, sg_cpu->bw_min, + cpufreq_driver_adjust_perf(sg_policy->policy, sg_cpu->bw_min, sg_cpu->util, max_cap); - sg_cpu->sg_policy->last_freq_update_time = time; + sg_policy->last_freq_update_time = time; } static unsigned int sugov_next_freq_shared(struct sugov_cpu *sg_cpu, u64 time) @@ -604,13 +633,13 @@ static const struct kobj_type sugov_tunables_ktype = { /********************** cpufreq governor interface *********************/ -struct cpufreq_governor schedutil_gov; +static struct cpufreq_governor schedutil_gov; static struct sugov_policy *sugov_policy_alloc(struct cpufreq_policy *policy) { struct sugov_policy *sg_policy; - sg_policy = kzalloc(sizeof(*sg_policy), GFP_KERNEL); + sg_policy = kzalloc_obj(*sg_policy); if (!sg_policy) return NULL; @@ -654,7 +683,7 @@ static int sugov_kthread_create(struct sugov_policy *sg_policy) "sugov:%d", cpumask_first(policy->related_cpus)); if (IS_ERR(thread)) { - pr_err("failed to create sugov thread: %ld\n", PTR_ERR(thread)); + pr_err("failed to create sugov thread: %pe\n", thread); return PTR_ERR(thread); } @@ -694,7 +723,7 @@ static struct sugov_tunables *sugov_tunables_alloc(struct sugov_policy *sg_polic { struct sugov_tunables *tunables; - tunables = kzalloc(sizeof(*tunables), GFP_KERNEL); + tunables = kzalloc_obj(*tunables); if (tunables) { gov_attr_set_init(&tunables->attr_set, &sg_policy->tunables_hook); if (!have_governor_per_policy()) @@ -871,10 +900,19 @@ static void sugov_limits(struct cpufreq_policy *policy) mutex_unlock(&sg_policy->work_lock); } - sg_policy->limits_changed = true; + /* + * The limits_changed update below must take place before the updates + * of policy limits in cpufreq_set_policy() or a policy limits update + * might be missed, so use a memory barrier to ensure it. + * + * This pairs with the memory barrier in sugov_should_update_freq(). + */ + smp_wmb(); + + WRITE_ONCE(sg_policy->limits_changed, true); } -struct cpufreq_governor schedutil_gov = { +static struct cpufreq_governor schedutil_gov = { .name = "schedutil", .owner = THIS_MODULE, .flags = CPUFREQ_GOV_DYNAMIC_SWITCHING, @@ -892,4 +930,9 @@ struct cpufreq_governor *cpufreq_default_governor(void) } #endif +bool sugov_is_governor(struct cpufreq_policy *policy) +{ + return policy->governor == &schedutil_gov; +} + cpufreq_governor_init(schedutil_gov); diff --git a/kernel/sched/cpupri.c b/kernel/sched/cpupri.c index 42c40cfdf836..8f2237e8b484 100644 --- a/kernel/sched/cpupri.c +++ b/kernel/sched/cpupri.c @@ -22,6 +22,7 @@ * worst case complexity of O(min(101, nr_domcpus)), though the scenario that * yields the worst case search is fairly contrived. */ +#include "sched.h" /* * p->rt_priority p->prio newpri cpupri @@ -287,7 +288,7 @@ int cpupri_init(struct cpupri *cp) goto cleanup; } - cp->cpu_to_pri = kcalloc(nr_cpu_ids, sizeof(int), GFP_KERNEL); + cp->cpu_to_pri = kzalloc_objs(int, nr_cpu_ids); if (!cp->cpu_to_pri) goto cleanup; diff --git a/kernel/sched/cpupri.h b/kernel/sched/cpupri.h index d6cba0020064..6f562088c056 100644 --- a/kernel/sched/cpupri.h +++ b/kernel/sched/cpupri.h @@ -1,4 +1,7 @@ /* SPDX-License-Identifier: GPL-2.0 */ +#include <linux/atomic.h> +#include <linux/cpumask.h> +#include <linux/sched/rt.h> #define CPUPRI_NR_PRIORITIES (MAX_RT_PRIO+1) @@ -17,7 +20,6 @@ struct cpupri { int *cpu_to_pri; }; -#ifdef CONFIG_SMP int cpupri_find(struct cpupri *cp, struct task_struct *p, struct cpumask *lowest_mask); int cpupri_find_fitness(struct cpupri *cp, struct task_struct *p, @@ -26,4 +28,3 @@ int cpupri_find_fitness(struct cpupri *cp, struct task_struct *p, void cpupri_set(struct cpupri *cp, int cpu, int pri); int cpupri_init(struct cpupri *cp); void cpupri_cleanup(struct cpupri *cp); -#endif diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 6dab4854c6c0..fbf31db0d2f3 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -2,6 +2,9 @@ /* * Simple CPU accounting cgroup controller */ +#include <linux/sched/cputime.h> +#include <linux/tsacct_kern.h> +#include "sched.h" #ifdef CONFIG_VIRT_CPU_ACCOUNTING_NATIVE #include <asm/cputime.h> @@ -9,6 +12,8 @@ #ifdef CONFIG_IRQ_TIME_ACCOUNTING +DEFINE_STATIC_KEY_FALSE(sched_clock_irqtime); + /* * There are no locks covering percpu hardirq/softirq time. * They are only modified in vtime_account, on corresponding CPU @@ -22,16 +27,15 @@ */ DEFINE_PER_CPU(struct irqtime, cpu_irqtime); -int sched_clock_irqtime; - void enable_sched_clock_irqtime(void) { - sched_clock_irqtime = 1; + static_branch_enable(&sched_clock_irqtime); } void disable_sched_clock_irqtime(void) { - sched_clock_irqtime = 0; + if (irqtime_enabled()) + static_branch_disable(&sched_clock_irqtime); } static void irqtime_account_delta(struct irqtime *irqtime, u64 delta, @@ -88,7 +92,7 @@ static u64 irqtime_tick_accounted(u64 maxtime) return delta; } -#else /* CONFIG_IRQ_TIME_ACCOUNTING */ +#else /* !CONFIG_IRQ_TIME_ACCOUNTING: */ static u64 irqtime_tick_accounted(u64 dummy) { @@ -241,13 +245,26 @@ void __account_forceidle_time(struct task_struct *p, u64 delta) task_group_account_field(p, CPUTIME_FORCEIDLE, delta); } -#endif +#endif /* CONFIG_SCHED_CORE */ /* * When a guest is interrupted for a longer amount of time, missed clock * ticks are not redelivered later. Due to that, this function may on * occasion account more time than the calling functions think elapsed. */ +#ifdef CONFIG_PARAVIRT +struct static_key paravirt_steal_enabled; + +#ifdef CONFIG_HAVE_PV_STEAL_CLOCK_GEN +static u64 native_steal_clock(int cpu) +{ + return 0; +} + +DEFINE_STATIC_CALL(pv_steal_clock, native_steal_clock); +#endif +#endif + static __always_inline u64 steal_account_process_time(u64 maxtime) { #ifdef CONFIG_PARAVIRT @@ -262,7 +279,7 @@ static __always_inline u64 steal_account_process_time(u64 maxtime) return steal; } -#endif +#endif /* CONFIG_PARAVIRT */ return 0; } @@ -288,7 +305,7 @@ static inline u64 read_sum_exec_runtime(struct task_struct *t) { return t->se.sum_exec_runtime; } -#else +#else /* !CONFIG_64BIT: */ static u64 read_sum_exec_runtime(struct task_struct *t) { u64 ns; @@ -301,7 +318,7 @@ static u64 read_sum_exec_runtime(struct task_struct *t) return ns; } -#endif +#endif /* !CONFIG_64BIT */ /* * Accumulate raw cputime values of dead tasks (sig->[us]time) and live @@ -310,10 +327,8 @@ static u64 read_sum_exec_runtime(struct task_struct *t) void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) { struct signal_struct *sig = tsk->signal; - u64 utime, stime; struct task_struct *t; - unsigned int seq, nextseq; - unsigned long flags; + u64 utime, stime; /* * Update current task runtime to account pending time since last @@ -326,27 +341,19 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) if (same_thread_group(current, tsk)) (void) task_sched_runtime(current); - rcu_read_lock(); - /* Attempt a lockless read on the first round. */ - nextseq = 0; - do { - seq = nextseq; - flags = read_seqbegin_or_lock_irqsave(&sig->stats_lock, &seq); + guard(rcu)(); + scoped_seqlock_read (&sig->stats_lock, ss_lock_irqsave) { times->utime = sig->utime; times->stime = sig->stime; times->sum_exec_runtime = sig->sum_sched_runtime; - for_each_thread(tsk, t) { + __for_each_thread(sig, t) { task_cputime(t, &utime, &stime); times->utime += utime; times->stime += stime; times->sum_exec_runtime += read_sum_exec_runtime(t); } - /* If lockless access failed, take the lock. */ - nextseq = 1; - } while (need_seqretry(&sig->stats_lock, seq)); - done_seqretry_irqrestore(&sig->stats_lock, seq, flags); - rcu_read_unlock(); + } } #ifdef CONFIG_IRQ_TIME_ACCOUNTING @@ -411,11 +418,11 @@ static void irqtime_account_idle_ticks(int ticks) { irqtime_account_process_tick(current, 0, ticks); } -#else /* CONFIG_IRQ_TIME_ACCOUNTING */ +#else /* !CONFIG_IRQ_TIME_ACCOUNTING: */ static inline void irqtime_account_idle_ticks(int ticks) { } static inline void irqtime_account_process_tick(struct task_struct *p, int user_tick, int nr_ticks) { } -#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ +#endif /* !CONFIG_IRQ_TIME_ACCOUNTING */ /* * Use precise platform statistics if available: diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index ff4df16b5186..7db4c87df83b 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -17,6 +17,11 @@ */ #include <linux/cpuset.h> +#include <linux/sched/clock.h> +#include <linux/sched/deadline.h> +#include <uapi/linux/sched/types.h> +#include "sched.h" +#include "pelt.h" /* * Default limits for DL period; on the top end we guard against small util @@ -51,18 +56,7 @@ static int __init sched_dl_sysctl_init(void) return 0; } late_initcall(sched_dl_sysctl_init); -#endif - -static bool dl_server(struct sched_dl_entity *dl_se) -{ - return dl_se->dl_server; -} - -static inline struct task_struct *dl_task_of(struct sched_dl_entity *dl_se) -{ - BUG_ON(dl_server(dl_se)); - return container_of(dl_se, struct task_struct, dl); -} +#endif /* CONFIG_SYSCTL */ static inline struct rq *rq_of_dl_rq(struct dl_rq *dl_rq) { @@ -99,7 +93,7 @@ static inline bool is_dl_boosted(struct sched_dl_entity *dl_se) { return pi_of(dl_se) != dl_se; } -#else +#else /* !CONFIG_RT_MUTEXES: */ static inline struct sched_dl_entity *pi_of(struct sched_dl_entity *dl_se) { return dl_se; @@ -109,9 +103,21 @@ static inline bool is_dl_boosted(struct sched_dl_entity *dl_se) { return false; } +#endif /* !CONFIG_RT_MUTEXES */ + +static inline u8 dl_get_type(struct sched_dl_entity *dl_se, struct rq *rq) +{ + if (!dl_server(dl_se)) + return DL_TASK; + if (dl_se == &rq->fair_server) + return DL_SERVER_FAIR; +#ifdef CONFIG_SCHED_CLASS_EXT + if (dl_se == &rq->ext_server) + return DL_SERVER_EXT; #endif + return DL_OTHER; +} -#ifdef CONFIG_SMP static inline struct dl_bw *dl_bw_of(int i) { RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), @@ -122,20 +128,11 @@ static inline struct dl_bw *dl_bw_of(int i) static inline int dl_bw_cpus(int i) { struct root_domain *rd = cpu_rq(i)->rd; - int cpus; RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held(), "sched RCU must be held"); - if (cpumask_subset(rd->span, cpu_active_mask)) - return cpumask_weight(rd->span); - - cpus = 0; - - for_each_cpu_and(i, rd->span, cpu_active_mask) - cpus++; - - return cpus; + return cpumask_weight_and(rd->span, cpu_active_mask); } static inline unsigned long __dl_bw_capacity(const struct cpumask *mask) @@ -166,14 +163,14 @@ static inline unsigned long dl_bw_capacity(int i) } } -static inline bool dl_bw_visited(int cpu, u64 gen) +bool dl_bw_visited(int cpu, u64 cookie) { struct root_domain *rd = cpu_rq(cpu)->rd; - if (rd->visit_gen == gen) + if (rd->visit_cookie == cookie) return true; - rd->visit_gen = gen; + rd->visit_cookie = cookie; return false; } @@ -191,35 +188,6 @@ void __dl_update(struct dl_bw *dl_b, s64 bw) rq->dl.extra_bw += bw; } } -#else -static inline struct dl_bw *dl_bw_of(int i) -{ - return &cpu_rq(i)->dl.dl_bw; -} - -static inline int dl_bw_cpus(int i) -{ - return 1; -} - -static inline unsigned long dl_bw_capacity(int i) -{ - return SCHED_CAPACITY_SCALE; -} - -static inline bool dl_bw_visited(int cpu, u64 gen) -{ - return false; -} - -static inline -void __dl_update(struct dl_bw *dl_b, s64 bw) -{ - struct dl_rq *dl = container_of(dl_b, struct dl_rq, dl_bw); - - dl->extra_bw += bw; -} -#endif static inline void __dl_sub(struct dl_bw *dl_b, u64 tsk_bw, int cpus) @@ -249,8 +217,8 @@ void __add_running_bw(u64 dl_bw, struct dl_rq *dl_rq) lockdep_assert_rq_held(rq_of_dl_rq(dl_rq)); dl_rq->running_bw += dl_bw; - SCHED_WARN_ON(dl_rq->running_bw < old); /* overflow */ - SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw); + WARN_ON_ONCE(dl_rq->running_bw < old); /* overflow */ + WARN_ON_ONCE(dl_rq->running_bw > dl_rq->this_bw); /* kick cpufreq (see the comment in kernel/sched/sched.h). */ cpufreq_update_util(rq_of_dl_rq(dl_rq), 0); } @@ -262,7 +230,7 @@ void __sub_running_bw(u64 dl_bw, struct dl_rq *dl_rq) lockdep_assert_rq_held(rq_of_dl_rq(dl_rq)); dl_rq->running_bw -= dl_bw; - SCHED_WARN_ON(dl_rq->running_bw > old); /* underflow */ + WARN_ON_ONCE(dl_rq->running_bw > old); /* underflow */ if (dl_rq->running_bw > old) dl_rq->running_bw = 0; /* kick cpufreq (see the comment in kernel/sched/sched.h). */ @@ -276,7 +244,7 @@ void __add_rq_bw(u64 dl_bw, struct dl_rq *dl_rq) lockdep_assert_rq_held(rq_of_dl_rq(dl_rq)); dl_rq->this_bw += dl_bw; - SCHED_WARN_ON(dl_rq->this_bw < old); /* overflow */ + WARN_ON_ONCE(dl_rq->this_bw < old); /* overflow */ } static inline @@ -286,10 +254,10 @@ void __sub_rq_bw(u64 dl_bw, struct dl_rq *dl_rq) lockdep_assert_rq_held(rq_of_dl_rq(dl_rq)); dl_rq->this_bw -= dl_bw; - SCHED_WARN_ON(dl_rq->this_bw > old); /* underflow */ + WARN_ON_ONCE(dl_rq->this_bw > old); /* underflow */ if (dl_rq->this_bw > old) dl_rq->this_bw = 0; - SCHED_WARN_ON(dl_rq->running_bw > dl_rq->this_bw); + WARN_ON_ONCE(dl_rq->running_bw > dl_rq->this_bw); } static inline @@ -431,7 +399,7 @@ static void __dl_clear_params(struct sched_dl_entity *dl_se); * up, and checks if the task is still in the "ACTIVE non contending" * state or not (in the second case, it updates running_bw). */ -static void task_non_contending(struct sched_dl_entity *dl_se) +static void task_non_contending(struct sched_dl_entity *dl_se, bool dl_task) { struct hrtimer *timer = &dl_se->inactive_timer; struct rq *rq = rq_of_dl_se(dl_se); @@ -470,10 +438,10 @@ static void task_non_contending(struct sched_dl_entity *dl_se) } else { struct task_struct *p = dl_task_of(dl_se); - if (dl_task(p)) + if (dl_task) sub_running_bw(dl_se, dl_rq); - if (!dl_task(p) || READ_ONCE(p->__state) == TASK_DEAD) { + if (!dl_task || READ_ONCE(p->__state) == TASK_DEAD) { struct dl_bw *dl_b = dl_bw_of(task_cpu(p)); if (READ_ONCE(p->__state) == TASK_DEAD) @@ -552,23 +520,17 @@ void init_dl_rq(struct dl_rq *dl_rq) { dl_rq->root = RB_ROOT_CACHED; -#ifdef CONFIG_SMP /* zero means no -deadline tasks */ dl_rq->earliest_dl.curr = dl_rq->earliest_dl.next = 0; dl_rq->overloaded = 0; dl_rq->pushable_dl_tasks_root = RB_ROOT_CACHED; -#else - init_dl_bw(&dl_rq->dl_bw); -#endif dl_rq->running_bw = 0; dl_rq->this_bw = 0; init_dl_rq_bw_ratio(dl_rq); } -#ifdef CONFIG_SMP - static inline int dl_overloaded(struct rq *rq) { return atomic_read(&rq->rd->dlo_count); @@ -753,37 +715,6 @@ static struct rq *dl_task_offline_migration(struct rq *rq, struct task_struct *p return later_rq; } -#else - -static inline -void enqueue_pushable_dl_task(struct rq *rq, struct task_struct *p) -{ -} - -static inline -void dequeue_pushable_dl_task(struct rq *rq, struct task_struct *p) -{ -} - -static inline -void inc_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) -{ -} - -static inline -void dec_dl_migration(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) -{ -} - -static inline void deadline_queue_push_tasks(struct rq *rq) -{ -} - -static inline void deadline_queue_pull_task(struct rq *rq) -{ -} -#endif /* CONFIG_SMP */ - static void enqueue_dl_entity(struct sched_dl_entity *dl_se, int flags); static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags); @@ -805,6 +736,7 @@ static inline void replenish_dl_new_period(struct sched_dl_entity *dl_se, dl_se->dl_throttled = 1; dl_se->dl_defer_armed = 1; } + trace_sched_dl_replenish_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); } /* @@ -920,6 +852,8 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se) if (dl_se->dl_throttled) dl_se->dl_throttled = 0; + trace_sched_dl_replenish_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); + /* * If this is the replenishment of a deferred reservation, * clear the flag and return. @@ -936,7 +870,7 @@ static void replenish_dl_entity(struct sched_dl_entity *dl_se) */ if (dl_se->dl_defer && !dl_se->dl_defer_running && dl_time_before(rq_clock(dl_se->rq), dl_se->deadline - dl_se->runtime)) { - if (!is_dl_boosted(dl_se) && dl_se->server_has_tasks(dl_se)) { + if (!is_dl_boosted(dl_se)) { /* * Set dl_se->dl_defer_armed and dl_throttled variables to @@ -1047,22 +981,6 @@ update_dl_revised_wakeup(struct sched_dl_entity *dl_se, struct rq *rq) } /* - * Regarding the deadline, a task with implicit deadline has a relative - * deadline == relative period. A task with constrained deadline has a - * relative deadline <= relative period. - * - * We support constrained deadline tasks. However, there are some restrictions - * applied only for tasks which do not have an implicit deadline. See - * update_dl_entity() to know more about such restrictions. - * - * The dl_is_implicit() returns true if the task has an implicit deadline. - */ -static inline bool dl_is_implicit(struct sched_dl_entity *dl_se) -{ - return dl_se->dl_deadline == dl_se->dl_period; -} - -/* * When a deadline entity is placed in the runqueue, its runtime and deadline * might need to be updated. This is done by a CBS wake up rule. There are two * different rules: 1) the original CBS; and 2) the Revisited CBS. @@ -1099,13 +1017,19 @@ static void update_dl_entity(struct sched_dl_entity *dl_se) if (dl_time_before(dl_se->deadline, rq_clock(rq)) || dl_entity_overflow(dl_se, rq_clock(rq))) { - if (unlikely(!dl_is_implicit(dl_se) && + if (unlikely((!dl_is_implicit(dl_se) || dl_se->dl_defer) && !dl_time_before(dl_se->deadline, rq_clock(rq)) && !is_dl_boosted(dl_se))) { update_dl_revised_wakeup(dl_se, rq); return; } + /* + * When [4] D->A is followed by [1] A->B, dl_defer_running + * needs to be cleared, otherwise it will fail to properly + * start the zero-laxity timer. + */ + dl_se->dl_defer_running = 0; replenish_dl_new_period(dl_se, rq); } else if (dl_server(dl_se) && dl_se->dl_defer) { /* @@ -1163,7 +1087,7 @@ static int start_dl_timer(struct sched_dl_entity *dl_se) act = ns_to_ktime(dl_next_period(dl_se)); } - now = hrtimer_cb_get_time(timer); + now = ktime_get(); delta = ktime_to_ns(now) - rq_clock(rq); act = ktime_add_ns(act, delta); @@ -1195,7 +1119,6 @@ static int start_dl_timer(struct sched_dl_entity *dl_se) static void __push_dl_task(struct rq *rq, struct rq_flags *rf) { -#ifdef CONFIG_SMP /* * Queueing this task back might have overloaded rq, check if we need * to kick someone away. @@ -1209,7 +1132,6 @@ static void __push_dl_task(struct rq *rq, struct rq_flags *rf) push_dl_task(rq); rq_repin_lock(rq, rf); } -#endif } /* a defer timer will not be reset if the runtime consumed was < dl_server_min_res */ @@ -1229,11 +1151,15 @@ static enum hrtimer_restart dl_server_timer(struct hrtimer *timer, struct sched_ sched_clock_tick(); update_rq_clock(rq); - if (!dl_se->dl_runtime) - return HRTIMER_NORESTART; + /* + * Make sure current has propagated its pending runtime into + * any relevant server through calling dl_server_update() and + * friends. + */ + rq->donor->sched_class->update_curr(rq); - if (!dl_se->server_has_tasks(dl_se)) { - replenish_dl_entity(dl_se); + if (dl_se->dl_defer_idle) { + dl_server_stop(dl_se); return HRTIMER_NORESTART; } @@ -1339,7 +1265,6 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) goto unlock; } -#ifdef CONFIG_SMP if (unlikely(!rq->online)) { /* * If the runqueue is no longer available, migrate the @@ -1356,7 +1281,6 @@ static enum hrtimer_restart dl_task_timer(struct hrtimer *timer) * there. */ } -#endif enqueue_task_dl(rq, p, ENQUEUE_REPLENISH); if (dl_task(rq->donor)) @@ -1382,8 +1306,7 @@ static void init_dl_task_timer(struct sched_dl_entity *dl_se) { struct hrtimer *timer = &dl_se->dl_timer; - hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); - timer->function = dl_task_timer; + hrtimer_setup(timer, dl_task_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); } /* @@ -1412,6 +1335,7 @@ static inline void dl_check_constrained_dl(struct sched_dl_entity *dl_se) dl_time_before(rq_clock(rq), dl_next_period(dl_se))) { if (unlikely(is_dl_boosted(dl_se) || !start_dl_timer(dl_se))) return; + trace_sched_dl_throttle_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); dl_se->dl_throttled = 1; if (dl_se->runtime > 0) dl_se->runtime = 0; @@ -1487,10 +1411,11 @@ s64 dl_scaled_delta_exec(struct rq *rq, struct sched_dl_entity *dl_se, s64 delta } static inline void -update_stats_dequeue_dl(struct dl_rq *dl_rq, struct sched_dl_entity *dl_se, - int flags); +update_stats_dequeue_dl(struct dl_rq *dl_rq, struct sched_dl_entity *dl_se, int flags); + static void update_curr_dl_se(struct rq *rq, struct sched_dl_entity *dl_se, s64 delta_exec) { + bool idle = idle_rq(rq); s64 scaled_delta_exec; if (unlikely(delta_exec <= 0)) { @@ -1505,13 +1430,18 @@ static void update_curr_dl_se(struct rq *rq, struct sched_dl_entity *dl_se, s64 if (dl_entity_is_special(dl_se)) return; - scaled_delta_exec = dl_scaled_delta_exec(rq, dl_se, delta_exec); + scaled_delta_exec = delta_exec; + if (!dl_server(dl_se)) + scaled_delta_exec = dl_scaled_delta_exec(rq, dl_se, delta_exec); dl_se->runtime -= scaled_delta_exec; + if (dl_se->dl_defer_idle && !idle) + dl_se->dl_defer_idle = 0; + /* - * The fair server can consume its runtime while throttled (not queued/ - * running as regular CFS). + * The DL server can consume its runtime while throttled (not + * queued / running as regular CFS). * * If the server consumes its entire runtime in this state. The server * is not required for the current period. Thus, reset the server by @@ -1519,6 +1449,29 @@ static void update_curr_dl_se(struct rq *rq, struct sched_dl_entity *dl_se, s64 */ if (dl_se->dl_defer && dl_se->dl_throttled && dl_runtime_exceeded(dl_se)) { /* + * Non-servers would never get time accounted while throttled. + */ + WARN_ON_ONCE(!dl_server(dl_se)); + + /* + * While the server is marked idle, do not push out the + * activation further, instead wait for the period timer + * to lapse and stop the server. + */ + if (dl_se->dl_defer_idle && idle) { + /* + * The timer is at the zero-laxity point, this means + * dl_server_stop() / dl_server_start() can happen + * while now < deadline. This means update_dl_entity() + * will not replenish. Additionally start_dl_timer() + * will be set for 'deadline - runtime'. Negative + * runtime will not do. + */ + dl_se->runtime = 0; + return; + } + + /* * If the server was previously activated - the starving condition * took place, it this point it went away because the fair scheduler * was able to get runtime in background. So return to the initial @@ -1530,6 +1483,9 @@ static void update_curr_dl_se(struct rq *rq, struct sched_dl_entity *dl_se, s64 replenish_dl_new_period(dl_se, dl_se->rq); + if (idle) + dl_se->dl_defer_idle = 1; + /* * Not being able to start the timer seems problematic. If it could not * be started for whatever reason, we need to "unthrottle" the DL server @@ -1543,6 +1499,7 @@ static void update_curr_dl_se(struct rq *rq, struct sched_dl_entity *dl_se, s64 throttle: if (dl_runtime_exceeded(dl_se) || dl_se->dl_yielded) { + trace_sched_dl_throttle_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); dl_se->dl_throttled = 1; /* If requested, inform the user about runtime overruns. */ @@ -1557,21 +1514,25 @@ throttle: } if (unlikely(is_dl_boosted(dl_se) || !start_dl_timer(dl_se))) { - if (dl_server(dl_se)) - enqueue_dl_entity(dl_se, ENQUEUE_REPLENISH); - else + if (dl_server(dl_se)) { + replenish_dl_new_period(dl_se, rq); + start_dl_timer(dl_se); + } else { enqueue_task_dl(rq, dl_task_of(dl_se), ENQUEUE_REPLENISH); + } } if (!is_leftmost(dl_se, &rq->dl)) resched_curr(rq); + } else { + trace_sched_dl_update_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); } /* - * The fair server (sole dl_server) does not account for real-time - * workload because it is running fair work. + * The dl_server does not account for real-time workload because it + * is running fair work. */ - if (dl_se == &rq->fair_server) + if (dl_se->dl_server) return; #ifdef CONFIG_RT_GROUP_SCHED @@ -1599,74 +1560,251 @@ throttle: rt_rq->rt_time += delta_exec; raw_spin_unlock(&rt_rq->rt_runtime_lock); } -#endif +#endif /* CONFIG_RT_GROUP_SCHED */ } /* * In the non-defer mode, the idle time is not accounted, as the * server provides a guarantee. * - * If the dl_server is in defer mode, the idle time is also considered - * as time available for the fair server, avoiding a penalty for the - * rt scheduler that did not consumed that time. + * If the dl_server is in defer mode, the idle time is also considered as + * time available for the dl_server, avoiding a penalty for the rt + * scheduler that did not consumed that time. */ -void dl_server_update_idle_time(struct rq *rq, struct task_struct *p) +void dl_server_update_idle(struct sched_dl_entity *dl_se, s64 delta_exec) { - s64 delta_exec, scaled_delta_exec; - - if (!rq->fair_server.dl_defer) - return; - - /* no need to discount more */ - if (rq->fair_server.runtime < 0) - return; - - delta_exec = rq_clock_task(rq) - p->se.exec_start; - if (delta_exec < 0) - return; - - scaled_delta_exec = dl_scaled_delta_exec(rq, &rq->fair_server, delta_exec); - - rq->fair_server.runtime -= scaled_delta_exec; - - if (rq->fair_server.runtime < 0) { - rq->fair_server.dl_defer_running = 0; - rq->fair_server.runtime = 0; - } - - p->se.exec_start = rq_clock_task(rq); + if (dl_se->dl_server_active && dl_se->dl_runtime && dl_se->dl_defer) + update_curr_dl_se(dl_se->rq, dl_se, delta_exec); } void dl_server_update(struct sched_dl_entity *dl_se, s64 delta_exec) { /* 0 runtime = fair server disabled */ - if (dl_se->dl_runtime) + if (dl_se->dl_server_active && dl_se->dl_runtime) update_curr_dl_se(dl_se->rq, dl_se, delta_exec); } +/* + * dl_server && dl_defer: + * + * 6 + * +--------------------+ + * v | + * +-------------+ 4 +-----------+ 5 +------------------+ + * +-> | A:init | <--- | D:running | -----> | E:replenish-wait | + * | +-------------+ +-----------+ +------------------+ + * | | | 1 ^ ^ | + * | | 1 +----------+ | 3 | + * | v | | + * | +--------------------------------+ 2 | + * | | | ----+ | + * | 8 | B:zero_laxity-wait | | | + * | | | <---+ | + * | +--------------------------------+ | + * | | ^ ^ 2 | + * | | 7 | 2, 1 +----------------+ + * | v | + * | +-------------+ | + * +-- | C:idle-wait | -+ + * +-------------+ + * ^ 7 | + * +---------+ + * + * + * [A] - init + * dl_server_active = 0 + * dl_throttled = 0 + * dl_defer_armed = 0 + * dl_defer_running = 0/1 + * dl_defer_idle = 0 + * + * [B] - zero_laxity-wait + * dl_server_active = 1 + * dl_throttled = 1 + * dl_defer_armed = 1 + * dl_defer_running = 0 + * dl_defer_idle = 0 + * + * [C] - idle-wait + * dl_server_active = 1 + * dl_throttled = 1 + * dl_defer_armed = 1 + * dl_defer_running = 0 + * dl_defer_idle = 1 + * + * [D] - running + * dl_server_active = 1 + * dl_throttled = 0 + * dl_defer_armed = 0 + * dl_defer_running = 1 + * dl_defer_idle = 0 + * + * [E] - replenish-wait + * dl_server_active = 1 + * dl_throttled = 1 + * dl_defer_armed = 0 + * dl_defer_running = 1 + * dl_defer_idle = 0 + * + * + * [1] A->B, A->D, C->B + * dl_server_start() + * dl_defer_idle = 0; + * if (dl_server_active) + * return; // [B] + * dl_server_active = 1; + * enqueue_dl_entity() + * update_dl_entity(WAKEUP) + * if (dl_time_before() || dl_entity_overflow()) + * dl_defer_running = 0; + * replenish_dl_new_period(); + * // fwd period + * dl_throttled = 1; + * dl_defer_armed = 1; + * if (!dl_defer_running) + * dl_defer_armed = 1; + * dl_throttled = 1; + * if (dl_throttled && start_dl_timer()) + * return; // [B] + * __enqueue_dl_entity(); + * // [D] + * + * // deplete server runtime from client-class + * [2] B->B, C->B, E->B + * dl_server_update() + * update_curr_dl_se() // idle = false + * if (dl_defer_idle) + * dl_defer_idle = 0; + * if (dl_defer && dl_throttled && dl_runtime_exceeded()) + * dl_defer_running = 0; + * hrtimer_try_to_cancel(); // stop timer + * replenish_dl_new_period() + * // fwd period + * dl_throttled = 1; + * dl_defer_armed = 1; + * start_dl_timer(); // restart timer + * // [B] + * + * // timer actually fires means we have runtime + * [3] B->D + * dl_server_timer() + * if (dl_defer_armed) + * dl_defer_running = 1; + * enqueue_dl_entity(REPLENISH) + * replenish_dl_entity() + * // fwd period + * if (dl_throttled) + * dl_throttled = 0; + * if (dl_defer_armed) + * dl_defer_armed = 0; + * __enqueue_dl_entity(); + * // [D] + * + * // schedule server + * [4] D->A + * pick_task_dl() + * p = server_pick_task(); + * if (!p) + * dl_server_stop() + * dequeue_dl_entity(); + * hrtimer_try_to_cancel(); + * dl_defer_armed = 0; + * dl_throttled = 0; + * dl_server_active = 0; + * // [A] + * return p; + * + * // server running + * [5] D->E + * update_curr_dl_se() + * if (dl_runtime_exceeded()) + * dl_throttled = 1; + * dequeue_dl_entity(); + * start_dl_timer(); + * // [E] + * + * // server replenished + * [6] E->D + * dl_server_timer() + * enqueue_dl_entity(REPLENISH) + * replenish_dl_entity() + * fwd-period + * if (dl_throttled) + * dl_throttled = 0; + * __enqueue_dl_entity(); + * // [D] + * + * // deplete server runtime from idle + * [7] B->C, C->C + * dl_server_update_idle() + * update_curr_dl_se() // idle = true + * if (dl_defer && dl_throttled && dl_runtime_exceeded()) + * if (dl_defer_idle) + * return; + * dl_defer_running = 0; + * hrtimer_try_to_cancel(); + * replenish_dl_new_period() + * // fwd period + * dl_throttled = 1; + * dl_defer_armed = 1; + * dl_defer_idle = 1; + * start_dl_timer(); // restart timer + * // [C] + * + * // stop idle server + * [8] C->A + * dl_server_timer() + * if (dl_defer_idle) + * dl_server_stop(); + * // [A] + * + * + * digraph dl_server { + * "A:init" -> "B:zero_laxity-wait" [label="1:dl_server_start"] + * "A:init" -> "D:running" [label="1:dl_server_start"] + * "B:zero_laxity-wait" -> "B:zero_laxity-wait" [label="2:dl_server_update"] + * "B:zero_laxity-wait" -> "C:idle-wait" [label="7:dl_server_update_idle"] + * "B:zero_laxity-wait" -> "D:running" [label="3:dl_server_timer"] + * "C:idle-wait" -> "A:init" [label="8:dl_server_timer"] + * "C:idle-wait" -> "B:zero_laxity-wait" [label="1:dl_server_start"] + * "C:idle-wait" -> "B:zero_laxity-wait" [label="2:dl_server_update"] + * "C:idle-wait" -> "C:idle-wait" [label="7:dl_server_update_idle"] + * "D:running" -> "A:init" [label="4:pick_task_dl"] + * "D:running" -> "E:replenish-wait" [label="5:update_curr_dl_se"] + * "E:replenish-wait" -> "B:zero_laxity-wait" [label="2:dl_server_update"] + * "E:replenish-wait" -> "D:running" [label="6:dl_server_timer"] + * } + * + * + * Notes: + * + * - When there are fair tasks running the most likely loop is [2]->[2]. + * the dl_server never actually runs, the timer never fires. + * + * - When there is actual fair starvation; the timer fires and starts the + * dl_server. This will then throttle and replenish like a normal DL + * task. Notably it will not 'defer' again. + * + * - When idle it will push the actication forward once, and then wait + * for the timer to hit or a non-idle update to restart things. + */ void dl_server_start(struct sched_dl_entity *dl_se) { struct rq *rq = dl_se->rq; + dl_se->dl_defer_idle = 0; + if (!dl_server(dl_se) || dl_se->dl_server_active || !dl_se->dl_runtime) + return; + /* - * XXX: the apply do not work fine at the init phase for the - * fair server because things are not yet set. We need to improve - * this before getting generic. + * Update the current task to 'now'. */ - if (!dl_server(dl_se)) { - u64 runtime = 50 * NSEC_PER_MSEC; - u64 period = 1000 * NSEC_PER_MSEC; + rq->donor->sched_class->update_curr(rq); - dl_server_apply_params(dl_se, runtime, period, 1); - - dl_se->dl_server = 1; - dl_se->dl_defer = 1; - setup_new_dl_entity(dl_se); - } - - if (!dl_se->dl_runtime) + if (WARN_ON_ONCE(!cpu_online(cpu_of(rq)))) return; + trace_sched_dl_server_start_tp(dl_se, cpu_of(rq), dl_get_type(dl_se, rq)); dl_se->dl_server_active = 1; enqueue_dl_entity(dl_se, ENQUEUE_WAKEUP); if (!dl_task(dl_se->rq->curr) || dl_entity_preempt(dl_se, &rq->curr->dl)) @@ -1675,25 +1813,65 @@ void dl_server_start(struct sched_dl_entity *dl_se) void dl_server_stop(struct sched_dl_entity *dl_se) { - if (!dl_se->dl_runtime) + if (!dl_server(dl_se) || !dl_server_active(dl_se)) return; + trace_sched_dl_server_stop_tp(dl_se, cpu_of(dl_se->rq), + dl_get_type(dl_se, dl_se->rq)); dequeue_dl_entity(dl_se, DEQUEUE_SLEEP); hrtimer_try_to_cancel(&dl_se->dl_timer); dl_se->dl_defer_armed = 0; dl_se->dl_throttled = 0; + dl_se->dl_defer_idle = 0; dl_se->dl_server_active = 0; } void dl_server_init(struct sched_dl_entity *dl_se, struct rq *rq, - dl_server_has_tasks_f has_tasks, dl_server_pick_f pick_task) { dl_se->rq = rq; - dl_se->server_has_tasks = has_tasks; dl_se->server_pick_task = pick_task; } +void sched_init_dl_servers(void) +{ + int cpu; + struct rq *rq; + struct sched_dl_entity *dl_se; + + for_each_online_cpu(cpu) { + u64 runtime = 50 * NSEC_PER_MSEC; + u64 period = 1000 * NSEC_PER_MSEC; + + rq = cpu_rq(cpu); + + guard(rq_lock_irq)(rq); + update_rq_clock(rq); + + dl_se = &rq->fair_server; + + WARN_ON(dl_server(dl_se)); + + dl_server_apply_params(dl_se, runtime, period, 1); + + dl_se->dl_server = 1; + dl_se->dl_defer = 1; + setup_new_dl_entity(dl_se); + +#ifdef CONFIG_SCHED_CLASS_EXT + dl_se = &rq->ext_server; + + WARN_ON(dl_server(dl_se)); + + dl_server_apply_params(dl_se, runtime, period, 1); + + dl_se->dl_server = 1; + dl_se->dl_defer = 1; + setup_new_dl_entity(dl_se); +#endif + } +} + void __dl_server_attach_root(struct sched_dl_entity *dl_se, struct rq *rq) { u64 new_bw = dl_se->dl_bw; @@ -1717,7 +1895,6 @@ int dl_server_apply_params(struct sched_dl_entity *dl_se, u64 runtime, u64 perio int cpu = cpu_of(rq); struct dl_bw *dl_b; unsigned long cap; - int retval = 0; int cpus; dl_b = dl_bw_of(cpu); @@ -1749,7 +1926,7 @@ int dl_server_apply_params(struct sched_dl_entity *dl_se, u64 runtime, u64 perio dl_se->dl_bw = to_ratio(dl_se->dl_period, dl_se->dl_runtime); dl_se->dl_density = to_ratio(dl_se->dl_deadline, dl_se->dl_runtime); - return retval; + return 0; } /* @@ -1839,15 +2016,12 @@ static void init_dl_inactive_task_timer(struct sched_dl_entity *dl_se) { struct hrtimer *timer = &dl_se->inactive_timer; - hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); - timer->function = inactive_task_timer; + hrtimer_setup(timer, inactive_task_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); } #define __node_2_dle(node) \ rb_entry((node), struct sched_dl_entity, rb_node) -#ifdef CONFIG_SMP - static void inc_dl_deadline(struct dl_rq *dl_rq, u64 deadline) { struct rq *rq = rq_of_dl_rq(dl_rq); @@ -1872,7 +2046,7 @@ static void dec_dl_deadline(struct dl_rq *dl_rq, u64 deadline) if (!dl_rq->dl_nr_running) { dl_rq->earliest_dl.curr = 0; dl_rq->earliest_dl.next = 0; - cpudl_clear(&rq->rd->cpudl, rq->cpu); + cpudl_clear(&rq->rd->cpudl, rq->cpu, rq->online); cpupri_set(&rq->rd->cpupri, rq->cpu, rq->rt.highest_prio.curr); } else { struct rb_node *leftmost = rb_first_cached(&dl_rq->root); @@ -1883,20 +2057,15 @@ static void dec_dl_deadline(struct dl_rq *dl_rq, u64 deadline) } } -#else - -static inline void inc_dl_deadline(struct dl_rq *dl_rq, u64 deadline) {} -static inline void dec_dl_deadline(struct dl_rq *dl_rq, u64 deadline) {} - -#endif /* CONFIG_SMP */ - static inline void inc_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) { u64 deadline = dl_se->deadline; dl_rq->dl_nr_running++; - add_nr_running(rq_of_dl_rq(dl_rq), 1); + + if (!dl_server(dl_se)) + add_nr_running(rq_of_dl_rq(dl_rq), 1); inc_dl_deadline(dl_rq, deadline); } @@ -1906,7 +2075,9 @@ void dec_dl_tasks(struct sched_dl_entity *dl_se, struct dl_rq *dl_rq) { WARN_ON(!dl_rq->dl_nr_running); dl_rq->dl_nr_running--; - sub_nr_running(rq_of_dl_rq(dl_rq), 1); + + if (!dl_server(dl_se)) + sub_nr_running(rq_of_dl_rq(dl_rq), 1); dec_dl_deadline(dl_rq, dl_se->deadline); } @@ -1968,10 +2139,14 @@ update_stats_dequeue_dl(struct dl_rq *dl_rq, struct sched_dl_entity *dl_se, int flags) { struct task_struct *p = dl_task_of(dl_se); + struct rq *rq = rq_of_dl_rq(dl_rq); if (!schedstat_enabled()) return; + if (p != rq->curr) + update_stats_wait_end_dl(dl_rq, dl_se); + if ((flags & DEQUEUE_SLEEP)) { unsigned int state; @@ -2063,7 +2238,7 @@ enqueue_dl_entity(struct sched_dl_entity *dl_se, int flags) update_dl_entity(dl_se); } else if (flags & ENQUEUE_REPLENISH) { replenish_dl_entity(dl_se); - } else if ((flags & ENQUEUE_RESTORE) && + } else if ((flags & ENQUEUE_MOVE) && !is_dl_boosted(dl_se) && dl_time_before(dl_se->deadline, rq_clock(rq_of_dl_se(dl_se)))) { setup_new_dl_entity(dl_se); @@ -2112,7 +2287,7 @@ static void dequeue_dl_entity(struct sched_dl_entity *dl_se, int flags) * or "inactive") */ if (flags & DEQUEUE_SLEEP) - task_non_contending(dl_se); + task_non_contending(dl_se, true); } static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags) @@ -2168,6 +2343,9 @@ static void enqueue_task_dl(struct rq *rq, struct task_struct *p, int flags) if (dl_server(&p->dl)) return; + if (task_is_blocked(p)) + return; + if (!task_current(rq, p) && !p->dl.dl_throttled && p->nr_cpus_allowed > 1) enqueue_pushable_dl_task(rq, p); } @@ -2204,7 +2382,7 @@ static void yield_task_dl(struct rq *rq) * it and the bandwidth timer will wake it up and will give it * new scheduling parameters (thanks to dl_yielded=1). */ - rq->curr->dl.dl_yielded = 1; + rq->donor->dl.dl_yielded = 1; update_rq_clock(rq); update_curr_dl(rq); @@ -2216,8 +2394,6 @@ static void yield_task_dl(struct rq *rq) rq_clock_skip_update(rq); } -#ifdef CONFIG_SMP - static inline bool dl_task_is_earliest_deadline(struct task_struct *p, struct rq *rq) { @@ -2236,7 +2412,7 @@ select_task_rq_dl(struct task_struct *p, int cpu, int flags) struct rq *rq; if (!(flags & WF_TTWU)) - goto out; + return cpu; rq = cpu_rq(cpu); @@ -2274,7 +2450,6 @@ select_task_rq_dl(struct task_struct *p, int cpu, int flags) } rcu_read_unlock(); -out: return cpu; } @@ -2347,21 +2522,26 @@ static int balance_dl(struct rq *rq, struct task_struct *p, struct rq_flags *rf) return sched_stop_runnable(rq) || sched_dl_runnable(rq); } -#endif /* CONFIG_SMP */ /* * Only called when both the current and waking task are -deadline * tasks. */ -static void wakeup_preempt_dl(struct rq *rq, struct task_struct *p, - int flags) +static void wakeup_preempt_dl(struct rq *rq, struct task_struct *p, int flags) { + /* + * Can only get preempted by stop-class, and those should be + * few and short lived, doesn't really make sense to push + * anything away for that. + */ + if (p->sched_class != &dl_sched_class) + return; + if (dl_entity_preempt(&p->dl, &rq->donor->dl)) { resched_curr(rq); return; } -#ifdef CONFIG_SMP /* * In the unlikely case current and p have the same deadline * let us try to decide what's the best thing to do... @@ -2369,7 +2549,6 @@ static void wakeup_preempt_dl(struct rq *rq, struct task_struct *p, if ((p->dl.deadline == rq->donor->dl.deadline) && !test_tsk_need_resched(rq->curr)) check_preempt_equal_dl(rq, p); -#endif /* CONFIG_SMP */ } #ifdef CONFIG_SCHED_HRTICK @@ -2377,11 +2556,11 @@ static void start_hrtick_dl(struct rq *rq, struct sched_dl_entity *dl_se) { hrtick_start(rq, dl_se->runtime); } -#else /* !CONFIG_SCHED_HRTICK */ +#else /* !CONFIG_SCHED_HRTICK: */ static void start_hrtick_dl(struct rq *rq, struct sched_dl_entity *dl_se) { } -#endif +#endif /* !CONFIG_SCHED_HRTICK */ static void set_next_task_dl(struct rq *rq, struct task_struct *p, bool first) { @@ -2421,7 +2600,7 @@ static struct sched_dl_entity *pick_next_dl_entity(struct dl_rq *dl_rq) * __pick_next_task_dl - Helper to pick the next -deadline task to run. * @rq: The runqueue to pick the next task from. */ -static struct task_struct *__pick_task_dl(struct rq *rq) +static struct task_struct *__pick_task_dl(struct rq *rq, struct rq_flags *rf) { struct sched_dl_entity *dl_se; struct dl_rq *dl_rq = &rq->dl; @@ -2435,12 +2614,9 @@ again: WARN_ON_ONCE(!dl_se); if (dl_server(dl_se)) { - p = dl_se->server_pick_task(dl_se); + p = dl_se->server_pick_task(dl_se, rf); if (!p) { - if (dl_server_active(dl_se)) { - dl_se->dl_yielded = 1; - update_curr_dl_se(rq, dl_se, 0); - } + dl_server_stop(dl_se); goto again; } rq->dl_server = dl_se; @@ -2451,9 +2627,9 @@ again: return p; } -static struct task_struct *pick_task_dl(struct rq *rq) +static struct task_struct *pick_task_dl(struct rq *rq, struct rq_flags *rf) { - return __pick_task_dl(rq); + return __pick_task_dl(rq, rf); } static void put_prev_task_dl(struct rq *rq, struct task_struct *p, struct task_struct *next) @@ -2467,6 +2643,10 @@ static void put_prev_task_dl(struct rq *rq, struct task_struct *p, struct task_s update_curr_dl(rq); update_dl_rq_load_avg(rq_clock_pelt(rq), rq, 1); + + if (task_is_blocked(p)) + return; + if (on_dl_rq(&p->dl) && p->nr_cpus_allowed > 1) enqueue_pushable_dl_task(rq, p); } @@ -2502,8 +2682,6 @@ static void task_fork_dl(struct task_struct *p) */ } -#ifdef CONFIG_SMP - /* Only try algorithms three times */ #define DL_MAX_TRIES 3 @@ -2532,6 +2710,7 @@ static struct task_struct *pick_earliest_pushable_dl_task(struct rq *rq, int cpu return NULL; } +/* Access rule: must be called on local CPU with preemption disabled */ static DEFINE_PER_CPU(cpumask_var_t, local_cpu_mask_dl); static int find_later_rq(struct task_struct *task) @@ -2621,6 +2800,39 @@ static int find_later_rq(struct task_struct *task) return -1; } +static struct task_struct *pick_next_pushable_dl_task(struct rq *rq) +{ + struct task_struct *i, *p = NULL; + struct rb_node *next_node; + + if (!has_pushable_dl_tasks(rq)) + return NULL; + + next_node = rb_first_cached(&rq->dl.pushable_dl_tasks_root); + while (next_node) { + i = __node_2_pdl(next_node); + /* make sure task isn't on_cpu (possible with proxy-exec) */ + if (!task_on_cpu(rq, i)) { + p = i; + break; + } + + next_node = rb_next(next_node); + } + + if (!p) + return NULL; + + WARN_ON_ONCE(rq->cpu != task_cpu(p)); + WARN_ON_ONCE(task_current(rq, p)); + WARN_ON_ONCE(p->nr_cpus_allowed <= 1); + + WARN_ON_ONCE(!task_on_rq_queued(p)); + WARN_ON_ONCE(!dl_task(p)); + + return p; +} + /* Locks the rq it finds */ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) { @@ -2648,12 +2860,37 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) /* Retry if something changed. */ if (double_lock_balance(rq, later_rq)) { - if (unlikely(task_rq(task) != rq || + /* + * double_lock_balance had to release rq->lock, in the + * meantime, task may no longer be fit to be migrated. + * Check the following to ensure that the task is + * still suitable for migration: + * 1. It is possible the task was scheduled, + * migrate_disabled was set and then got preempted, + * so we must check the task migration disable + * flag. + * 2. The CPU picked is in the task's affinity. + * 3. For throttled task (dl_task_offline_migration), + * check the following: + * - the task is not on the rq anymore (it was + * migrated) + * - the task is not on CPU anymore + * - the task is still a dl task + * - the task is not queued on the rq anymore + * 4. For the non-throttled task (push_dl_task), the + * check to ensure that this task is still at the + * head of the pushable tasks list is enough. + */ + if (unlikely(is_migration_disabled(task) || !cpumask_test_cpu(later_rq->cpu, &task->cpus_mask) || - task_on_cpu(rq, task) || - !dl_task(task) || - is_migration_disabled(task) || - !task_on_rq_queued(task))) { + (task->dl.dl_throttled && + (task_rq(task) != rq || + task_on_cpu(rq, task) || + !dl_task(task) || + !task_on_rq_queued(task))) || + (!task->dl.dl_throttled && + task != pick_next_pushable_dl_task(rq)))) { + double_unlock_balance(rq, later_rq); later_rq = NULL; break; @@ -2676,25 +2913,6 @@ static struct rq *find_lock_later_rq(struct task_struct *task, struct rq *rq) return later_rq; } -static struct task_struct *pick_next_pushable_dl_task(struct rq *rq) -{ - struct task_struct *p; - - if (!has_pushable_dl_tasks(rq)) - return NULL; - - p = __node_2_pdl(rb_first_cached(&rq->dl.pushable_dl_tasks_root)); - - WARN_ON_ONCE(rq->cpu != task_cpu(p)); - WARN_ON_ONCE(task_current(rq, p)); - WARN_ON_ONCE(p->nr_cpus_allowed <= 1); - - WARN_ON_ONCE(!task_on_rq_queued(p)); - WARN_ON_ONCE(!dl_task(p)); - - return p; -} - /* * See if the non running -deadline tasks on this rq * can be sent to some other CPU where they can preempt @@ -2889,20 +3107,18 @@ static void task_woken_dl(struct rq *rq, struct task_struct *p) static void set_cpus_allowed_dl(struct task_struct *p, struct affinity_context *ctx) { - struct root_domain *src_rd; struct rq *rq; WARN_ON_ONCE(!dl_task(p)); rq = task_rq(p); - src_rd = rq->rd; /* * Migrating a SCHED_DEADLINE task between exclusive * cpusets (different root_domains) entails a bandwidth * update. We already made space for us in the destination * domain (see cpuset_can_attach()). */ - if (!cpumask_intersects(src_rd->span, ctx->new_mask)) { + if (dl_task_needs_bw_move(p, ctx->new_mask)) { struct dl_bw *src_dl_b; src_dl_b = dl_bw_of(cpu_of(rq)); @@ -2919,15 +3135,25 @@ static void set_cpus_allowed_dl(struct task_struct *p, set_cpus_allowed_common(p, ctx); } +bool dl_task_needs_bw_move(struct task_struct *p, + const struct cpumask *new_mask) +{ + if (!dl_task(p)) + return false; + + return !cpumask_intersects(task_rq(p)->rd->span, new_mask); +} + /* Assumes rq->lock is held */ static void rq_online_dl(struct rq *rq) { if (rq->dl.overloaded) dl_set_overload(rq); - cpudl_set_freecpu(&rq->rd->cpudl, rq->cpu); if (rq->dl.dl_nr_running > 0) cpudl_set(&rq->rd->cpudl, rq->cpu, rq->dl.earliest_dl.curr); + else + cpudl_clear(&rq->rd->cpudl, rq->cpu, true); } /* Assumes rq->lock is held */ @@ -2936,8 +3162,7 @@ static void rq_offline_dl(struct rq *rq) if (rq->dl.overloaded) dl_clear_overload(rq); - cpudl_clear(&rq->rd->cpudl, rq->cpu); - cpudl_clear_freecpu(&rq->rd->cpudl, rq->cpu); + cpudl_clear(&rq->rd->cpudl, rq->cpu, false); } void __init init_sched_dl_class(void) @@ -2949,28 +3174,91 @@ void __init init_sched_dl_class(void) GFP_KERNEL, cpu_to_node(i)); } +/* + * This function always returns a non-empty bitmap in @cpus. This is because + * if a root domain has reserved bandwidth for DL tasks, the DL bandwidth + * check will prevent CPU hotplug from deactivating all CPUs in that domain. + */ +static void dl_get_task_effective_cpus(struct task_struct *p, struct cpumask *cpus) +{ + const struct cpumask *hk_msk; + + hk_msk = housekeeping_cpumask(HK_TYPE_DOMAIN); + if (housekeeping_enabled(HK_TYPE_DOMAIN)) { + if (!cpumask_intersects(p->cpus_ptr, hk_msk)) { + /* + * CPUs isolated by isolcpu="domain" always belong to + * def_root_domain. + */ + cpumask_andnot(cpus, cpu_active_mask, hk_msk); + return; + } + } + + /* + * If a root domain holds a DL task, it must have active CPUs. So + * active CPUs can always be found by walking up the task's cpuset + * hierarchy up to the partition root. + */ + cpuset_cpus_allowed_locked(p, cpus); +} + +/* The caller should hold cpuset_mutex */ void dl_add_task_root_domain(struct task_struct *p) { struct rq_flags rf; struct rq *rq; struct dl_bw *dl_b; + unsigned int cpu; + struct cpumask *msk; raw_spin_lock_irqsave(&p->pi_lock, rf.flags); - if (!dl_task(p)) { + if (!dl_task(p) || dl_entity_is_special(&p->dl)) { raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); return; } - rq = __task_rq_lock(p, &rf); - + msk = this_cpu_cpumask_var_ptr(local_cpu_mask_dl); + dl_get_task_effective_cpus(p, msk); + cpu = cpumask_first_and(cpu_active_mask, msk); + BUG_ON(cpu >= nr_cpu_ids); + rq = cpu_rq(cpu); dl_b = &rq->rd->dl_bw; - raw_spin_lock(&dl_b->lock); + raw_spin_lock(&dl_b->lock); __dl_add(dl_b, p->dl.dl_bw, cpumask_weight(rq->rd->span)); - raw_spin_unlock(&dl_b->lock); + raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); +} - task_rq_unlock(rq, p, &rf); +static void dl_server_add_bw(struct root_domain *rd, int cpu) +{ + struct sched_dl_entity *dl_se; + + dl_se = &cpu_rq(cpu)->fair_server; + if (dl_server(dl_se) && cpu_active(cpu)) + __dl_add(&rd->dl_bw, dl_se->dl_bw, dl_bw_cpus(cpu)); + +#ifdef CONFIG_SCHED_CLASS_EXT + dl_se = &cpu_rq(cpu)->ext_server; + if (dl_server(dl_se) && cpu_active(cpu)) + __dl_add(&rd->dl_bw, dl_se->dl_bw, dl_bw_cpus(cpu)); +#endif +} + +static u64 dl_server_read_bw(int cpu) +{ + u64 dl_bw = 0; + + if (cpu_rq(cpu)->fair_server.dl_server) + dl_bw += cpu_rq(cpu)->fair_server.dl_bw; + +#ifdef CONFIG_SCHED_CLASS_EXT + if (cpu_rq(cpu)->ext_server.dl_server) + dl_bw += cpu_rq(cpu)->ext_server.dl_bw; +#endif + + return dl_bw; } void dl_clear_root_domain(struct root_domain *rd) @@ -2978,22 +3266,27 @@ void dl_clear_root_domain(struct root_domain *rd) int i; guard(raw_spinlock_irqsave)(&rd->dl_bw.lock); - rd->dl_bw.total_bw = 0; /* - * dl_server bandwidth is only restored when CPUs are attached to root - * domains (after domains are created or CPUs moved back to the - * default root doamin). + * Reset total_bw to zero and extra_bw to max_bw so that next + * loop will add dl-servers contributions back properly, */ - for_each_cpu(i, rd->span) { - struct sched_dl_entity *dl_se = &cpu_rq(i)->fair_server; + rd->dl_bw.total_bw = 0; + for_each_cpu(i, rd->span) + cpu_rq(i)->dl.extra_bw = cpu_rq(i)->dl.max_bw; - if (dl_server(dl_se) && cpu_active(i)) - rd->dl_bw.total_bw += dl_se->dl_bw; - } + /* + * dl_servers are not tasks. Since dl_add_task_root_domain ignores + * them, we need to account for them here explicitly. + */ + for_each_cpu(i, rd->span) + dl_server_add_bw(rd, i); } -#endif /* CONFIG_SMP */ +void dl_clear_root_domain_cpu(int cpu) +{ + dl_clear_root_domain(cpu_rq(cpu)->rd); +} static void switched_from_dl(struct rq *rq, struct task_struct *p) { @@ -3006,7 +3299,7 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p) * will reset the task parameters. */ if (task_on_rq_queued(p) && p->dl.dl_runtime) - task_non_contending(&p->dl); + task_non_contending(&p->dl, false); /* * In case a task is setscheduled out from SCHED_DEADLINE we need to @@ -3067,10 +3360,8 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) } if (rq->donor != p) { -#ifdef CONFIG_SMP if (p->nr_cpus_allowed > 1 && rq->dl.overloaded) deadline_queue_push_tasks(rq); -#endif if (dl_task(rq->donor)) wakeup_preempt_dl(rq, p, 0); else @@ -3080,24 +3371,30 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) } } +static u64 get_prio_dl(struct rq *rq, struct task_struct *p) +{ + /* + * Make sure to update current so we don't return a stale value. + */ + if (task_current_donor(rq, p)) + update_curr_dl(rq); + + return p->dl.deadline; +} + /* * If the scheduling parameters of a -deadline task changed, * a push or pull operation might be needed. */ -static void prio_changed_dl(struct rq *rq, struct task_struct *p, - int oldprio) +static void prio_changed_dl(struct rq *rq, struct task_struct *p, u64 old_deadline) { if (!task_on_rq_queued(p)) return; -#ifdef CONFIG_SMP - /* - * This might be too much, but unfortunately - * we don't have the old deadline value, and - * we can't argue if the task is increasing - * or lowering its prio, so... - */ - if (!rq->dl.overloaded) + if (p->dl.deadline == old_deadline) + return; + + if (dl_time_before(old_deadline, p->dl.deadline)) deadline_queue_pull_task(rq); if (task_current_donor(rq, p)) { @@ -3119,13 +3416,6 @@ static void prio_changed_dl(struct rq *rq, struct task_struct *p, dl_time_before(p->dl.deadline, rq->curr->dl.deadline)) resched_curr(rq); } -#else - /* - * We don't know if p has a earlier or later deadline, so let's blindly - * set a (maybe not needed) rescheduling point. - */ - resched_curr(rq); -#endif } #ifdef CONFIG_SCHED_CORE @@ -3136,7 +3426,6 @@ static int task_is_throttled_dl(struct task_struct *p, int cpu) #endif DEFINE_SCHED_CLASS(dl) = { - .enqueue_task = enqueue_task_dl, .dequeue_task = dequeue_task_dl, .yield_task = yield_task_dl, @@ -3147,7 +3436,6 @@ DEFINE_SCHED_CLASS(dl) = { .put_prev_task = put_prev_task_dl, .set_next_task = set_next_task_dl, -#ifdef CONFIG_SMP .balance = balance_dl, .select_task_rq = select_task_rq_dl, .migrate_task_rq = migrate_task_rq_dl, @@ -3156,11 +3444,11 @@ DEFINE_SCHED_CLASS(dl) = { .rq_offline = rq_offline_dl, .task_woken = task_woken_dl, .find_lock_rq = find_lock_later_rq, -#endif .task_tick = task_tick_dl, .task_fork = task_fork_dl, + .get_prio = get_prio_dl, .prio_changed = prio_changed_dl, .switched_from = switched_from_dl, .switched_to = switched_to_dl, @@ -3171,15 +3459,18 @@ DEFINE_SCHED_CLASS(dl) = { #endif }; -/* Used for dl_bw check and update, used under sched_rt_handler()::mutex */ -static u64 dl_generation; +/* + * Used for dl_bw check and update, used under sched_rt_handler()::mutex and + * sched_domains_mutex. + */ +u64 dl_cookie; int sched_dl_global_validate(void) { u64 runtime = global_rt_runtime(); u64 period = global_rt_period(); u64 new_bw = to_ratio(period, runtime); - u64 gen = ++dl_generation; + u64 cookie = ++dl_cookie; struct dl_bw *dl_b; int cpu, cpus, ret = 0; unsigned long flags; @@ -3192,7 +3483,7 @@ int sched_dl_global_validate(void) for_each_online_cpu(cpu) { rcu_read_lock_sched(); - if (dl_bw_visited(cpu, gen)) + if (dl_bw_visited(cpu, cookie)) goto next; dl_b = dl_bw_of(cpu); @@ -3229,7 +3520,7 @@ static void init_dl_rq_bw_ratio(struct dl_rq *dl_rq) void sched_dl_do_global(void) { u64 new_bw = -1; - u64 gen = ++dl_generation; + u64 cookie = ++dl_cookie; struct dl_bw *dl_b; int cpu; unsigned long flags; @@ -3237,10 +3528,13 @@ void sched_dl_do_global(void) if (global_rt_runtime() != RUNTIME_INF) new_bw = to_ratio(global_rt_period(), global_rt_runtime()); + for_each_possible_cpu(cpu) + init_dl_rq_bw_ratio(&cpu_rq(cpu)->dl); + for_each_possible_cpu(cpu) { rcu_read_lock_sched(); - if (dl_bw_visited(cpu, gen)) { + if (dl_bw_visited(cpu, cookie)) { rcu_read_unlock_sched(); continue; } @@ -3252,7 +3546,6 @@ void sched_dl_do_global(void) raw_spin_unlock_irqrestore(&dl_b->lock, flags); rcu_read_unlock_sched(); - init_dl_rq_bw_ratio(&cpu_rq(cpu)->dl); } } @@ -3342,13 +3635,26 @@ void __setparam_dl(struct task_struct *p, const struct sched_attr *attr) dl_se->dl_density = to_ratio(dl_se->dl_deadline, dl_se->dl_runtime); } -void __getparam_dl(struct task_struct *p, struct sched_attr *attr) +void __getparam_dl(struct task_struct *p, struct sched_attr *attr, unsigned int flags) { struct sched_dl_entity *dl_se = &p->dl; + struct rq *rq = task_rq(p); + u64 adj_deadline; attr->sched_priority = p->rt_priority; - attr->sched_runtime = dl_se->dl_runtime; - attr->sched_deadline = dl_se->dl_deadline; + if (flags & SCHED_GETATTR_FLAG_DL_DYNAMIC) { + guard(raw_spinlock_irq)(&rq->__lock); + update_rq_clock(rq); + if (task_current(rq, p)) + update_curr_dl(rq); + + attr->sched_runtime = dl_se->runtime; + adj_deadline = dl_se->deadline - rq_clock(rq) + ktime_get_ns(); + attr->sched_deadline = adj_deadline; + } else { + attr->sched_runtime = dl_se->dl_runtime; + attr->sched_deadline = dl_se->dl_deadline; + } attr->sched_period = dl_se->dl_period; attr->sched_flags &= ~SCHED_DL_FLAGS; attr->sched_flags |= dl_se->flags; @@ -3426,6 +3732,9 @@ static void __dl_clear_params(struct sched_dl_entity *dl_se) dl_se->dl_non_contending = 0; dl_se->dl_overrun = 0; dl_se->dl_server = 0; + dl_se->dl_defer = 0; + dl_se->dl_defer_running = 0; + dl_se->dl_defer_armed = 0; #ifdef CONFIG_RT_MUTEXES dl_se->pi_se = dl_se; @@ -3453,7 +3762,6 @@ bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr) return false; } -#ifdef CONFIG_SMP int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial) { @@ -3484,7 +3792,7 @@ static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) unsigned long flags, cap; struct dl_bw *dl_b; bool overflow = 0; - u64 fair_server_bw = 0; + u64 dl_server_bw = 0; rcu_read_lock_sched(); dl_b = dl_bw_of(cpu); @@ -3517,27 +3825,26 @@ static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) cap -= arch_scale_cpu_capacity(cpu); /* - * cpu is going offline and NORMAL tasks will be moved away - * from it. We can thus discount dl_server bandwidth - * contribution as it won't need to be servicing tasks after - * the cpu is off. + * cpu is going offline and NORMAL and EXT tasks will be + * moved away from it. We can thus discount dl_server + * bandwidth contribution as it won't need to be servicing + * tasks after the cpu is off. */ - if (cpu_rq(cpu)->fair_server.dl_server) - fair_server_bw = cpu_rq(cpu)->fair_server.dl_bw; + dl_server_bw = dl_server_read_bw(cpu); /* * Not much to check if no DEADLINE bandwidth is present. * dl_servers we can discount, as tasks will be moved out the * offlined CPUs anyway. */ - if (dl_b->total_bw - fair_server_bw > 0) { + if (dl_b->total_bw - dl_server_bw > 0) { /* * Leaving at least one CPU for DEADLINE tasks seems a * wise thing to do. As said above, cpu is not offline * yet, so account for that. */ if (dl_bw_cpus(cpu) - 1) - overflow = __dl_overflow(dl_b, cap, fair_server_bw, 0); + overflow = __dl_overflow(dl_b, cap, dl_server_bw, 0); else overflow = 1; } @@ -3565,11 +3872,8 @@ void dl_bw_free(int cpu, u64 dl_bw) { dl_bw_manage(dl_bw_req_free, cpu, dl_bw); } -#endif -#ifdef CONFIG_SCHED_DEBUG void print_dl_stats(struct seq_file *m, int cpu) { print_dl_rq(m, cpu, &cpu_rq(cpu)->dl); } -#endif /* CONFIG_SCHED_DEBUG */ diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index ef047add7f9e..74c1617cf652 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -6,6 +6,10 @@ * * Copyright(C) 2007, Red Hat, Inc., Ingo Molnar */ +#include <linux/debugfs.h> +#include <linux/nmi.h> +#include <linux/log2.h> +#include "sched.h" /* * This allows printing both to /sys/kernel/debug/sched/debug and @@ -90,10 +94,10 @@ static void sched_feat_enable(int i) { static_key_enable_cpuslocked(&sched_feat_keys[i]); } -#else +#else /* !CONFIG_JUMP_LABEL: */ static void sched_feat_disable(int i) { }; static void sched_feat_enable(int i) { }; -#endif /* CONFIG_JUMP_LABEL */ +#endif /* !CONFIG_JUMP_LABEL */ static int sched_feat_set(char *cmp) { @@ -166,23 +170,15 @@ static const struct file_operations sched_feat_fops = { .release = single_release, }; -#ifdef CONFIG_SMP - static ssize_t sched_scaling_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - char buf[16]; unsigned int scaling; + int ret; - if (cnt > 15) - cnt = 15; - - if (copy_from_user(&buf, ubuf, cnt)) - return -EFAULT; - buf[cnt] = '\0'; - - if (kstrtouint(buf, 10, &scaling)) - return -EINVAL; + ret = kstrtouint_from_user(ubuf, cnt, 10, &scaling); + if (ret) + return ret; if (scaling >= SCHED_TUNABLESCALING_END) return -EINVAL; @@ -214,8 +210,6 @@ static const struct file_operations sched_scaling_fops = { .release = single_release, }; -#endif /* SMP */ - #ifdef CONFIG_PREEMPT_DYNAMIC static ssize_t sched_dynamic_write(struct file *filp, const char __user *ubuf, @@ -244,11 +238,13 @@ static ssize_t sched_dynamic_write(struct file *filp, const char __user *ubuf, static int sched_dynamic_show(struct seq_file *m, void *v) { - static const char * preempt_modes[] = { - "none", "voluntary", "full", "lazy", - }; - int j = ARRAY_SIZE(preempt_modes) - !IS_ENABLED(CONFIG_ARCH_HAS_PREEMPT_LAZY); - int i = IS_ENABLED(CONFIG_PREEMPT_RT) * 2; + int i = (IS_ENABLED(CONFIG_PREEMPT_RT) || IS_ENABLED(CONFIG_ARCH_HAS_PREEMPT_LAZY)) * 2; + int j; + + /* Count entries in NULL terminated preempt_modes */ + for (j = 0; preempt_modes[j]; j++) + ; + j -= !IS_ENABLED(CONFIG_ARCH_HAS_PREEMPT_LAZY); for (; i < j; i++) { if (preempt_dynamic_mode == i) @@ -281,7 +277,6 @@ static const struct file_operations sched_dynamic_fops = { __read_mostly bool sched_debug_verbose; -#ifdef CONFIG_SMP static struct dentry *sd_dentry; @@ -292,7 +287,7 @@ static ssize_t sched_verbose_write(struct file *filp, const char __user *ubuf, bool orig; cpus_read_lock(); - mutex_lock(&sched_domains_mutex); + sched_domains_mutex_lock(); orig = sched_debug_verbose; result = debugfs_write_file_bool(filp, ubuf, cnt, ppos); @@ -304,14 +299,11 @@ static ssize_t sched_verbose_write(struct file *filp, const char __user *ubuf, sd_dentry = NULL; } - mutex_unlock(&sched_domains_mutex); + sched_domains_mutex_unlock(); cpus_read_unlock(); return result; } -#else -#define sched_verbose_write debugfs_write_file_bool -#endif static const struct file_operations sched_verbose_fops = { .read = debugfs_read_file_bool, @@ -339,17 +331,19 @@ enum dl_param { DL_PERIOD, }; -static unsigned long fair_server_period_max = (1UL << 22) * NSEC_PER_USEC; /* ~4 seconds */ -static unsigned long fair_server_period_min = (100) * NSEC_PER_USEC; /* 100 us */ +static unsigned long dl_server_period_max = (1UL << 22) * NSEC_PER_USEC; /* ~4 seconds */ +static unsigned long dl_server_period_min = (100) * NSEC_PER_USEC; /* 100 us */ -static ssize_t sched_fair_server_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos, enum dl_param param) +static ssize_t sched_server_write_common(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos, enum dl_param param, + void *server) { long cpu = (long) ((struct seq_file *) filp->private_data)->private; + struct sched_dl_entity *dl_se = (struct sched_dl_entity *)server; + u64 old_runtime, runtime, period; struct rq *rq = cpu_rq(cpu); - u64 runtime, period; + int retval = 0; size_t err; - int retval; u64 value; err = kstrtoull_from_user(ubuf, cnt, 10, &value); @@ -357,8 +351,8 @@ static ssize_t sched_fair_server_write(struct file *filp, const char __user *ubu return err; scoped_guard (rq_lock_irqsave, rq) { - runtime = rq->fair_server.dl_runtime; - period = rq->fair_server.dl_period; + old_runtime = runtime = dl_se->dl_runtime; + period = dl_se->dl_period; switch (param) { case DL_RUNTIME: @@ -374,62 +368,68 @@ static ssize_t sched_fair_server_write(struct file *filp, const char __user *ubu } if (runtime > period || - period > fair_server_period_max || - period < fair_server_period_min) { + period > dl_server_period_max || + period < dl_server_period_min) { return -EINVAL; } - if (rq->cfs.h_nr_queued) { - update_rq_clock(rq); - dl_server_stop(&rq->fair_server); - } - - retval = dl_server_apply_params(&rq->fair_server, runtime, period, 0); - if (retval) - cnt = retval; + update_rq_clock(rq); + dl_server_stop(dl_se); + retval = dl_server_apply_params(dl_se, runtime, period, 0); + dl_server_start(dl_se); - if (!runtime) - printk_deferred("Fair server disabled in CPU %d, system may crash due to starvation.\n", - cpu_of(rq)); + if (retval < 0) + return retval; + } - if (rq->cfs.h_nr_queued) - dl_server_start(&rq->fair_server); + if (!!old_runtime ^ !!runtime) { + pr_info("%s server %sabled on CPU %d%s.\n", + server == &rq->fair_server ? "Fair" : "Ext", + runtime ? "en" : "dis", + cpu_of(rq), + runtime ? "" : ", system may malfunction due to starvation"); } *ppos += cnt; return cnt; } -static size_t sched_fair_server_show(struct seq_file *m, void *v, enum dl_param param) +static size_t sched_server_show_common(struct seq_file *m, void *v, enum dl_param param, + void *server) { - unsigned long cpu = (unsigned long) m->private; - struct rq *rq = cpu_rq(cpu); + struct sched_dl_entity *dl_se = (struct sched_dl_entity *)server; u64 value; switch (param) { case DL_RUNTIME: - value = rq->fair_server.dl_runtime; + value = dl_se->dl_runtime; break; case DL_PERIOD: - value = rq->fair_server.dl_period; + value = dl_se->dl_period; break; } seq_printf(m, "%llu\n", value); return 0; - } static ssize_t sched_fair_server_runtime_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - return sched_fair_server_write(filp, ubuf, cnt, ppos, DL_RUNTIME); + long cpu = (long) ((struct seq_file *) filp->private_data)->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_write_common(filp, ubuf, cnt, ppos, DL_RUNTIME, + &rq->fair_server); } static int sched_fair_server_runtime_show(struct seq_file *m, void *v) { - return sched_fair_server_show(m, v, DL_RUNTIME); + unsigned long cpu = (unsigned long) m->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_show_common(m, v, DL_RUNTIME, &rq->fair_server); } static int sched_fair_server_runtime_open(struct inode *inode, struct file *filp) @@ -445,16 +445,57 @@ static const struct file_operations fair_server_runtime_fops = { .release = single_release, }; +#ifdef CONFIG_SCHED_CLASS_EXT +static ssize_t +sched_ext_server_runtime_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + long cpu = (long) ((struct seq_file *) filp->private_data)->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_write_common(filp, ubuf, cnt, ppos, DL_RUNTIME, + &rq->ext_server); +} + +static int sched_ext_server_runtime_show(struct seq_file *m, void *v) +{ + unsigned long cpu = (unsigned long) m->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_show_common(m, v, DL_RUNTIME, &rq->ext_server); +} + +static int sched_ext_server_runtime_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, sched_ext_server_runtime_show, inode->i_private); +} + +static const struct file_operations ext_server_runtime_fops = { + .open = sched_ext_server_runtime_open, + .write = sched_ext_server_runtime_write, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif /* CONFIG_SCHED_CLASS_EXT */ + static ssize_t sched_fair_server_period_write(struct file *filp, const char __user *ubuf, size_t cnt, loff_t *ppos) { - return sched_fair_server_write(filp, ubuf, cnt, ppos, DL_PERIOD); + long cpu = (long) ((struct seq_file *) filp->private_data)->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_write_common(filp, ubuf, cnt, ppos, DL_PERIOD, + &rq->fair_server); } static int sched_fair_server_period_show(struct seq_file *m, void *v) { - return sched_fair_server_show(m, v, DL_PERIOD); + unsigned long cpu = (unsigned long) m->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_show_common(m, v, DL_PERIOD, &rq->fair_server); } static int sched_fair_server_period_open(struct inode *inode, struct file *filp) @@ -470,6 +511,40 @@ static const struct file_operations fair_server_period_fops = { .release = single_release, }; +#ifdef CONFIG_SCHED_CLASS_EXT +static ssize_t +sched_ext_server_period_write(struct file *filp, const char __user *ubuf, + size_t cnt, loff_t *ppos) +{ + long cpu = (long) ((struct seq_file *) filp->private_data)->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_write_common(filp, ubuf, cnt, ppos, DL_PERIOD, + &rq->ext_server); +} + +static int sched_ext_server_period_show(struct seq_file *m, void *v) +{ + unsigned long cpu = (unsigned long) m->private; + struct rq *rq = cpu_rq(cpu); + + return sched_server_show_common(m, v, DL_PERIOD, &rq->ext_server); +} + +static int sched_ext_server_period_open(struct inode *inode, struct file *filp) +{ + return single_open(filp, sched_ext_server_period_show, inode->i_private); +} + +static const struct file_operations ext_server_period_fops = { + .open = sched_ext_server_period_open, + .write = sched_ext_server_period_write, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; +#endif /* CONFIG_SCHED_CLASS_EXT */ + static struct dentry *debugfs_sched; static void debugfs_fair_server_init(void) @@ -493,6 +568,29 @@ static void debugfs_fair_server_init(void) } } +#ifdef CONFIG_SCHED_CLASS_EXT +static void debugfs_ext_server_init(void) +{ + struct dentry *d_ext; + unsigned long cpu; + + d_ext = debugfs_create_dir("ext_server", debugfs_sched); + if (!d_ext) + return; + + for_each_possible_cpu(cpu) { + struct dentry *d_cpu; + char buf[32]; + + snprintf(buf, sizeof(buf), "cpu%lu", cpu); + d_cpu = debugfs_create_dir(buf, d_ext); + + debugfs_create_file("runtime", 0644, d_cpu, (void *) cpu, &ext_server_runtime_fops); + debugfs_create_file("period", 0644, d_cpu, (void *) cpu, &ext_server_period_fops); + } +} +#endif /* CONFIG_SCHED_CLASS_EXT */ + static __init int sched_init_debug(void) { struct dentry __maybe_unused *numa; @@ -510,15 +608,13 @@ static __init int sched_init_debug(void) debugfs_create_u32("latency_warn_ms", 0644, debugfs_sched, &sysctl_resched_latency_warn_ms); debugfs_create_u32("latency_warn_once", 0644, debugfs_sched, &sysctl_resched_latency_warn_once); -#ifdef CONFIG_SMP debugfs_create_file("tunable_scaling", 0644, debugfs_sched, NULL, &sched_scaling_fops); debugfs_create_u32("migration_cost_ns", 0644, debugfs_sched, &sysctl_sched_migration_cost); debugfs_create_u32("nr_migrate", 0644, debugfs_sched, &sysctl_sched_nr_migrate); - mutex_lock(&sched_domains_mutex); + sched_domains_mutex_lock(); update_sched_domain_debugfs(); - mutex_unlock(&sched_domains_mutex); -#endif + sched_domains_mutex_unlock(); #ifdef CONFIG_NUMA_BALANCING numa = debugfs_create_dir("numa_balancing", debugfs_sched); @@ -528,18 +624,19 @@ static __init int sched_init_debug(void) debugfs_create_u32("scan_period_max_ms", 0644, numa, &sysctl_numa_balancing_scan_period_max); debugfs_create_u32("scan_size_mb", 0644, numa, &sysctl_numa_balancing_scan_size); debugfs_create_u32("hot_threshold_ms", 0644, numa, &sysctl_numa_balancing_hot_threshold); -#endif +#endif /* CONFIG_NUMA_BALANCING */ debugfs_create_file("debug", 0444, debugfs_sched, NULL, &sched_debug_fops); debugfs_fair_server_init(); +#ifdef CONFIG_SCHED_CLASS_EXT + debugfs_ext_server_init(); +#endif return 0; } late_initcall(sched_init_debug); -#ifdef CONFIG_SMP - static cpumask_var_t sd_sysctl_cpus; static int sd_flags_show(struct seq_file *m, void *v) @@ -586,6 +683,10 @@ static void register_sd(struct sched_domain *sd, struct dentry *parent) debugfs_create_file("flags", 0444, parent, &sd->flags, &sd_flags_fops); debugfs_create_file("groups_flags", 0444, parent, &sd->groups->flags, &sd_flags_fops); debugfs_create_u32("level", 0444, parent, (u32 *)&sd->level); + + if (sd->flags & SD_ASYM_PACKING) + debugfs_create_u32("group_asym_prefer_cpu", 0444, parent, + (u32 *)&sd->groups->asym_prefer_cpu); } void update_sched_domain_debugfs(void) @@ -646,8 +747,6 @@ void dirty_sched_domain_sysctl(int cpu) __cpumask_set_cpu(cpu, sd_sysctl_cpus); } -#endif /* CONFIG_SMP */ - #ifdef CONFIG_FAIR_GROUP_SCHED static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group *tg) { @@ -684,18 +783,16 @@ static void print_cfs_group_stats(struct seq_file *m, int cpu, struct task_group } P(se->load.weight); -#ifdef CONFIG_SMP P(se->avg.load_avg); P(se->avg.util_avg); P(se->avg.runnable_avg); -#endif #undef PN_SCHEDSTAT #undef PN #undef P_SCHEDSTAT #undef P } -#endif +#endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_CGROUP_SCHED static DEFINE_SPINLOCK(sched_debug_lock); @@ -805,10 +902,14 @@ static void print_rq(struct seq_file *m, struct rq *rq, int rq_cpu) void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) { - s64 left_vruntime = -1, min_vruntime, right_vruntime = -1, left_deadline = -1, spread; + s64 left_vruntime = -1, right_vruntime = -1, left_deadline = -1, spread; + s64 zero_vruntime = -1, sum_w_vruntime = -1; + u64 avruntime; struct sched_entity *last, *first, *root; struct rq *rq = cpu_rq(cpu); + unsigned int sum_shift; unsigned long flags; + u64 sum_weight; #ifdef CONFIG_FAIR_GROUP_SCHED SEQ_printf(m, "\n"); @@ -828,17 +929,26 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) last = __pick_last_entity(cfs_rq); if (last) right_vruntime = last->vruntime; - min_vruntime = cfs_rq->min_vruntime; + zero_vruntime = cfs_rq->zero_vruntime; + sum_w_vruntime = cfs_rq->sum_w_vruntime; + sum_weight = cfs_rq->sum_weight; + sum_shift = cfs_rq->sum_shift; + avruntime = avg_vruntime(cfs_rq); raw_spin_rq_unlock_irqrestore(rq, flags); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "left_deadline", SPLIT_NS(left_deadline)); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "left_vruntime", SPLIT_NS(left_vruntime)); - SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "min_vruntime", - SPLIT_NS(min_vruntime)); + SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "zero_vruntime", + SPLIT_NS(zero_vruntime)); + SEQ_printf(m, " .%-30s: %Ld (%d bits)\n", "sum_w_vruntime", + sum_w_vruntime, ilog2(abs(sum_w_vruntime))); + SEQ_printf(m, " .%-30s: %Lu\n", "sum_weight", + sum_weight); + SEQ_printf(m, " .%-30s: %u\n", "sum_shift", sum_shift); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "avg_vruntime", - SPLIT_NS(avg_vruntime(cfs_rq))); + SPLIT_NS(avruntime)); SEQ_printf(m, " .%-30s: %Ld.%06ld\n", "right_vruntime", SPLIT_NS(right_vruntime)); spread = right_vruntime - left_vruntime; @@ -848,7 +958,6 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) SEQ_printf(m, " .%-30s: %d\n", "h_nr_queued", cfs_rq->h_nr_queued); SEQ_printf(m, " .%-30s: %d\n", "h_nr_idle", cfs_rq->h_nr_idle); SEQ_printf(m, " .%-30s: %ld\n", "load", cfs_rq->load.weight); -#ifdef CONFIG_SMP SEQ_printf(m, " .%-30s: %lu\n", "load_avg", cfs_rq->avg.load_avg); SEQ_printf(m, " .%-30s: %lu\n", "runnable_avg", @@ -868,8 +977,7 @@ void print_cfs_rq(struct seq_file *m, int cpu, struct cfs_rq *cfs_rq) cfs_rq->tg_load_avg_contrib); SEQ_printf(m, " .%-30s: %ld\n", "tg_load_avg", atomic_long_read(&cfs_rq->tg->load_avg)); -#endif -#endif +#endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_CFS_BANDWIDTH SEQ_printf(m, " .%-30s: %d\n", "throttled", cfs_rq->throttled); @@ -923,11 +1031,7 @@ void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq) SEQ_printf(m, " .%-30s: %lu\n", #x, (unsigned long)(dl_rq->x)) PU(dl_nr_running); -#ifdef CONFIG_SMP dl_bw = &cpu_rq(cpu)->rd->dl_bw; -#else - dl_bw = &dl_rq->dl_bw; -#endif SEQ_printf(m, " .%-30s: %lld\n", "dl_bw->bw", dl_bw->bw); SEQ_printf(m, " .%-30s: %lld\n", "dl_bw->total_bw", dl_bw->total_bw); @@ -945,9 +1049,9 @@ static void print_cpu(struct seq_file *m, int cpu) SEQ_printf(m, "cpu#%d, %u.%03u MHz\n", cpu, freq / 1000, (freq % 1000)); } -#else +#else /* !CONFIG_X86: */ SEQ_printf(m, "cpu#%d\n", cpu); -#endif +#endif /* !CONFIG_X86 */ #define P(x) \ do { \ @@ -970,12 +1074,10 @@ do { \ #undef P #undef PN -#ifdef CONFIG_SMP #define P64(n) SEQ_printf(m, " .%-30s: %Ld\n", #n, rq->n); P64(avg_idle); P64(max_idle_balance_cost); #undef P64 -#endif #define P(n) SEQ_printf(m, " .%-30s: %d\n", #n, schedstat_val(rq->n)); if (schedstat_enabled()) { @@ -1157,7 +1259,7 @@ static void sched_show_numa(struct task_struct *p, struct seq_file *m) SEQ_printf(m, "current_node=%d, numa_group_id=%d\n", task_node(p), task_numa_group_id(p)); show_numa_stats(p, m); -#endif +#endif /* CONFIG_NUMA_BALANCING */ } void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, @@ -1241,7 +1343,6 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, __PS("nr_involuntary_switches", p->nivcsw); P(se.load.weight); -#ifdef CONFIG_SMP P(se.avg.load_sum); P(se.avg.runnable_sum); P(se.avg.util_sum); @@ -1250,13 +1351,12 @@ void proc_sched_show_task(struct task_struct *p, struct pid_namespace *ns, P(se.avg.util_avg); P(se.avg.last_update_time); PM(se.avg.util_est, ~UTIL_AVG_UNCHANGED); -#endif #ifdef CONFIG_UCLAMP_TASK __PS("uclamp.min", p->uclamp_req[UCLAMP_MIN].value); __PS("uclamp.max", p->uclamp_req[UCLAMP_MAX].value); __PS("effective uclamp.min", uclamp_eff_value(p, UCLAMP_MIN)); __PS("effective uclamp.max", uclamp_eff_value(p, UCLAMP_MAX)); -#endif +#endif /* CONFIG_UCLAMP_TASK */ P(policy); P(prio); if (task_has_dl_policy(p)) { diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c index 7b9dfee858e7..65631e577ee9 100644 --- a/kernel/sched/ext.c +++ b/kernel/sched/ext.c @@ -6,868 +6,37 @@ * Copyright (c) 2022 Tejun Heo <tj@kernel.org> * Copyright (c) 2022 David Vernet <dvernet@meta.com> */ -#define SCX_OP_IDX(op) (offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void))) - -enum scx_consts { - SCX_DSP_DFL_MAX_BATCH = 32, - SCX_DSP_MAX_LOOPS = 32, - SCX_WATCHDOG_MAX_TIMEOUT = 30 * HZ, - - SCX_EXIT_BT_LEN = 64, - SCX_EXIT_MSG_LEN = 1024, - SCX_EXIT_DUMP_DFL_LEN = 32768, - - SCX_CPUPERF_ONE = SCHED_CAPACITY_SCALE, - - /* - * Iterating all tasks may take a while. Periodically drop - * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls. - */ - SCX_OPS_TASK_ITER_BATCH = 32, -}; - -enum scx_exit_kind { - SCX_EXIT_NONE, - SCX_EXIT_DONE, - - SCX_EXIT_UNREG = 64, /* user-space initiated unregistration */ - SCX_EXIT_UNREG_BPF, /* BPF-initiated unregistration */ - SCX_EXIT_UNREG_KERN, /* kernel-initiated unregistration */ - SCX_EXIT_SYSRQ, /* requested by 'S' sysrq */ +#include <linux/btf_ids.h> +#include "ext_idle.h" - SCX_EXIT_ERROR = 1024, /* runtime error, error msg contains details */ - SCX_EXIT_ERROR_BPF, /* ERROR but triggered through scx_bpf_error() */ - SCX_EXIT_ERROR_STALL, /* watchdog detected stalled runnable tasks */ -}; +static DEFINE_RAW_SPINLOCK(scx_sched_lock); /* - * An exit code can be specified when exiting with scx_bpf_exit() or - * scx_ops_exit(), corresponding to exit_kind UNREG_BPF and UNREG_KERN - * respectively. The codes are 64bit of the format: - * - * Bits: [63 .. 48 47 .. 32 31 .. 0] - * [ SYS ACT ] [ SYS RSN ] [ USR ] - * - * SYS ACT: System-defined exit actions - * SYS RSN: System-defined exit reasons - * USR : User-defined exit codes and reasons - * - * Using the above, users may communicate intention and context by ORing system - * actions and/or system reasons with a user-defined exit code. + * NOTE: sched_ext is in the process of growing multiple scheduler support and + * scx_root usage is in a transitional state. Naked dereferences are safe if the + * caller is one of the tasks attached to SCX and explicit RCU dereference is + * necessary otherwise. Naked scx_root dereferences trigger sparse warnings but + * are used as temporary markers to indicate that the dereferences need to be + * updated to point to the associated scheduler instances rather than scx_root. */ -enum scx_exit_code { - /* Reasons */ - SCX_ECODE_RSN_HOTPLUG = 1LLU << 32, - - /* Actions */ - SCX_ECODE_ACT_RESTART = 1LLU << 48, -}; +struct scx_sched __rcu *scx_root; /* - * scx_exit_info is passed to ops.exit() to describe why the BPF scheduler is - * being disabled. + * All scheds, writers must hold both scx_enable_mutex and scx_sched_lock. + * Readers can hold either or rcu_read_lock(). */ -struct scx_exit_info { - /* %SCX_EXIT_* - broad category of the exit reason */ - enum scx_exit_kind kind; - - /* exit code if gracefully exiting */ - s64 exit_code; - - /* textual representation of the above */ - const char *reason; - - /* backtrace if exiting due to an error */ - unsigned long *bt; - u32 bt_len; - - /* informational message */ - char *msg; - - /* debug dump */ - char *dump; -}; - -/* sched_ext_ops.flags */ -enum scx_ops_flags { - /* - * Keep built-in idle tracking even if ops.update_idle() is implemented. - */ - SCX_OPS_KEEP_BUILTIN_IDLE = 1LLU << 0, - - /* - * By default, if there are no other task to run on the CPU, ext core - * keeps running the current task even after its slice expires. If this - * flag is specified, such tasks are passed to ops.enqueue() with - * %SCX_ENQ_LAST. See the comment above %SCX_ENQ_LAST for more info. - */ - SCX_OPS_ENQ_LAST = 1LLU << 1, - - /* - * An exiting task may schedule after PF_EXITING is set. In such cases, - * bpf_task_from_pid() may not be able to find the task and if the BPF - * scheduler depends on pid lookup for dispatching, the task will be - * lost leading to various issues including RCU grace period stalls. - * - * To mask this problem, by default, unhashed tasks are automatically - * dispatched to the local DSQ on enqueue. If the BPF scheduler doesn't - * depend on pid lookups and wants to handle these tasks directly, the - * following flag can be used. - */ - SCX_OPS_ENQ_EXITING = 1LLU << 2, - - /* - * If set, only tasks with policy set to SCHED_EXT are attached to - * sched_ext. If clear, SCHED_NORMAL tasks are also included. - */ - SCX_OPS_SWITCH_PARTIAL = 1LLU << 3, - - /* - * A migration disabled task can only execute on its current CPU. By - * default, such tasks are automatically put on the CPU's local DSQ with - * the default slice on enqueue. If this ops flag is set, they also go - * through ops.enqueue(). - * - * A migration disabled task never invokes ops.select_cpu() as it can - * only select the current CPU. Also, p->cpus_ptr will only contain its - * current CPU while p->nr_cpus_allowed keeps tracking p->user_cpus_ptr - * and thus may disagree with cpumask_weight(p->cpus_ptr). - */ - SCX_OPS_ENQ_MIGRATION_DISABLED = 1LLU << 4, - - /* - * CPU cgroup support flags - */ - SCX_OPS_HAS_CGROUP_WEIGHT = 1LLU << 16, /* cpu.weight */ - - SCX_OPS_ALL_FLAGS = SCX_OPS_KEEP_BUILTIN_IDLE | - SCX_OPS_ENQ_LAST | - SCX_OPS_ENQ_EXITING | - SCX_OPS_ENQ_MIGRATION_DISABLED | - SCX_OPS_SWITCH_PARTIAL | - SCX_OPS_HAS_CGROUP_WEIGHT, +static LIST_HEAD(scx_sched_all); + +#ifdef CONFIG_EXT_SUB_SCHED +static const struct rhashtable_params scx_sched_hash_params = { + .key_len = sizeof_field(struct scx_sched, ops.sub_cgroup_id), + .key_offset = offsetof(struct scx_sched, ops.sub_cgroup_id), + .head_offset = offsetof(struct scx_sched, hash_node), + .insecure_elasticity = true, /* inserted under scx_sched_lock */ }; -/* argument container for ops.init_task() */ -struct scx_init_task_args { - /* - * Set if ops.init_task() is being invoked on the fork path, as opposed - * to the scheduler transition path. - */ - bool fork; -#ifdef CONFIG_EXT_GROUP_SCHED - /* the cgroup the task is joining */ - struct cgroup *cgroup; +static struct rhashtable scx_sched_hash; #endif -}; - -/* argument container for ops.exit_task() */ -struct scx_exit_task_args { - /* Whether the task exited before running on sched_ext. */ - bool cancelled; -}; - -/* argument container for ops->cgroup_init() */ -struct scx_cgroup_init_args { - /* the weight of the cgroup [1..10000] */ - u32 weight; -}; - -enum scx_cpu_preempt_reason { - /* next task is being scheduled by &sched_class_rt */ - SCX_CPU_PREEMPT_RT, - /* next task is being scheduled by &sched_class_dl */ - SCX_CPU_PREEMPT_DL, - /* next task is being scheduled by &sched_class_stop */ - SCX_CPU_PREEMPT_STOP, - /* unknown reason for SCX being preempted */ - SCX_CPU_PREEMPT_UNKNOWN, -}; - -/* - * Argument container for ops->cpu_acquire(). Currently empty, but may be - * expanded in the future. - */ -struct scx_cpu_acquire_args {}; - -/* argument container for ops->cpu_release() */ -struct scx_cpu_release_args { - /* the reason the CPU was preempted */ - enum scx_cpu_preempt_reason reason; - - /* the task that's going to be scheduled on the CPU */ - struct task_struct *task; -}; - -/* - * Informational context provided to dump operations. - */ -struct scx_dump_ctx { - enum scx_exit_kind kind; - s64 exit_code; - const char *reason; - u64 at_ns; - u64 at_jiffies; -}; - -/** - * struct sched_ext_ops - Operation table for BPF scheduler implementation - * - * A BPF scheduler can implement an arbitrary scheduling policy by - * implementing and loading operations in this table. Note that a userland - * scheduling policy can also be implemented using the BPF scheduler - * as a shim layer. - */ -struct sched_ext_ops { - /** - * @select_cpu: Pick the target CPU for a task which is being woken up - * @p: task being woken up - * @prev_cpu: the cpu @p was on before sleeping - * @wake_flags: SCX_WAKE_* - * - * Decision made here isn't final. @p may be moved to any CPU while it - * is getting dispatched for execution later. However, as @p is not on - * the rq at this point, getting the eventual execution CPU right here - * saves a small bit of overhead down the line. - * - * If an idle CPU is returned, the CPU is kicked and will try to - * dispatch. While an explicit custom mechanism can be added, - * select_cpu() serves as the default way to wake up idle CPUs. - * - * @p may be inserted into a DSQ directly by calling - * scx_bpf_dsq_insert(). If so, the ops.enqueue() will be skipped. - * Directly inserting into %SCX_DSQ_LOCAL will put @p in the local DSQ - * of the CPU returned by this operation. - * - * Note that select_cpu() is never called for tasks that can only run - * on a single CPU or tasks with migration disabled, as they don't have - * the option to select a different CPU. See select_task_rq() for - * details. - */ - s32 (*select_cpu)(struct task_struct *p, s32 prev_cpu, u64 wake_flags); - - /** - * @enqueue: Enqueue a task on the BPF scheduler - * @p: task being enqueued - * @enq_flags: %SCX_ENQ_* - * - * @p is ready to run. Insert directly into a DSQ by calling - * scx_bpf_dsq_insert() or enqueue on the BPF scheduler. If not directly - * inserted, the bpf scheduler owns @p and if it fails to dispatch @p, - * the task will stall. - * - * If @p was inserted into a DSQ from ops.select_cpu(), this callback is - * skipped. - */ - void (*enqueue)(struct task_struct *p, u64 enq_flags); - - /** - * @dequeue: Remove a task from the BPF scheduler - * @p: task being dequeued - * @deq_flags: %SCX_DEQ_* - * - * Remove @p from the BPF scheduler. This is usually called to isolate - * the task while updating its scheduling properties (e.g. priority). - * - * The ext core keeps track of whether the BPF side owns a given task or - * not and can gracefully ignore spurious dispatches from BPF side, - * which makes it safe to not implement this method. However, depending - * on the scheduling logic, this can lead to confusing behaviors - e.g. - * scheduling position not being updated across a priority change. - */ - void (*dequeue)(struct task_struct *p, u64 deq_flags); - - /** - * @dispatch: Dispatch tasks from the BPF scheduler and/or user DSQs - * @cpu: CPU to dispatch tasks for - * @prev: previous task being switched out - * - * Called when a CPU's local dsq is empty. The operation should dispatch - * one or more tasks from the BPF scheduler into the DSQs using - * scx_bpf_dsq_insert() and/or move from user DSQs into the local DSQ - * using scx_bpf_dsq_move_to_local(). - * - * The maximum number of times scx_bpf_dsq_insert() can be called - * without an intervening scx_bpf_dsq_move_to_local() is specified by - * ops.dispatch_max_batch. See the comments on top of the two functions - * for more details. - * - * When not %NULL, @prev is an SCX task with its slice depleted. If - * @prev is still runnable as indicated by set %SCX_TASK_QUEUED in - * @prev->scx.flags, it is not enqueued yet and will be enqueued after - * ops.dispatch() returns. To keep executing @prev, return without - * dispatching or moving any tasks. Also see %SCX_OPS_ENQ_LAST. - */ - void (*dispatch)(s32 cpu, struct task_struct *prev); - - /** - * @tick: Periodic tick - * @p: task running currently - * - * This operation is called every 1/HZ seconds on CPUs which are - * executing an SCX task. Setting @p->scx.slice to 0 will trigger an - * immediate dispatch cycle on the CPU. - */ - void (*tick)(struct task_struct *p); - - /** - * @runnable: A task is becoming runnable on its associated CPU - * @p: task becoming runnable - * @enq_flags: %SCX_ENQ_* - * - * This and the following three functions can be used to track a task's - * execution state transitions. A task becomes ->runnable() on a CPU, - * and then goes through one or more ->running() and ->stopping() pairs - * as it runs on the CPU, and eventually becomes ->quiescent() when it's - * done running on the CPU. - * - * @p is becoming runnable on the CPU because it's - * - * - waking up (%SCX_ENQ_WAKEUP) - * - being moved from another CPU - * - being restored after temporarily taken off the queue for an - * attribute change. - * - * This and ->enqueue() are related but not coupled. This operation - * notifies @p's state transition and may not be followed by ->enqueue() - * e.g. when @p is being dispatched to a remote CPU, or when @p is - * being enqueued on a CPU experiencing a hotplug event. Likewise, a - * task may be ->enqueue()'d without being preceded by this operation - * e.g. after exhausting its slice. - */ - void (*runnable)(struct task_struct *p, u64 enq_flags); - - /** - * @running: A task is starting to run on its associated CPU - * @p: task starting to run - * - * See ->runnable() for explanation on the task state notifiers. - */ - void (*running)(struct task_struct *p); - - /** - * @stopping: A task is stopping execution - * @p: task stopping to run - * @runnable: is task @p still runnable? - * - * See ->runnable() for explanation on the task state notifiers. If - * !@runnable, ->quiescent() will be invoked after this operation - * returns. - */ - void (*stopping)(struct task_struct *p, bool runnable); - - /** - * @quiescent: A task is becoming not runnable on its associated CPU - * @p: task becoming not runnable - * @deq_flags: %SCX_DEQ_* - * - * See ->runnable() for explanation on the task state notifiers. - * - * @p is becoming quiescent on the CPU because it's - * - * - sleeping (%SCX_DEQ_SLEEP) - * - being moved to another CPU - * - being temporarily taken off the queue for an attribute change - * (%SCX_DEQ_SAVE) - * - * This and ->dequeue() are related but not coupled. This operation - * notifies @p's state transition and may not be preceded by ->dequeue() - * e.g. when @p is being dispatched to a remote CPU. - */ - void (*quiescent)(struct task_struct *p, u64 deq_flags); - - /** - * @yield: Yield CPU - * @from: yielding task - * @to: optional yield target task - * - * If @to is NULL, @from is yielding the CPU to other runnable tasks. - * The BPF scheduler should ensure that other available tasks are - * dispatched before the yielding task. Return value is ignored in this - * case. - * - * If @to is not-NULL, @from wants to yield the CPU to @to. If the bpf - * scheduler can implement the request, return %true; otherwise, %false. - */ - bool (*yield)(struct task_struct *from, struct task_struct *to); - - /** - * @core_sched_before: Task ordering for core-sched - * @a: task A - * @b: task B - * - * Used by core-sched to determine the ordering between two tasks. See - * Documentation/admin-guide/hw-vuln/core-scheduling.rst for details on - * core-sched. - * - * Both @a and @b are runnable and may or may not currently be queued on - * the BPF scheduler. Should return %true if @a should run before @b. - * %false if there's no required ordering or @b should run before @a. - * - * If not specified, the default is ordering them according to when they - * became runnable. - */ - bool (*core_sched_before)(struct task_struct *a, struct task_struct *b); - - /** - * @set_weight: Set task weight - * @p: task to set weight for - * @weight: new weight [1..10000] - * - * Update @p's weight to @weight. - */ - void (*set_weight)(struct task_struct *p, u32 weight); - - /** - * @set_cpumask: Set CPU affinity - * @p: task to set CPU affinity for - * @cpumask: cpumask of cpus that @p can run on - * - * Update @p's CPU affinity to @cpumask. - */ - void (*set_cpumask)(struct task_struct *p, - const struct cpumask *cpumask); - - /** - * @update_idle: Update the idle state of a CPU - * @cpu: CPU to update the idle state for - * @idle: whether entering or exiting the idle state - * - * This operation is called when @rq's CPU goes or leaves the idle - * state. By default, implementing this operation disables the built-in - * idle CPU tracking and the following helpers become unavailable: - * - * - scx_bpf_select_cpu_dfl() - * - scx_bpf_test_and_clear_cpu_idle() - * - scx_bpf_pick_idle_cpu() - * - * The user also must implement ops.select_cpu() as the default - * implementation relies on scx_bpf_select_cpu_dfl(). - * - * Specify the %SCX_OPS_KEEP_BUILTIN_IDLE flag to keep the built-in idle - * tracking. - */ - void (*update_idle)(s32 cpu, bool idle); - - /** - * @cpu_acquire: A CPU is becoming available to the BPF scheduler - * @cpu: The CPU being acquired by the BPF scheduler. - * @args: Acquire arguments, see the struct definition. - * - * A CPU that was previously released from the BPF scheduler is now once - * again under its control. - */ - void (*cpu_acquire)(s32 cpu, struct scx_cpu_acquire_args *args); - - /** - * @cpu_release: A CPU is taken away from the BPF scheduler - * @cpu: The CPU being released by the BPF scheduler. - * @args: Release arguments, see the struct definition. - * - * The specified CPU is no longer under the control of the BPF - * scheduler. This could be because it was preempted by a higher - * priority sched_class, though there may be other reasons as well. The - * caller should consult @args->reason to determine the cause. - */ - void (*cpu_release)(s32 cpu, struct scx_cpu_release_args *args); - - /** - * @init_task: Initialize a task to run in a BPF scheduler - * @p: task to initialize for BPF scheduling - * @args: init arguments, see the struct definition - * - * Either we're loading a BPF scheduler or a new task is being forked. - * Initialize @p for BPF scheduling. This operation may block and can - * be used for allocations, and is called exactly once for a task. - * - * Return 0 for success, -errno for failure. An error return while - * loading will abort loading of the BPF scheduler. During a fork, it - * will abort that specific fork. - */ - s32 (*init_task)(struct task_struct *p, struct scx_init_task_args *args); - - /** - * @exit_task: Exit a previously-running task from the system - * @p: task to exit - * @args: exit arguments, see the struct definition - * - * @p is exiting or the BPF scheduler is being unloaded. Perform any - * necessary cleanup for @p. - */ - void (*exit_task)(struct task_struct *p, struct scx_exit_task_args *args); - - /** - * @enable: Enable BPF scheduling for a task - * @p: task to enable BPF scheduling for - * - * Enable @p for BPF scheduling. enable() is called on @p any time it - * enters SCX, and is always paired with a matching disable(). - */ - void (*enable)(struct task_struct *p); - - /** - * @disable: Disable BPF scheduling for a task - * @p: task to disable BPF scheduling for - * - * @p is exiting, leaving SCX or the BPF scheduler is being unloaded. - * Disable BPF scheduling for @p. A disable() call is always matched - * with a prior enable() call. - */ - void (*disable)(struct task_struct *p); - - /** - * @dump: Dump BPF scheduler state on error - * @ctx: debug dump context - * - * Use scx_bpf_dump() to generate BPF scheduler specific debug dump. - */ - void (*dump)(struct scx_dump_ctx *ctx); - - /** - * @dump_cpu: Dump BPF scheduler state for a CPU on error - * @ctx: debug dump context - * @cpu: CPU to generate debug dump for - * @idle: @cpu is currently idle without any runnable tasks - * - * Use scx_bpf_dump() to generate BPF scheduler specific debug dump for - * @cpu. If @idle is %true and this operation doesn't produce any - * output, @cpu is skipped for dump. - */ - void (*dump_cpu)(struct scx_dump_ctx *ctx, s32 cpu, bool idle); - - /** - * @dump_task: Dump BPF scheduler state for a runnable task on error - * @ctx: debug dump context - * @p: runnable task to generate debug dump for - * - * Use scx_bpf_dump() to generate BPF scheduler specific debug dump for - * @p. - */ - void (*dump_task)(struct scx_dump_ctx *ctx, struct task_struct *p); - -#ifdef CONFIG_EXT_GROUP_SCHED - /** - * @cgroup_init: Initialize a cgroup - * @cgrp: cgroup being initialized - * @args: init arguments, see the struct definition - * - * Either the BPF scheduler is being loaded or @cgrp created, initialize - * @cgrp for sched_ext. This operation may block. - * - * Return 0 for success, -errno for failure. An error return while - * loading will abort loading of the BPF scheduler. During cgroup - * creation, it will abort the specific cgroup creation. - */ - s32 (*cgroup_init)(struct cgroup *cgrp, - struct scx_cgroup_init_args *args); - - /** - * @cgroup_exit: Exit a cgroup - * @cgrp: cgroup being exited - * - * Either the BPF scheduler is being unloaded or @cgrp destroyed, exit - * @cgrp for sched_ext. This operation my block. - */ - void (*cgroup_exit)(struct cgroup *cgrp); - - /** - * @cgroup_prep_move: Prepare a task to be moved to a different cgroup - * @p: task being moved - * @from: cgroup @p is being moved from - * @to: cgroup @p is being moved to - * - * Prepare @p for move from cgroup @from to @to. This operation may - * block and can be used for allocations. - * - * Return 0 for success, -errno for failure. An error return aborts the - * migration. - */ - s32 (*cgroup_prep_move)(struct task_struct *p, - struct cgroup *from, struct cgroup *to); - - /** - * @cgroup_move: Commit cgroup move - * @p: task being moved - * @from: cgroup @p is being moved from - * @to: cgroup @p is being moved to - * - * Commit the move. @p is dequeued during this operation. - */ - void (*cgroup_move)(struct task_struct *p, - struct cgroup *from, struct cgroup *to); - - /** - * @cgroup_cancel_move: Cancel cgroup move - * @p: task whose cgroup move is being canceled - * @from: cgroup @p was being moved from - * @to: cgroup @p was being moved to - * - * @p was cgroup_prep_move()'d but failed before reaching cgroup_move(). - * Undo the preparation. - */ - void (*cgroup_cancel_move)(struct task_struct *p, - struct cgroup *from, struct cgroup *to); - - /** - * @cgroup_set_weight: A cgroup's weight is being changed - * @cgrp: cgroup whose weight is being updated - * @weight: new weight [1..10000] - * - * Update @tg's weight to @weight. - */ - void (*cgroup_set_weight)(struct cgroup *cgrp, u32 weight); -#endif /* CONFIG_EXT_GROUP_SCHED */ - - /* - * All online ops must come before ops.cpu_online(). - */ - - /** - * @cpu_online: A CPU became online - * @cpu: CPU which just came up - * - * @cpu just came online. @cpu will not call ops.enqueue() or - * ops.dispatch(), nor run tasks associated with other CPUs beforehand. - */ - void (*cpu_online)(s32 cpu); - - /** - * @cpu_offline: A CPU is going offline - * @cpu: CPU which is going offline - * - * @cpu is going offline. @cpu will not call ops.enqueue() or - * ops.dispatch(), nor run tasks associated with other CPUs afterwards. - */ - void (*cpu_offline)(s32 cpu); - - /* - * All CPU hotplug ops must come before ops.init(). - */ - - /** - * @init: Initialize the BPF scheduler - */ - s32 (*init)(void); - - /** - * @exit: Clean up after the BPF scheduler - * @info: Exit info - * - * ops.exit() is also called on ops.init() failure, which is a bit - * unusual. This is to allow rich reporting through @info on how - * ops.init() failed. - */ - void (*exit)(struct scx_exit_info *info); - - /** - * @dispatch_max_batch: Max nr of tasks that dispatch() can dispatch - */ - u32 dispatch_max_batch; - - /** - * @flags: %SCX_OPS_* flags - */ - u64 flags; - - /** - * @timeout_ms: The maximum amount of time, in milliseconds, that a - * runnable task should be able to wait before being scheduled. The - * maximum timeout may not exceed the default timeout of 30 seconds. - * - * Defaults to the maximum allowed timeout value of 30 seconds. - */ - u32 timeout_ms; - - /** - * @exit_dump_len: scx_exit_info.dump buffer length. If 0, the default - * value of 32768 is used. - */ - u32 exit_dump_len; - - /** - * @hotplug_seq: A sequence number that may be set by the scheduler to - * detect when a hotplug event has occurred during the loading process. - * If 0, no detection occurs. Otherwise, the scheduler will fail to - * load if the sequence number does not match @scx_hotplug_seq on the - * enable path. - */ - u64 hotplug_seq; - - /** - * @name: BPF scheduler's name - * - * Must be a non-zero valid BPF object name including only isalnum(), - * '_' and '.' chars. Shows up in kernel.sched_ext_ops sysctl while the - * BPF scheduler is enabled. - */ - char name[SCX_OPS_NAME_LEN]; -}; - -enum scx_opi { - SCX_OPI_BEGIN = 0, - SCX_OPI_NORMAL_BEGIN = 0, - SCX_OPI_NORMAL_END = SCX_OP_IDX(cpu_online), - SCX_OPI_CPU_HOTPLUG_BEGIN = SCX_OP_IDX(cpu_online), - SCX_OPI_CPU_HOTPLUG_END = SCX_OP_IDX(init), - SCX_OPI_END = SCX_OP_IDX(init), -}; - -enum scx_wake_flags { - /* expose select WF_* flags as enums */ - SCX_WAKE_FORK = WF_FORK, - SCX_WAKE_TTWU = WF_TTWU, - SCX_WAKE_SYNC = WF_SYNC, -}; - -enum scx_enq_flags { - /* expose select ENQUEUE_* flags as enums */ - SCX_ENQ_WAKEUP = ENQUEUE_WAKEUP, - SCX_ENQ_HEAD = ENQUEUE_HEAD, - SCX_ENQ_CPU_SELECTED = ENQUEUE_RQ_SELECTED, - - /* high 32bits are SCX specific */ - - /* - * Set the following to trigger preemption when calling - * scx_bpf_dsq_insert() with a local dsq as the target. The slice of the - * current task is cleared to zero and the CPU is kicked into the - * scheduling path. Implies %SCX_ENQ_HEAD. - */ - SCX_ENQ_PREEMPT = 1LLU << 32, - - /* - * The task being enqueued was previously enqueued on the current CPU's - * %SCX_DSQ_LOCAL, but was removed from it in a call to the - * bpf_scx_reenqueue_local() kfunc. If bpf_scx_reenqueue_local() was - * invoked in a ->cpu_release() callback, and the task is again - * dispatched back to %SCX_LOCAL_DSQ by this current ->enqueue(), the - * task will not be scheduled on the CPU until at least the next invocation - * of the ->cpu_acquire() callback. - */ - SCX_ENQ_REENQ = 1LLU << 40, - - /* - * The task being enqueued is the only task available for the cpu. By - * default, ext core keeps executing such tasks but when - * %SCX_OPS_ENQ_LAST is specified, they're ops.enqueue()'d with the - * %SCX_ENQ_LAST flag set. - * - * The BPF scheduler is responsible for triggering a follow-up - * scheduling event. Otherwise, Execution may stall. - */ - SCX_ENQ_LAST = 1LLU << 41, - - /* high 8 bits are internal */ - __SCX_ENQ_INTERNAL_MASK = 0xffLLU << 56, - - SCX_ENQ_CLEAR_OPSS = 1LLU << 56, - SCX_ENQ_DSQ_PRIQ = 1LLU << 57, -}; - -enum scx_deq_flags { - /* expose select DEQUEUE_* flags as enums */ - SCX_DEQ_SLEEP = DEQUEUE_SLEEP, - - /* high 32bits are SCX specific */ - - /* - * The generic core-sched layer decided to execute the task even though - * it hasn't been dispatched yet. Dequeue from the BPF side. - */ - SCX_DEQ_CORE_SCHED_EXEC = 1LLU << 32, -}; - -enum scx_pick_idle_cpu_flags { - SCX_PICK_IDLE_CORE = 1LLU << 0, /* pick a CPU whose SMT siblings are also idle */ -}; - -enum scx_kick_flags { - /* - * Kick the target CPU if idle. Guarantees that the target CPU goes - * through at least one full scheduling cycle before going idle. If the - * target CPU can be determined to be currently not idle and going to go - * through a scheduling cycle before going idle, noop. - */ - SCX_KICK_IDLE = 1LLU << 0, - - /* - * Preempt the current task and execute the dispatch path. If the - * current task of the target CPU is an SCX task, its ->scx.slice is - * cleared to zero before the scheduling path is invoked so that the - * task expires and the dispatch path is invoked. - */ - SCX_KICK_PREEMPT = 1LLU << 1, - - /* - * Wait for the CPU to be rescheduled. The scx_bpf_kick_cpu() call will - * return after the target CPU finishes picking the next task. - */ - SCX_KICK_WAIT = 1LLU << 2, -}; - -enum scx_tg_flags { - SCX_TG_ONLINE = 1U << 0, - SCX_TG_INITED = 1U << 1, -}; - -enum scx_ops_enable_state { - SCX_OPS_ENABLING, - SCX_OPS_ENABLED, - SCX_OPS_DISABLING, - SCX_OPS_DISABLED, -}; - -static const char *scx_ops_enable_state_str[] = { - [SCX_OPS_ENABLING] = "enabling", - [SCX_OPS_ENABLED] = "enabled", - [SCX_OPS_DISABLING] = "disabling", - [SCX_OPS_DISABLED] = "disabled", -}; - -/* - * sched_ext_entity->ops_state - * - * Used to track the task ownership between the SCX core and the BPF scheduler. - * State transitions look as follows: - * - * NONE -> QUEUEING -> QUEUED -> DISPATCHING - * ^ | | - * | v v - * \-------------------------------/ - * - * QUEUEING and DISPATCHING states can be waited upon. See wait_ops_state() call - * sites for explanations on the conditions being waited upon and why they are - * safe. Transitions out of them into NONE or QUEUED must store_release and the - * waiters should load_acquire. - * - * Tracking scx_ops_state enables sched_ext core to reliably determine whether - * any given task can be dispatched by the BPF scheduler at all times and thus - * relaxes the requirements on the BPF scheduler. This allows the BPF scheduler - * to try to dispatch any task anytime regardless of its state as the SCX core - * can safely reject invalid dispatches. - */ -enum scx_ops_state { - SCX_OPSS_NONE, /* owned by the SCX core */ - SCX_OPSS_QUEUEING, /* in transit to the BPF scheduler */ - SCX_OPSS_QUEUED, /* owned by the BPF scheduler */ - SCX_OPSS_DISPATCHING, /* in transit back to the SCX core */ - - /* - * QSEQ brands each QUEUED instance so that, when dispatch races - * dequeue/requeue, the dispatcher can tell whether it still has a claim - * on the task being dispatched. - * - * As some 32bit archs can't do 64bit store_release/load_acquire, - * p->scx.ops_state is atomic_long_t which leaves 30 bits for QSEQ on - * 32bit machines. The dispatch race window QSEQ protects is very narrow - * and runs with IRQ disabled. 30 bits should be sufficient. - */ - SCX_OPSS_QSEQ_SHIFT = 2, -}; - -/* Use macros to ensure that the type is unsigned long for the masks */ -#define SCX_OPSS_STATE_MASK ((1LU << SCX_OPSS_QSEQ_SHIFT) - 1) -#define SCX_OPSS_QSEQ_MASK (~SCX_OPSS_STATE_MASK) /* * During exit, a task may schedule after losing its PIDs. When disabling the @@ -875,58 +44,45 @@ enum scx_ops_state { * guarantee system safety. Maintain a dedicated task list which contains every * task between its fork and eventual free. */ -static DEFINE_SPINLOCK(scx_tasks_lock); +static DEFINE_RAW_SPINLOCK(scx_tasks_lock); static LIST_HEAD(scx_tasks); /* ops enable/disable */ -static struct kthread_worker *scx_ops_helper; -static DEFINE_MUTEX(scx_ops_enable_mutex); -DEFINE_STATIC_KEY_FALSE(__scx_ops_enabled); +static DEFINE_MUTEX(scx_enable_mutex); +DEFINE_STATIC_KEY_FALSE(__scx_enabled); DEFINE_STATIC_PERCPU_RWSEM(scx_fork_rwsem); -static atomic_t scx_ops_enable_state_var = ATOMIC_INIT(SCX_OPS_DISABLED); -static unsigned long scx_in_softlockup; -static atomic_t scx_ops_breather_depth = ATOMIC_INIT(0); -static int scx_ops_bypass_depth; -static bool scx_ops_init_task_enabled; +static atomic_t scx_enable_state_var = ATOMIC_INIT(SCX_DISABLED); +static DEFINE_RAW_SPINLOCK(scx_bypass_lock); +static bool scx_init_task_enabled; static bool scx_switching_all; DEFINE_STATIC_KEY_FALSE(__scx_switched_all); -static struct sched_ext_ops scx_ops; -static bool scx_warned_zero_slice; - -static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_last); -static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_exiting); -static DEFINE_STATIC_KEY_FALSE(scx_ops_enq_migration_disabled); -static DEFINE_STATIC_KEY_FALSE(scx_ops_cpu_preempt); -static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_enabled); - -#ifdef CONFIG_SMP -static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_llc); -static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_numa); -#endif - -static struct static_key_false scx_has_op[SCX_OPI_END] = - { [0 ... SCX_OPI_END-1] = STATIC_KEY_FALSE_INIT }; - -static atomic_t scx_exit_kind = ATOMIC_INIT(SCX_EXIT_DONE); -static struct scx_exit_info *scx_exit_info; - static atomic_long_t scx_nr_rejected = ATOMIC_LONG_INIT(0); static atomic_long_t scx_hotplug_seq = ATOMIC_LONG_INIT(0); +#ifdef CONFIG_EXT_SUB_SCHED +/* + * The sub sched being enabled. Used by scx_disable_and_exit_task() to exit + * tasks for the sub-sched being enabled. Use a global variable instead of a + * per-task field as all enables are serialized. + */ +static struct scx_sched *scx_enabling_sub_sched; +#else +#define scx_enabling_sub_sched (struct scx_sched *)NULL +#endif /* CONFIG_EXT_SUB_SCHED */ + /* - * A monotically increasing sequence number that is incremented every time a - * scheduler is enabled. This can be used by to check if any custom sched_ext + * A monotonically increasing sequence number that is incremented every time a + * scheduler is enabled. This can be used to check if any custom sched_ext * scheduler has ever been used in the system. */ static atomic_long_t scx_enable_seq = ATOMIC_LONG_INIT(0); /* - * The maximum amount of time in jiffies that a task may be runnable without - * being scheduled on a CPU. If this timeout is exceeded, it will trigger - * scx_ops_error(). + * Watchdog interval. All scx_sched's share a single watchdog timer and the + * interval is half of the shortest sch->watchdog_timeout. */ -static unsigned long scx_watchdog_timeout; +static unsigned long scx_watchdog_interval; /* * The last time the delayed work was run. This delayed work relies on @@ -938,23 +94,19 @@ static unsigned long scx_watchdog_timestamp = INITIAL_JIFFIES; static struct delayed_work scx_watchdog_work; -/* idle tracking */ -#ifdef CONFIG_SMP -#ifdef CONFIG_CPUMASK_OFFSTACK -#define CL_ALIGNED_IF_ONSTACK -#else -#define CL_ALIGNED_IF_ONSTACK __cacheline_aligned_in_smp -#endif - -static struct { - cpumask_var_t cpu; - cpumask_var_t smt; -} idle_masks CL_ALIGNED_IF_ONSTACK; - -#endif /* CONFIG_SMP */ +/* + * For %SCX_KICK_WAIT: Each CPU has a pointer to an array of kick_sync sequence + * numbers. The arrays are allocated with kvzalloc() as size can exceed percpu + * allocator limits on large machines. O(nr_cpu_ids^2) allocation, allocated + * lazily when enabling and freed when disabling to avoid waste when sched_ext + * isn't active. + */ +struct scx_kick_syncs { + struct rcu_head rcu; + unsigned long syncs[]; +}; -/* for %SCX_KICK_WAIT */ -static unsigned long __percpu *scx_kick_cpus_pnt_seqs; +static DEFINE_PER_CPU(struct scx_kick_syncs __rcu *, scx_kick_syncs); /* * Direct dispatch marker. @@ -965,44 +117,14 @@ static unsigned long __percpu *scx_kick_cpus_pnt_seqs; */ static DEFINE_PER_CPU(struct task_struct *, direct_dispatch_task); -/* - * Dispatch queues. - * - * The global DSQ (%SCX_DSQ_GLOBAL) is split per-node for scalability. This is - * to avoid live-locking in bypass mode where all tasks are dispatched to - * %SCX_DSQ_GLOBAL and all CPUs consume from it. If per-node split isn't - * sufficient, it can be further split. - */ -static struct scx_dispatch_q **global_dsqs; - static const struct rhashtable_params dsq_hash_params = { .key_len = sizeof_field(struct scx_dispatch_q, id), .key_offset = offsetof(struct scx_dispatch_q, id), .head_offset = offsetof(struct scx_dispatch_q, hash_node), }; -static struct rhashtable dsq_hash; static LLIST_HEAD(dsqs_to_free); -/* dispatch buf */ -struct scx_dsp_buf_ent { - struct task_struct *task; - unsigned long qseq; - u64 dsq_id; - u64 enq_flags; -}; - -static u32 scx_dsp_max_batch; - -struct scx_dsp_ctx { - struct rq *rq; - u32 cursor; - u32 nr_tasks; - struct scx_dsp_buf_ent buf[]; -}; - -static struct scx_dsp_ctx __percpu *scx_dsp_ctx; - /* string formatting from BPF */ struct scx_bstr_buf { u64 data[MAX_BPRINTF_VARARGS]; @@ -1013,6 +135,8 @@ static DEFINE_RAW_SPINLOCK(scx_exit_bstr_buf_lock); static struct scx_bstr_buf scx_exit_bstr_buf; /* ops debug dump */ +static DEFINE_RAW_SPINLOCK(scx_dump_lock); + struct scx_dump_data { s32 cpu; bool first; @@ -1028,27 +152,73 @@ static struct scx_dump_data scx_dump_data = { /* /sys/kernel/sched_ext interface */ static struct kset *scx_kset; -static struct kobject *scx_root_kobj; + +/* + * Parameters that can be adjusted through /sys/module/sched_ext/parameters. + * There usually is no reason to modify these as normal scheduler operation + * shouldn't be affected by them. The knobs are primarily for debugging. + */ +static unsigned int scx_slice_bypass_us = SCX_SLICE_BYPASS / NSEC_PER_USEC; +static unsigned int scx_bypass_lb_intv_us = SCX_BYPASS_LB_DFL_INTV_US; + +static int set_slice_us(const char *val, const struct kernel_param *kp) +{ + return param_set_uint_minmax(val, kp, 100, 100 * USEC_PER_MSEC); +} + +static const struct kernel_param_ops slice_us_param_ops = { + .set = set_slice_us, + .get = param_get_uint, +}; + +static int set_bypass_lb_intv_us(const char *val, const struct kernel_param *kp) +{ + return param_set_uint_minmax(val, kp, 0, 10 * USEC_PER_SEC); +} + +static const struct kernel_param_ops bypass_lb_intv_us_param_ops = { + .set = set_bypass_lb_intv_us, + .get = param_get_uint, +}; + +#undef MODULE_PARAM_PREFIX +#define MODULE_PARAM_PREFIX "sched_ext." + +module_param_cb(slice_bypass_us, &slice_us_param_ops, &scx_slice_bypass_us, 0600); +MODULE_PARM_DESC(slice_bypass_us, "bypass slice in microseconds, applied on [un]load (100us to 100ms)"); +module_param_cb(bypass_lb_intv_us, &bypass_lb_intv_us_param_ops, &scx_bypass_lb_intv_us, 0600); +MODULE_PARM_DESC(bypass_lb_intv_us, "bypass load balance interval in microseconds (0 (disable) to 10s)"); + +#undef MODULE_PARAM_PREFIX #define CREATE_TRACE_POINTS #include <trace/events/sched_ext.h> -static void process_ddsp_deferred_locals(struct rq *rq); -static void scx_bpf_kick_cpu(s32 cpu, u64 flags); -static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind, - s64 exit_code, - const char *fmt, ...); +static void run_deferred(struct rq *rq); +static bool task_dead_and_done(struct task_struct *p); +static void scx_kick_cpu(struct scx_sched *sch, s32 cpu, u64 flags); +static void scx_disable(struct scx_sched *sch, enum scx_exit_kind kind); +static bool scx_vexit(struct scx_sched *sch, enum scx_exit_kind kind, + s64 exit_code, const char *fmt, va_list args); -#define scx_ops_error_kind(err, fmt, args...) \ - scx_ops_exit_kind((err), 0, fmt, ##args) +static __printf(4, 5) bool scx_exit(struct scx_sched *sch, + enum scx_exit_kind kind, s64 exit_code, + const char *fmt, ...) +{ + va_list args; + bool ret; + + va_start(args, fmt); + ret = scx_vexit(sch, kind, exit_code, fmt, args); + va_end(args); -#define scx_ops_exit(code, fmt, args...) \ - scx_ops_exit_kind(SCX_EXIT_UNREG_KERN, (code), fmt, ##args) + return ret; +} -#define scx_ops_error(fmt, args...) \ - scx_ops_error_kind(SCX_EXIT_ERROR, fmt, ##args) +#define scx_error(sch, fmt, args...) scx_exit((sch), SCX_EXIT_ERROR, 0, fmt, ##args) +#define scx_verror(sch, fmt, args) scx_vexit((sch), SCX_EXIT_ERROR, 0, fmt, args) -#define SCX_HAS_OP(op) static_branch_likely(&scx_has_op[SCX_OP_IDX(op)]) +#define SCX_HAS_OP(sch, op) test_bit(SCX_OP_IDX(op), (sch)->has_op) static long jiffies_delta_msecs(unsigned long at, unsigned long now) { @@ -1058,178 +228,352 @@ static long jiffies_delta_msecs(unsigned long at, unsigned long now) return -(long)jiffies_to_msecs(now - at); } -/* if the highest set bit is N, return a mask with bits [N+1, 31] set */ -static u32 higher_bits(u32 flags) +static bool u32_before(u32 a, u32 b) { - return ~((1 << fls(flags)) - 1); + return (s32)(a - b) < 0; } -/* return the mask with only the highest bit set */ -static u32 highest_bit(u32 flags) +#ifdef CONFIG_EXT_SUB_SCHED +/** + * scx_parent - Find the parent sched + * @sch: sched to find the parent of + * + * Returns the parent scheduler or %NULL if @sch is root. + */ +static struct scx_sched *scx_parent(struct scx_sched *sch) { - int bit = fls(flags); - return ((u64)1 << bit) >> 1; + if (sch->level) + return sch->ancestors[sch->level - 1]; + else + return NULL; } -static bool u32_before(u32 a, u32 b) +/** + * scx_next_descendant_pre - find the next descendant for pre-order walk + * @pos: the current position (%NULL to initiate traversal) + * @root: sched whose descendants to walk + * + * To be used by scx_for_each_descendant_pre(). Find the next descendant to + * visit for pre-order traversal of @root's descendants. @root is included in + * the iteration and the first node to be visited. + */ +static struct scx_sched *scx_next_descendant_pre(struct scx_sched *pos, + struct scx_sched *root) { - return (s32)(a - b) < 0; + struct scx_sched *next; + + lockdep_assert(lockdep_is_held(&scx_enable_mutex) || + lockdep_is_held(&scx_sched_lock)); + + /* if first iteration, visit @root */ + if (!pos) + return root; + + /* visit the first child if exists */ + next = list_first_entry_or_null(&pos->children, struct scx_sched, sibling); + if (next) + return next; + + /* no child, visit my or the closest ancestor's next sibling */ + while (pos != root) { + if (!list_is_last(&pos->sibling, &scx_parent(pos)->children)) + return list_next_entry(pos, sibling); + pos = scx_parent(pos); + } + + return NULL; } -static struct scx_dispatch_q *find_global_dsq(struct task_struct *p) +static struct scx_sched *scx_find_sub_sched(u64 cgroup_id) { - return global_dsqs[cpu_to_node(task_cpu(p))]; + return rhashtable_lookup(&scx_sched_hash, &cgroup_id, + scx_sched_hash_params); } -static struct scx_dispatch_q *find_user_dsq(u64 dsq_id) +static void scx_set_task_sched(struct task_struct *p, struct scx_sched *sch) { - return rhashtable_lookup_fast(&dsq_hash, &dsq_id, dsq_hash_params); + rcu_assign_pointer(p->scx.sched, sch); } +#else /* CONFIG_EXT_SUB_SCHED */ +static struct scx_sched *scx_parent(struct scx_sched *sch) { return NULL; } +static struct scx_sched *scx_next_descendant_pre(struct scx_sched *pos, struct scx_sched *root) { return pos ? NULL : root; } +static void scx_set_task_sched(struct task_struct *p, struct scx_sched *sch) {} +#endif /* CONFIG_EXT_SUB_SCHED */ -/* - * scx_kf_mask enforcement. Some kfuncs can only be called from specific SCX - * ops. When invoking SCX ops, SCX_CALL_OP[_RET]() should be used to indicate - * the allowed kfuncs and those kfuncs should use scx_kf_allowed() to check - * whether it's running from an allowed context. +/** + * scx_is_descendant - Test whether sched is a descendant + * @sch: sched to test + * @ancestor: ancestor sched to test against * - * @mask is constant, always inline to cull the mask calculations. + * Test whether @sch is a descendant of @ancestor. */ -static __always_inline void scx_kf_allow(u32 mask) +static bool scx_is_descendant(struct scx_sched *sch, struct scx_sched *ancestor) { - /* nesting is allowed only in increasing scx_kf_mask order */ - WARN_ONCE((mask | higher_bits(mask)) & current->scx.kf_mask, - "invalid nesting current->scx.kf_mask=0x%x mask=0x%x\n", - current->scx.kf_mask, mask); - current->scx.kf_mask |= mask; - barrier(); + if (sch->level < ancestor->level) + return false; + return sch->ancestors[ancestor->level] == ancestor; +} + +/** + * scx_for_each_descendant_pre - pre-order walk of a sched's descendants + * @pos: iteration cursor + * @root: sched to walk the descendants of + * + * Walk @root's descendants. @root is included in the iteration and the first + * node to be visited. Must be called with either scx_enable_mutex or + * scx_sched_lock held. + */ +#define scx_for_each_descendant_pre(pos, root) \ + for ((pos) = scx_next_descendant_pre(NULL, (root)); (pos); \ + (pos) = scx_next_descendant_pre((pos), (root))) + +static struct scx_dispatch_q *find_global_dsq(struct scx_sched *sch, s32 cpu) +{ + return &sch->pnode[cpu_to_node(cpu)]->global_dsq; +} + +static struct scx_dispatch_q *find_user_dsq(struct scx_sched *sch, u64 dsq_id) +{ + return rhashtable_lookup(&sch->dsq_hash, &dsq_id, dsq_hash_params); +} + +static const struct sched_class *scx_setscheduler_class(struct task_struct *p) +{ + if (p->sched_class == &stop_sched_class) + return &stop_sched_class; + + return __setscheduler_class(p->policy, p->prio); +} + +static struct scx_dispatch_q *bypass_dsq(struct scx_sched *sch, s32 cpu) +{ + return &per_cpu_ptr(sch->pcpu, cpu)->bypass_dsq; +} + +static struct scx_dispatch_q *bypass_enq_target_dsq(struct scx_sched *sch, s32 cpu) +{ +#ifdef CONFIG_EXT_SUB_SCHED + /* + * If @sch is a sub-sched which is bypassing, its tasks should go into + * the bypass DSQs of the nearest ancestor which is not bypassing. The + * not-bypassing ancestor is responsible for scheduling all tasks from + * bypassing sub-trees. If all ancestors including root are bypassing, + * all tasks should go to the root's bypass DSQs. + * + * Whenever a sched starts bypassing, all runnable tasks in its subtree + * are re-enqueued after scx_bypassing() is turned on, guaranteeing that + * all tasks are transferred to the right DSQs. + */ + while (scx_parent(sch) && scx_bypassing(sch, cpu)) + sch = scx_parent(sch); +#endif /* CONFIG_EXT_SUB_SCHED */ + + return bypass_dsq(sch, cpu); +} + +/** + * bypass_dsp_enabled - Check if bypass dispatch path is enabled + * @sch: scheduler to check + * + * When a descendant scheduler enters bypass mode, bypassed tasks are scheduled + * by the nearest non-bypassing ancestor, or the root scheduler if all ancestors + * are bypassing. In the former case, the ancestor is not itself bypassing but + * its bypass DSQs will be populated with bypassed tasks from descendants. Thus, + * the ancestor's bypass dispatch path must be active even though its own + * bypass_depth remains zero. + * + * This function checks bypass_dsp_enable_depth which is managed separately from + * bypass_depth to enable this decoupling. See enable_bypass_dsp() and + * disable_bypass_dsp(). + */ +static bool bypass_dsp_enabled(struct scx_sched *sch) +{ + return unlikely(atomic_read(&sch->bypass_dsp_enable_depth)); +} + +/** + * rq_is_open - Is the rq available for immediate execution of an SCX task? + * @rq: rq to test + * @enq_flags: optional %SCX_ENQ_* of the task being enqueued + * + * Returns %true if @rq is currently open for executing an SCX task. After a + * %false return, @rq is guaranteed to invoke SCX dispatch path at least once + * before going to idle and not inserting a task into @rq's local DSQ after a + * %false return doesn't cause @rq to stall. + */ +static bool rq_is_open(struct rq *rq, u64 enq_flags) +{ + lockdep_assert_rq_held(rq); + + /* + * A higher-priority class task is either running or in the process of + * waking up on @rq. + */ + if (sched_class_above(rq->next_class, &ext_sched_class)) + return false; + + /* + * @rq is either in transition to or in idle and there is no + * higher-priority class task waking up on it. + */ + if (sched_class_above(&ext_sched_class, rq->next_class)) + return true; + + /* + * @rq is either picking, in transition to, or running an SCX task. + */ + + /* + * If we're in the dispatch path holding rq lock, $curr may or may not + * be ready depending on whether the on-going dispatch decides to extend + * $curr's slice. We say yes here and resolve it at the end of dispatch. + * See balance_one(). + */ + if (rq->scx.flags & SCX_RQ_IN_BALANCE) + return true; + + /* + * %SCX_ENQ_PREEMPT clears $curr's slice if on SCX and kicks dispatch, + * so allow it to avoid spuriously triggering reenq on a combined + * PREEMPT|IMMED insertion. + */ + if (enq_flags & SCX_ENQ_PREEMPT) + return true; + + /* + * @rq is either in transition to or running an SCX task and can't go + * idle without another SCX dispatch cycle. + */ + return false; } -static void scx_kf_disallow(u32 mask) +/* + * Track the rq currently locked. + * + * This allows kfuncs to safely operate on rq from any scx ops callback, + * knowing which rq is already locked. + */ +DEFINE_PER_CPU(struct rq *, scx_locked_rq_state); + +static inline void update_locked_rq(struct rq *rq) { - barrier(); - current->scx.kf_mask &= ~mask; + /* + * Check whether @rq is actually locked. This can help expose bugs + * or incorrect assumptions about the context in which a kfunc or + * callback is executed. + */ + if (rq) + lockdep_assert_rq_held(rq); + __this_cpu_write(scx_locked_rq_state, rq); } -#define SCX_CALL_OP(mask, op, args...) \ +/* + * SCX ops can recurse via scx_bpf_sub_dispatch() - the inner call must not + * clobber the outer's scx_locked_rq_state. Save it on entry, restore on exit. + */ +#define SCX_CALL_OP(sch, op, locked_rq, args...) \ do { \ - if (mask) { \ - scx_kf_allow(mask); \ - scx_ops.op(args); \ - scx_kf_disallow(mask); \ - } else { \ - scx_ops.op(args); \ + struct rq *__prev_locked_rq; \ + \ + if (locked_rq) { \ + __prev_locked_rq = scx_locked_rq(); \ + update_locked_rq(locked_rq); \ } \ + (sch)->ops.op(args); \ + if (locked_rq) \ + update_locked_rq(__prev_locked_rq); \ } while (0) -#define SCX_CALL_OP_RET(mask, op, args...) \ +#define SCX_CALL_OP_RET(sch, op, locked_rq, args...) \ ({ \ - __typeof__(scx_ops.op(args)) __ret; \ - if (mask) { \ - scx_kf_allow(mask); \ - __ret = scx_ops.op(args); \ - scx_kf_disallow(mask); \ - } else { \ - __ret = scx_ops.op(args); \ + struct rq *__prev_locked_rq; \ + __typeof__((sch)->ops.op(args)) __ret; \ + \ + if (locked_rq) { \ + __prev_locked_rq = scx_locked_rq(); \ + update_locked_rq(locked_rq); \ } \ + __ret = (sch)->ops.op(args); \ + if (locked_rq) \ + update_locked_rq(__prev_locked_rq); \ __ret; \ }) /* - * Some kfuncs are allowed only on the tasks that are subjects of the - * in-progress scx_ops operation for, e.g., locking guarantees. To enforce such - * restrictions, the following SCX_CALL_OP_*() variants should be used when - * invoking scx_ops operations that take task arguments. These can only be used - * for non-nesting operations due to the way the tasks are tracked. - * - * kfuncs which can only operate on such tasks can in turn use - * scx_kf_allowed_on_arg_tasks() to test whether the invocation is allowed on - * the specific task. + * SCX_CALL_OP_TASK*() invokes an SCX op that takes one or two task arguments + * and records them in current->scx.kf_tasks[] for the duration of the call. A + * kfunc invoked from inside such an op can then use + * scx_kf_arg_task_ok() to verify that its task argument is one of + * those subject tasks. + * + * Every SCX_CALL_OP_TASK*() call site invokes its op with @p's rq lock held - + * either via the @locked_rq argument here, or (for ops.select_cpu()) via @p's + * pi_lock held by try_to_wake_up() with rq tracking via scx_rq.in_select_cpu. + * So if kf_tasks[] is set, @p's scheduler-protected fields are stable. + * + * kf_tasks[] can not stack, so task-based SCX ops must not nest. The + * WARN_ON_ONCE() in each macro catches a re-entry of any of the three variants + * while a previous one is still in progress. */ -#define SCX_CALL_OP_TASK(mask, op, task, args...) \ +#define SCX_CALL_OP_TASK(sch, op, locked_rq, task, args...) \ do { \ - BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ + WARN_ON_ONCE(current->scx.kf_tasks[0]); \ current->scx.kf_tasks[0] = task; \ - SCX_CALL_OP(mask, op, task, ##args); \ + SCX_CALL_OP((sch), op, locked_rq, task, ##args); \ current->scx.kf_tasks[0] = NULL; \ } while (0) -#define SCX_CALL_OP_TASK_RET(mask, op, task, args...) \ +#define SCX_CALL_OP_TASK_RET(sch, op, locked_rq, task, args...) \ ({ \ - __typeof__(scx_ops.op(task, ##args)) __ret; \ - BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ + __typeof__((sch)->ops.op(task, ##args)) __ret; \ + WARN_ON_ONCE(current->scx.kf_tasks[0]); \ current->scx.kf_tasks[0] = task; \ - __ret = SCX_CALL_OP_RET(mask, op, task, ##args); \ + __ret = SCX_CALL_OP_RET((sch), op, locked_rq, task, ##args); \ current->scx.kf_tasks[0] = NULL; \ __ret; \ }) -#define SCX_CALL_OP_2TASKS_RET(mask, op, task0, task1, args...) \ +#define SCX_CALL_OP_2TASKS_RET(sch, op, locked_rq, task0, task1, args...) \ ({ \ - __typeof__(scx_ops.op(task0, task1, ##args)) __ret; \ - BUILD_BUG_ON((mask) & ~__SCX_KF_TERMINAL); \ + __typeof__((sch)->ops.op(task0, task1, ##args)) __ret; \ + WARN_ON_ONCE(current->scx.kf_tasks[0]); \ current->scx.kf_tasks[0] = task0; \ current->scx.kf_tasks[1] = task1; \ - __ret = SCX_CALL_OP_RET(mask, op, task0, task1, ##args); \ + __ret = SCX_CALL_OP_RET((sch), op, locked_rq, task0, task1, ##args); \ current->scx.kf_tasks[0] = NULL; \ current->scx.kf_tasks[1] = NULL; \ __ret; \ }) -/* @mask is constant, always inline to cull unnecessary branches */ -static __always_inline bool scx_kf_allowed(u32 mask) -{ - if (unlikely(!(current->scx.kf_mask & mask))) { - scx_ops_error("kfunc with mask 0x%x called from an operation only allowing 0x%x", - mask, current->scx.kf_mask); - return false; - } - - /* - * Enforce nesting boundaries. e.g. A kfunc which can be called from - * DISPATCH must not be called if we're running DEQUEUE which is nested - * inside ops.dispatch(). We don't need to check boundaries for any - * blocking kfuncs as the verifier ensures they're only called from - * sleepable progs. - */ - if (unlikely(highest_bit(mask) == SCX_KF_CPU_RELEASE && - (current->scx.kf_mask & higher_bits(SCX_KF_CPU_RELEASE)))) { - scx_ops_error("cpu_release kfunc called from a nested operation"); - return false; - } - - if (unlikely(highest_bit(mask) == SCX_KF_DISPATCH && - (current->scx.kf_mask & higher_bits(SCX_KF_DISPATCH)))) { - scx_ops_error("dispatch kfunc called from a nested operation"); - return false; - } - - return true; -} - /* see SCX_CALL_OP_TASK() */ -static __always_inline bool scx_kf_allowed_on_arg_tasks(u32 mask, +static __always_inline bool scx_kf_arg_task_ok(struct scx_sched *sch, struct task_struct *p) { - if (!scx_kf_allowed(mask)) - return false; - if (unlikely((p != current->scx.kf_tasks[0] && p != current->scx.kf_tasks[1]))) { - scx_ops_error("called on a task not being operated on"); + scx_error(sch, "called on a task not being operated on"); return false; } return true; } -static bool scx_kf_allowed_if_unlocked(void) -{ - return !current->scx.kf_mask; -} +enum scx_dsq_iter_flags { + /* iterate in the reverse dispatch order */ + SCX_DSQ_ITER_REV = 1U << 16, + + __SCX_DSQ_ITER_HAS_SLICE = 1U << 30, + __SCX_DSQ_ITER_HAS_VTIME = 1U << 31, + + __SCX_DSQ_ITER_USER_FLAGS = SCX_DSQ_ITER_REV, + __SCX_DSQ_ITER_ALL_FLAGS = __SCX_DSQ_ITER_USER_FLAGS | + __SCX_DSQ_ITER_HAS_SLICE | + __SCX_DSQ_ITER_HAS_VTIME, +}; /** * nldsq_next_task - Iterate to the next task in a non-local DSQ - * @dsq: user dsq being iterated + * @dsq: non-local dsq being iterated * @cur: current position, %NULL to start iteration * @rev: walk backwards * @@ -1269,6 +613,85 @@ static struct task_struct *nldsq_next_task(struct scx_dispatch_q *dsq, for ((p) = nldsq_next_task((dsq), NULL, false); (p); \ (p) = nldsq_next_task((dsq), (p), false)) +/** + * nldsq_cursor_next_task - Iterate to the next task given a cursor in a non-local DSQ + * @cursor: scx_dsq_list_node initialized with INIT_DSQ_LIST_CURSOR() + * @dsq: non-local dsq being iterated + * + * Find the next task in a cursor based iteration. The caller must have + * initialized @cursor using INIT_DSQ_LIST_CURSOR() and can release the DSQ lock + * between the iteration steps. + * + * Only tasks which were queued before @cursor was initialized are visible. This + * bounds the iteration and guarantees that vtime never jumps in the other + * direction while iterating. + */ +static struct task_struct *nldsq_cursor_next_task(struct scx_dsq_list_node *cursor, + struct scx_dispatch_q *dsq) +{ + bool rev = cursor->flags & SCX_DSQ_ITER_REV; + struct task_struct *p; + + lockdep_assert_held(&dsq->lock); + BUG_ON(!(cursor->flags & SCX_DSQ_LNODE_ITER_CURSOR)); + + if (list_empty(&cursor->node)) + p = NULL; + else + p = container_of(cursor, struct task_struct, scx.dsq_list); + + /* skip cursors and tasks that were queued after @cursor init */ + do { + p = nldsq_next_task(dsq, p, rev); + } while (p && unlikely(u32_before(cursor->priv, p->scx.dsq_seq))); + + if (p) { + if (rev) + list_move_tail(&cursor->node, &p->scx.dsq_list.node); + else + list_move(&cursor->node, &p->scx.dsq_list.node); + } else { + list_del_init(&cursor->node); + } + + return p; +} + +/** + * nldsq_cursor_lost_task - Test whether someone else took the task since iteration + * @cursor: scx_dsq_list_node initialized with INIT_DSQ_LIST_CURSOR() + * @rq: rq @p was on + * @dsq: dsq @p was on + * @p: target task + * + * @p is a task returned by nldsq_cursor_next_task(). The locks may have been + * dropped and re-acquired inbetween. Verify that no one else took or is in the + * process of taking @p from @dsq. + * + * On %false return, the caller can assume full ownership of @p. + */ +static bool nldsq_cursor_lost_task(struct scx_dsq_list_node *cursor, + struct rq *rq, struct scx_dispatch_q *dsq, + struct task_struct *p) +{ + lockdep_assert_rq_held(rq); + lockdep_assert_held(&dsq->lock); + + /* + * @p could have already left $src_dsq, got re-enqueud, or be in the + * process of being consumed by someone else. + */ + if (unlikely(p->scx.dsq != dsq || + u32_before(cursor->priv, p->scx.dsq_seq) || + p->scx.holding_cpu >= 0)) + return true; + + /* if @p has stayed on @dsq, its rq couldn't have changed */ + if (WARN_ON_ONCE(rq != task_rq(p))) + return true; + + return false; +} /* * BPF DSQ iterator. Tasks in a non-local DSQ can be iterated in [reverse] @@ -1276,19 +699,6 @@ static struct task_struct *nldsq_next_task(struct scx_dispatch_q *dsq, * changes without breaking backward compatibility. Can be used with * bpf_for_each(). See bpf_iter_scx_dsq_*(). */ -enum scx_dsq_iter_flags { - /* iterate in the reverse dispatch order */ - SCX_DSQ_ITER_REV = 1U << 16, - - __SCX_DSQ_ITER_HAS_SLICE = 1U << 30, - __SCX_DSQ_ITER_HAS_VTIME = 1U << 31, - - __SCX_DSQ_ITER_USER_FLAGS = SCX_DSQ_ITER_REV, - __SCX_DSQ_ITER_ALL_FLAGS = __SCX_DSQ_ITER_USER_FLAGS | - __SCX_DSQ_ITER_HAS_SLICE | - __SCX_DSQ_ITER_HAS_VTIME, -}; - struct bpf_iter_scx_dsq_kern { struct scx_dsq_list_node cursor; struct scx_dispatch_q *dsq; @@ -1301,23 +711,86 @@ struct bpf_iter_scx_dsq { } __attribute__((aligned(8))); +static u32 scx_get_task_state(const struct task_struct *p) +{ + return p->scx.flags & SCX_TASK_STATE_MASK; +} + +static void scx_set_task_state(struct task_struct *p, u32 state) +{ + u32 prev_state = scx_get_task_state(p); + bool warn = false; + + switch (state) { + case SCX_TASK_NONE: + warn = prev_state == SCX_TASK_DEAD; + break; + case SCX_TASK_INIT_BEGIN: + warn = prev_state != SCX_TASK_NONE; + break; + case SCX_TASK_INIT: + warn = prev_state != SCX_TASK_INIT_BEGIN; + p->scx.flags |= SCX_TASK_RESET_RUNNABLE_AT; + break; + case SCX_TASK_READY: + warn = !(prev_state == SCX_TASK_INIT || + prev_state == SCX_TASK_ENABLED); + break; + case SCX_TASK_ENABLED: + warn = prev_state != SCX_TASK_READY; + break; + case SCX_TASK_DEAD: + warn = !(prev_state == SCX_TASK_NONE || + prev_state == SCX_TASK_INIT_BEGIN); + break; + default: + WARN_ONCE(1, "sched_ext: Invalid task state %d -> %d for %s[%d]", + prev_state, state, p->comm, p->pid); + return; + } + + WARN_ONCE(warn, "sched_ext: Invalid task state transition 0x%x -> 0x%x for %s[%d]", + prev_state, state, p->comm, p->pid); + + p->scx.flags &= ~SCX_TASK_STATE_MASK; + p->scx.flags |= state; +} + /* * SCX task iterator. */ struct scx_task_iter { struct sched_ext_entity cursor; - struct task_struct *locked; + struct task_struct *locked_task; struct rq *rq; struct rq_flags rf; u32 cnt; + bool list_locked; +#ifdef CONFIG_EXT_SUB_SCHED + struct cgroup *cgrp; + struct cgroup_subsys_state *css_pos; + struct css_task_iter css_iter; +#endif }; /** * scx_task_iter_start - Lock scx_tasks_lock and start a task iteration * @iter: iterator to init + * @cgrp: Optional root of cgroup subhierarchy to iterate * - * Initialize @iter and return with scx_tasks_lock held. Once initialized, @iter - * must eventually be stopped with scx_task_iter_stop(). + * Initialize @iter. Once initialized, @iter must eventually be stopped with + * scx_task_iter_stop(). + * + * If @cgrp is %NULL, scx_tasks is used for iteration and this function returns + * with scx_tasks_lock held and @iter->cursor inserted into scx_tasks. + * + * If @cgrp is not %NULL, @cgrp and its descendants' tasks are walked using + * @iter->css_iter. The caller must be holding cgroup_lock() to prevent cgroup + * task migrations. + * + * The two modes of iterations are largely independent and it's likely that + * scx_tasks can be removed in favor of always using cgroup iteration if + * CONFIG_SCHED_CLASS_EXT depends on CONFIG_CGROUPS. * * scx_tasks_lock and the rq lock may be released using scx_task_iter_unlock() * between this and the first next() call or between any two next() calls. If @@ -1326,26 +799,35 @@ struct scx_task_iter { * RCU read lock or obtaining a reference count. * * All tasks which existed when the iteration started are guaranteed to be - * visited as long as they still exist. + * visited as long as they are not dead. */ -static void scx_task_iter_start(struct scx_task_iter *iter) +static void scx_task_iter_start(struct scx_task_iter *iter, struct cgroup *cgrp) { - BUILD_BUG_ON(__SCX_DSQ_ITER_ALL_FLAGS & - ((1U << __SCX_DSQ_LNODE_PRIV_SHIFT) - 1)); + memset(iter, 0, sizeof(*iter)); - spin_lock_irq(&scx_tasks_lock); +#ifdef CONFIG_EXT_SUB_SCHED + if (cgrp) { + lockdep_assert_held(&cgroup_mutex); + iter->cgrp = cgrp; + iter->css_pos = css_next_descendant_pre(NULL, &iter->cgrp->self); + css_task_iter_start(iter->css_pos, CSS_TASK_ITER_WITH_DEAD, + &iter->css_iter); + return; + } +#endif + raw_spin_lock_irq(&scx_tasks_lock); iter->cursor = (struct sched_ext_entity){ .flags = SCX_TASK_CURSOR }; list_add(&iter->cursor.tasks_node, &scx_tasks); - iter->locked = NULL; - iter->cnt = 0; + iter->list_locked = true; } static void __scx_task_iter_rq_unlock(struct scx_task_iter *iter) { - if (iter->locked) { - task_rq_unlock(iter->rq, iter->locked, &iter->rf); - iter->locked = NULL; + if (iter->locked_task) { + __balance_callbacks(iter->rq, &iter->rf); + task_rq_unlock(iter->rq, iter->locked_task, &iter->rf); + iter->locked_task = NULL; } } @@ -1355,24 +837,24 @@ static void __scx_task_iter_rq_unlock(struct scx_task_iter *iter) * * If @iter is in the middle of a locked iteration, it may be locking the rq of * the task currently being visited in addition to scx_tasks_lock. Unlock both. - * This function can be safely called anytime during an iteration. + * This function can be safely called anytime during an iteration. The next + * iterator operation will automatically restore the necessary locking. */ static void scx_task_iter_unlock(struct scx_task_iter *iter) { __scx_task_iter_rq_unlock(iter); - spin_unlock_irq(&scx_tasks_lock); + if (iter->list_locked) { + iter->list_locked = false; + raw_spin_unlock_irq(&scx_tasks_lock); + } } -/** - * scx_task_iter_relock - Lock scx_tasks_lock released by scx_task_iter_unlock() - * @iter: iterator to re-lock - * - * Re-lock scx_tasks_lock unlocked by scx_task_iter_unlock(). Note that it - * doesn't re-lock the rq lock. Must be called before other iterator operations. - */ -static void scx_task_iter_relock(struct scx_task_iter *iter) +static void __scx_task_iter_maybe_relock(struct scx_task_iter *iter) { - spin_lock_irq(&scx_tasks_lock); + if (!iter->list_locked) { + raw_spin_lock_irq(&scx_tasks_lock); + iter->list_locked = true; + } } /** @@ -1385,6 +867,15 @@ static void scx_task_iter_relock(struct scx_task_iter *iter) */ static void scx_task_iter_stop(struct scx_task_iter *iter) { +#ifdef CONFIG_EXT_SUB_SCHED + if (iter->cgrp) { + if (iter->css_pos) + css_task_iter_end(&iter->css_iter); + __scx_task_iter_rq_unlock(iter); + return; + } +#endif + __scx_task_iter_maybe_relock(iter); list_del_init(&iter->cursor.tasks_node); scx_task_iter_unlock(iter); } @@ -1394,20 +885,40 @@ static void scx_task_iter_stop(struct scx_task_iter *iter) * @iter: iterator to walk * * Visit the next task. See scx_task_iter_start() for details. Locks are dropped - * and re-acquired every %SCX_OPS_TASK_ITER_BATCH iterations to avoid causing - * stalls by holding scx_tasks_lock for too long. + * and re-acquired every %SCX_TASK_ITER_BATCH iterations to avoid causing stalls + * by holding scx_tasks_lock for too long. */ static struct task_struct *scx_task_iter_next(struct scx_task_iter *iter) { struct list_head *cursor = &iter->cursor.tasks_node; struct sched_ext_entity *pos; - if (!(++iter->cnt % SCX_OPS_TASK_ITER_BATCH)) { + if (!(++iter->cnt % SCX_TASK_ITER_BATCH)) { scx_task_iter_unlock(iter); cond_resched(); - scx_task_iter_relock(iter); } +#ifdef CONFIG_EXT_SUB_SCHED + if (iter->cgrp) { + while (iter->css_pos) { + struct task_struct *p; + + p = css_task_iter_next(&iter->css_iter); + if (p) + return p; + + css_task_iter_end(&iter->css_iter); + iter->css_pos = css_next_descendant_pre(iter->css_pos, + &iter->cgrp->self); + if (iter->css_pos) + css_task_iter_start(iter->css_pos, CSS_TASK_ITER_WITH_DEAD, + &iter->css_iter); + } + return NULL; + } +#endif + __scx_task_iter_maybe_relock(iter); + list_for_each_entry(pos, cursor, tasks_node) { if (&pos->tasks_node == &scx_tasks) return NULL; @@ -1461,40 +972,95 @@ static struct task_struct *scx_task_iter_next_locked(struct scx_task_iter *iter) * * Test for idle_sched_class as only init_tasks are on it. */ - if (p->sched_class != &idle_sched_class) - break; - } - if (!p) - return NULL; + if (p->sched_class == &idle_sched_class) + continue; - iter->rq = task_rq_lock(p, &iter->rf); - iter->locked = p; + iter->rq = task_rq_lock(p, &iter->rf); + iter->locked_task = p; - return p; + /* + * cgroup_task_dead() removes the dead tasks from cset->tasks + * after sched_ext_dead() and cgroup iteration may see tasks + * which already finished sched_ext_dead(). %SCX_TASK_DEAD is + * set by sched_ext_dead() under @p's rq lock. Test it to + * avoid visiting tasks which are already dead from SCX POV. + */ + if (scx_get_task_state(p) == SCX_TASK_DEAD) { + __scx_task_iter_rq_unlock(iter); + continue; + } + + return p; + } + return NULL; } -static enum scx_ops_enable_state scx_ops_enable_state(void) +/** + * scx_add_event - Increase an event counter for 'name' by 'cnt' + * @sch: scx_sched to account events for + * @name: an event name defined in struct scx_event_stats + * @cnt: the number of the event occurred + * + * This can be used when preemption is not disabled. + */ +#define scx_add_event(sch, name, cnt) do { \ + this_cpu_add((sch)->pcpu->event_stats.name, (cnt)); \ + trace_sched_ext_event(#name, (cnt)); \ +} while(0) + +/** + * __scx_add_event - Increase an event counter for 'name' by 'cnt' + * @sch: scx_sched to account events for + * @name: an event name defined in struct scx_event_stats + * @cnt: the number of the event occurred + * + * This should be used only when preemption is disabled. + */ +#define __scx_add_event(sch, name, cnt) do { \ + __this_cpu_add((sch)->pcpu->event_stats.name, (cnt)); \ + trace_sched_ext_event(#name, cnt); \ +} while(0) + +/** + * scx_agg_event - Aggregate an event counter 'kind' from 'src_e' to 'dst_e' + * @dst_e: destination event stats + * @src_e: source event stats + * @kind: a kind of event to be aggregated + */ +#define scx_agg_event(dst_e, src_e, kind) do { \ + (dst_e)->kind += READ_ONCE((src_e)->kind); \ +} while(0) + +/** + * scx_dump_event - Dump an event 'kind' in 'events' to 's' + * @s: output seq_buf + * @events: event stats + * @kind: a kind of event to dump + */ +#define scx_dump_event(s, events, kind) do { \ + dump_line(&(s), "%40s: %16lld", #kind, (events)->kind); \ +} while (0) + + +static void scx_read_events(struct scx_sched *sch, + struct scx_event_stats *events); + +static enum scx_enable_state scx_enable_state(void) { - return atomic_read(&scx_ops_enable_state_var); + return atomic_read(&scx_enable_state_var); } -static enum scx_ops_enable_state -scx_ops_set_enable_state(enum scx_ops_enable_state to) +static enum scx_enable_state scx_set_enable_state(enum scx_enable_state to) { - return atomic_xchg(&scx_ops_enable_state_var, to); + return atomic_xchg(&scx_enable_state_var, to); } -static bool scx_ops_tryset_enable_state(enum scx_ops_enable_state to, - enum scx_ops_enable_state from) +static bool scx_tryset_enable_state(enum scx_enable_state to, + enum scx_enable_state from) { int from_v = from; - return atomic_try_cmpxchg(&scx_ops_enable_state_var, &from_v, to); -} - -static bool scx_rq_bypassing(struct rq *rq) -{ - return unlikely(rq->scx.flags & SCX_RQ_BYPASSING); + return atomic_try_cmpxchg(&scx_enable_state_var, &from_v, to); } /** @@ -1514,8 +1080,14 @@ static void wait_ops_state(struct task_struct *p, unsigned long opss) } while (atomic_long_read_acquire(&p->scx.ops_state) == opss); } +static inline bool __cpu_valid(s32 cpu) +{ + return likely(cpu >= 0 && cpu < nr_cpu_ids && cpu_possible(cpu)); +} + /** - * ops_cpu_valid - Verify a cpu number + * ops_cpu_valid - Verify a cpu number, to be used on ops input args + * @sch: scx_sched to abort on error * @cpu: cpu number which came from a BPF ops * @where: extra information reported on error * @@ -1523,49 +1095,42 @@ static void wait_ops_state(struct task_struct *p, unsigned long opss) * Verify that it is in range and one of the possible cpus. If invalid, trigger * an ops error. */ -static bool ops_cpu_valid(s32 cpu, const char *where) +static bool ops_cpu_valid(struct scx_sched *sch, s32 cpu, const char *where) { - if (likely(cpu >= 0 && cpu < nr_cpu_ids && cpu_possible(cpu))) { + if (__cpu_valid(cpu)) { return true; } else { - scx_ops_error("invalid CPU %d%s%s", cpu, - where ? " " : "", where ?: ""); + scx_error(sch, "invalid CPU %d%s%s", cpu, where ? " " : "", where ?: ""); return false; } } /** * ops_sanitize_err - Sanitize a -errno value + * @sch: scx_sched to error out on error * @ops_name: operation to blame on failure * @err: -errno value to sanitize * - * Verify @err is a valid -errno. If not, trigger scx_ops_error() and return + * Verify @err is a valid -errno. If not, trigger scx_error() and return * -%EPROTO. This is necessary because returning a rogue -errno up the chain can * cause misbehaviors. For an example, a large negative return from * ops.init_task() triggers an oops when passed up the call chain because the * value fails IS_ERR() test after being encoded with ERR_PTR() and then is * handled as a pointer. */ -static int ops_sanitize_err(const char *ops_name, s32 err) +static int ops_sanitize_err(struct scx_sched *sch, const char *ops_name, s32 err) { if (err < 0 && err >= -MAX_ERRNO) return err; - scx_ops_error("ops.%s() returned an invalid errno %d", ops_name, err); + scx_error(sch, "ops.%s() returned an invalid errno %d", ops_name, err); return -EPROTO; } -static void run_deferred(struct rq *rq) -{ - process_ddsp_deferred_locals(rq); -} - -#ifdef CONFIG_SMP static void deferred_bal_cb_workfn(struct rq *rq) { run_deferred(rq); } -#endif static void deferred_irq_workfn(struct irq_work *irq_work) { @@ -1580,15 +1145,38 @@ static void deferred_irq_workfn(struct irq_work *irq_work) * schedule_deferred - Schedule execution of deferred actions on an rq * @rq: target rq * - * Schedule execution of deferred actions on @rq. Must be called with @rq - * locked. Deferred actions are executed with @rq locked but unpinned, and thus - * can unlock @rq to e.g. migrate tasks to other rqs. + * Schedule execution of deferred actions on @rq. Deferred actions are executed + * with @rq locked but unpinned, and thus can unlock @rq to e.g. migrate tasks + * to other rqs. */ static void schedule_deferred(struct rq *rq) { + /* + * This is the fallback when schedule_deferred_locked() can't use + * the cheaper balance callback or wakeup hook paths (the target + * CPU is not in balance or wakeup). Currently, this is primarily + * hit by reenqueue operations targeting a remote CPU. + * + * Queue on the target CPU. The deferred work can run from any CPU + * correctly - the _locked() path already processes remote rqs from + * the calling CPU - but targeting the owning CPU allows IPI delivery + * without waiting for the calling CPU to re-enable IRQs and is + * cheaper as the reenqueue runs locally. + */ + irq_work_queue_on(&rq->scx.deferred_irq_work, cpu_of(rq)); +} + +/** + * schedule_deferred_locked - Schedule execution of deferred actions on an rq + * @rq: target rq + * + * Schedule execution of deferred actions on @rq. Equivalent to + * schedule_deferred() but requires @rq to be locked and can be more efficient. + */ +static void schedule_deferred_locked(struct rq *rq) +{ lockdep_assert_rq_held(rq); -#ifdef CONFIG_SMP /* * If in the middle of waking up a task, task_woken_scx() will be called * afterwards which will then run the deferred actions, no need to @@ -1597,23 +1185,107 @@ static void schedule_deferred(struct rq *rq) if (rq->scx.flags & SCX_RQ_IN_WAKEUP) return; + /* Don't do anything if there already is a deferred operation. */ + if (rq->scx.flags & SCX_RQ_BAL_CB_PENDING) + return; + /* * If in balance, the balance callbacks will be called before rq lock is * released. Schedule one. + * + * + * We can't directly insert the callback into the + * rq's list: The call can drop its lock and make the pending balance + * callback visible to unrelated code paths that call rq_pin_lock(). + * + * Just let balance_one() know that it must do it itself. */ if (rq->scx.flags & SCX_RQ_IN_BALANCE) { - queue_balance_callback(rq, &rq->scx.deferred_bal_cb, - deferred_bal_cb_workfn); + rq->scx.flags |= SCX_RQ_BAL_CB_PENDING; return; } -#endif + /* - * No scheduler hooks available. Queue an irq work. They are executed on - * IRQ re-enable which may take a bit longer than the scheduler hooks. - * The above WAKEUP and BALANCE paths should cover most of the cases and - * the time to IRQ re-enable shouldn't be long. + * No scheduler hooks available. Use the generic irq_work path. The + * above WAKEUP and BALANCE paths should cover most of the cases and the + * time to IRQ re-enable shouldn't be long. */ - irq_work_queue(&rq->scx.deferred_irq_work); + schedule_deferred(rq); +} + +static void schedule_dsq_reenq(struct scx_sched *sch, struct scx_dispatch_q *dsq, + u64 reenq_flags, struct rq *locked_rq) +{ + struct rq *rq; + + /* + * Allowing reenqueues doesn't make sense while bypassing. This also + * blocks from new reenqueues to be scheduled on dead scheds. + */ + if (unlikely(READ_ONCE(sch->bypass_depth))) + return; + + if (dsq->id == SCX_DSQ_LOCAL) { + rq = container_of(dsq, struct rq, scx.local_dsq); + + struct scx_sched_pcpu *sch_pcpu = per_cpu_ptr(sch->pcpu, cpu_of(rq)); + struct scx_deferred_reenq_local *drl = &sch_pcpu->deferred_reenq_local; + + /* + * Pairs with smp_mb() in process_deferred_reenq_locals() and + * guarantees that there is a reenq_local() afterwards. + */ + smp_mb(); + + if (list_empty(&drl->node) || + (READ_ONCE(drl->flags) & reenq_flags) != reenq_flags) { + + guard(raw_spinlock_irqsave)(&rq->scx.deferred_reenq_lock); + + if (list_empty(&drl->node)) + list_move_tail(&drl->node, &rq->scx.deferred_reenq_locals); + WRITE_ONCE(drl->flags, drl->flags | reenq_flags); + } + } else if (!(dsq->id & SCX_DSQ_FLAG_BUILTIN)) { + rq = this_rq(); + + struct scx_dsq_pcpu *dsq_pcpu = per_cpu_ptr(dsq->pcpu, cpu_of(rq)); + struct scx_deferred_reenq_user *dru = &dsq_pcpu->deferred_reenq_user; + + /* + * Pairs with smp_mb() in process_deferred_reenq_users() and + * guarantees that there is a reenq_user() afterwards. + */ + smp_mb(); + + if (list_empty(&dru->node) || + (READ_ONCE(dru->flags) & reenq_flags) != reenq_flags) { + + guard(raw_spinlock_irqsave)(&rq->scx.deferred_reenq_lock); + + if (list_empty(&dru->node)) + list_move_tail(&dru->node, &rq->scx.deferred_reenq_users); + WRITE_ONCE(dru->flags, dru->flags | reenq_flags); + } + } else { + scx_error(sch, "DSQ 0x%llx not allowed for reenq", dsq->id); + return; + } + + if (rq == locked_rq) + schedule_deferred_locked(rq); + else + schedule_deferred(rq); +} + +static void schedule_reenq_local(struct rq *rq, u64 reenq_flags) +{ + struct scx_sched *root = rcu_dereference_sched(scx_root); + + if (WARN_ON_ONCE(!root)) + return; + + schedule_dsq_reenq(root, &rq->scx.local_dsq, reenq_flags, rq); } /** @@ -1658,7 +1330,7 @@ static void touch_core_sched_dispatch(struct rq *rq, struct task_struct *p) lockdep_assert_rq_held(rq); #ifdef CONFIG_SCHED_CORE - if (SCX_HAS_OP(core_sched_before)) + if (unlikely(SCX_HAS_OP(scx_root, core_sched_before))) touch_core_sched(rq, p); #endif } @@ -1677,6 +1349,8 @@ static void update_curr_scx(struct rq *rq) if (!curr->scx.slice) touch_core_sched(rq, curr); } + + dl_server_update(&rq->ext_server, delta_exec); } static bool scx_dsq_priq_less(struct rb_node *node_a, @@ -1690,13 +1364,159 @@ static bool scx_dsq_priq_less(struct rb_node *node_a, return time_before64(a->scx.dsq_vtime, b->scx.dsq_vtime); } -static void dsq_mod_nr(struct scx_dispatch_q *dsq, s32 delta) +static void dsq_inc_nr(struct scx_dispatch_q *dsq, struct task_struct *p, u64 enq_flags) { /* scx_bpf_dsq_nr_queued() reads ->nr without locking, use WRITE_ONCE() */ - WRITE_ONCE(dsq->nr, dsq->nr + delta); + WRITE_ONCE(dsq->nr, dsq->nr + 1); + + /* + * Once @p reaches a local DSQ, it can only leave it by being dispatched + * to the CPU or dequeued. In both cases, the only way @p can go back to + * the BPF sched is through enqueueing. If being inserted into a local + * DSQ with IMMED, persist the state until the next enqueueing event in + * do_enqueue_task() so that we can maintain IMMED protection through + * e.g. SAVE/RESTORE cycles and slice extensions. + */ + if (enq_flags & SCX_ENQ_IMMED) { + if (unlikely(dsq->id != SCX_DSQ_LOCAL)) { + WARN_ON_ONCE(!(enq_flags & SCX_ENQ_GDSQ_FALLBACK)); + return; + } + p->scx.flags |= SCX_TASK_IMMED; + } + + if (p->scx.flags & SCX_TASK_IMMED) { + struct rq *rq = container_of(dsq, struct rq, scx.local_dsq); + + if (WARN_ON_ONCE(dsq->id != SCX_DSQ_LOCAL)) + return; + + rq->scx.nr_immed++; + + /* + * If @rq already had other tasks or the current task is not + * done yet, @p can't go on the CPU immediately. Re-enqueue. + */ + if (unlikely(dsq->nr > 1 || !rq_is_open(rq, enq_flags))) + schedule_reenq_local(rq, 0); + } +} + +static void dsq_dec_nr(struct scx_dispatch_q *dsq, struct task_struct *p) +{ + /* see dsq_inc_nr() */ + WRITE_ONCE(dsq->nr, dsq->nr - 1); + + if (p->scx.flags & SCX_TASK_IMMED) { + struct rq *rq = container_of(dsq, struct rq, scx.local_dsq); + + if (WARN_ON_ONCE(dsq->id != SCX_DSQ_LOCAL) || + WARN_ON_ONCE(rq->scx.nr_immed <= 0)) + return; + + rq->scx.nr_immed--; + } +} + +static void refill_task_slice_dfl(struct scx_sched *sch, struct task_struct *p) +{ + p->scx.slice = READ_ONCE(sch->slice_dfl); + __scx_add_event(sch, SCX_EV_REFILL_SLICE_DFL, 1); +} + +/* + * Return true if @p is moving due to an internal SCX migration, false + * otherwise. + */ +static inline bool task_scx_migrating(struct task_struct *p) +{ + /* + * We only need to check sticky_cpu: it is set to the destination + * CPU in move_remote_task_to_local_dsq() before deactivate_task() + * and cleared when the task is enqueued on the destination, so it + * is only non-negative during an internal SCX migration. + */ + return p->scx.sticky_cpu >= 0; +} + +/* + * Call ops.dequeue() if the task is in BPF custody and not migrating. + * Clears %SCX_TASK_IN_CUSTODY when the callback is invoked. + */ +static void call_task_dequeue(struct scx_sched *sch, struct rq *rq, + struct task_struct *p, u64 deq_flags) +{ + if (!(p->scx.flags & SCX_TASK_IN_CUSTODY) || task_scx_migrating(p)) + return; + + if (SCX_HAS_OP(sch, dequeue)) + SCX_CALL_OP_TASK(sch, dequeue, rq, p, deq_flags); + + p->scx.flags &= ~SCX_TASK_IN_CUSTODY; +} + +static void local_dsq_post_enq(struct scx_sched *sch, struct scx_dispatch_q *dsq, + struct task_struct *p, u64 enq_flags) +{ + struct rq *rq = container_of(dsq, struct rq, scx.local_dsq); + + call_task_dequeue(sch, rq, p, 0); + + /* + * Note that @rq's lock may be dropped between this enqueue and @p + * actually getting on CPU. This gives higher-class tasks (e.g. RT) + * an opportunity to wake up on @rq and prevent @p from running. + * Here are some concrete examples: + * + * Example 1: + * + * We dispatch two tasks from a single ops.dispatch(): + * - First, a local task to this CPU's local DSQ; + * - Second, a local/remote task to a remote CPU's local DSQ. + * We must drop the local rq lock in order to finish the second + * dispatch. In that time, an RT task can wake up on the local rq. + * + * Example 2: + * + * We dispatch a local/remote task to a remote CPU's local DSQ. + * We must drop the remote rq lock before the dispatched task can run, + * which gives an RT task an opportunity to wake up on the remote rq. + * + * Both examples work the same if we replace dispatching with moving + * the tasks from a user-created DSQ. + * + * We must detect these wakeups so that we can re-enqueue IMMED tasks + * from @rq's local DSQ. scx_wakeup_preempt() serves exactly this + * purpose, but for it to be invoked, we must ensure that we bump + * @rq->next_class to &ext_sched_class if it's currently idle. + * + * wakeup_preempt() does the bumping, and since we only invoke it if + * @rq->next_class is below &ext_sched_class, it will also + * resched_curr(rq). + */ + if (sched_class_above(p->sched_class, rq->next_class)) + wakeup_preempt(rq, p, 0); + + /* + * If @rq is in balance, the CPU is already vacant and looking for the + * next task to run. No need to preempt or trigger resched after moving + * @p into its local DSQ. + * Note that the wakeup_preempt() above may have already triggered + * a resched if @rq->next_class was idle. It's harmless, since + * need_resched is cleared immediately after task pick. + */ + if (rq->scx.flags & SCX_RQ_IN_BALANCE) + return; + + if ((enq_flags & SCX_ENQ_PREEMPT) && p != rq->curr && + rq->curr->sched_class == &ext_sched_class) { + rq->curr->scx.slice = 0; + resched_curr(rq); + } } -static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, +static void dispatch_enqueue(struct scx_sched *sch, struct rq *rq, + struct scx_dispatch_q *dsq, struct task_struct *p, u64 enq_flags) { bool is_local = dsq->id == SCX_DSQ_LOCAL; @@ -1706,12 +1526,14 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, !RB_EMPTY_NODE(&p->scx.dsq_priq)); if (!is_local) { - raw_spin_lock(&dsq->lock); + raw_spin_lock_nested(&dsq->lock, + (enq_flags & SCX_ENQ_NESTED) ? SINGLE_DEPTH_NESTING : 0); + if (unlikely(dsq->id == SCX_DSQ_INVALID)) { - scx_ops_error("attempting to dispatch to a destroyed dsq"); + scx_error(sch, "attempting to dispatch to a destroyed dsq"); /* fall back to the global dsq */ raw_spin_unlock(&dsq->lock); - dsq = find_global_dsq(p); + dsq = find_global_dsq(sch, task_cpu(p)); raw_spin_lock(&dsq->lock); } } @@ -1725,7 +1547,7 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, * disallow any internal DSQ from doing vtime ordering of * tasks. */ - scx_ops_error("cannot use vtime ordering for built-in DSQs"); + scx_error(sch, "cannot use vtime ordering for built-in DSQs"); enq_flags &= ~SCX_ENQ_DSQ_PRIQ; } @@ -1739,8 +1561,8 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, */ if (unlikely(RB_EMPTY_ROOT(&dsq->priq) && nldsq_next_task(dsq, NULL, false))) - scx_ops_error("DSQ ID 0x%016llx already had FIFO-enqueued tasks", - dsq->id); + scx_error(sch, "DSQ ID 0x%016llx already had FIFO-enqueued tasks", + dsq->id); p->scx.dsq_flags |= SCX_TASK_DSQ_ON_PRIQ; rb_add(&p->scx.dsq_priq, &dsq->priq, scx_dsq_priq_less); @@ -1755,36 +1577,63 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, container_of(rbp, struct task_struct, scx.dsq_priq); list_add(&p->scx.dsq_list.node, &prev->scx.dsq_list.node); + /* first task unchanged - no update needed */ } else { list_add(&p->scx.dsq_list.node, &dsq->list); + /* not builtin and new task is at head - use fastpath */ + rcu_assign_pointer(dsq->first_task, p); } } else { /* a FIFO DSQ shouldn't be using PRIQ enqueuing */ if (unlikely(!RB_EMPTY_ROOT(&dsq->priq))) - scx_ops_error("DSQ ID 0x%016llx already had PRIQ-enqueued tasks", - dsq->id); + scx_error(sch, "DSQ ID 0x%016llx already had PRIQ-enqueued tasks", + dsq->id); - if (enq_flags & (SCX_ENQ_HEAD | SCX_ENQ_PREEMPT)) + if (enq_flags & (SCX_ENQ_HEAD | SCX_ENQ_PREEMPT)) { list_add(&p->scx.dsq_list.node, &dsq->list); - else + /* new task inserted at head - use fastpath */ + if (!(dsq->id & SCX_DSQ_FLAG_BUILTIN)) + rcu_assign_pointer(dsq->first_task, p); + } else { + /* + * dsq->list can contain parked BPF iterator cursors, so + * list_empty() here isn't a reliable proxy for "no real + * task in the DSQ". Test dsq->first_task directly. + */ list_add_tail(&p->scx.dsq_list.node, &dsq->list); + if (!dsq->first_task && !(dsq->id & SCX_DSQ_FLAG_BUILTIN)) + rcu_assign_pointer(dsq->first_task, p); + } } /* seq records the order tasks are queued, used by BPF DSQ iterator */ - dsq->seq++; + WRITE_ONCE(dsq->seq, dsq->seq + 1); p->scx.dsq_seq = dsq->seq; - dsq_mod_nr(dsq, 1); + dsq_inc_nr(dsq, p, enq_flags); p->scx.dsq = dsq; /* - * scx.ddsp_dsq_id and scx.ddsp_enq_flags are only relevant on the - * direct dispatch path, but we clear them here because the direct - * dispatch verdict may be overridden on the enqueue path during e.g. - * bypass. + * Update custody and call ops.dequeue() before clearing ops_state: + * once ops_state is cleared, waiters in ops_dequeue() can proceed + * and dequeue_task_scx() will RMW p->scx.flags. If we clear + * ops_state first, both sides would modify p->scx.flags + * concurrently in a non-atomic way. */ - p->scx.ddsp_dsq_id = SCX_DSQ_INVALID; - p->scx.ddsp_enq_flags = 0; + if (is_local) { + local_dsq_post_enq(sch, dsq, p, enq_flags); + } else { + /* + * Task on global/bypass DSQ: leave custody, task on + * non-terminal DSQ: enter custody. + */ + if (dsq->id == SCX_DSQ_GLOBAL || dsq->id == SCX_DSQ_BYPASS) + call_task_dequeue(sch, rq, p, 0); + else + p->scx.flags |= SCX_TASK_IN_CUSTODY; + + raw_spin_unlock(&dsq->lock); + } /* * We're transitioning out of QUEUEING or DISPATCHING. store_release to @@ -1792,23 +1641,6 @@ static void dispatch_enqueue(struct scx_dispatch_q *dsq, struct task_struct *p, */ if (enq_flags & SCX_ENQ_CLEAR_OPSS) atomic_long_set_release(&p->scx.ops_state, SCX_OPSS_NONE); - - if (is_local) { - struct rq *rq = container_of(dsq, struct rq, scx.local_dsq); - bool preempt = false; - - if ((enq_flags & SCX_ENQ_PREEMPT) && p != rq->curr && - rq->curr->sched_class == &ext_sched_class) { - rq->curr->scx.slice = 0; - preempt = true; - } - - if (preempt || sched_class_above(&ext_sched_class, - rq->curr->sched_class)) - resched_curr(rq); - } else { - raw_spin_unlock(&dsq->lock); - } } static void task_unlink_from_dsq(struct task_struct *p, @@ -1823,7 +1655,14 @@ static void task_unlink_from_dsq(struct task_struct *p, } list_del_init(&p->scx.dsq_list.node); - dsq_mod_nr(dsq, -1); + dsq_dec_nr(dsq, p); + + if (!(dsq->id & SCX_DSQ_FLAG_BUILTIN) && dsq->first_task == p) { + struct task_struct *first_task; + + first_task = nldsq_next_task(dsq, NULL, false); + rcu_assign_pointer(dsq->first_task, first_task); + } } static void dispatch_dequeue(struct rq *rq, struct task_struct *p) @@ -1831,6 +1670,8 @@ static void dispatch_dequeue(struct rq *rq, struct task_struct *p) struct scx_dispatch_q *dsq = p->scx.dsq; bool is_local = dsq == &rq->scx.local_dsq; + lockdep_assert_rq_held(rq); + if (!dsq) { /* * If !dsq && on-list, @p is on @rq's ddsp_deferred_locals. @@ -1877,8 +1718,23 @@ static void dispatch_dequeue(struct rq *rq, struct task_struct *p) raw_spin_unlock(&dsq->lock); } -static struct scx_dispatch_q *find_dsq_for_dispatch(struct rq *rq, u64 dsq_id, - struct task_struct *p) +/* + * Abbreviated version of dispatch_dequeue() that can be used when both @p's rq + * and dsq are locked. + */ +static void dispatch_dequeue_locked(struct task_struct *p, + struct scx_dispatch_q *dsq) +{ + lockdep_assert_rq_held(task_rq(p)); + lockdep_assert_held(&dsq->lock); + + task_unlink_from_dsq(p, dsq); + p->scx.dsq = NULL; +} + +static struct scx_dispatch_q *find_dsq_for_dispatch(struct scx_sched *sch, + struct rq *rq, u64 dsq_id, + s32 tcpu) { struct scx_dispatch_q *dsq; @@ -1888,27 +1744,27 @@ static struct scx_dispatch_q *find_dsq_for_dispatch(struct rq *rq, u64 dsq_id, if ((dsq_id & SCX_DSQ_LOCAL_ON) == SCX_DSQ_LOCAL_ON) { s32 cpu = dsq_id & SCX_DSQ_LOCAL_CPU_MASK; - if (!ops_cpu_valid(cpu, "in SCX_DSQ_LOCAL_ON dispatch verdict")) - return find_global_dsq(p); + if (!ops_cpu_valid(sch, cpu, "in SCX_DSQ_LOCAL_ON dispatch verdict")) + return find_global_dsq(sch, tcpu); return &cpu_rq(cpu)->scx.local_dsq; } if (dsq_id == SCX_DSQ_GLOBAL) - dsq = find_global_dsq(p); + dsq = find_global_dsq(sch, tcpu); else - dsq = find_user_dsq(dsq_id); + dsq = find_user_dsq(sch, dsq_id); if (unlikely(!dsq)) { - scx_ops_error("non-existent DSQ 0x%llx for %s[%d]", - dsq_id, p->comm, p->pid); - return find_global_dsq(p); + scx_error(sch, "non-existent DSQ 0x%llx", dsq_id); + return find_global_dsq(sch, tcpu); } return dsq; } -static void mark_direct_dispatch(struct task_struct *ddsp_task, +static void mark_direct_dispatch(struct scx_sched *sch, + struct task_struct *ddsp_task, struct task_struct *p, u64 dsq_id, u64 enq_flags) { @@ -1922,12 +1778,12 @@ static void mark_direct_dispatch(struct task_struct *ddsp_task, /* @p must match the task on the enqueue path */ if (unlikely(p != ddsp_task)) { if (IS_ERR(ddsp_task)) - scx_ops_error("%s[%d] already direct-dispatched", - p->comm, p->pid); + scx_error(sch, "%s[%d] already direct-dispatched", + p->comm, p->pid); else - scx_ops_error("scheduling for %s[%d] but trying to direct-dispatch %s[%d]", - ddsp_task->comm, ddsp_task->pid, - p->comm, p->pid); + scx_error(sch, "scheduling for %s[%d] but trying to direct-dispatch %s[%d]", + ddsp_task->comm, ddsp_task->pid, + p->comm, p->pid); return; } @@ -1938,11 +1794,34 @@ static void mark_direct_dispatch(struct task_struct *ddsp_task, p->scx.ddsp_enq_flags = enq_flags; } -static void direct_dispatch(struct task_struct *p, u64 enq_flags) +/* + * Clear @p direct dispatch state when leaving the scheduler. + * + * Direct dispatch state must be cleared in the following cases: + * - direct_dispatch(): cleared on the synchronous enqueue path, deferred + * dispatch keeps the state until consumed + * - process_ddsp_deferred_locals(): cleared after consuming deferred state, + * - do_enqueue_task(): cleared on enqueue fallbacks where the dispatch + * verdict is ignored (local/global/bypass) + * - dequeue_task_scx(): cleared after dispatch_dequeue(), covering deferred + * cancellation and holding_cpu races + * - scx_disable_task(): cleared for queued wakeup tasks, which are excluded by + * the scx_bypass() loop, so that stale state is not reused by a subsequent + * scheduler instance + */ +static inline void clear_direct_dispatch(struct task_struct *p) +{ + p->scx.ddsp_dsq_id = SCX_DSQ_INVALID; + p->scx.ddsp_enq_flags = 0; +} + +static void direct_dispatch(struct scx_sched *sch, struct task_struct *p, + u64 enq_flags) { struct rq *rq = task_rq(p); struct scx_dispatch_q *dsq = - find_dsq_for_dispatch(rq, p->scx.ddsp_dsq_id, p); + find_dsq_for_dispatch(sch, rq, p->scx.ddsp_dsq_id, task_cpu(p)); + u64 ddsp_enq_flags; touch_core_sched_dispatch(rq, p); @@ -1979,11 +1858,14 @@ static void direct_dispatch(struct task_struct *p, u64 enq_flags) WARN_ON_ONCE(p->scx.dsq || !list_empty(&p->scx.dsq_list.node)); list_add_tail(&p->scx.dsq_list.node, &rq->scx.ddsp_deferred_locals); - schedule_deferred(rq); + schedule_deferred_locked(rq); return; } - dispatch_enqueue(dsq, p, p->scx.ddsp_enq_flags | SCX_ENQ_CLEAR_OPSS); + ddsp_enq_flags = p->scx.ddsp_enq_flags; + clear_direct_dispatch(p); + + dispatch_enqueue(sch, rq, dsq, p, ddsp_enq_flags | SCX_ENQ_CLEAR_OPSS); } static bool scx_rq_online(struct rq *rq) @@ -2001,16 +1883,26 @@ static bool scx_rq_online(struct rq *rq) static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, int sticky_cpu) { + struct scx_sched *sch = scx_task_sched(p); struct task_struct **ddsp_taskp; + struct scx_dispatch_q *dsq; unsigned long qseq; WARN_ON_ONCE(!(p->scx.flags & SCX_TASK_QUEUED)); - /* rq migration */ + /* internal movements - rq migration / RESTORE */ if (sticky_cpu == cpu_of(rq)) goto local_norefill; /* + * Clear persistent TASK_IMMED for fresh enqueues, see dsq_inc_nr(). + * Note that exiting and migration-disabled tasks that skip + * ops.enqueue() below will lose IMMED protection unless + * %SCX_OPS_ENQ_EXITING / %SCX_OPS_ENQ_MIGRATION_DISABLED are set. + */ + p->scx.flags &= ~SCX_TASK_IMMED; + + /* * If !scx_rq_online(), we already told the BPF scheduler that the CPU * is offline and are just running the hotplug path. Don't bother the * BPF scheduler. @@ -2018,23 +1910,29 @@ static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, if (!scx_rq_online(rq)) goto local; - if (scx_rq_bypassing(rq)) - goto global; + if (scx_bypassing(sch, cpu_of(rq))) { + __scx_add_event(sch, SCX_EV_BYPASS_DISPATCH, 1); + goto bypass; + } if (p->scx.ddsp_dsq_id != SCX_DSQ_INVALID) goto direct; /* see %SCX_OPS_ENQ_EXITING */ - if (!static_branch_unlikely(&scx_ops_enq_exiting) && - unlikely(p->flags & PF_EXITING)) + if (!(sch->ops.flags & SCX_OPS_ENQ_EXITING) && + unlikely(p->flags & PF_EXITING)) { + __scx_add_event(sch, SCX_EV_ENQ_SKIP_EXITING, 1); goto local; + } /* see %SCX_OPS_ENQ_MIGRATION_DISABLED */ - if (!static_branch_unlikely(&scx_ops_enq_migration_disabled) && - is_migration_disabled(p)) + if (!(sch->ops.flags & SCX_OPS_ENQ_MIGRATION_DISABLED) && + is_migration_disabled(p)) { + __scx_add_event(sch, SCX_EV_ENQ_SKIP_MIGRATION_DISABLED, 1); goto local; + } - if (!SCX_HAS_OP(enqueue)) + if (unlikely(!SCX_HAS_OP(sch, enqueue))) goto global; /* DSQ bypass didn't trigger, enqueue on the BPF scheduler */ @@ -2047,13 +1945,19 @@ static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, WARN_ON_ONCE(*ddsp_taskp); *ddsp_taskp = p; - SCX_CALL_OP_TASK(SCX_KF_ENQUEUE, enqueue, p, enq_flags); + SCX_CALL_OP_TASK(sch, enqueue, rq, p, enq_flags); *ddsp_taskp = NULL; if (p->scx.ddsp_dsq_id != SCX_DSQ_INVALID) goto direct; /* + * Task is now in BPF scheduler's custody. Set %SCX_TASK_IN_CUSTODY + * so ops.dequeue() is called when it leaves custody. + */ + p->scx.flags |= SCX_TASK_IN_CUSTODY; + + /* * If not directly dispatched, QUEUEING isn't clear yet and dispatch or * dequeue may be waiting. The store_release matches their load_acquire. */ @@ -2061,25 +1965,31 @@ static void do_enqueue_task(struct rq *rq, struct task_struct *p, u64 enq_flags, return; direct: - direct_dispatch(p, enq_flags); + direct_dispatch(sch, p, enq_flags); + return; +local_norefill: + dispatch_enqueue(sch, rq, &rq->scx.local_dsq, p, enq_flags); return; - local: + dsq = &rq->scx.local_dsq; + goto enqueue; +global: + dsq = find_global_dsq(sch, task_cpu(p)); + goto enqueue; +bypass: + dsq = bypass_enq_target_dsq(sch, task_cpu(p)); + goto enqueue; + +enqueue: /* * For task-ordering, slice refill must be treated as implying the end * of the current slice. Otherwise, the longer @p stays on the CPU, the * higher priority it becomes from scx_prio_less()'s POV. */ touch_core_sched(rq, p); - p->scx.slice = SCX_SLICE_DFL; -local_norefill: - dispatch_enqueue(&rq->scx.local_dsq, p, enq_flags); - return; - -global: - touch_core_sched(rq, p); /* see the comment in local: */ - p->scx.slice = SCX_SLICE_DFL; - dispatch_enqueue(find_global_dsq(p), p, enq_flags); + refill_task_slice_dfl(sch, p); + clear_direct_dispatch(p); + dispatch_enqueue(sch, rq, dsq, p, enq_flags); } static bool task_runnable(const struct task_struct *p) @@ -2097,7 +2007,7 @@ static void set_task_runnable(struct rq *rq, struct task_struct *p) } /* - * list_add_tail() must be used. scx_ops_bypass() depends on tasks being + * list_add_tail() must be used. scx_bypass() depends on tasks being * appended to the runnable_list. */ list_add_tail(&p->scx.runnable_node, &rq->scx.runnable_list); @@ -2110,18 +2020,15 @@ static void clr_task_runnable(struct task_struct *p, bool reset_runnable_at) p->scx.flags |= SCX_TASK_RESET_RUNNABLE_AT; } -static void enqueue_task_scx(struct rq *rq, struct task_struct *p, int enq_flags) +static void enqueue_task_scx(struct rq *rq, struct task_struct *p, int core_enq_flags) { + struct scx_sched *sch = scx_task_sched(p); int sticky_cpu = p->scx.sticky_cpu; + u64 enq_flags = core_enq_flags | rq->scx.extra_enq_flags; if (enq_flags & ENQUEUE_WAKEUP) rq->scx.flags |= SCX_RQ_IN_WAKEUP; - enq_flags |= rq->scx.extra_enq_flags; - - if (sticky_cpu >= 0) - p->scx.sticky_cpu = -1; - /* * Restoring a running task will be immediately followed by * set_next_task_scx() which expects the task to not be on the BPF @@ -2141,24 +2048,37 @@ static void enqueue_task_scx(struct rq *rq, struct task_struct *p, int enq_flags rq->scx.nr_running++; add_nr_running(rq, 1); - if (SCX_HAS_OP(runnable) && !task_on_rq_migrating(p)) - SCX_CALL_OP_TASK(SCX_KF_REST, runnable, p, enq_flags); + if (SCX_HAS_OP(sch, runnable) && !task_on_rq_migrating(p)) + SCX_CALL_OP_TASK(sch, runnable, rq, p, enq_flags); if (enq_flags & SCX_ENQ_WAKEUP) touch_core_sched(rq, p); + /* Start dl_server if this is the first task being enqueued */ + if (rq->scx.nr_running == 1) + dl_server_start(&rq->ext_server); + do_enqueue_task(rq, p, enq_flags, sticky_cpu); + + if (sticky_cpu >= 0) + p->scx.sticky_cpu = -1; out: rq->scx.flags &= ~SCX_RQ_IN_WAKEUP; + + if ((enq_flags & SCX_ENQ_CPU_SELECTED) && + unlikely(cpu_of(rq) != p->scx.selected_cpu)) + __scx_add_event(sch, SCX_EV_SELECT_CPU_FALLBACK, 1); } -static void ops_dequeue(struct task_struct *p, u64 deq_flags) +static void ops_dequeue(struct rq *rq, struct task_struct *p, u64 deq_flags) { + struct scx_sched *sch = scx_task_sched(p); unsigned long opss; /* dequeue is always temporary, don't reset runnable_at */ clr_task_runnable(p, false); +retry: /* acquire ensures that we see the preceding updates on QUEUED */ opss = atomic_long_read_acquire(&p->scx.ops_state); @@ -2172,8 +2092,19 @@ static void ops_dequeue(struct task_struct *p, u64 deq_flags) */ BUG(); case SCX_OPSS_QUEUED: - if (SCX_HAS_OP(dequeue)) - SCX_CALL_OP_TASK(SCX_KF_REST, dequeue, p, deq_flags); + /* + * A queued task must always be in BPF scheduler's custody. If + * SCX_TASK_IN_CUSTODY is clear, finish_dispatch() on another + * CPU has already passed call_task_dequeue() (which clears the + * flag), but has not yet written SCX_OPSS_NONE. That final + * store does not require this rq's lock, so retrying with + * cpu_relax() is bounded: we will observe NONE (or DISPATCHING, + * handled by the fallthrough) on a subsequent iteration. + */ + if (unlikely(!(READ_ONCE(p->scx.flags) & SCX_TASK_IN_CUSTODY))) { + cpu_relax(); + goto retry; + } if (atomic_long_try_cmpxchg(&p->scx.ops_state, &opss, SCX_OPSS_NONE)) @@ -2197,16 +2128,42 @@ static void ops_dequeue(struct task_struct *p, u64 deq_flags) BUG_ON(atomic_long_read(&p->scx.ops_state) != SCX_OPSS_NONE); break; } + + /* + * Call ops.dequeue() if the task is still in BPF custody. + * + * The code that clears ops_state to %SCX_OPSS_NONE does not always + * clear %SCX_TASK_IN_CUSTODY: in dispatch_to_local_dsq(), when + * we're moving a task that was in %SCX_OPSS_DISPATCHING to a + * remote CPU's local DSQ, we only set ops_state to %SCX_OPSS_NONE + * so that a concurrent dequeue can proceed, but we clear + * %SCX_TASK_IN_CUSTODY only when we later enqueue or move the + * task. So we can see NONE + IN_CUSTODY here and we must handle + * it. Similarly, after waiting on %SCX_OPSS_DISPATCHING we see + * NONE but the task may still have %SCX_TASK_IN_CUSTODY set until + * it is enqueued on the destination. + */ + call_task_dequeue(sch, rq, p, deq_flags); } -static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int deq_flags) +static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int core_deq_flags) { + struct scx_sched *sch = scx_task_sched(p); + u64 deq_flags = core_deq_flags; + + /* + * Set %SCX_DEQ_SCHED_CHANGE when the dequeue is due to a property + * change (not sleep or core-sched pick). + */ + if (!(deq_flags & (DEQUEUE_SLEEP | SCX_DEQ_CORE_SCHED_EXEC))) + deq_flags |= SCX_DEQ_SCHED_CHANGE; + if (!(p->scx.flags & SCX_TASK_QUEUED)) { WARN_ON_ONCE(task_runnable(p)); return true; } - ops_dequeue(p, deq_flags); + ops_dequeue(rq, p, deq_flags); /* * A currently running task which is going off @rq first gets dequeued @@ -2216,17 +2173,17 @@ static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int deq_flags * * @p may go through multiple stopping <-> running transitions between * here and put_prev_task_scx() if task attribute changes occur while - * balance_scx() leaves @rq unlocked. However, they don't contain any + * balance_one() leaves @rq unlocked. However, they don't contain any * information meaningful to the BPF scheduler and can be suppressed by * skipping the callbacks if the task is !QUEUED. */ - if (SCX_HAS_OP(stopping) && task_current(rq, p)) { + if (SCX_HAS_OP(sch, stopping) && task_current(rq, p)) { update_curr_scx(rq); - SCX_CALL_OP_TASK(SCX_KF_REST, stopping, p, false); + SCX_CALL_OP_TASK(sch, stopping, rq, p, false); } - if (SCX_HAS_OP(quiescent) && !task_on_rq_migrating(p)) - SCX_CALL_OP_TASK(SCX_KF_REST, quiescent, p, deq_flags); + if (SCX_HAS_OP(sch, quiescent) && !task_on_rq_migrating(p)) + SCX_CALL_OP_TASK(sch, quiescent, rq, p, deq_flags); if (deq_flags & SCX_DEQ_SLEEP) p->scx.flags |= SCX_TASK_DEQD_FOR_SLEEP; @@ -2238,30 +2195,58 @@ static bool dequeue_task_scx(struct rq *rq, struct task_struct *p, int deq_flags sub_nr_running(rq, 1); dispatch_dequeue(rq, p); + clear_direct_dispatch(p); return true; } static void yield_task_scx(struct rq *rq) { - struct task_struct *p = rq->curr; + struct task_struct *p = rq->donor; + struct scx_sched *sch = scx_task_sched(p); - if (SCX_HAS_OP(yield)) - SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, p, NULL); + if (SCX_HAS_OP(sch, yield)) + SCX_CALL_OP_2TASKS_RET(sch, yield, rq, p, NULL); else p->scx.slice = 0; } static bool yield_to_task_scx(struct rq *rq, struct task_struct *to) { - struct task_struct *from = rq->curr; + struct task_struct *from = rq->donor; + struct scx_sched *sch = scx_task_sched(from); - if (SCX_HAS_OP(yield)) - return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, yield, from, to); + if (SCX_HAS_OP(sch, yield) && sch == scx_task_sched(to)) + return SCX_CALL_OP_2TASKS_RET(sch, yield, rq, from, to); else return false; } -static void move_local_task_to_local_dsq(struct task_struct *p, u64 enq_flags, +static void wakeup_preempt_scx(struct rq *rq, struct task_struct *p, int wake_flags) +{ + /* + * Preemption between SCX tasks is implemented by resetting the victim + * task's slice to 0 and triggering reschedule on the target CPU. + * Nothing to do. + */ + if (p->sched_class == &ext_sched_class) + return; + + /* + * Getting preempted by a higher-priority class. Reenqueue IMMED tasks. + * This captures all preemption cases including: + * + * - A SCX task is currently running. + * + * - @rq is waking from idle due to a SCX task waking to it. + * + * - A higher-priority wakes up while SCX dispatch is in progress. + */ + if (rq->scx.nr_immed) + schedule_reenq_local(rq, 0); +} + +static void move_local_task_to_local_dsq(struct scx_sched *sch, + struct task_struct *p, u64 enq_flags, struct scx_dispatch_q *src_dsq, struct rq *dst_rq) { @@ -2278,11 +2263,12 @@ static void move_local_task_to_local_dsq(struct task_struct *p, u64 enq_flags, else list_add_tail(&p->scx.dsq_list.node, &dst_dsq->list); - dsq_mod_nr(dst_dsq, 1); + dsq_inc_nr(dst_dsq, p, enq_flags); p->scx.dsq = dst_dsq; + + local_dsq_post_enq(sch, dst_dsq, p, enq_flags); } -#ifdef CONFIG_SMP /** * move_remote_task_to_local_dsq - Move a task from a foreign rq to a local DSQ * @p: task to move @@ -2297,10 +2283,13 @@ static void move_remote_task_to_local_dsq(struct task_struct *p, u64 enq_flags, { lockdep_assert_rq_held(src_rq); - /* the following marks @p MIGRATING which excludes dequeue */ + /* + * Set sticky_cpu before deactivate_task() to properly mark the + * beginning of an SCX-internal migration. + */ + p->scx.sticky_cpu = cpu_of(dst_rq); deactivate_task(src_rq, p, 0); set_task_cpu(p, cpu_of(dst_rq)); - p->scx.sticky_cpu = cpu_of(dst_rq); raw_spin_rq_unlock(src_rq); raw_spin_rq_lock(dst_rq); @@ -2336,12 +2325,13 @@ static void move_remote_task_to_local_dsq(struct task_struct *p, u64 enq_flags, * * The caller must ensure that @p and @rq are on different CPUs. */ -static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, - bool trigger_error) +static bool task_can_run_on_remote_rq(struct scx_sched *sch, + struct task_struct *p, struct rq *rq, + bool enforce) { - int cpu = cpu_of(rq); + s32 cpu = cpu_of(rq); - SCHED_WARN_ON(task_cpu(p) == cpu); + WARN_ON_ONCE(task_cpu(p) == cpu); /* * If @p has migration disabled, @p->cpus_ptr is updated to contain only @@ -2356,9 +2346,9 @@ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, * easily be masked if task_allowed_on_cpu() is done first. */ if (unlikely(is_migration_disabled(p))) { - if (trigger_error) - scx_ops_error("SCX_DSQ_LOCAL[_ON] cannot move migration disabled %s[%d] from CPU %d to %d", - p->comm, p->pid, task_cpu(p), cpu); + if (enforce) + scx_error(sch, "SCX_DSQ_LOCAL[_ON] cannot move migration disabled %s[%d] from CPU %d to %d", + p->comm, p->pid, task_cpu(p), cpu); return false; } @@ -2369,14 +2359,17 @@ static bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, * picked CPU is outside the allowed mask. */ if (!task_allowed_on_cpu(p, cpu)) { - if (trigger_error) - scx_ops_error("SCX_DSQ_LOCAL[_ON] target CPU %d not allowed for %s[%d]", - cpu, p->comm, p->pid); + if (enforce) + scx_error(sch, "SCX_DSQ_LOCAL[_ON] target CPU %d not allowed for %s[%d]", + cpu, p->comm, p->pid); return false; } - if (!scx_rq_online(rq)) + if (!scx_rq_online(rq)) { + if (enforce) + __scx_add_event(sch, SCX_EV_DISPATCH_LOCAL_DSQ_OFFLINE, 1); return false; + } return true; } @@ -2430,13 +2423,14 @@ static bool unlink_dsq_and_lock_src_rq(struct task_struct *p, !WARN_ON_ONCE(src_rq != task_rq(p)); } -static bool consume_remote_task(struct rq *this_rq, struct task_struct *p, +static bool consume_remote_task(struct rq *this_rq, + struct task_struct *p, u64 enq_flags, struct scx_dispatch_q *dsq, struct rq *src_rq) { raw_spin_rq_unlock(this_rq); if (unlink_dsq_and_lock_src_rq(p, dsq, src_rq)) { - move_remote_task_to_local_dsq(p, 0, src_rq, this_rq); + move_remote_task_to_local_dsq(p, enq_flags, src_rq, this_rq); return true; } else { raw_spin_rq_unlock(src_rq); @@ -2444,14 +2438,10 @@ static bool consume_remote_task(struct rq *this_rq, struct task_struct *p, return false; } } -#else /* CONFIG_SMP */ -static inline void move_remote_task_to_local_dsq(struct task_struct *p, u64 enq_flags, struct rq *src_rq, struct rq *dst_rq) { WARN_ON_ONCE(1); } -static inline bool task_can_run_on_remote_rq(struct task_struct *p, struct rq *rq, bool trigger_error) { return false; } -static inline bool consume_remote_task(struct rq *this_rq, struct task_struct *p, struct scx_dispatch_q *dsq, struct rq *task_rq) { return false; } -#endif /* CONFIG_SMP */ /** * move_task_between_dsqs() - Move a task from one DSQ to another + * @sch: scx_sched being operated on * @p: target task * @enq_flags: %SCX_ENQ_* * @src_dsq: DSQ @p is currently on, must not be a local DSQ @@ -2465,7 +2455,8 @@ static inline bool consume_remote_task(struct rq *this_rq, struct task_struct *p * On return, @src_dsq is unlocked and only @p's new task_rq, which is the * return value, is locked. */ -static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, +static struct rq *move_task_between_dsqs(struct scx_sched *sch, + struct task_struct *p, u64 enq_flags, struct scx_dispatch_q *src_dsq, struct scx_dispatch_q *dst_dsq) { @@ -2478,9 +2469,10 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, if (dst_dsq->id == SCX_DSQ_LOCAL) { dst_rq = container_of(dst_dsq, struct rq, scx.local_dsq); if (src_rq != dst_rq && - unlikely(!task_can_run_on_remote_rq(p, dst_rq, true))) { - dst_dsq = find_global_dsq(p); + unlikely(!task_can_run_on_remote_rq(sch, p, dst_rq, true))) { + dst_dsq = find_global_dsq(sch, task_cpu(p)); dst_rq = src_rq; + enq_flags |= SCX_ENQ_GDSQ_FALLBACK; } } else { /* no need to migrate if destination is a non-local DSQ */ @@ -2495,7 +2487,7 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, /* @p is going from a non-local DSQ to a local DSQ */ if (src_rq == dst_rq) { task_unlink_from_dsq(p, src_dsq); - move_local_task_to_local_dsq(p, enq_flags, + move_local_task_to_local_dsq(sch, p, enq_flags, src_dsq, dst_rq); raw_spin_unlock(&src_dsq->lock); } else { @@ -2508,58 +2500,21 @@ static struct rq *move_task_between_dsqs(struct task_struct *p, u64 enq_flags, * @p is going from a non-local DSQ to a non-local DSQ. As * $src_dsq is already locked, do an abbreviated dequeue. */ - task_unlink_from_dsq(p, src_dsq); - p->scx.dsq = NULL; + dispatch_dequeue_locked(p, src_dsq); raw_spin_unlock(&src_dsq->lock); - dispatch_enqueue(dst_dsq, p, enq_flags); + dispatch_enqueue(sch, dst_rq, dst_dsq, p, enq_flags); } return dst_rq; } -/* - * A poorly behaving BPF scheduler can live-lock the system by e.g. incessantly - * banging on the same DSQ on a large NUMA system to the point where switching - * to the bypass mode can take a long time. Inject artificial delays while the - * bypass mode is switching to guarantee timely completion. - */ -static void scx_ops_breather(struct rq *rq) -{ - u64 until; - - lockdep_assert_rq_held(rq); - - if (likely(!atomic_read(&scx_ops_breather_depth))) - return; - - raw_spin_rq_unlock(rq); - - until = ktime_get_ns() + NSEC_PER_MSEC; - - do { - int cnt = 1024; - while (atomic_read(&scx_ops_breather_depth) && --cnt) - cpu_relax(); - } while (atomic_read(&scx_ops_breather_depth) && - time_before64(ktime_get_ns(), until)); - - raw_spin_rq_lock(rq); -} - -static bool consume_dispatch_q(struct rq *rq, struct scx_dispatch_q *dsq) +static bool consume_dispatch_q(struct scx_sched *sch, struct rq *rq, + struct scx_dispatch_q *dsq, u64 enq_flags) { struct task_struct *p; retry: /* - * This retry loop can repeatedly race against scx_ops_bypass() - * dequeueing tasks from @dsq trying to put the system into the bypass - * mode. On some multi-socket machines (e.g. 2x Intel 8480c), this can - * live-lock the machine into soft lockups. Give a breather. - */ - scx_ops_breather(rq); - - /* * The caller can't expect to successfully consume a task if the task's * addition to @dsq isn't guaranteed to be visible somehow. Test * @dsq->list without locking and skip if it seems empty. @@ -2572,15 +2527,26 @@ retry: nldsq_for_each_task(p, dsq) { struct rq *task_rq = task_rq(p); + /* + * This loop can lead to multiple lockup scenarios, e.g. the BPF + * scheduler can put an enormous number of affinitized tasks into + * a contended DSQ, or the outer retry loop can repeatedly race + * against scx_bypass() dequeueing tasks from @dsq trying to put + * the system into the bypass mode. This can easily live-lock the + * machine. If aborting, exit from all non-bypass DSQs. + */ + if (unlikely(READ_ONCE(sch->aborting)) && dsq->id != SCX_DSQ_BYPASS) + break; + if (rq == task_rq) { task_unlink_from_dsq(p, dsq); - move_local_task_to_local_dsq(p, 0, dsq, rq); + move_local_task_to_local_dsq(sch, p, enq_flags, dsq, rq); raw_spin_unlock(&dsq->lock); return true; } - if (task_can_run_on_remote_rq(p, rq, false)) { - if (likely(consume_remote_task(rq, p, dsq, task_rq))) + if (task_can_run_on_remote_rq(sch, p, rq, false)) { + if (likely(consume_remote_task(rq, p, enq_flags, dsq, task_rq))) return true; goto retry; } @@ -2590,15 +2556,16 @@ retry: return false; } -static bool consume_global_dsq(struct rq *rq) +static bool consume_global_dsq(struct scx_sched *sch, struct rq *rq) { int node = cpu_to_node(cpu_of(rq)); - return consume_dispatch_q(rq, global_dsqs[node]); + return consume_dispatch_q(sch, rq, &sch->pnode[node]->global_dsq, 0); } /** * dispatch_to_local_dsq - Dispatch a task to a local dsq + * @sch: scx_sched being operated on * @rq: current rq which is locked * @dst_dsq: destination DSQ * @p: task to dispatch @@ -2611,14 +2578,13 @@ static bool consume_global_dsq(struct rq *rq) * The caller must have exclusive ownership of @p (e.g. through * %SCX_OPSS_DISPATCHING). */ -static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, +static void dispatch_to_local_dsq(struct scx_sched *sch, struct rq *rq, + struct scx_dispatch_q *dst_dsq, struct task_struct *p, u64 enq_flags) { struct rq *src_rq = task_rq(p); struct rq *dst_rq = container_of(dst_dsq, struct rq, scx.local_dsq); -#ifdef CONFIG_SMP struct rq *locked_rq = rq; -#endif /* * We're synchronized against dequeue through DISPATCHING. As @p can't @@ -2627,15 +2593,15 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, * If dispatching to @rq that @p is already on, no lock dancing needed. */ if (rq == src_rq && rq == dst_rq) { - dispatch_enqueue(dst_dsq, p, enq_flags | SCX_ENQ_CLEAR_OPSS); + dispatch_enqueue(sch, rq, dst_dsq, p, + enq_flags | SCX_ENQ_CLEAR_OPSS); return; } -#ifdef CONFIG_SMP if (src_rq != dst_rq && - unlikely(!task_can_run_on_remote_rq(p, dst_rq, true))) { - dispatch_enqueue(find_global_dsq(p), p, - enq_flags | SCX_ENQ_CLEAR_OPSS); + unlikely(!task_can_run_on_remote_rq(sch, p, dst_rq, true))) { + dispatch_enqueue(sch, rq, find_global_dsq(sch, task_cpu(p)), p, + enq_flags | SCX_ENQ_CLEAR_OPSS | SCX_ENQ_GDSQ_FALLBACK); return; } @@ -2672,7 +2638,8 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, */ if (src_rq == dst_rq) { p->scx.holding_cpu = -1; - dispatch_enqueue(&dst_rq->scx.local_dsq, p, enq_flags); + dispatch_enqueue(sch, dst_rq, &dst_rq->scx.local_dsq, p, + enq_flags); } else { move_remote_task_to_local_dsq(p, enq_flags, src_rq, dst_rq); @@ -2690,9 +2657,6 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, raw_spin_rq_unlock(locked_rq); raw_spin_rq_lock(rq); } -#else /* CONFIG_SMP */ - BUG(); /* control can not reach here on UP */ -#endif /* CONFIG_SMP */ } /** @@ -2714,7 +2678,8 @@ static void dispatch_to_local_dsq(struct rq *rq, struct scx_dispatch_q *dst_dsq, * was valid in the first place. Make sure that the task is still owned by the * BPF scheduler and claim the ownership before dispatching. */ -static void finish_dispatch(struct rq *rq, struct task_struct *p, +static void finish_dispatch(struct scx_sched *sch, struct rq *rq, + struct task_struct *p, unsigned long qseq_at_dispatch, u64 dsq_id, u64 enq_flags) { @@ -2743,6 +2708,12 @@ retry: if ((opss & SCX_OPSS_QSEQ_MASK) != qseq_at_dispatch) return; + /* see SCX_EV_INSERT_NOT_OWNED definition */ + if (unlikely(!scx_task_on_sched(sch, p))) { + __scx_add_event(sch, SCX_EV_INSERT_NOT_OWNED, 1); + return; + } + /* * While we know @p is accessible, we don't yet have a claim on * it - the BPF scheduler is allowed to dispatch tasks @@ -2767,23 +2738,23 @@ retry: BUG_ON(!(p->scx.flags & SCX_TASK_QUEUED)); - dsq = find_dsq_for_dispatch(this_rq(), dsq_id, p); + dsq = find_dsq_for_dispatch(sch, this_rq(), dsq_id, task_cpu(p)); if (dsq->id == SCX_DSQ_LOCAL) - dispatch_to_local_dsq(rq, dsq, p, enq_flags); + dispatch_to_local_dsq(sch, rq, dsq, p, enq_flags); else - dispatch_enqueue(dsq, p, enq_flags | SCX_ENQ_CLEAR_OPSS); + dispatch_enqueue(sch, rq, dsq, p, enq_flags | SCX_ENQ_CLEAR_OPSS); } -static void flush_dispatch_buf(struct rq *rq) +static void flush_dispatch_buf(struct scx_sched *sch, struct rq *rq) { - struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); + struct scx_dsp_ctx *dspc = &this_cpu_ptr(sch->pcpu)->dsp_ctx; u32 u; for (u = 0; u < dspc->cursor; u++) { struct scx_dsp_buf_ent *ent = &dspc->buf[u]; - finish_dispatch(rq, ent->task, ent->qseq, ent->dsq_id, + finish_dispatch(sch, rq, ent->task, ent->qseq, ent->dsq_id, ent->enq_flags); } @@ -2791,58 +2762,67 @@ static void flush_dispatch_buf(struct rq *rq) dspc->cursor = 0; } -static int balance_one(struct rq *rq, struct task_struct *prev) +static inline void maybe_queue_balance_callback(struct rq *rq) { - struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); - bool prev_on_scx = prev->sched_class == &ext_sched_class; - bool prev_on_rq = prev->scx.flags & SCX_TASK_QUEUED; - int nr_loops = SCX_DSP_MAX_LOOPS; - lockdep_assert_rq_held(rq); - rq->scx.flags |= SCX_RQ_IN_BALANCE; - rq->scx.flags &= ~(SCX_RQ_BAL_PENDING | SCX_RQ_BAL_KEEP); - if (static_branch_unlikely(&scx_ops_cpu_preempt) && - unlikely(rq->scx.cpu_released)) { - /* - * If the previous sched_class for the current CPU was not SCX, - * notify the BPF scheduler that it again has control of the - * core. This callback complements ->cpu_release(), which is - * emitted in switch_class(). - */ - if (SCX_HAS_OP(cpu_acquire)) - SCX_CALL_OP(SCX_KF_REST, cpu_acquire, cpu_of(rq), NULL); - rq->scx.cpu_released = false; - } + if (!(rq->scx.flags & SCX_RQ_BAL_CB_PENDING)) + return; - if (prev_on_scx) { - update_curr_scx(rq); + queue_balance_callback(rq, &rq->scx.deferred_bal_cb, + deferred_bal_cb_workfn); + + rq->scx.flags &= ~SCX_RQ_BAL_CB_PENDING; +} + +/* + * One user of this function is scx_bpf_dispatch() which can be called + * recursively as sub-sched dispatches nest. Always inline to reduce stack usage + * from the call frame. + */ +static __always_inline bool +scx_dispatch_sched(struct scx_sched *sch, struct rq *rq, + struct task_struct *prev, bool nested) +{ + struct scx_dsp_ctx *dspc = &this_cpu_ptr(sch->pcpu)->dsp_ctx; + int nr_loops = SCX_DSP_MAX_LOOPS; + s32 cpu = cpu_of(rq); + bool prev_on_sch = (prev->sched_class == &ext_sched_class) && + scx_task_on_sched(sch, prev); + + if (consume_global_dsq(sch, rq)) + return true; + if (bypass_dsp_enabled(sch)) { + /* if @sch is bypassing, only the bypass DSQs are active */ + if (scx_bypassing(sch, cpu)) + return consume_dispatch_q(sch, rq, bypass_dsq(sch, cpu), 0); + +#ifdef CONFIG_EXT_SUB_SCHED /* - * If @prev is runnable & has slice left, it has priority and - * fetching more just increases latency for the fetched tasks. - * Tell pick_task_scx() to keep running @prev. If the BPF - * scheduler wants to handle this explicitly, it should - * implement ->cpu_release(). + * If @sch isn't bypassing but its children are, @sch is + * responsible for making forward progress for both its own + * tasks that aren't bypassing and the bypassing descendants' + * tasks. The following implements a simple built-in behavior - + * let each CPU try to run the bypass DSQ every Nth time. * - * See scx_ops_disable_workfn() for the explanation on the - * bypassing test. + * Later, if necessary, we can add an ops flag to suppress the + * auto-consumption and a kfunc to consume the bypass DSQ and, + * so that the BPF scheduler can fully control scheduling of + * bypassed tasks. */ - if (prev_on_rq && prev->scx.slice && !scx_rq_bypassing(rq)) { - rq->scx.flags |= SCX_RQ_BAL_KEEP; - goto has_tasks; + struct scx_sched_pcpu *pcpu = per_cpu_ptr(sch->pcpu, cpu); + + if (!(pcpu->bypass_host_seq++ % SCX_BYPASS_HOST_NTH) && + consume_dispatch_q(sch, rq, bypass_dsq(sch, cpu), 0)) { + __scx_add_event(sch, SCX_EV_SUB_BYPASS_DISPATCH, 1); + return true; } +#endif /* CONFIG_EXT_SUB_SCHED */ } - /* if there already are tasks to run, nothing to do */ - if (rq->scx.local_dsq.nr) - goto has_tasks; - - if (consume_global_dsq(rq)) - goto has_tasks; - - if (!SCX_HAS_OP(dispatch) || scx_rq_bypassing(rq) || !scx_rq_online(rq)) - goto no_tasks; + if (unlikely(!SCX_HAS_OP(sch, dispatch)) || !scx_rq_online(rq)) + return false; dspc->rq = rq; @@ -2856,19 +2836,25 @@ static int balance_one(struct rq *rq, struct task_struct *prev) do { dspc->nr_tasks = 0; - SCX_CALL_OP(SCX_KF_DISPATCH, dispatch, cpu_of(rq), - prev_on_scx ? prev : NULL); + if (nested) { + SCX_CALL_OP(sch, dispatch, rq, cpu, prev_on_sch ? prev : NULL); + } else { + /* stash @prev so that nested invocations can access it */ + rq->scx.sub_dispatch_prev = prev; + SCX_CALL_OP(sch, dispatch, rq, cpu, prev_on_sch ? prev : NULL); + rq->scx.sub_dispatch_prev = NULL; + } - flush_dispatch_buf(rq); + flush_dispatch_buf(sch, rq); - if (prev_on_rq && prev->scx.slice) { + if ((prev->scx.flags & SCX_TASK_QUEUED) && prev->scx.slice) { rq->scx.flags |= SCX_RQ_BAL_KEEP; - goto has_tasks; + return true; } if (rq->scx.local_dsq.nr) - goto has_tasks; - if (consume_global_dsq(rq)) - goto has_tasks; + return true; + if (consume_global_dsq(sch, rq)) + return true; /* * ops.dispatch() can trap us in this loop by repeatedly @@ -2877,107 +2863,123 @@ static int balance_one(struct rq *rq, struct task_struct *prev) * balance(), we want to complete this scheduling cycle and then * start a new one. IOW, we want to call resched_curr() on the * next, most likely idle, task, not the current one. Use - * scx_bpf_kick_cpu() for deferred kicking. + * __scx_bpf_kick_cpu() for deferred kicking. */ if (unlikely(!--nr_loops)) { - scx_bpf_kick_cpu(cpu_of(rq), 0); + scx_kick_cpu(sch, cpu, 0); break; } } while (dspc->nr_tasks); -no_tasks: /* - * Didn't find another task to run. Keep running @prev unless - * %SCX_OPS_ENQ_LAST is in effect. + * Prevent the CPU from going idle while bypassed descendants have tasks + * queued. Without this fallback, bypassed tasks could stall if the host + * scheduler's ops.dispatch() doesn't yield any tasks. */ - if (prev_on_rq && (!static_branch_unlikely(&scx_ops_enq_last) || - scx_rq_bypassing(rq))) { - rq->scx.flags |= SCX_RQ_BAL_KEEP; - goto has_tasks; - } - rq->scx.flags &= ~SCX_RQ_IN_BALANCE; - return false; + if (bypass_dsp_enabled(sch)) + return consume_dispatch_q(sch, rq, bypass_dsq(sch, cpu), 0); -has_tasks: - rq->scx.flags &= ~SCX_RQ_IN_BALANCE; - return true; + return false; } -static int balance_scx(struct rq *rq, struct task_struct *prev, - struct rq_flags *rf) +static int balance_one(struct rq *rq, struct task_struct *prev) { - int ret; - - rq_unpin_lock(rq, rf); + struct scx_sched *sch = scx_root; + s32 cpu = cpu_of(rq); - ret = balance_one(rq, prev); + lockdep_assert_rq_held(rq); + rq->scx.flags |= SCX_RQ_IN_BALANCE; + rq->scx.flags &= ~SCX_RQ_BAL_KEEP; -#ifdef CONFIG_SCHED_SMT - /* - * When core-sched is enabled, this ops.balance() call will be followed - * by pick_task_scx() on this CPU and the SMT siblings. Balance the - * siblings too. - */ - if (sched_core_enabled(rq)) { - const struct cpumask *smt_mask = cpu_smt_mask(cpu_of(rq)); - int scpu; + if ((sch->ops.flags & SCX_OPS_HAS_CPU_PREEMPT) && + unlikely(rq->scx.cpu_released)) { + /* + * If the previous sched_class for the current CPU was not SCX, + * notify the BPF scheduler that it again has control of the + * core. This callback complements ->cpu_release(), which is + * emitted in switch_class(). + */ + if (SCX_HAS_OP(sch, cpu_acquire)) + SCX_CALL_OP(sch, cpu_acquire, rq, cpu, NULL); + rq->scx.cpu_released = false; + } - for_each_cpu_andnot(scpu, smt_mask, cpumask_of(cpu_of(rq))) { - struct rq *srq = cpu_rq(scpu); - struct task_struct *sprev = srq->curr; + if (prev->sched_class == &ext_sched_class) { + update_curr_scx(rq); - WARN_ON_ONCE(__rq_lockp(rq) != __rq_lockp(srq)); - update_rq_clock(srq); - balance_one(srq, sprev); + /* + * If @prev is runnable & has slice left, it has priority and + * fetching more just increases latency for the fetched tasks. + * Tell pick_task_scx() to keep running @prev. If the BPF + * scheduler wants to handle this explicitly, it should + * implement ->cpu_release(). + * + * See scx_disable_workfn() for the explanation on the bypassing + * test. + */ + if ((prev->scx.flags & SCX_TASK_QUEUED) && prev->scx.slice && + !scx_bypassing(sch, cpu)) { + rq->scx.flags |= SCX_RQ_BAL_KEEP; + goto has_tasks; } } -#endif - rq_repin_lock(rq, rf); - - return ret; -} -static void process_ddsp_deferred_locals(struct rq *rq) -{ - struct task_struct *p; + /* if there already are tasks to run, nothing to do */ + if (rq->scx.local_dsq.nr) + goto has_tasks; - lockdep_assert_rq_held(rq); + if (scx_dispatch_sched(sch, rq, prev, false)) + goto has_tasks; /* - * Now that @rq can be unlocked, execute the deferred enqueueing of - * tasks directly dispatched to the local DSQs of other CPUs. See - * direct_dispatch(). Keep popping from the head instead of using - * list_for_each_entry_safe() as dispatch_local_dsq() may unlock @rq - * temporarily. + * Didn't find another task to run. Keep running @prev unless + * %SCX_OPS_ENQ_LAST is in effect. */ - while ((p = list_first_entry_or_null(&rq->scx.ddsp_deferred_locals, - struct task_struct, scx.dsq_list.node))) { - struct scx_dispatch_q *dsq; + if ((prev->scx.flags & SCX_TASK_QUEUED) && + (!(sch->ops.flags & SCX_OPS_ENQ_LAST) || scx_bypassing(sch, cpu))) { + rq->scx.flags |= SCX_RQ_BAL_KEEP; + __scx_add_event(sch, SCX_EV_DISPATCH_KEEP_LAST, 1); + goto has_tasks; + } + rq->scx.flags &= ~SCX_RQ_IN_BALANCE; + return false; - list_del_init(&p->scx.dsq_list.node); +has_tasks: + /* + * @rq may have extra IMMED tasks without reenq scheduled: + * + * - rq_is_open() can't reliably tell when and how slice is going to be + * modified for $curr and allows IMMED tasks to be queued while + * dispatch is in progress. + * + * - A non-IMMED HEAD task can get queued in front of an IMMED task + * between the IMMED queueing and the subsequent scheduling event. + */ + if (unlikely(rq->scx.local_dsq.nr > 1 && rq->scx.nr_immed)) + schedule_reenq_local(rq, 0); - dsq = find_dsq_for_dispatch(rq, p->scx.ddsp_dsq_id, p); - if (!WARN_ON_ONCE(dsq->id != SCX_DSQ_LOCAL)) - dispatch_to_local_dsq(rq, dsq, p, p->scx.ddsp_enq_flags); - } + rq->scx.flags &= ~SCX_RQ_IN_BALANCE; + return true; } static void set_next_task_scx(struct rq *rq, struct task_struct *p, bool first) { + struct scx_sched *sch = scx_task_sched(p); + if (p->scx.flags & SCX_TASK_QUEUED) { /* * Core-sched might decide to execute @p before it is * dispatched. Call ops_dequeue() to notify the BPF scheduler. */ - ops_dequeue(p, SCX_DEQ_CORE_SCHED_EXEC); + ops_dequeue(rq, p, SCX_DEQ_CORE_SCHED_EXEC); dispatch_dequeue(rq, p); } p->se.exec_start = rq_clock_task(rq); /* see dequeue_task_scx() on why we skip when !QUEUED */ - if (SCX_HAS_OP(running) && (p->scx.flags & SCX_TASK_QUEUED)) - SCX_CALL_OP_TASK(SCX_KF_REST, running, p); + if (SCX_HAS_OP(sch, running) && (p->scx.flags & SCX_TASK_QUEUED)) + SCX_CALL_OP_TASK(sch, running, rq, p); clr_task_runnable(p, true); @@ -3007,10 +3009,8 @@ static void set_next_task_scx(struct rq *rq, struct task_struct *p, bool first) static enum scx_cpu_preempt_reason preempt_reason_from_class(const struct sched_class *class) { -#ifdef CONFIG_SMP if (class == &stop_sched_class) return SCX_CPU_PREEMPT_STOP; -#endif if (class == &dl_sched_class) return SCX_CPU_PREEMPT_DL; if (class == &rt_sched_class) @@ -3020,17 +3020,10 @@ preempt_reason_from_class(const struct sched_class *class) static void switch_class(struct rq *rq, struct task_struct *next) { + struct scx_sched *sch = scx_root; const struct sched_class *next_class = next->sched_class; -#ifdef CONFIG_SMP - /* - * Pairs with the smp_load_acquire() issued by a CPU in - * kick_cpus_irq_workfn() who is waiting for this CPU to perform a - * resched. - */ - smp_store_release(&rq->scx.pnt_seq, rq->scx.pnt_seq + 1); -#endif - if (!static_branch_unlikely(&scx_ops_cpu_preempt)) + if (!(sch->ops.flags & SCX_OPS_HAS_CPU_PREEMPT)) return; /* @@ -3049,17 +3042,16 @@ static void switch_class(struct rq *rq, struct task_struct *next) * preempted, and it regaining control of the CPU. * * ->cpu_release() complements ->cpu_acquire(), which is emitted the - * next time that balance_scx() is invoked. + * next time that balance_one() is invoked. */ if (!rq->scx.cpu_released) { - if (SCX_HAS_OP(cpu_release)) { + if (SCX_HAS_OP(sch, cpu_release)) { struct scx_cpu_release_args args = { .reason = preempt_reason_from_class(next_class), .task = next, }; - SCX_CALL_OP(SCX_KF_CPU_RELEASE, - cpu_release, cpu_of(rq), &args); + SCX_CALL_OP(sch, cpu_release, rq, cpu_of(rq), &args); } rq->scx.cpu_released = true; } @@ -3068,11 +3060,16 @@ static void switch_class(struct rq *rq, struct task_struct *next) static void put_prev_task_scx(struct rq *rq, struct task_struct *p, struct task_struct *next) { + struct scx_sched *sch = scx_task_sched(p); + + /* see kick_sync_wait_bal_cb() */ + smp_store_release(&rq->scx.kick_sync, rq->scx.kick_sync + 1); + update_curr_scx(rq); /* see dequeue_task_scx() on why we skip when !QUEUED */ - if (SCX_HAS_OP(stopping) && (p->scx.flags & SCX_TASK_QUEUED)) - SCX_CALL_OP_TASK(SCX_KF_REST, stopping, p, true); + if (SCX_HAS_OP(sch, stopping) && (p->scx.flags & SCX_TASK_QUEUED)) + SCX_CALL_OP_TASK(sch, stopping, rq, p, true); if (p->scx.flags & SCX_TASK_QUEUED) { set_task_runnable(rq, p); @@ -3081,10 +3078,17 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, * If @p has slice left and is being put, @p is getting * preempted by a higher priority scheduler class or core-sched * forcing a different task. Leave it at the head of the local - * DSQ. + * DSQ unless it was an IMMED task. IMMED tasks should not + * linger on a busy CPU, reenqueue them to the BPF scheduler. */ - if (p->scx.slice && !scx_rq_bypassing(rq)) { - dispatch_enqueue(&rq->scx.local_dsq, p, SCX_ENQ_HEAD); + if (p->scx.slice && !scx_bypassing(sch, cpu_of(rq))) { + if (p->scx.flags & SCX_TASK_IMMED) { + p->scx.flags |= SCX_TASK_REENQ_PREEMPTED; + do_enqueue_task(rq, p, SCX_ENQ_REENQ, -1); + p->scx.flags &= ~SCX_TASK_REENQ_REASON_MASK; + } else { + dispatch_enqueue(sch, rq, &rq->scx.local_dsq, p, SCX_ENQ_HEAD); + } goto switch_class; } @@ -3094,8 +3098,8 @@ static void put_prev_task_scx(struct rq *rq, struct task_struct *p, * ops.enqueue() that @p is the only one available for this cpu, * which should trigger an explicit follow-up scheduling event. */ - if (sched_class_above(&ext_sched_class, next->sched_class)) { - WARN_ON_ONCE(!static_branch_unlikely(&scx_ops_enq_last)); + if (next && sched_class_above(&ext_sched_class, next->sched_class)) { + WARN_ON_ONCE(!(sch->ops.flags & SCX_OPS_ENQ_LAST)); do_enqueue_task(rq, p, SCX_ENQ_LAST, -1); } else { do_enqueue_task(rq, p, 0, -1); @@ -3107,81 +3111,162 @@ switch_class: switch_class(rq, next); } +static void kick_sync_wait_bal_cb(struct rq *rq) +{ + struct scx_kick_syncs __rcu *ks = __this_cpu_read(scx_kick_syncs); + unsigned long *ksyncs = rcu_dereference_sched(ks)->syncs; + bool waited; + s32 cpu; + + /* + * Drop rq lock and enable IRQs while waiting. IRQs must be enabled + * — a target CPU may be waiting for us to process an IPI (e.g. TLB + * flush) while we wait for its kick_sync to advance. + * + * Also, keep advancing our own kick_sync so that new kick_sync waits + * targeting us, which can start after we drop the lock, cannot form + * cyclic dependencies. + */ +retry: + waited = false; + for_each_cpu(cpu, rq->scx.cpus_to_sync) { + /* + * smp_load_acquire() pairs with smp_store_release() on + * kick_sync updates on the target CPUs. + */ + if (cpu == cpu_of(rq) || + smp_load_acquire(&cpu_rq(cpu)->scx.kick_sync) != ksyncs[cpu]) { + cpumask_clear_cpu(cpu, rq->scx.cpus_to_sync); + continue; + } + + raw_spin_rq_unlock_irq(rq); + while (READ_ONCE(cpu_rq(cpu)->scx.kick_sync) == ksyncs[cpu]) { + smp_store_release(&rq->scx.kick_sync, rq->scx.kick_sync + 1); + cpu_relax(); + } + raw_spin_rq_lock_irq(rq); + waited = true; + } + + if (waited) + goto retry; +} + static struct task_struct *first_local_task(struct rq *rq) { return list_first_entry_or_null(&rq->scx.local_dsq.list, struct task_struct, scx.dsq_list.node); } -static struct task_struct *pick_task_scx(struct rq *rq) +static struct task_struct * +do_pick_task_scx(struct rq *rq, struct rq_flags *rf, bool force_scx) { struct task_struct *prev = rq->curr; + bool keep_prev; struct task_struct *p; - bool keep_prev = rq->scx.flags & SCX_RQ_BAL_KEEP; - bool kick_idle = false; + + /* see kick_sync_wait_bal_cb() */ + smp_store_release(&rq->scx.kick_sync, rq->scx.kick_sync + 1); + + rq_modified_begin(rq, &ext_sched_class); + + rq_unpin_lock(rq, rf); + balance_one(rq, prev); + rq_repin_lock(rq, rf); + maybe_queue_balance_callback(rq); /* - * WORKAROUND: - * - * %SCX_RQ_BAL_KEEP should be set iff $prev is on SCX as it must just - * have gone through balance_scx(). Unfortunately, there currently is a - * bug where fair could say yes on balance() but no on pick_task(), - * which then ends up calling pick_task_scx() without preceding - * balance_scx(). - * - * Keep running @prev if possible and avoid stalling from entering idle - * without balancing. + * Defer to a balance callback which can drop rq lock and enable + * IRQs. Waiting directly in the pick path would deadlock against + * CPUs sending us IPIs (e.g. TLB flushes) while we wait for them. + */ + if (unlikely(rq->scx.kick_sync_pending)) { + rq->scx.kick_sync_pending = false; + queue_balance_callback(rq, &rq->scx.kick_sync_bal_cb, + kick_sync_wait_bal_cb); + } + + /* + * If any higher-priority sched class enqueued a runnable task on + * this rq during balance_one(), abort and return RETRY_TASK, so + * that the scheduler loop can restart. * - * Once fair is fixed, remove the workaround and trigger WARN_ON_ONCE() - * if pick_task_scx() is called without preceding balance_scx(). + * If @force_scx is true, always try to pick a SCHED_EXT task, + * regardless of any higher-priority sched classes activity. */ - if (unlikely(rq->scx.flags & SCX_RQ_BAL_PENDING)) { - if (prev->scx.flags & SCX_TASK_QUEUED) { - keep_prev = true; - } else { - keep_prev = false; - kick_idle = true; - } - } else if (unlikely(keep_prev && - prev->sched_class != &ext_sched_class)) { - /* - * Can happen while enabling as SCX_RQ_BAL_PENDING assertion is - * conditional on scx_enabled() and may have been skipped. - */ - WARN_ON_ONCE(scx_ops_enable_state() == SCX_OPS_ENABLED); + if (!force_scx && rq_modified_above(rq, &ext_sched_class)) + return RETRY_TASK; + + keep_prev = rq->scx.flags & SCX_RQ_BAL_KEEP; + if (unlikely(keep_prev && + prev->sched_class != &ext_sched_class)) { + WARN_ON_ONCE(scx_enable_state() == SCX_ENABLED); keep_prev = false; } /* - * If balance_scx() is telling us to keep running @prev, replenish slice + * If balance_one() is telling us to keep running @prev, replenish slice * if necessary and keep running @prev. Otherwise, pop the first one * from the local DSQ. */ if (keep_prev) { p = prev; if (!p->scx.slice) - p->scx.slice = SCX_SLICE_DFL; + refill_task_slice_dfl(scx_task_sched(p), p); } else { p = first_local_task(rq); - if (!p) { - if (kick_idle) - scx_bpf_kick_cpu(cpu_of(rq), SCX_KICK_IDLE); + if (!p) return NULL; - } if (unlikely(!p->scx.slice)) { - if (!scx_rq_bypassing(rq) && !scx_warned_zero_slice) { + struct scx_sched *sch = scx_task_sched(p); + + if (!scx_bypassing(sch, cpu_of(rq)) && + !sch->warned_zero_slice) { printk_deferred(KERN_WARNING "sched_ext: %s[%d] has zero slice in %s()\n", p->comm, p->pid, __func__); - scx_warned_zero_slice = true; + sch->warned_zero_slice = true; } - p->scx.slice = SCX_SLICE_DFL; + refill_task_slice_dfl(sch, p); } } return p; } +static struct task_struct *pick_task_scx(struct rq *rq, struct rq_flags *rf) +{ + return do_pick_task_scx(rq, rf, false); +} + +/* + * Select the next task to run from the ext scheduling class. + * + * Use do_pick_task_scx() directly with @force_scx enabled, since the + * dl_server must always select a sched_ext task. + */ +static struct task_struct * +ext_server_pick_task(struct sched_dl_entity *dl_se, struct rq_flags *rf) +{ + if (!scx_enabled()) + return NULL; + + return do_pick_task_scx(dl_se->rq, rf, true); +} + +/* + * Initialize the ext server deadline entity. + */ +void ext_server_init(struct rq *rq) +{ + struct sched_dl_entity *dl_se = &rq->ext_server; + + init_dl_entity(dl_se); + + dl_server_init(dl_se, rq, ext_server_pick_task); +} + #ifdef CONFIG_SCHED_CORE /** * scx_prio_less - Task ordering for core-sched @@ -3204,13 +3289,18 @@ static struct task_struct *pick_task_scx(struct rq *rq) bool scx_prio_less(const struct task_struct *a, const struct task_struct *b, bool in_fi) { + struct scx_sched *sch_a = scx_task_sched(a); + struct scx_sched *sch_b = scx_task_sched(b); + /* * The const qualifiers are dropped from task_struct pointers when * calling ops.core_sched_before(). Accesses are controlled by the * verifier. */ - if (SCX_HAS_OP(core_sched_before) && !scx_rq_bypassing(task_rq(a))) - return SCX_CALL_OP_2TASKS_RET(SCX_KF_REST, core_sched_before, + if (sch_a == sch_b && SCX_HAS_OP(sch_a, core_sched_before) && + !scx_bypassing(sch_a, task_cpu(a))) + return SCX_CALL_OP_2TASKS_RET(sch_a, core_sched_before, + task_rq(a), (struct task_struct *)a, (struct task_struct *)b); else @@ -3218,420 +3308,11 @@ bool scx_prio_less(const struct task_struct *a, const struct task_struct *b, } #endif /* CONFIG_SCHED_CORE */ -#ifdef CONFIG_SMP - -static bool test_and_clear_cpu_idle(int cpu) -{ -#ifdef CONFIG_SCHED_SMT - /* - * SMT mask should be cleared whether we can claim @cpu or not. The SMT - * cluster is not wholly idle either way. This also prevents - * scx_pick_idle_cpu() from getting caught in an infinite loop. - */ - if (sched_smt_active()) { - const struct cpumask *smt = cpu_smt_mask(cpu); - - /* - * If offline, @cpu is not its own sibling and - * scx_pick_idle_cpu() can get caught in an infinite loop as - * @cpu is never cleared from idle_masks.smt. Ensure that @cpu - * is eventually cleared. - * - * NOTE: Use cpumask_intersects() and cpumask_test_cpu() to - * reduce memory writes, which may help alleviate cache - * coherence pressure. - */ - if (cpumask_intersects(smt, idle_masks.smt)) - cpumask_andnot(idle_masks.smt, idle_masks.smt, smt); - else if (cpumask_test_cpu(cpu, idle_masks.smt)) - __cpumask_clear_cpu(cpu, idle_masks.smt); - } -#endif - return cpumask_test_and_clear_cpu(cpu, idle_masks.cpu); -} - -static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags) -{ - int cpu; - -retry: - if (sched_smt_active()) { - cpu = cpumask_any_and_distribute(idle_masks.smt, cpus_allowed); - if (cpu < nr_cpu_ids) - goto found; - - if (flags & SCX_PICK_IDLE_CORE) - return -EBUSY; - } - - cpu = cpumask_any_and_distribute(idle_masks.cpu, cpus_allowed); - if (cpu >= nr_cpu_ids) - return -EBUSY; - -found: - if (test_and_clear_cpu_idle(cpu)) - return cpu; - else - goto retry; -} - -/* - * Return the amount of CPUs in the same LLC domain of @cpu (or zero if the LLC - * domain is not defined). - */ -static unsigned int llc_weight(s32 cpu) -{ - struct sched_domain *sd; - - sd = rcu_dereference(per_cpu(sd_llc, cpu)); - if (!sd) - return 0; - - return sd->span_weight; -} - -/* - * Return the cpumask representing the LLC domain of @cpu (or NULL if the LLC - * domain is not defined). - */ -static struct cpumask *llc_span(s32 cpu) -{ - struct sched_domain *sd; - - sd = rcu_dereference(per_cpu(sd_llc, cpu)); - if (!sd) - return 0; - - return sched_domain_span(sd); -} - -/* - * Return the amount of CPUs in the same NUMA domain of @cpu (or zero if the - * NUMA domain is not defined). - */ -static unsigned int numa_weight(s32 cpu) -{ - struct sched_domain *sd; - struct sched_group *sg; - - sd = rcu_dereference(per_cpu(sd_numa, cpu)); - if (!sd) - return 0; - sg = sd->groups; - if (!sg) - return 0; - - return sg->group_weight; -} - -/* - * Return the cpumask representing the NUMA domain of @cpu (or NULL if the NUMA - * domain is not defined). - */ -static struct cpumask *numa_span(s32 cpu) -{ - struct sched_domain *sd; - struct sched_group *sg; - - sd = rcu_dereference(per_cpu(sd_numa, cpu)); - if (!sd) - return NULL; - sg = sd->groups; - if (!sg) - return NULL; - - return sched_group_span(sg); -} - -/* - * Return true if the LLC domains do not perfectly overlap with the NUMA - * domains, false otherwise. - */ -static bool llc_numa_mismatch(void) -{ - int cpu; - - /* - * We need to scan all online CPUs to verify whether their scheduling - * domains overlap. - * - * While it is rare to encounter architectures with asymmetric NUMA - * topologies, CPU hotplugging or virtualized environments can result - * in asymmetric configurations. - * - * For example: - * - * NUMA 0: - * - LLC 0: cpu0..cpu7 - * - LLC 1: cpu8..cpu15 [offline] - * - * NUMA 1: - * - LLC 0: cpu16..cpu23 - * - LLC 1: cpu24..cpu31 - * - * In this case, if we only check the first online CPU (cpu0), we might - * incorrectly assume that the LLC and NUMA domains are fully - * overlapping, which is incorrect (as NUMA 1 has two distinct LLC - * domains). - */ - for_each_online_cpu(cpu) - if (llc_weight(cpu) != numa_weight(cpu)) - return true; - - return false; -} - -/* - * Initialize topology-aware scheduling. - * - * Detect if the system has multiple LLC or multiple NUMA domains and enable - * cache-aware / NUMA-aware scheduling optimizations in the default CPU idle - * selection policy. - * - * Assumption: the kernel's internal topology representation assumes that each - * CPU belongs to a single LLC domain, and that each LLC domain is entirely - * contained within a single NUMA node. - */ -static void update_selcpu_topology(void) -{ - bool enable_llc = false, enable_numa = false; - unsigned int nr_cpus; - s32 cpu = cpumask_first(cpu_online_mask); - - /* - * Enable LLC domain optimization only when there are multiple LLC - * domains among the online CPUs. If all online CPUs are part of a - * single LLC domain, the idle CPU selection logic can choose any - * online CPU without bias. - * - * Note that it is sufficient to check the LLC domain of the first - * online CPU to determine whether a single LLC domain includes all - * CPUs. - */ - rcu_read_lock(); - nr_cpus = llc_weight(cpu); - if (nr_cpus > 0) { - if (nr_cpus < num_online_cpus()) - enable_llc = true; - pr_debug("sched_ext: LLC=%*pb weight=%u\n", - cpumask_pr_args(llc_span(cpu)), llc_weight(cpu)); - } - - /* - * Enable NUMA optimization only when there are multiple NUMA domains - * among the online CPUs and the NUMA domains don't perfectly overlaps - * with the LLC domains. - * - * If all CPUs belong to the same NUMA node and the same LLC domain, - * enabling both NUMA and LLC optimizations is unnecessary, as checking - * for an idle CPU in the same domain twice is redundant. - */ - nr_cpus = numa_weight(cpu); - if (nr_cpus > 0) { - if (nr_cpus < num_online_cpus() && llc_numa_mismatch()) - enable_numa = true; - pr_debug("sched_ext: NUMA=%*pb weight=%u\n", - cpumask_pr_args(numa_span(cpu)), numa_weight(cpu)); - } - rcu_read_unlock(); - - pr_debug("sched_ext: LLC idle selection %s\n", - str_enabled_disabled(enable_llc)); - pr_debug("sched_ext: NUMA idle selection %s\n", - str_enabled_disabled(enable_numa)); - - if (enable_llc) - static_branch_enable_cpuslocked(&scx_selcpu_topo_llc); - else - static_branch_disable_cpuslocked(&scx_selcpu_topo_llc); - if (enable_numa) - static_branch_enable_cpuslocked(&scx_selcpu_topo_numa); - else - static_branch_disable_cpuslocked(&scx_selcpu_topo_numa); -} - -/* - * Built-in CPU idle selection policy: - * - * 1. Prioritize full-idle cores: - * - always prioritize CPUs from fully idle cores (both logical CPUs are - * idle) to avoid interference caused by SMT. - * - * 2. Reuse the same CPU: - * - prefer the last used CPU to take advantage of cached data (L1, L2) and - * branch prediction optimizations. - * - * 3. Pick a CPU within the same LLC (Last-Level Cache): - * - if the above conditions aren't met, pick a CPU that shares the same LLC - * to maintain cache locality. - * - * 4. Pick a CPU within the same NUMA node, if enabled: - * - choose a CPU from the same NUMA node to reduce memory access latency. - * - * 5. Pick any idle CPU usable by the task. - * - * Step 3 and 4 are performed only if the system has, respectively, multiple - * LLC domains / multiple NUMA nodes (see scx_selcpu_topo_llc and - * scx_selcpu_topo_numa). - * - * NOTE: tasks that can only run on 1 CPU are excluded by this logic, because - * we never call ops.select_cpu() for them, see select_task_rq(). - */ -static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, - u64 wake_flags, bool *found) -{ - const struct cpumask *llc_cpus = NULL; - const struct cpumask *numa_cpus = NULL; - s32 cpu; - - *found = false; - - /* - * This is necessary to protect llc_cpus. - */ - rcu_read_lock(); - - /* - * Determine the scheduling domain only if the task is allowed to run - * on all CPUs. - * - * This is done primarily for efficiency, as it avoids the overhead of - * updating a cpumask every time we need to select an idle CPU (which - * can be costly in large SMP systems), but it also aligns logically: - * if a task's scheduling domain is restricted by user-space (through - * CPU affinity), the task will simply use the flat scheduling domain - * defined by user-space. - */ - if (p->nr_cpus_allowed >= num_possible_cpus()) { - if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) - numa_cpus = numa_span(prev_cpu); - - if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) - llc_cpus = llc_span(prev_cpu); - } - - /* - * If WAKE_SYNC, try to migrate the wakee to the waker's CPU. - */ - if (wake_flags & SCX_WAKE_SYNC) { - cpu = smp_processor_id(); - - /* - * If the waker's CPU is cache affine and prev_cpu is idle, - * then avoid a migration. - */ - if (cpus_share_cache(cpu, prev_cpu) && - test_and_clear_cpu_idle(prev_cpu)) { - cpu = prev_cpu; - goto cpu_found; - } - - /* - * If the waker's local DSQ is empty, and the system is under - * utilized, try to wake up @p to the local DSQ of the waker. - * - * Checking only for an empty local DSQ is insufficient as it - * could give the wakee an unfair advantage when the system is - * oversaturated. - * - * Checking only for the presence of idle CPUs is also - * insufficient as the local DSQ of the waker could have tasks - * piled up on it even if there is an idle core elsewhere on - * the system. - */ - if (!cpumask_empty(idle_masks.cpu) && - !(current->flags & PF_EXITING) && - cpu_rq(cpu)->scx.local_dsq.nr == 0) { - if (cpumask_test_cpu(cpu, p->cpus_ptr)) - goto cpu_found; - } - } - - /* - * If CPU has SMT, any wholly idle CPU is likely a better pick than - * partially idle @prev_cpu. - */ - if (sched_smt_active()) { - /* - * Keep using @prev_cpu if it's part of a fully idle core. - */ - if (cpumask_test_cpu(prev_cpu, idle_masks.smt) && - test_and_clear_cpu_idle(prev_cpu)) { - cpu = prev_cpu; - goto cpu_found; - } - - /* - * Search for any fully idle core in the same LLC domain. - */ - if (llc_cpus) { - cpu = scx_pick_idle_cpu(llc_cpus, SCX_PICK_IDLE_CORE); - if (cpu >= 0) - goto cpu_found; - } - - /* - * Search for any fully idle core in the same NUMA node. - */ - if (numa_cpus) { - cpu = scx_pick_idle_cpu(numa_cpus, SCX_PICK_IDLE_CORE); - if (cpu >= 0) - goto cpu_found; - } - - /* - * Search for any full idle core usable by the task. - */ - cpu = scx_pick_idle_cpu(p->cpus_ptr, SCX_PICK_IDLE_CORE); - if (cpu >= 0) - goto cpu_found; - } - - /* - * Use @prev_cpu if it's idle. - */ - if (test_and_clear_cpu_idle(prev_cpu)) { - cpu = prev_cpu; - goto cpu_found; - } - - /* - * Search for any idle CPU in the same LLC domain. - */ - if (llc_cpus) { - cpu = scx_pick_idle_cpu(llc_cpus, 0); - if (cpu >= 0) - goto cpu_found; - } - - /* - * Search for any idle CPU in the same NUMA node. - */ - if (numa_cpus) { - cpu = scx_pick_idle_cpu(numa_cpus, 0); - if (cpu >= 0) - goto cpu_found; - } - - /* - * Search for any idle CPU usable by the task. - */ - cpu = scx_pick_idle_cpu(p->cpus_ptr, 0); - if (cpu >= 0) - goto cpu_found; - - rcu_read_unlock(); - return prev_cpu; - -cpu_found: - rcu_read_unlock(); - - *found = true; - return cpu; -} - static int select_task_rq_scx(struct task_struct *p, int prev_cpu, int wake_flags) { + struct scx_sched *sch = scx_task_sched(p); + bool bypassing; + /* * sched_exec() calls with %WF_EXEC when @p is about to exec(2) as it * can be a good migration opportunity with low cache and memory @@ -3645,7 +3326,8 @@ static int select_task_rq_scx(struct task_struct *p, int prev_cpu, int wake_flag if (unlikely(wake_flags & WF_EXEC)) return prev_cpu; - if (SCX_HAS_OP(select_cpu) && !scx_rq_bypassing(task_rq(p))) { + bypassing = scx_bypassing(sch, task_cpu(p)); + if (likely(SCX_HAS_OP(sch, select_cpu)) && !bypassing) { s32 cpu; struct task_struct **ddsp_taskp; @@ -3653,22 +3335,29 @@ static int select_task_rq_scx(struct task_struct *p, int prev_cpu, int wake_flag WARN_ON_ONCE(*ddsp_taskp); *ddsp_taskp = p; - cpu = SCX_CALL_OP_TASK_RET(SCX_KF_ENQUEUE | SCX_KF_SELECT_CPU, - select_cpu, p, prev_cpu, wake_flags); + this_rq()->scx.in_select_cpu = true; + cpu = SCX_CALL_OP_TASK_RET(sch, select_cpu, NULL, p, prev_cpu, wake_flags); + this_rq()->scx.in_select_cpu = false; + p->scx.selected_cpu = cpu; *ddsp_taskp = NULL; - if (ops_cpu_valid(cpu, "from ops.select_cpu()")) + if (ops_cpu_valid(sch, cpu, "from ops.select_cpu()")) return cpu; else return prev_cpu; } else { - bool found; s32 cpu; - cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, &found); - if (found) { - p->scx.slice = SCX_SLICE_DFL; + cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, NULL, 0); + if (cpu >= 0) { + refill_task_slice_dfl(sch, p); p->scx.ddsp_dsq_id = SCX_DSQ_LOCAL; + } else { + cpu = prev_cpu; } + p->scx.selected_cpu = cpu; + + if (bypassing) + __scx_add_event(sch, SCX_EV_BYPASS_DISPATCH, 1); return cpu; } } @@ -3681,8 +3370,13 @@ static void task_woken_scx(struct rq *rq, struct task_struct *p) static void set_cpus_allowed_scx(struct task_struct *p, struct affinity_context *ac) { + struct scx_sched *sch = scx_task_sched(p); + set_cpus_allowed_common(p, ac); + if (task_dead_and_done(p)) + return; + /* * The effective cpumask is stored in @p->cpus_ptr which may temporarily * differ from the configured one in @p->cpus_mask. Always tell the bpf @@ -3691,112 +3385,37 @@ static void set_cpus_allowed_scx(struct task_struct *p, * Fine-grained memory write control is enforced by BPF making the const * designation pointless. Cast it away when calling the operation. */ - if (SCX_HAS_OP(set_cpumask)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, p, - (struct cpumask *)p->cpus_ptr); + if (SCX_HAS_OP(sch, set_cpumask)) + SCX_CALL_OP_TASK(sch, set_cpumask, task_rq(p), p, (struct cpumask *)p->cpus_ptr); } -static void reset_idle_masks(void) -{ - /* - * Consider all online cpus idle. Should converge to the actual state - * quickly. - */ - cpumask_copy(idle_masks.cpu, cpu_online_mask); - cpumask_copy(idle_masks.smt, cpu_online_mask); -} - -static void update_builtin_idle(int cpu, bool idle) -{ - assign_cpu(cpu, idle_masks.cpu, idle); - -#ifdef CONFIG_SCHED_SMT - if (sched_smt_active()) { - const struct cpumask *smt = cpu_smt_mask(cpu); - - if (idle) { - /* - * idle_masks.smt handling is racy but that's fine as - * it's only for optimization and self-correcting. - */ - if (!cpumask_subset(smt, idle_masks.cpu)) - return; - cpumask_or(idle_masks.smt, idle_masks.smt, smt); - } else { - cpumask_andnot(idle_masks.smt, idle_masks.smt, smt); - } - } -#endif -} - -/* - * Update the idle state of a CPU to @idle. - * - * If @do_notify is true, ops.update_idle() is invoked to notify the scx - * scheduler of an actual idle state transition (idle to busy or vice - * versa). If @do_notify is false, only the idle state in the idle masks is - * refreshed without invoking ops.update_idle(). - * - * This distinction is necessary, because an idle CPU can be "reserved" and - * awakened via scx_bpf_pick_idle_cpu() + scx_bpf_kick_cpu(), marking it as - * busy even if no tasks are dispatched. In this case, the CPU may return - * to idle without a true state transition. Refreshing the idle masks - * without invoking ops.update_idle() ensures accurate idle state tracking - * while avoiding unnecessary updates and maintaining balanced state - * transitions. - */ -void __scx_update_idle(struct rq *rq, bool idle, bool do_notify) +static void handle_hotplug(struct rq *rq, bool online) { - int cpu = cpu_of(rq); + struct scx_sched *sch = scx_root; + s32 cpu = cpu_of(rq); - lockdep_assert_rq_held(rq); - - /* - * Trigger ops.update_idle() only when transitioning from a task to - * the idle thread and vice versa. - * - * Idle transitions are indicated by do_notify being set to true, - * managed by put_prev_task_idle()/set_next_task_idle(). - */ - if (SCX_HAS_OP(update_idle) && do_notify && !scx_rq_bypassing(rq)) - SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle); + atomic_long_inc(&scx_hotplug_seq); /* - * Update the idle masks: - * - for real idle transitions (do_notify == true) - * - for idle-to-idle transitions (indicated by the previous task - * being the idle thread, managed by pick_task_idle()) - * - * Skip updating idle masks if the previous task is not the idle - * thread, since set_next_task_idle() has already handled it when - * transitioning from a task to the idle thread (calling this - * function with do_notify == true). - * - * In this way we can avoid updating the idle masks twice, - * unnecessarily. + * scx_root updates are protected by cpus_read_lock() and will stay + * stable here. Note that we can't depend on scx_enabled() test as the + * hotplug ops need to be enabled before __scx_enabled is set. */ - if (static_branch_likely(&scx_builtin_idle_enabled)) - if (do_notify || is_idle_task(rq->curr)) - update_builtin_idle(cpu, idle); -} - -static void handle_hotplug(struct rq *rq, bool online) -{ - int cpu = cpu_of(rq); - - atomic_long_inc(&scx_hotplug_seq); + if (unlikely(!sch)) + return; if (scx_enabled()) - update_selcpu_topology(); + scx_idle_update_selcpu_topology(&sch->ops); - if (online && SCX_HAS_OP(cpu_online)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, cpu); - else if (!online && SCX_HAS_OP(cpu_offline)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, cpu); + if (online && SCX_HAS_OP(sch, cpu_online)) + SCX_CALL_OP(sch, cpu_online, NULL, cpu); + else if (!online && SCX_HAS_OP(sch, cpu_offline)) + SCX_CALL_OP(sch, cpu_offline, NULL, cpu); else - scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, - "cpu %d going %s, exiting scheduler", cpu, - online ? "online" : "offline"); + scx_exit(sch, SCX_EXIT_UNREG_KERN, + SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, + "cpu %d going %s, exiting scheduler", cpu, + online ? "online" : "offline"); } void scx_rq_activate(struct rq *rq) @@ -3819,43 +3438,41 @@ static void rq_offline_scx(struct rq *rq) rq->scx.flags &= ~SCX_RQ_ONLINE; } -#else /* CONFIG_SMP */ - -static bool test_and_clear_cpu_idle(int cpu) { return false; } -static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags) { return -EBUSY; } -static void reset_idle_masks(void) {} - -#endif /* CONFIG_SMP */ - static bool check_rq_for_timeouts(struct rq *rq) { + struct scx_sched *sch; struct task_struct *p; struct rq_flags rf; bool timed_out = false; rq_lock_irqsave(rq, &rf); + sch = rcu_dereference_bh(scx_root); + if (unlikely(!sch)) + goto out_unlock; + list_for_each_entry(p, &rq->scx.runnable_list, scx.runnable_node) { + struct scx_sched *sch = scx_task_sched(p); unsigned long last_runnable = p->scx.runnable_at; if (unlikely(time_after(jiffies, - last_runnable + scx_watchdog_timeout))) { + last_runnable + READ_ONCE(sch->watchdog_timeout)))) { u32 dur_ms = jiffies_to_msecs(jiffies - last_runnable); - scx_ops_error_kind(SCX_EXIT_ERROR_STALL, - "%s[%d] failed to run for %u.%03us", - p->comm, p->pid, - dur_ms / 1000, dur_ms % 1000); + scx_exit(sch, SCX_EXIT_ERROR_STALL, 0, + "%s[%d] failed to run for %u.%03us", + p->comm, p->pid, dur_ms / 1000, dur_ms % 1000); timed_out = true; break; } } +out_unlock: rq_unlock_irqrestore(rq, &rf); - return timed_out; } static void scx_watchdog_workfn(struct work_struct *work) { + unsigned long intv; int cpu; WRITE_ONCE(scx_watchdog_timestamp, jiffies); @@ -3866,25 +3483,32 @@ static void scx_watchdog_workfn(struct work_struct *work) cond_resched(); } - queue_delayed_work(system_unbound_wq, to_delayed_work(work), - scx_watchdog_timeout / 2); + + intv = READ_ONCE(scx_watchdog_interval); + if (intv < ULONG_MAX) + queue_delayed_work(system_dfl_wq, to_delayed_work(work), intv); } void scx_tick(struct rq *rq) { + struct scx_sched *root; unsigned long last_check; if (!scx_enabled()) return; + root = rcu_dereference_bh(scx_root); + if (unlikely(!root)) + return; + last_check = READ_ONCE(scx_watchdog_timestamp); if (unlikely(time_after(jiffies, - last_check + READ_ONCE(scx_watchdog_timeout)))) { + last_check + READ_ONCE(root->watchdog_timeout)))) { u32 dur_ms = jiffies_to_msecs(jiffies - last_check); - scx_ops_error_kind(SCX_EXIT_ERROR_STALL, - "watchdog failed to check in for %u.%03us", - dur_ms / 1000, dur_ms % 1000); + scx_exit(root, SCX_EXIT_ERROR_STALL, 0, + "watchdog failed to check in for %u.%03us", + dur_ms / 1000, dur_ms % 1000); } update_other_load_avgs(rq); @@ -3892,17 +3516,19 @@ void scx_tick(struct rq *rq) static void task_tick_scx(struct rq *rq, struct task_struct *curr, int queued) { + struct scx_sched *sch = scx_task_sched(curr); + update_curr_scx(rq); /* * While disabling, always resched and refresh core-sched timestamp as * we can't trust the slice management or ops.core_sched_before(). */ - if (scx_rq_bypassing(rq)) { + if (scx_bypassing(sch, cpu_of(rq))) { curr->scx.slice = 0; touch_core_sched(rq, curr); - } else if (SCX_HAS_OP(tick)) { - SCX_CALL_OP_TASK(SCX_KF_REST, tick, curr); + } else if (SCX_HAS_OP(sch, tick)) { + SCX_CALL_OP_TASK(sch, tick, rq, curr); } if (!curr->scx.slice) @@ -3931,65 +3557,33 @@ static struct cgroup *tg_cgrp(struct task_group *tg) #endif /* CONFIG_EXT_GROUP_SCHED */ -static enum scx_task_state scx_get_task_state(const struct task_struct *p) -{ - return (p->scx.flags & SCX_TASK_STATE_MASK) >> SCX_TASK_STATE_SHIFT; -} - -static void scx_set_task_state(struct task_struct *p, enum scx_task_state state) -{ - enum scx_task_state prev_state = scx_get_task_state(p); - bool warn = false; - - BUILD_BUG_ON(SCX_TASK_NR_STATES > (1 << SCX_TASK_STATE_BITS)); - - switch (state) { - case SCX_TASK_NONE: - break; - case SCX_TASK_INIT: - warn = prev_state != SCX_TASK_NONE; - break; - case SCX_TASK_READY: - warn = prev_state == SCX_TASK_NONE; - break; - case SCX_TASK_ENABLED: - warn = prev_state != SCX_TASK_READY; - break; - default: - warn = true; - return; - } - - WARN_ONCE(warn, "sched_ext: Invalid task state transition %d -> %d for %s[%d]", - prev_state, state, p->comm, p->pid); - - p->scx.flags &= ~SCX_TASK_STATE_MASK; - p->scx.flags |= state << SCX_TASK_STATE_SHIFT; -} - -static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool fork) +static int __scx_init_task(struct scx_sched *sch, struct task_struct *p, bool fork) { int ret; p->scx.disallow = false; - if (SCX_HAS_OP(init_task)) { + if (SCX_HAS_OP(sch, init_task)) { struct scx_init_task_args args = { - SCX_INIT_TASK_ARGS_CGROUP(tg) + SCX_INIT_TASK_ARGS_CGROUP(task_group(p)) .fork = fork, }; - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init_task, p, &args); + ret = SCX_CALL_OP_RET(sch, init_task, NULL, p, &args); if (unlikely(ret)) { - ret = ops_sanitize_err("init_task", ret); + ret = ops_sanitize_err(sch, "init_task", ret); return ret; } } - scx_set_task_state(p, SCX_TASK_INIT); - if (p->scx.disallow) { - if (!fork) { + if (unlikely(scx_parent(sch))) { + scx_error(sch, "non-root ops.init_task() set task->scx.disallow for %s[%d]", + p->comm, p->pid); + } else if (unlikely(fork)) { + scx_error(sch, "ops.init_task() set task->scx.disallow for %s[%d] during fork", + p->comm, p->pid); + } else { struct rq *rq; struct rq_flags rf; @@ -4008,21 +3602,25 @@ static int scx_ops_init_task(struct task_struct *p, struct task_group *tg, bool } task_rq_unlock(rq, p, &rf); - } else if (p->policy == SCHED_EXT) { - scx_ops_error("ops.init_task() set task->scx.disallow for %s[%d] during fork", - p->comm, p->pid); } } - p->scx.flags |= SCX_TASK_RESET_RUNNABLE_AT; return 0; } -static void scx_ops_enable_task(struct task_struct *p) +static void __scx_enable_task(struct scx_sched *sch, struct task_struct *p) { + struct rq *rq = task_rq(p); u32 weight; - lockdep_assert_rq_held(task_rq(p)); + lockdep_assert_rq_held(rq); + + /* + * Verify the task is not in BPF scheduler's custody. If flag + * transitions are consistent, the flag should always be clear + * here. + */ + WARN_ON_ONCE(p->scx.flags & SCX_TASK_IN_CUSTODY); /* * Set the weight before calling ops.enable() so that the scheduler @@ -4035,30 +3633,48 @@ static void scx_ops_enable_task(struct task_struct *p) p->scx.weight = sched_weight_to_cgroup(weight); - if (SCX_HAS_OP(enable)) - SCX_CALL_OP_TASK(SCX_KF_REST, enable, p); - scx_set_task_state(p, SCX_TASK_ENABLED); + if (SCX_HAS_OP(sch, enable)) + SCX_CALL_OP_TASK(sch, enable, rq, p); - if (SCX_HAS_OP(set_weight)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, p, p->scx.weight); + if (SCX_HAS_OP(sch, set_weight)) + SCX_CALL_OP_TASK(sch, set_weight, rq, p, p->scx.weight); } -static void scx_ops_disable_task(struct task_struct *p) +static void scx_enable_task(struct scx_sched *sch, struct task_struct *p) { - lockdep_assert_rq_held(task_rq(p)); + __scx_enable_task(sch, p); + scx_set_task_state(p, SCX_TASK_ENABLED); +} + +static void scx_disable_task(struct scx_sched *sch, struct task_struct *p) +{ + struct rq *rq = task_rq(p); + + lockdep_assert_rq_held(rq); WARN_ON_ONCE(scx_get_task_state(p) != SCX_TASK_ENABLED); - if (SCX_HAS_OP(disable)) - SCX_CALL_OP_TASK(SCX_KF_REST, disable, p); + clear_direct_dispatch(p); + + if (SCX_HAS_OP(sch, disable)) + SCX_CALL_OP_TASK(sch, disable, rq, p); scx_set_task_state(p, SCX_TASK_READY); + + /* + * Verify the task is not in BPF scheduler's custody. If flag + * transitions are consistent, the flag should always be clear + * here. + */ + WARN_ON_ONCE(p->scx.flags & SCX_TASK_IN_CUSTODY); } -static void scx_ops_exit_task(struct task_struct *p) +static void __scx_disable_and_exit_task(struct scx_sched *sch, + struct task_struct *p) { struct scx_exit_task_args args = { .cancelled = false, }; + lockdep_assert_held(&p->pi_lock); lockdep_assert_rq_held(task_rq(p)); switch (scx_get_task_state(p)) { @@ -4070,15 +3686,52 @@ static void scx_ops_exit_task(struct task_struct *p) case SCX_TASK_READY: break; case SCX_TASK_ENABLED: - scx_ops_disable_task(p); + scx_disable_task(sch, p); break; default: WARN_ON_ONCE(true); return; } - if (SCX_HAS_OP(exit_task)) - SCX_CALL_OP_TASK(SCX_KF_REST, exit_task, p, &args); + if (SCX_HAS_OP(sch, exit_task)) + SCX_CALL_OP_TASK(sch, exit_task, task_rq(p), p, &args); +} + +/* + * Undo a completed __scx_init_task(sch, p, false) when scx_enable_task() never + * ran. The task state has not been transitioned, so this mirrors the + * SCX_TASK_INIT branch in __scx_disable_and_exit_task(). + */ +static void scx_sub_init_cancel_task(struct scx_sched *sch, struct task_struct *p) +{ + struct scx_exit_task_args args = { .cancelled = true }; + + lockdep_assert_held(&p->pi_lock); + lockdep_assert_rq_held(task_rq(p)); + + if (SCX_HAS_OP(sch, exit_task)) + SCX_CALL_OP_TASK(sch, exit_task, task_rq(p), p, &args); +} + +static void scx_disable_and_exit_task(struct scx_sched *sch, + struct task_struct *p) +{ + __scx_disable_and_exit_task(sch, p); + + /* + * If set, @p exited between __scx_init_task() and scx_enable_task() in + * scx_sub_enable() and is initialized for both the associated sched and + * its parent. Exit for the child too - scx_enable_task() never ran for + * it, so undo only init_task. The flag is only set on the sub-enable + * path, so it's always clear when @p arrives here in %SCX_TASK_NONE. + */ + if (p->scx.flags & SCX_TASK_SUB_INIT) { + if (!WARN_ON_ONCE(!scx_enabling_sub_sched)) + scx_sub_init_cancel_task(scx_enabling_sub_sched, p); + p->scx.flags &= ~SCX_TASK_SUB_INIT; + } + + scx_set_task_sched(p, NULL); scx_set_task_state(p, SCX_TASK_NONE); } @@ -4106,19 +3759,34 @@ void scx_pre_fork(struct task_struct *p) percpu_down_read(&scx_fork_rwsem); } -int scx_fork(struct task_struct *p) +int scx_fork(struct task_struct *p, struct kernel_clone_args *kargs) { + s32 ret; + percpu_rwsem_assert_held(&scx_fork_rwsem); - if (scx_ops_init_task_enabled) - return scx_ops_init_task(p, task_group(p), true); - else - return 0; + if (scx_init_task_enabled) { +#ifdef CONFIG_EXT_SUB_SCHED + struct scx_sched *sch = kargs->cset->dfl_cgrp->scx_sched; +#else + struct scx_sched *sch = scx_root; +#endif + scx_set_task_state(p, SCX_TASK_INIT_BEGIN); + ret = __scx_init_task(sch, p, true); + if (unlikely(ret)) { + scx_set_task_state(p, SCX_TASK_NONE); + return ret; + } + scx_set_task_state(p, SCX_TASK_INIT); + scx_set_task_sched(p, sch); + } + + return 0; } void scx_post_fork(struct task_struct *p) { - if (scx_ops_init_task_enabled) { + if (scx_init_task_enabled) { scx_set_task_state(p, SCX_TASK_READY); /* @@ -4131,14 +3799,14 @@ void scx_post_fork(struct task_struct *p) struct rq *rq; rq = task_rq_lock(p, &rf); - scx_ops_enable_task(p); + scx_enable_task(scx_task_sched(p), p); task_rq_unlock(rq, p, &rf); } } - spin_lock_irq(&scx_tasks_lock); + raw_spin_lock_irq(&scx_tasks_lock); list_add_tail(&p->scx.tasks_node, &scx_tasks); - spin_unlock_irq(&scx_tasks_lock); + raw_spin_unlock_irq(&scx_tasks_lock); percpu_up_read(&scx_fork_rwsem); } @@ -4151,31 +3819,77 @@ void scx_cancel_fork(struct task_struct *p) rq = task_rq_lock(p, &rf); WARN_ON_ONCE(scx_get_task_state(p) >= SCX_TASK_READY); - scx_ops_exit_task(p); + scx_disable_and_exit_task(scx_task_sched(p), p); task_rq_unlock(rq, p, &rf); } percpu_up_read(&scx_fork_rwsem); } -void sched_ext_free(struct task_struct *p) +/** + * task_dead_and_done - Is a task dead and done running? + * @p: target task + * + * Once sched_ext_dead() removes the dead task from scx_tasks and exits it, the + * task no longer exists from SCX's POV. However, certain sched_class ops may be + * invoked on these dead tasks leading to failures - e.g. sched_setscheduler() + * may try to switch a task which finished sched_ext_dead() back into SCX + * triggering invalid SCX task state transitions and worse. + * + * Once a task has finished the final switch, sched_ext_dead() is the only thing + * that needs to happen on the task. Use this test to short-circuit sched_class + * operations which may be called on dead tasks. + */ +static bool task_dead_and_done(struct task_struct *p) +{ + struct rq *rq = task_rq(p); + + lockdep_assert_rq_held(rq); + + /* + * In do_task_dead(), a dying task sets %TASK_DEAD with preemption + * disabled and __schedule(). If @p has %TASK_DEAD set and off CPU, @p + * won't ever run again. + */ + return unlikely(READ_ONCE(p->__state) == TASK_DEAD) && + !task_on_cpu(rq, p); +} + +void sched_ext_dead(struct task_struct *p) { unsigned long flags; - spin_lock_irqsave(&scx_tasks_lock, flags); + /* + * By the time control reaches here, @p has %TASK_DEAD set, switched out + * for the last time and then dropped the rq lock - task_dead_and_done() + * should be returning %true nullifying the straggling sched_class ops. + * Remove from scx_tasks and exit @p. + */ + raw_spin_lock_irqsave(&scx_tasks_lock, flags); list_del_init(&p->scx.tasks_node); - spin_unlock_irqrestore(&scx_tasks_lock, flags); + raw_spin_unlock_irqrestore(&scx_tasks_lock, flags); /* - * @p is off scx_tasks and wholly ours. scx_ops_enable()'s READY -> + * @p is off scx_tasks and wholly ours. scx_root_enable()'s READY -> * ENABLED transitions can't race us. Disable ops for @p. + * + * %SCX_TASK_DEAD synchronizes against cgroup task iteration - see + * scx_task_iter_next_locked(). NONE tasks need no marking: cgroup + * iteration is only used from sub-sched paths, which require root + * enabled. Root enable transitions every live task to at least READY. + * + * %INIT_BEGIN means ops.init_task() is running for @p. Don't call + * into ops; transition to %DEAD so the post-init recheck unwinds + * via scx_sub_init_cancel_task(). */ if (scx_get_task_state(p) != SCX_TASK_NONE) { struct rq_flags rf; struct rq *rq; rq = task_rq_lock(p, &rf); - scx_ops_exit_task(p); + if (scx_get_task_state(p) != SCX_TASK_INIT_BEGIN) + scx_disable_and_exit_task(scx_task_sched(p), p); + scx_set_task_state(p, SCX_TASK_DEAD); task_rq_unlock(rq, p, &rf); } } @@ -4183,36 +3897,57 @@ void sched_ext_free(struct task_struct *p) static void reweight_task_scx(struct rq *rq, struct task_struct *p, const struct load_weight *lw) { + struct scx_sched *sch = scx_task_sched(p); + lockdep_assert_rq_held(task_rq(p)); + if (task_dead_and_done(p)) + return; + p->scx.weight = sched_weight_to_cgroup(scale_load_down(lw->weight)); - if (SCX_HAS_OP(set_weight)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_weight, p, p->scx.weight); + if (SCX_HAS_OP(sch, set_weight)) + SCX_CALL_OP_TASK(sch, set_weight, rq, p, p->scx.weight); } -static void prio_changed_scx(struct rq *rq, struct task_struct *p, int oldprio) +static void prio_changed_scx(struct rq *rq, struct task_struct *p, u64 oldprio) { } static void switching_to_scx(struct rq *rq, struct task_struct *p) { - scx_ops_enable_task(p); + struct scx_sched *sch = scx_task_sched(p); + + if (task_dead_and_done(p)) + return; + + scx_enable_task(sch, p); /* * set_cpus_allowed_scx() is not called while @p is associated with a * different scheduler class. Keep the BPF scheduler up-to-date. */ - if (SCX_HAS_OP(set_cpumask)) - SCX_CALL_OP_TASK(SCX_KF_REST, set_cpumask, p, - (struct cpumask *)p->cpus_ptr); + if (SCX_HAS_OP(sch, set_cpumask)) + SCX_CALL_OP_TASK(sch, set_cpumask, rq, p, (struct cpumask *)p->cpus_ptr); } static void switched_from_scx(struct rq *rq, struct task_struct *p) { - scx_ops_disable_task(p); + if (task_dead_and_done(p)) + return; + + /* + * %NONE means SCX is no longer tracking @p at the task level (e.g. + * scx_fail_parent() handed @p back to the parent at NONE pending the + * parent's own teardown). There is nothing to disable; calling + * scx_disable_task() would WARN on the non-%ENABLED state and trigger a + * NONE -> READY validation failure. + */ + if (scx_get_task_state(p) == SCX_TASK_NONE) + return; + + scx_disable_task(scx_task_sched(p), p); } -static void wakeup_preempt_scx(struct rq *rq, struct task_struct *p,int wake_flags) {} static void switched_to_scx(struct rq *rq, struct task_struct *p) {} int scx_check_setscheduler(struct task_struct *p, int policy) @@ -4227,17 +3962,327 @@ int scx_check_setscheduler(struct task_struct *p, int policy) return 0; } +static void process_ddsp_deferred_locals(struct rq *rq) +{ + struct task_struct *p; + + lockdep_assert_rq_held(rq); + + /* + * Now that @rq can be unlocked, execute the deferred enqueueing of + * tasks directly dispatched to the local DSQs of other CPUs. See + * direct_dispatch(). Keep popping from the head instead of using + * list_for_each_entry_safe() as dispatch_local_dsq() may unlock @rq + * temporarily. + */ + while ((p = list_first_entry_or_null(&rq->scx.ddsp_deferred_locals, + struct task_struct, scx.dsq_list.node))) { + struct scx_sched *sch = scx_task_sched(p); + struct scx_dispatch_q *dsq; + u64 dsq_id = p->scx.ddsp_dsq_id; + u64 enq_flags = p->scx.ddsp_enq_flags; + + list_del_init(&p->scx.dsq_list.node); + clear_direct_dispatch(p); + + dsq = find_dsq_for_dispatch(sch, rq, dsq_id, task_cpu(p)); + if (!WARN_ON_ONCE(dsq->id != SCX_DSQ_LOCAL)) + dispatch_to_local_dsq(sch, rq, dsq, p, enq_flags); + } +} + +/* + * Determine whether @p should be reenqueued from a local DSQ. + * + * @reenq_flags is mutable and accumulates state across the DSQ walk: + * + * - %SCX_REENQ_TSR_NOT_FIRST: Set after the first task is visited. "First" + * tracks position in the DSQ list, not among IMMED tasks. A non-IMMED task at + * the head consumes the first slot. + * + * - %SCX_REENQ_TSR_RQ_OPEN: Set by reenq_local() before the walk if + * rq_is_open() is true. + * + * An IMMED task is kept (returns %false) only if it's the first task in the DSQ + * AND the current task is done — i.e. it will execute immediately. All other + * IMMED tasks are reenqueued. This means if a non-IMMED task sits at the head, + * every IMMED task behind it gets reenqueued. + * + * Reenqueued tasks go through ops.enqueue() with %SCX_ENQ_REENQ | + * %SCX_TASK_REENQ_IMMED. If the BPF scheduler dispatches back to the same local + * DSQ with %SCX_ENQ_IMMED while the CPU is still unavailable, this triggers + * another reenq cycle. Repetitions are bounded by %SCX_REENQ_LOCAL_MAX_REPEAT + * in process_deferred_reenq_locals(). + */ +static bool local_task_should_reenq(struct task_struct *p, u64 *reenq_flags, u32 *reason) +{ + bool first; + + first = !(*reenq_flags & SCX_REENQ_TSR_NOT_FIRST); + *reenq_flags |= SCX_REENQ_TSR_NOT_FIRST; + + *reason = SCX_TASK_REENQ_KFUNC; + + if ((p->scx.flags & SCX_TASK_IMMED) && + (!first || !(*reenq_flags & SCX_REENQ_TSR_RQ_OPEN))) { + __scx_add_event(scx_task_sched(p), SCX_EV_REENQ_IMMED, 1); + *reason = SCX_TASK_REENQ_IMMED; + return true; + } + + return *reenq_flags & SCX_REENQ_ANY; +} + +static u32 reenq_local(struct scx_sched *sch, struct rq *rq, u64 reenq_flags) +{ + LIST_HEAD(tasks); + u32 nr_enqueued = 0; + struct task_struct *p, *n; + + lockdep_assert_rq_held(rq); + + if (WARN_ON_ONCE(reenq_flags & __SCX_REENQ_TSR_MASK)) + reenq_flags &= ~__SCX_REENQ_TSR_MASK; + if (rq_is_open(rq, 0)) + reenq_flags |= SCX_REENQ_TSR_RQ_OPEN; + + /* + * The BPF scheduler may choose to dispatch tasks back to + * @rq->scx.local_dsq. Move all candidate tasks off to a private list + * first to avoid processing the same tasks repeatedly. + */ + list_for_each_entry_safe(p, n, &rq->scx.local_dsq.list, + scx.dsq_list.node) { + struct scx_sched *task_sch = scx_task_sched(p); + u32 reason; + + /* + * If @p is being migrated, @p's current CPU may not agree with + * its allowed CPUs and the migration_cpu_stop is about to + * deactivate and re-activate @p anyway. Skip re-enqueueing. + * + * While racing sched property changes may also dequeue and + * re-enqueue a migrating task while its current CPU and allowed + * CPUs disagree, they use %ENQUEUE_RESTORE which is bypassed to + * the current local DSQ for running tasks and thus are not + * visible to the BPF scheduler. + */ + if (p->migration_pending) + continue; + + if (!scx_is_descendant(task_sch, sch)) + continue; + + if (!local_task_should_reenq(p, &reenq_flags, &reason)) + continue; + + dispatch_dequeue(rq, p); + + if (WARN_ON_ONCE(p->scx.flags & SCX_TASK_REENQ_REASON_MASK)) + p->scx.flags &= ~SCX_TASK_REENQ_REASON_MASK; + p->scx.flags |= reason; + + list_add_tail(&p->scx.dsq_list.node, &tasks); + } + + list_for_each_entry_safe(p, n, &tasks, scx.dsq_list.node) { + list_del_init(&p->scx.dsq_list.node); + + do_enqueue_task(rq, p, SCX_ENQ_REENQ, -1); + + p->scx.flags &= ~SCX_TASK_REENQ_REASON_MASK; + nr_enqueued++; + } + + return nr_enqueued; +} + +static void process_deferred_reenq_locals(struct rq *rq) +{ + u64 seq = ++rq->scx.deferred_reenq_locals_seq; + + lockdep_assert_rq_held(rq); + + while (true) { + struct scx_sched *sch; + u64 reenq_flags; + bool skip = false; + + scoped_guard (raw_spinlock, &rq->scx.deferred_reenq_lock) { + struct scx_deferred_reenq_local *drl = + list_first_entry_or_null(&rq->scx.deferred_reenq_locals, + struct scx_deferred_reenq_local, + node); + struct scx_sched_pcpu *sch_pcpu; + + if (!drl) + return; + + sch_pcpu = container_of(drl, struct scx_sched_pcpu, + deferred_reenq_local); + sch = sch_pcpu->sch; + + reenq_flags = drl->flags; + WRITE_ONCE(drl->flags, 0); + list_del_init(&drl->node); + + if (likely(drl->seq != seq)) { + drl->seq = seq; + drl->cnt = 0; + } else { + if (unlikely(++drl->cnt > SCX_REENQ_LOCAL_MAX_REPEAT)) { + scx_error(sch, "SCX_ENQ_REENQ on SCX_DSQ_LOCAL repeated %u times", + drl->cnt); + skip = true; + } + + __scx_add_event(sch, SCX_EV_REENQ_LOCAL_REPEAT, 1); + } + } + + if (!skip) { + /* see schedule_dsq_reenq() */ + smp_mb(); + + reenq_local(sch, rq, reenq_flags); + } + } +} + +static bool user_task_should_reenq(struct task_struct *p, u64 reenq_flags, u32 *reason) +{ + *reason = SCX_TASK_REENQ_KFUNC; + return reenq_flags & SCX_REENQ_ANY; +} + +static void reenq_user(struct rq *rq, struct scx_dispatch_q *dsq, u64 reenq_flags) +{ + struct rq *locked_rq = rq; + struct scx_sched *sch = dsq->sched; + struct scx_dsq_list_node cursor = INIT_DSQ_LIST_CURSOR(cursor, dsq, 0); + struct task_struct *p; + s32 nr_enqueued = 0; + + lockdep_assert_rq_held(rq); + + raw_spin_lock(&dsq->lock); + + while (likely(!READ_ONCE(sch->bypass_depth))) { + struct rq *task_rq; + u32 reason; + + p = nldsq_cursor_next_task(&cursor, dsq); + if (!p) + break; + + if (!user_task_should_reenq(p, reenq_flags, &reason)) + continue; + + task_rq = task_rq(p); + + if (locked_rq != task_rq) { + if (locked_rq) + raw_spin_rq_unlock(locked_rq); + if (unlikely(!raw_spin_rq_trylock(task_rq))) { + raw_spin_unlock(&dsq->lock); + raw_spin_rq_lock(task_rq); + raw_spin_lock(&dsq->lock); + } + locked_rq = task_rq; + + /* did we lose @p while switching locks? */ + if (nldsq_cursor_lost_task(&cursor, task_rq, dsq, p)) + continue; + } + + /* @p is on @dsq, its rq and @dsq are locked */ + dispatch_dequeue_locked(p, dsq); + raw_spin_unlock(&dsq->lock); + + if (WARN_ON_ONCE(p->scx.flags & SCX_TASK_REENQ_REASON_MASK)) + p->scx.flags &= ~SCX_TASK_REENQ_REASON_MASK; + p->scx.flags |= reason; + + do_enqueue_task(task_rq, p, SCX_ENQ_REENQ, -1); + + p->scx.flags &= ~SCX_TASK_REENQ_REASON_MASK; + + if (!(++nr_enqueued % SCX_TASK_ITER_BATCH)) { + raw_spin_rq_unlock(locked_rq); + locked_rq = NULL; + cpu_relax(); + } + + raw_spin_lock(&dsq->lock); + } + + list_del_init(&cursor.node); + raw_spin_unlock(&dsq->lock); + + if (locked_rq != rq) { + if (locked_rq) + raw_spin_rq_unlock(locked_rq); + raw_spin_rq_lock(rq); + } +} + +static void process_deferred_reenq_users(struct rq *rq) +{ + lockdep_assert_rq_held(rq); + + while (true) { + struct scx_dispatch_q *dsq; + u64 reenq_flags; + + scoped_guard (raw_spinlock, &rq->scx.deferred_reenq_lock) { + struct scx_deferred_reenq_user *dru = + list_first_entry_or_null(&rq->scx.deferred_reenq_users, + struct scx_deferred_reenq_user, + node); + struct scx_dsq_pcpu *dsq_pcpu; + + if (!dru) + return; + + dsq_pcpu = container_of(dru, struct scx_dsq_pcpu, + deferred_reenq_user); + dsq = dsq_pcpu->dsq; + reenq_flags = dru->flags; + WRITE_ONCE(dru->flags, 0); + list_del_init(&dru->node); + } + + /* see schedule_dsq_reenq() */ + smp_mb(); + + BUG_ON(dsq->id & SCX_DSQ_FLAG_BUILTIN); + reenq_user(rq, dsq, reenq_flags); + } +} + +static void run_deferred(struct rq *rq) +{ + process_ddsp_deferred_locals(rq); + + if (!list_empty(&rq->scx.deferred_reenq_locals)) + process_deferred_reenq_locals(rq); + + if (!list_empty(&rq->scx.deferred_reenq_users)) + process_deferred_reenq_users(rq); +} + #ifdef CONFIG_NO_HZ_FULL bool scx_can_stop_tick(struct rq *rq) { struct task_struct *p = rq->curr; - - if (scx_rq_bypassing(rq)) - return false; + struct scx_sched *sch = scx_task_sched(p); if (p->sched_class != &ext_sched_class) return true; + if (scx_bypassing(sch, cpu_of(rq))) + return false; + /* * @rq can dispatch from different DSQs, so we can't tell whether it * needs the tick or not by looking at nr_running. Allow stopping ticks @@ -4249,90 +4294,65 @@ bool scx_can_stop_tick(struct rq *rq) #ifdef CONFIG_EXT_GROUP_SCHED -DEFINE_STATIC_PERCPU_RWSEM(scx_cgroup_rwsem); +DEFINE_STATIC_PERCPU_RWSEM(scx_cgroup_ops_rwsem); static bool scx_cgroup_enabled; -static bool cgroup_warned_missing_weight; -static bool cgroup_warned_missing_idle; - -static void scx_cgroup_warn_missing_weight(struct task_group *tg) -{ - if (scx_ops_enable_state() == SCX_OPS_DISABLED || - cgroup_warned_missing_weight) - return; - if ((scx_ops.flags & SCX_OPS_HAS_CGROUP_WEIGHT) || !tg->css.parent) - return; - - pr_warn("sched_ext: \"%s\" does not implement cgroup cpu.weight\n", - scx_ops.name); - cgroup_warned_missing_weight = true; -} - -static void scx_cgroup_warn_missing_idle(struct task_group *tg) +void scx_tg_init(struct task_group *tg) { - if (!scx_cgroup_enabled || cgroup_warned_missing_idle) - return; - - if (!tg->idle) - return; - - pr_warn("sched_ext: \"%s\" does not implement cgroup cpu.idle\n", - scx_ops.name); - cgroup_warned_missing_idle = true; + tg->scx.weight = CGROUP_WEIGHT_DFL; + tg->scx.bw_period_us = default_bw_period_us(); + tg->scx.bw_quota_us = RUNTIME_INF; + tg->scx.idle = false; } int scx_tg_online(struct task_group *tg) { + struct scx_sched *sch = scx_root; int ret = 0; - WARN_ON_ONCE(tg->scx_flags & (SCX_TG_ONLINE | SCX_TG_INITED)); - - percpu_down_read(&scx_cgroup_rwsem); - - scx_cgroup_warn_missing_weight(tg); + WARN_ON_ONCE(tg->scx.flags & (SCX_TG_ONLINE | SCX_TG_INITED)); if (scx_cgroup_enabled) { - if (SCX_HAS_OP(cgroup_init)) { + if (SCX_HAS_OP(sch, cgroup_init)) { struct scx_cgroup_init_args args = - { .weight = tg->scx_weight }; + { .weight = tg->scx.weight, + .bw_period_us = tg->scx.bw_period_us, + .bw_quota_us = tg->scx.bw_quota_us, + .bw_burst_us = tg->scx.bw_burst_us }; - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, - tg->css.cgroup, &args); + ret = SCX_CALL_OP_RET(sch, cgroup_init, + NULL, tg->css.cgroup, &args); if (ret) - ret = ops_sanitize_err("cgroup_init", ret); + ret = ops_sanitize_err(sch, "cgroup_init", ret); } if (ret == 0) - tg->scx_flags |= SCX_TG_ONLINE | SCX_TG_INITED; + tg->scx.flags |= SCX_TG_ONLINE | SCX_TG_INITED; } else { - tg->scx_flags |= SCX_TG_ONLINE; + tg->scx.flags |= SCX_TG_ONLINE; } - percpu_up_read(&scx_cgroup_rwsem); return ret; } void scx_tg_offline(struct task_group *tg) { - WARN_ON_ONCE(!(tg->scx_flags & SCX_TG_ONLINE)); - - percpu_down_read(&scx_cgroup_rwsem); + struct scx_sched *sch = scx_root; - if (SCX_HAS_OP(cgroup_exit) && (tg->scx_flags & SCX_TG_INITED)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, tg->css.cgroup); - tg->scx_flags &= ~(SCX_TG_ONLINE | SCX_TG_INITED); + WARN_ON_ONCE(!(tg->scx.flags & SCX_TG_ONLINE)); - percpu_up_read(&scx_cgroup_rwsem); + if (scx_cgroup_enabled && SCX_HAS_OP(sch, cgroup_exit) && + (tg->scx.flags & SCX_TG_INITED)) + SCX_CALL_OP(sch, cgroup_exit, NULL, tg->css.cgroup); + tg->scx.flags &= ~(SCX_TG_ONLINE | SCX_TG_INITED); } int scx_cgroup_can_attach(struct cgroup_taskset *tset) { + struct scx_sched *sch = scx_root; struct cgroup_subsys_state *css; struct task_struct *p; int ret; - /* released in scx_finish/cancel_attach() */ - percpu_down_read(&scx_cgroup_rwsem); - if (!scx_cgroup_enabled) return 0; @@ -4350,8 +4370,8 @@ int scx_cgroup_can_attach(struct cgroup_taskset *tset) if (from == to) continue; - if (SCX_HAS_OP(cgroup_prep_move)) { - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_prep_move, + if (SCX_HAS_OP(sch, cgroup_prep_move)) { + ret = SCX_CALL_OP_RET(sch, cgroup_prep_move, NULL, p, from, css->cgroup); if (ret) goto err; @@ -4364,18 +4384,20 @@ int scx_cgroup_can_attach(struct cgroup_taskset *tset) err: cgroup_taskset_for_each(p, css, tset) { - if (SCX_HAS_OP(cgroup_cancel_move) && p->scx.cgrp_moving_from) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, p, - p->scx.cgrp_moving_from, css->cgroup); + if (SCX_HAS_OP(sch, cgroup_cancel_move) && + p->scx.cgrp_moving_from) + SCX_CALL_OP(sch, cgroup_cancel_move, NULL, + p, p->scx.cgrp_moving_from, css->cgroup); p->scx.cgrp_moving_from = NULL; } - percpu_up_read(&scx_cgroup_rwsem); - return ops_sanitize_err("cgroup_prep_move", ret); + return ops_sanitize_err(sch, "cgroup_prep_move", ret); } void scx_cgroup_move_task(struct task_struct *p) { + struct scx_sched *sch = scx_root; + if (!scx_cgroup_enabled) return; @@ -4383,80 +4405,137 @@ void scx_cgroup_move_task(struct task_struct *p) * @p must have ops.cgroup_prep_move() called on it and thus * cgrp_moving_from set. */ - if (SCX_HAS_OP(cgroup_move) && !WARN_ON_ONCE(!p->scx.cgrp_moving_from)) - SCX_CALL_OP_TASK(SCX_KF_UNLOCKED, cgroup_move, p, - p->scx.cgrp_moving_from, tg_cgrp(task_group(p))); + if (SCX_HAS_OP(sch, cgroup_move) && + !WARN_ON_ONCE(!p->scx.cgrp_moving_from)) + SCX_CALL_OP_TASK(sch, cgroup_move, task_rq(p), + p, p->scx.cgrp_moving_from, + tg_cgrp(task_group(p))); p->scx.cgrp_moving_from = NULL; } -void scx_cgroup_finish_attach(void) -{ - percpu_up_read(&scx_cgroup_rwsem); -} - void scx_cgroup_cancel_attach(struct cgroup_taskset *tset) { + struct scx_sched *sch = scx_root; struct cgroup_subsys_state *css; struct task_struct *p; if (!scx_cgroup_enabled) - goto out_unlock; + return; cgroup_taskset_for_each(p, css, tset) { - if (SCX_HAS_OP(cgroup_cancel_move) && p->scx.cgrp_moving_from) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_cancel_move, p, - p->scx.cgrp_moving_from, css->cgroup); + if (SCX_HAS_OP(sch, cgroup_cancel_move) && + p->scx.cgrp_moving_from) + SCX_CALL_OP(sch, cgroup_cancel_move, NULL, + p, p->scx.cgrp_moving_from, css->cgroup); p->scx.cgrp_moving_from = NULL; } -out_unlock: - percpu_up_read(&scx_cgroup_rwsem); } void scx_group_set_weight(struct task_group *tg, unsigned long weight) { - percpu_down_read(&scx_cgroup_rwsem); + struct scx_sched *sch; - if (scx_cgroup_enabled && tg->scx_weight != weight) { - if (SCX_HAS_OP(cgroup_set_weight)) - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_set_weight, - tg_cgrp(tg), weight); - tg->scx_weight = weight; - } + percpu_down_read(&scx_cgroup_ops_rwsem); + sch = scx_root; + + if (scx_cgroup_enabled && SCX_HAS_OP(sch, cgroup_set_weight) && + tg->scx.weight != weight) + SCX_CALL_OP(sch, cgroup_set_weight, NULL, tg_cgrp(tg), weight); - percpu_up_read(&scx_cgroup_rwsem); + tg->scx.weight = weight; + + percpu_up_read(&scx_cgroup_ops_rwsem); } void scx_group_set_idle(struct task_group *tg, bool idle) { - percpu_down_read(&scx_cgroup_rwsem); - scx_cgroup_warn_missing_idle(tg); - percpu_up_read(&scx_cgroup_rwsem); + struct scx_sched *sch; + + percpu_down_read(&scx_cgroup_ops_rwsem); + sch = scx_root; + + if (scx_cgroup_enabled && SCX_HAS_OP(sch, cgroup_set_idle)) + SCX_CALL_OP(sch, cgroup_set_idle, NULL, tg_cgrp(tg), idle); + + /* Update the task group's idle state */ + tg->scx.idle = idle; + + percpu_up_read(&scx_cgroup_ops_rwsem); +} + +void scx_group_set_bandwidth(struct task_group *tg, + u64 period_us, u64 quota_us, u64 burst_us) +{ + struct scx_sched *sch; + + percpu_down_read(&scx_cgroup_ops_rwsem); + sch = scx_root; + + if (scx_cgroup_enabled && SCX_HAS_OP(sch, cgroup_set_bandwidth) && + (tg->scx.bw_period_us != period_us || + tg->scx.bw_quota_us != quota_us || + tg->scx.bw_burst_us != burst_us)) + SCX_CALL_OP(sch, cgroup_set_bandwidth, NULL, + tg_cgrp(tg), period_us, quota_us, burst_us); + + tg->scx.bw_period_us = period_us; + tg->scx.bw_quota_us = quota_us; + tg->scx.bw_burst_us = burst_us; + + percpu_up_read(&scx_cgroup_ops_rwsem); +} +#endif /* CONFIG_EXT_GROUP_SCHED */ + +#if defined(CONFIG_EXT_GROUP_SCHED) || defined(CONFIG_EXT_SUB_SCHED) +static struct cgroup *root_cgroup(void) +{ + return &cgrp_dfl_root.cgrp; } static void scx_cgroup_lock(void) { - percpu_down_write(&scx_cgroup_rwsem); +#ifdef CONFIG_EXT_GROUP_SCHED + percpu_down_write(&scx_cgroup_ops_rwsem); +#endif + cgroup_lock(); } static void scx_cgroup_unlock(void) { - percpu_up_write(&scx_cgroup_rwsem); + cgroup_unlock(); +#ifdef CONFIG_EXT_GROUP_SCHED + percpu_up_write(&scx_cgroup_ops_rwsem); +#endif } +#else /* CONFIG_EXT_GROUP_SCHED || CONFIG_EXT_SUB_SCHED */ +static struct cgroup *root_cgroup(void) { return NULL; } +static void scx_cgroup_lock(void) {} +static void scx_cgroup_unlock(void) {} +#endif /* CONFIG_EXT_GROUP_SCHED || CONFIG_EXT_SUB_SCHED */ -#else /* CONFIG_EXT_GROUP_SCHED */ +#ifdef CONFIG_EXT_SUB_SCHED +static struct cgroup *sch_cgroup(struct scx_sched *sch) +{ + return sch->cgrp; +} -static inline void scx_cgroup_lock(void) {} -static inline void scx_cgroup_unlock(void) {} +/* for each descendant of @cgrp including self, set ->scx_sched to @sch */ +static void set_cgroup_sched(struct cgroup *cgrp, struct scx_sched *sch) +{ + struct cgroup *pos; + struct cgroup_subsys_state *css; -#endif /* CONFIG_EXT_GROUP_SCHED */ + cgroup_for_each_live_descendant_pre(pos, css, cgrp) + rcu_assign_pointer(pos->scx_sched, sch); +} +#else /* CONFIG_EXT_SUB_SCHED */ +static struct cgroup *sch_cgroup(struct scx_sched *sch) { return NULL; } +static void set_cgroup_sched(struct cgroup *cgrp, struct scx_sched *sch) {} +#endif /* CONFIG_EXT_SUB_SCHED */ /* * Omitted operations: * - * - wakeup_preempt: NOOP as it isn't useful in the wakeup path because the task - * isn't tied to the CPU at that point. Preemption is implemented by resetting - * the victim task's slice to 0 and triggering reschedule on the target CPU. - * * - migrate_task_rq: Unnecessary as task to cpu mapping is transient. * * - task_fork/dead: We need fork/dead notifications for all tasks regardless of @@ -4470,20 +4549,17 @@ DEFINE_SCHED_CLASS(ext) = { .wakeup_preempt = wakeup_preempt_scx, - .balance = balance_scx, .pick_task = pick_task_scx, .put_prev_task = put_prev_task_scx, .set_next_task = set_next_task_scx, -#ifdef CONFIG_SMP .select_task_rq = select_task_rq_scx, .task_woken = task_woken_scx, .set_cpus_allowed = set_cpus_allowed_scx, .rq_online = rq_online_scx, .rq_offline = rq_offline_scx, -#endif .task_tick = task_tick_scx, @@ -4500,36 +4576,60 @@ DEFINE_SCHED_CLASS(ext) = { #endif }; -static void init_dsq(struct scx_dispatch_q *dsq, u64 dsq_id) +static s32 init_dsq(struct scx_dispatch_q *dsq, u64 dsq_id, + struct scx_sched *sch) { + s32 cpu; + memset(dsq, 0, sizeof(*dsq)); raw_spin_lock_init(&dsq->lock); INIT_LIST_HEAD(&dsq->list); dsq->id = dsq_id; + dsq->sched = sch; + + dsq->pcpu = alloc_percpu(struct scx_dsq_pcpu); + if (!dsq->pcpu) + return -ENOMEM; + + for_each_possible_cpu(cpu) { + struct scx_dsq_pcpu *pcpu = per_cpu_ptr(dsq->pcpu, cpu); + + pcpu->dsq = dsq; + INIT_LIST_HEAD(&pcpu->deferred_reenq_user.node); + } + + return 0; } -static struct scx_dispatch_q *create_dsq(u64 dsq_id, int node) +static void exit_dsq(struct scx_dispatch_q *dsq) { - struct scx_dispatch_q *dsq; - int ret; + s32 cpu; - if (dsq_id & SCX_DSQ_FLAG_BUILTIN) - return ERR_PTR(-EINVAL); + for_each_possible_cpu(cpu) { + struct scx_dsq_pcpu *pcpu = per_cpu_ptr(dsq->pcpu, cpu); + struct scx_deferred_reenq_user *dru = &pcpu->deferred_reenq_user; + struct rq *rq = cpu_rq(cpu); - dsq = kmalloc_node(sizeof(*dsq), GFP_KERNEL, node); - if (!dsq) - return ERR_PTR(-ENOMEM); + /* + * There must have been a RCU grace period since the last + * insertion and @dsq should be off the deferred list by now. + */ + if (WARN_ON_ONCE(!list_empty(&dru->node))) { + guard(raw_spinlock_irqsave)(&rq->scx.deferred_reenq_lock); + list_del_init(&dru->node); + } + } - init_dsq(dsq, dsq_id); + free_percpu(dsq->pcpu); +} - ret = rhashtable_insert_fast(&dsq_hash, &dsq->hash_node, - dsq_hash_params); - if (ret) { - kfree(dsq); - return ERR_PTR(ret); - } - return dsq; +static void free_dsq_rcufn(struct rcu_head *rcu) +{ + struct scx_dispatch_q *dsq = container_of(rcu, struct scx_dispatch_q, rcu); + + exit_dsq(dsq); + kfree(dsq); } static void free_dsq_irq_workfn(struct irq_work *irq_work) @@ -4538,31 +4638,32 @@ static void free_dsq_irq_workfn(struct irq_work *irq_work) struct scx_dispatch_q *dsq, *tmp_dsq; llist_for_each_entry_safe(dsq, tmp_dsq, to_free, free_node) - kfree_rcu(dsq, rcu); + call_rcu(&dsq->rcu, free_dsq_rcufn); } static DEFINE_IRQ_WORK(free_dsq_irq_work, free_dsq_irq_workfn); -static void destroy_dsq(u64 dsq_id) +static void destroy_dsq(struct scx_sched *sch, u64 dsq_id) { struct scx_dispatch_q *dsq; unsigned long flags; rcu_read_lock(); - dsq = find_user_dsq(dsq_id); + dsq = find_user_dsq(sch, dsq_id); if (!dsq) goto out_unlock_rcu; raw_spin_lock_irqsave(&dsq->lock, flags); if (dsq->nr) { - scx_ops_error("attempting to destroy in-use dsq 0x%016llx (nr=%u)", - dsq->id, dsq->nr); + scx_error(sch, "attempting to destroy in-use dsq 0x%016llx (nr=%u)", + dsq->id, dsq->nr); goto out_unlock_dsq; } - if (rhashtable_remove_fast(&dsq_hash, &dsq->hash_node, dsq_hash_params)) + if (rhashtable_remove_fast(&sch->dsq_hash, &dsq->hash_node, + dsq_hash_params)) goto out_unlock_dsq; /* @@ -4572,8 +4673,8 @@ static void destroy_dsq(u64 dsq_id) * operations inside scheduler locks. */ dsq->id = SCX_DSQ_INVALID; - llist_add(&dsq->free_node, &dsqs_to_free); - irq_work_queue(&free_dsq_irq_work); + if (llist_add(&dsq->free_node, &dsqs_to_free)) + irq_work_queue(&free_dsq_irq_work); out_unlock_dsq: raw_spin_unlock_irqrestore(&dsq->lock, flags); @@ -4582,89 +4683,65 @@ out_unlock_rcu: } #ifdef CONFIG_EXT_GROUP_SCHED -static void scx_cgroup_exit(void) +static void scx_cgroup_exit(struct scx_sched *sch) { struct cgroup_subsys_state *css; - percpu_rwsem_assert_held(&scx_cgroup_rwsem); - scx_cgroup_enabled = false; /* - * scx_tg_on/offline() are excluded through scx_cgroup_rwsem. If we walk + * scx_tg_on/offline() are excluded through cgroup_lock(). If we walk * cgroups and exit all the inited ones, all online cgroups are exited. */ - rcu_read_lock(); css_for_each_descendant_post(css, &root_task_group.css) { struct task_group *tg = css_tg(css); - if (!(tg->scx_flags & SCX_TG_INITED)) - continue; - tg->scx_flags &= ~SCX_TG_INITED; - - if (!scx_ops.cgroup_exit) + if (!(tg->scx.flags & SCX_TG_INITED)) continue; + tg->scx.flags &= ~SCX_TG_INITED; - if (WARN_ON_ONCE(!css_tryget(css))) + if (!sch->ops.cgroup_exit) continue; - rcu_read_unlock(); - SCX_CALL_OP(SCX_KF_UNLOCKED, cgroup_exit, css->cgroup); - - rcu_read_lock(); - css_put(css); + SCX_CALL_OP(sch, cgroup_exit, NULL, css->cgroup); } - rcu_read_unlock(); } -static int scx_cgroup_init(void) +static int scx_cgroup_init(struct scx_sched *sch) { struct cgroup_subsys_state *css; int ret; - percpu_rwsem_assert_held(&scx_cgroup_rwsem); - - cgroup_warned_missing_weight = false; - cgroup_warned_missing_idle = false; - /* - * scx_tg_on/offline() are excluded through scx_cgroup_rwsem. If we walk + * scx_tg_on/offline() are excluded through cgroup_lock(). If we walk * cgroups and init, all online cgroups are initialized. */ - rcu_read_lock(); css_for_each_descendant_pre(css, &root_task_group.css) { struct task_group *tg = css_tg(css); - struct scx_cgroup_init_args args = { .weight = tg->scx_weight }; - - scx_cgroup_warn_missing_weight(tg); - scx_cgroup_warn_missing_idle(tg); + struct scx_cgroup_init_args args = { + .weight = tg->scx.weight, + .bw_period_us = tg->scx.bw_period_us, + .bw_quota_us = tg->scx.bw_quota_us, + .bw_burst_us = tg->scx.bw_burst_us, + }; - if ((tg->scx_flags & + if ((tg->scx.flags & (SCX_TG_ONLINE | SCX_TG_INITED)) != SCX_TG_ONLINE) continue; - if (!scx_ops.cgroup_init) { - tg->scx_flags |= SCX_TG_INITED; + if (!sch->ops.cgroup_init) { + tg->scx.flags |= SCX_TG_INITED; continue; } - if (WARN_ON_ONCE(!css_tryget(css))) - continue; - rcu_read_unlock(); - - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, cgroup_init, + ret = SCX_CALL_OP_RET(sch, cgroup_init, NULL, css->cgroup, &args); if (ret) { - css_put(css); - scx_ops_error("ops.cgroup_init() failed (%d)", ret); + scx_error(sch, "ops.cgroup_init() failed (%d)", ret); return ret; } - tg->scx_flags |= SCX_TG_INITED; - - rcu_read_lock(); - css_put(css); + tg->scx.flags |= SCX_TG_INITED; } - rcu_read_unlock(); WARN_ON_ONCE(scx_cgroup_enabled); scx_cgroup_enabled = true; @@ -4673,8 +4750,8 @@ static int scx_cgroup_init(void) } #else -static void scx_cgroup_exit(void) {} -static int scx_cgroup_init(void) { return 0; } +static void scx_cgroup_exit(struct scx_sched *sch) {} +static int scx_cgroup_init(struct scx_sched *sch) { return 0; } #endif @@ -4691,8 +4768,7 @@ static int scx_cgroup_init(void) { return 0; } static ssize_t scx_attr_state_show(struct kobject *kobj, struct kobj_attribute *ka, char *buf) { - return sysfs_emit(buf, "%s\n", - scx_ops_enable_state_str[scx_ops_enable_state()]); + return sysfs_emit(buf, "%s\n", scx_enable_state_str[scx_enable_state()]); } SCX_ATTR(state); @@ -4737,20 +4813,115 @@ static const struct attribute_group scx_global_attr_group = { .attrs = scx_global_attrs, }; +static void free_pnode(struct scx_sched_pnode *pnode); +static void free_exit_info(struct scx_exit_info *ei); + +static void scx_sched_free_rcu_work(struct work_struct *work) +{ + struct rcu_work *rcu_work = to_rcu_work(work); + struct scx_sched *sch = container_of(rcu_work, struct scx_sched, rcu_work); + struct rhashtable_iter rht_iter; + struct scx_dispatch_q *dsq; + int cpu, node; + + irq_work_sync(&sch->disable_irq_work); + kthread_destroy_worker(sch->helper); + timer_shutdown_sync(&sch->bypass_lb_timer); + free_cpumask_var(sch->bypass_lb_donee_cpumask); + free_cpumask_var(sch->bypass_lb_resched_cpumask); + +#ifdef CONFIG_EXT_SUB_SCHED + kfree(sch->cgrp_path); + if (sch_cgroup(sch)) + cgroup_put(sch_cgroup(sch)); + if (sch->sub_kset) + kobject_put(&sch->sub_kset->kobj); +#endif /* CONFIG_EXT_SUB_SCHED */ + + for_each_possible_cpu(cpu) { + struct scx_sched_pcpu *pcpu = per_cpu_ptr(sch->pcpu, cpu); + + /* + * $sch would have entered bypass mode before the RCU grace + * period. As that blocks new deferrals, all + * deferred_reenq_local_node's must be off-list by now. + */ + WARN_ON_ONCE(!list_empty(&pcpu->deferred_reenq_local.node)); + + exit_dsq(bypass_dsq(sch, cpu)); + } + + free_percpu(sch->pcpu); + + for_each_node_state(node, N_POSSIBLE) + free_pnode(sch->pnode[node]); + kfree(sch->pnode); + + rhashtable_walk_enter(&sch->dsq_hash, &rht_iter); + do { + rhashtable_walk_start(&rht_iter); + + while (!IS_ERR_OR_NULL((dsq = rhashtable_walk_next(&rht_iter)))) + destroy_dsq(sch, dsq->id); + + rhashtable_walk_stop(&rht_iter); + } while (dsq == ERR_PTR(-EAGAIN)); + rhashtable_walk_exit(&rht_iter); + + rhashtable_free_and_destroy(&sch->dsq_hash, NULL, NULL); + free_exit_info(sch->exit_info); + kfree(sch); +} + static void scx_kobj_release(struct kobject *kobj) { - kfree(kobj); + struct scx_sched *sch = container_of(kobj, struct scx_sched, kobj); + + INIT_RCU_WORK(&sch->rcu_work, scx_sched_free_rcu_work); + queue_rcu_work(system_dfl_wq, &sch->rcu_work); } static ssize_t scx_attr_ops_show(struct kobject *kobj, struct kobj_attribute *ka, char *buf) { - return sysfs_emit(buf, "%s\n", scx_ops.name); + struct scx_sched *sch = container_of(kobj, struct scx_sched, kobj); + + return sysfs_emit(buf, "%s\n", sch->ops.name); } SCX_ATTR(ops); +#define scx_attr_event_show(buf, at, events, kind) ({ \ + sysfs_emit_at(buf, at, "%s %llu\n", #kind, (events)->kind); \ +}) + +static ssize_t scx_attr_events_show(struct kobject *kobj, + struct kobj_attribute *ka, char *buf) +{ + struct scx_sched *sch = container_of(kobj, struct scx_sched, kobj); + struct scx_event_stats events; + int at = 0; + + scx_read_events(sch, &events); + at += scx_attr_event_show(buf, at, &events, SCX_EV_SELECT_CPU_FALLBACK); + at += scx_attr_event_show(buf, at, &events, SCX_EV_DISPATCH_LOCAL_DSQ_OFFLINE); + at += scx_attr_event_show(buf, at, &events, SCX_EV_DISPATCH_KEEP_LAST); + at += scx_attr_event_show(buf, at, &events, SCX_EV_ENQ_SKIP_EXITING); + at += scx_attr_event_show(buf, at, &events, SCX_EV_ENQ_SKIP_MIGRATION_DISABLED); + at += scx_attr_event_show(buf, at, &events, SCX_EV_REENQ_IMMED); + at += scx_attr_event_show(buf, at, &events, SCX_EV_REENQ_LOCAL_REPEAT); + at += scx_attr_event_show(buf, at, &events, SCX_EV_REFILL_SLICE_DFL); + at += scx_attr_event_show(buf, at, &events, SCX_EV_BYPASS_DURATION); + at += scx_attr_event_show(buf, at, &events, SCX_EV_BYPASS_DISPATCH); + at += scx_attr_event_show(buf, at, &events, SCX_EV_BYPASS_ACTIVATE); + at += scx_attr_event_show(buf, at, &events, SCX_EV_INSERT_NOT_OWNED); + at += scx_attr_event_show(buf, at, &events, SCX_EV_SUB_BYPASS_DISPATCH); + return at; +} +SCX_ATTR(events); + static struct attribute *scx_sched_attrs[] = { &scx_attr_ops.attr, + &scx_attr_events.attr, NULL, }; ATTRIBUTE_GROUPS(scx_sched); @@ -4763,7 +4934,19 @@ static const struct kobj_type scx_ktype = { static int scx_uevent(const struct kobject *kobj, struct kobj_uevent_env *env) { - return add_uevent_var(env, "SCXOPS=%s", scx_ops.name); + const struct scx_sched *sch; + + /* + * scx_uevent() can be reached by both scx_sched kobjects (scx_ktype) + * and sub-scheduler kset kobjects (kset_ktype) through the parent + * chain walk. Filter out the latter to avoid invalid casts. + */ + if (kobj->ktype != &scx_ktype) + return 0; + + sch = container_of(kobj, struct scx_sched, kobj); + + return add_uevent_var(env, "SCXOPS=%s", sch->ops.name); } static const struct kset_uevent_ops scx_uevent_ops = { @@ -4776,14 +4959,105 @@ static const struct kset_uevent_ops scx_uevent_ops = { */ bool task_should_scx(int policy) { - if (!scx_enabled() || - unlikely(scx_ops_enable_state() == SCX_OPS_DISABLING)) + /* if disabled, nothing should be on it */ + if (!scx_enabled()) return false; + + /* scx is taking over all SCHED_OTHER and SCHED_EXT tasks */ if (READ_ONCE(scx_switching_all)) return true; + + /* + * scx is tearing down - keep new SCHED_EXT tasks out. + * + * Must come after scx_switching_all test, which serves as a proxy + * for __scx_switched_all. While __scx_switched_all is set, we must + * return true via the branch above: a fork routed to fair would + * stall because next_active_class() skips fair. + * + * This can develop into a deadlock - scx holds scx_enable_mutex across + * kthread_create() in scx_alloc_and_add_sched(); if the new kthread is + * the stalled task, the disable path can never grab the mutex to clear + * scx_switching_all. + */ + if (unlikely(scx_enable_state() == SCX_DISABLING)) + return false; + return policy == SCHED_EXT; } +bool scx_allow_ttwu_queue(const struct task_struct *p) +{ + struct scx_sched *sch; + + if (!scx_enabled()) + return true; + + sch = scx_task_sched(p); + if (unlikely(!sch)) + return true; + + if (sch->ops.flags & SCX_OPS_ALLOW_QUEUED_WAKEUP) + return true; + + if (unlikely(p->sched_class != &ext_sched_class)) + return true; + + return false; +} + +/** + * handle_lockup - sched_ext common lockup handler + * @fmt: format string + * + * Called on system stall or lockup condition and initiates abort of sched_ext + * if enabled, which may resolve the reported lockup. + * + * Returns %true if sched_ext is enabled and abort was initiated, which may + * resolve the lockup. %false if sched_ext is not enabled or abort was already + * initiated by someone else. + */ +static __printf(1, 2) bool handle_lockup(const char *fmt, ...) +{ + struct scx_sched *sch; + va_list args; + bool ret; + + guard(rcu)(); + + sch = rcu_dereference(scx_root); + if (unlikely(!sch)) + return false; + + switch (scx_enable_state()) { + case SCX_ENABLING: + case SCX_ENABLED: + va_start(args, fmt); + ret = scx_verror(sch, fmt, args); + va_end(args); + return ret; + default: + return false; + } +} + +/** + * scx_rcu_cpu_stall - sched_ext RCU CPU stall handler + * + * While there are various reasons why RCU CPU stalls can occur on a system + * that may not be caused by the current BPF scheduler, try kicking out the + * current scheduler in an attempt to recover the system to a good state before + * issuing panics. + * + * Returns %true if sched_ext is enabled and abort was initiated, which may + * resolve the reported RCU stall. %false if sched_ext is not enabled or someone + * else already initiated abort. + */ +bool scx_rcu_cpu_stall(void) +{ + return handle_lockup("RCU CPU stall detected!"); +} + /** * scx_softlockup - sched_ext softlockup handler * @dur_s: number of seconds of CPU stuck due to soft lockup @@ -4792,44 +5066,354 @@ bool task_should_scx(int policy) * live-lock the system by making many CPUs target the same DSQ to the point * where soft-lockup detection triggers. This function is called from * soft-lockup watchdog when the triggering point is close and tries to unjam - * the system by enabling the breather and aborting the BPF scheduler. + * the system and aborting the BPF scheduler. */ void scx_softlockup(u32 dur_s) { - switch (scx_ops_enable_state()) { - case SCX_OPS_ENABLING: - case SCX_OPS_ENABLED: - break; - default: + if (!handle_lockup("soft lockup - CPU %d stuck for %us", smp_processor_id(), dur_s)) return; + + printk_deferred(KERN_ERR "sched_ext: Soft lockup - CPU %d stuck for %us, disabling BPF scheduler\n", + smp_processor_id(), dur_s); +} + +/* + * scx_hardlockup() runs from NMI and eventually calls scx_claim_exit(), + * which takes scx_sched_lock. scx_sched_lock isn't NMI-safe and grabbing + * it from NMI context can lead to deadlocks. Defer via irq_work; the + * disable path runs off irq_work anyway. + */ +static atomic_t scx_hardlockup_cpu = ATOMIC_INIT(-1); + +static void scx_hardlockup_irq_workfn(struct irq_work *work) +{ + int cpu = atomic_xchg(&scx_hardlockup_cpu, -1); + + if (cpu >= 0 && handle_lockup("hard lockup - CPU %d", cpu)) + printk_deferred(KERN_ERR "sched_ext: Hard lockup - CPU %d, disabling BPF scheduler\n", + cpu); +} + +static DEFINE_IRQ_WORK(scx_hardlockup_irq_work, scx_hardlockup_irq_workfn); + +/** + * scx_hardlockup - sched_ext hardlockup handler + * + * A poorly behaving BPF scheduler can trigger hard lockup by e.g. putting + * numerous affinitized tasks in a single queue and directing all CPUs at it. + * Try kicking out the current scheduler in an attempt to recover the system to + * a good state before taking more drastic actions. + * + * Queues an irq_work; the handle_lockup() call happens in IRQ context (see + * scx_hardlockup_irq_workfn). + * + * Returns %true if sched_ext is enabled and the work was queued, %false + * otherwise. + */ +bool scx_hardlockup(int cpu) +{ + if (!rcu_access_pointer(scx_root)) + return false; + + atomic_cmpxchg(&scx_hardlockup_cpu, -1, cpu); + irq_work_queue(&scx_hardlockup_irq_work); + return true; +} + +static u32 bypass_lb_cpu(struct scx_sched *sch, s32 donor, + struct cpumask *donee_mask, struct cpumask *resched_mask, + u32 nr_donor_target, u32 nr_donee_target) +{ + struct rq *donor_rq = cpu_rq(donor); + struct scx_dispatch_q *donor_dsq = bypass_dsq(sch, donor); + struct task_struct *p, *n; + struct scx_dsq_list_node cursor = INIT_DSQ_LIST_CURSOR(cursor, donor_dsq, 0); + s32 delta = READ_ONCE(donor_dsq->nr) - nr_donor_target; + u32 nr_balanced = 0, min_delta_us; + + /* + * All we want to guarantee is reasonable forward progress. No reason to + * fine tune. Assuming every task on @donor_dsq runs their full slice, + * consider offloading iff the total queued duration is over the + * threshold. + */ + min_delta_us = READ_ONCE(scx_bypass_lb_intv_us) / SCX_BYPASS_LB_MIN_DELTA_DIV; + if (delta < DIV_ROUND_UP(min_delta_us, READ_ONCE(scx_slice_bypass_us))) + return 0; + + raw_spin_rq_lock_irq(donor_rq); + raw_spin_lock(&donor_dsq->lock); + list_add(&cursor.node, &donor_dsq->list); +resume: + n = container_of(&cursor, struct task_struct, scx.dsq_list); + n = nldsq_next_task(donor_dsq, n, false); + + while ((p = n)) { + struct scx_dispatch_q *donee_dsq; + int donee; + + n = nldsq_next_task(donor_dsq, n, false); + + if (donor_dsq->nr <= nr_donor_target) + break; + + if (cpumask_empty(donee_mask)) + break; + + /* + * If an earlier pass placed @p on @donor_dsq from a different + * CPU and the donee hasn't consumed it yet, @p is still on the + * previous CPU and task_rq(@p) != @donor_rq. @p can't be moved + * without its rq locked. Skip. + */ + if (task_rq(p) != donor_rq) + continue; + + donee = cpumask_any_and_distribute(donee_mask, p->cpus_ptr); + if (donee >= nr_cpu_ids) + continue; + + donee_dsq = bypass_dsq(sch, donee); + + /* + * $p's rq is not locked but $p's DSQ lock protects its + * scheduling properties making this test safe. + */ + if (!task_can_run_on_remote_rq(sch, p, cpu_rq(donee), false)) + continue; + + /* + * Moving $p from one non-local DSQ to another. The source rq + * and DSQ are already locked. Do an abbreviated dequeue and + * then perform enqueue without unlocking $donor_dsq. + * + * We don't want to drop and reacquire the lock on each + * iteration as @donor_dsq can be very long and potentially + * highly contended. Donee DSQs are less likely to be contended. + * The nested locking is safe as only this LB moves tasks + * between bypass DSQs. + */ + dispatch_dequeue_locked(p, donor_dsq); + dispatch_enqueue(sch, cpu_rq(donee), donee_dsq, p, SCX_ENQ_NESTED); + + /* + * $donee might have been idle and need to be woken up. No need + * to be clever. Kick every CPU that receives tasks. + */ + cpumask_set_cpu(donee, resched_mask); + + if (READ_ONCE(donee_dsq->nr) >= nr_donee_target) + cpumask_clear_cpu(donee, donee_mask); + + nr_balanced++; + if (!(nr_balanced % SCX_BYPASS_LB_BATCH) && n) { + list_move_tail(&cursor.node, &n->scx.dsq_list.node); + raw_spin_unlock(&donor_dsq->lock); + raw_spin_rq_unlock_irq(donor_rq); + cpu_relax(); + raw_spin_rq_lock_irq(donor_rq); + raw_spin_lock(&donor_dsq->lock); + goto resume; + } } - /* allow only one instance, cleared at the end of scx_ops_bypass() */ - if (test_and_set_bit(0, &scx_in_softlockup)) + list_del_init(&cursor.node); + raw_spin_unlock(&donor_dsq->lock); + raw_spin_rq_unlock_irq(donor_rq); + + return nr_balanced; +} + +static void bypass_lb_node(struct scx_sched *sch, int node) +{ + const struct cpumask *node_mask = cpumask_of_node(node); + struct cpumask *donee_mask = sch->bypass_lb_donee_cpumask; + struct cpumask *resched_mask = sch->bypass_lb_resched_cpumask; + u32 nr_tasks = 0, nr_cpus = 0, nr_balanced = 0; + u32 nr_target, nr_donor_target; + u32 before_min = U32_MAX, before_max = 0; + u32 after_min = U32_MAX, after_max = 0; + int cpu; + + /* count the target tasks and CPUs */ + for_each_cpu_and(cpu, cpu_online_mask, node_mask) { + u32 nr = READ_ONCE(bypass_dsq(sch, cpu)->nr); + + nr_tasks += nr; + nr_cpus++; + + before_min = min(nr, before_min); + before_max = max(nr, before_max); + } + + if (!nr_cpus) + return; + + /* + * We don't want CPUs to have more than $nr_donor_target tasks and + * balancing to fill donee CPUs upto $nr_target. Once targets are + * calculated, find the donee CPUs. + */ + nr_target = DIV_ROUND_UP(nr_tasks, nr_cpus); + nr_donor_target = DIV_ROUND_UP(nr_target * SCX_BYPASS_LB_DONOR_PCT, 100); + + cpumask_clear(donee_mask); + for_each_cpu_and(cpu, cpu_online_mask, node_mask) { + if (READ_ONCE(bypass_dsq(sch, cpu)->nr) < nr_target) + cpumask_set_cpu(cpu, donee_mask); + } + + /* iterate !donee CPUs and see if they should be offloaded */ + cpumask_clear(resched_mask); + for_each_cpu_and(cpu, cpu_online_mask, node_mask) { + if (cpumask_empty(donee_mask)) + break; + if (cpumask_test_cpu(cpu, donee_mask)) + continue; + if (READ_ONCE(bypass_dsq(sch, cpu)->nr) <= nr_donor_target) + continue; + + nr_balanced += bypass_lb_cpu(sch, cpu, donee_mask, resched_mask, + nr_donor_target, nr_target); + } + + for_each_cpu(cpu, resched_mask) + resched_cpu(cpu); + + for_each_cpu_and(cpu, cpu_online_mask, node_mask) { + u32 nr = READ_ONCE(bypass_dsq(sch, cpu)->nr); + + after_min = min(nr, after_min); + after_max = max(nr, after_max); + + } + + trace_sched_ext_bypass_lb(node, nr_cpus, nr_tasks, nr_balanced, + before_min, before_max, after_min, after_max); +} + +/* + * In bypass mode, all tasks are put on the per-CPU bypass DSQs. If the machine + * is over-saturated and the BPF scheduler skewed tasks into few CPUs, some + * bypass DSQs can be overloaded. If there are enough tasks to saturate other + * lightly loaded CPUs, such imbalance can lead to very high execution latency + * on the overloaded CPUs and thus to hung tasks and RCU stalls. To avoid such + * outcomes, a simple load balancing mechanism is implemented by the following + * timer which runs periodically while bypass mode is in effect. + */ +static void scx_bypass_lb_timerfn(struct timer_list *timer) +{ + struct scx_sched *sch = container_of(timer, struct scx_sched, bypass_lb_timer); + int node; + u32 intv_us; + + if (!bypass_dsp_enabled(sch)) return; - printk_deferred(KERN_ERR "sched_ext: Soft lockup - CPU%d stuck for %us, disabling \"%s\"\n", - smp_processor_id(), dur_s, scx_ops.name); + for_each_node_with_cpus(node) + bypass_lb_node(sch, node); + + intv_us = READ_ONCE(scx_bypass_lb_intv_us); + if (intv_us) + mod_timer(timer, jiffies + usecs_to_jiffies(intv_us)); +} + +static bool inc_bypass_depth(struct scx_sched *sch) +{ + lockdep_assert_held(&scx_bypass_lock); + + WARN_ON_ONCE(sch->bypass_depth < 0); + WRITE_ONCE(sch->bypass_depth, sch->bypass_depth + 1); + if (sch->bypass_depth != 1) + return false; + + WRITE_ONCE(sch->slice_dfl, READ_ONCE(scx_slice_bypass_us) * NSEC_PER_USEC); + sch->bypass_timestamp = ktime_get_ns(); + scx_add_event(sch, SCX_EV_BYPASS_ACTIVATE, 1); + return true; +} + +static bool dec_bypass_depth(struct scx_sched *sch) +{ + lockdep_assert_held(&scx_bypass_lock); + + WARN_ON_ONCE(sch->bypass_depth < 1); + WRITE_ONCE(sch->bypass_depth, sch->bypass_depth - 1); + if (sch->bypass_depth != 0) + return false; + + WRITE_ONCE(sch->slice_dfl, SCX_SLICE_DFL); + scx_add_event(sch, SCX_EV_BYPASS_DURATION, + ktime_get_ns() - sch->bypass_timestamp); + return true; +} + +static void enable_bypass_dsp(struct scx_sched *sch) +{ + struct scx_sched *host = scx_parent(sch) ?: sch; + u32 intv_us = READ_ONCE(scx_bypass_lb_intv_us); + s32 ret; + + /* + * @sch->bypass_depth transitioning from 0 to 1 triggers enabling. + * Shouldn't stagger. + */ + if (WARN_ON_ONCE(test_and_set_bit(0, &sch->bypass_dsp_claim))) + return; /* - * Some CPUs may be trapped in the dispatch paths. Enable breather - * immediately; otherwise, we might even be able to get to - * scx_ops_bypass(). + * When a sub-sched bypasses, its tasks are queued on the bypass DSQs of + * the nearest non-bypassing ancestor or root. As enable_bypass_dsp() is + * called iff @sch is not already bypassed due to an ancestor bypassing, + * we can assume that the parent is not bypassing and thus will be the + * host of the bypass DSQs. + * + * While the situation may change in the future, the following + * guarantees that the nearest non-bypassing ancestor or root has bypass + * dispatch enabled while a descendant is bypassing, which is all that's + * required. + * + * bypass_dsp_enabled() test is used to determine whether to enter the + * bypass dispatch handling path from both bypassing and hosting scheds. + * Bump enable depth on both @sch and bypass dispatch host. */ - atomic_inc(&scx_ops_breather_depth); + ret = atomic_inc_return(&sch->bypass_dsp_enable_depth); + WARN_ON_ONCE(ret <= 0); + + if (host != sch) { + ret = atomic_inc_return(&host->bypass_dsp_enable_depth); + WARN_ON_ONCE(ret <= 0); + } - scx_ops_error("soft lockup - CPU#%d stuck for %us", - smp_processor_id(), dur_s); + /* + * The LB timer will stop running if bypass dispatch is disabled. Start + * after enabling bypass dispatch. + */ + if (intv_us && !timer_pending(&host->bypass_lb_timer)) + mod_timer(&host->bypass_lb_timer, + jiffies + usecs_to_jiffies(intv_us)); } -static void scx_clear_softlockup(void) +/* may be called without holding scx_bypass_lock */ +static void disable_bypass_dsp(struct scx_sched *sch) { - if (test_and_clear_bit(0, &scx_in_softlockup)) - atomic_dec(&scx_ops_breather_depth); + s32 ret; + + if (!test_and_clear_bit(0, &sch->bypass_dsp_claim)) + return; + + ret = atomic_dec_return(&sch->bypass_dsp_enable_depth); + WARN_ON_ONCE(ret < 0); + + if (scx_parent(sch)) { + ret = atomic_dec_return(&scx_parent(sch)->bypass_dsp_enable_depth); + WARN_ON_ONCE(ret < 0); + } } /** - * scx_ops_bypass - [Un]bypass scx_ops and guarantee forward progress + * scx_bypass - [Un]bypass scx_ops and guarantee forward progress + * @sch: sched to bypass * @bypass: true for bypass, false for unbypass * * Bypassing guarantees that all runnable tasks make forward progress without @@ -4848,41 +5432,53 @@ static void scx_clear_softlockup(void) * * - ops.dispatch() is ignored. * - * - balance_scx() does not set %SCX_RQ_BAL_KEEP on non-zero slice as slice + * - balance_one() does not set %SCX_RQ_BAL_KEEP on non-zero slice as slice * can't be trusted. Whenever a tick triggers, the running task is rotated to * the tail of the queue with core_sched_at touched. * * - pick_next_task() suppresses zero slice warning. * - * - scx_bpf_kick_cpu() is disabled to avoid irq_work malfunction during PM + * - scx_kick_cpu() is disabled to avoid irq_work malfunction during PM * operations. * * - scx_prio_less() reverts to the default core_sched_at order. */ -static void scx_ops_bypass(bool bypass) +static void scx_bypass(struct scx_sched *sch, bool bypass) { - static DEFINE_RAW_SPINLOCK(bypass_lock); - int cpu; + struct scx_sched *pos; unsigned long flags; + int cpu; + + raw_spin_lock_irqsave(&scx_bypass_lock, flags); - raw_spin_lock_irqsave(&bypass_lock, flags); if (bypass) { - scx_ops_bypass_depth++; - WARN_ON_ONCE(scx_ops_bypass_depth <= 0); - if (scx_ops_bypass_depth != 1) + if (!inc_bypass_depth(sch)) goto unlock; + + enable_bypass_dsp(sch); } else { - scx_ops_bypass_depth--; - WARN_ON_ONCE(scx_ops_bypass_depth < 0); - if (scx_ops_bypass_depth != 0) + if (!dec_bypass_depth(sch)) goto unlock; } - atomic_inc(&scx_ops_breather_depth); + /* + * Bypass state is propagated to all descendants - an scx_sched bypasses + * if itself or any of its ancestors are in bypass mode. + */ + raw_spin_lock(&scx_sched_lock); + scx_for_each_descendant_pre(pos, sch) { + if (pos == sch) + continue; + if (bypass) + inc_bypass_depth(pos); + else + dec_bypass_depth(pos); + } + raw_spin_unlock(&scx_sched_lock); /* * No task property is changing. We just need to make sure all currently - * queued tasks are re-queued according to the new scx_rq_bypassing() + * queued tasks are re-queued according to the new scx_bypassing() * state. As an optimization, walk each rq's runnable_list instead of * the scx_tasks list. * @@ -4894,19 +5490,23 @@ static void scx_ops_bypass(bool bypass) struct task_struct *p, *n; raw_spin_rq_lock(rq); + raw_spin_lock(&scx_sched_lock); - if (bypass) { - WARN_ON_ONCE(rq->scx.flags & SCX_RQ_BYPASSING); - rq->scx.flags |= SCX_RQ_BYPASSING; - } else { - WARN_ON_ONCE(!(rq->scx.flags & SCX_RQ_BYPASSING)); - rq->scx.flags &= ~SCX_RQ_BYPASSING; + scx_for_each_descendant_pre(pos, sch) { + struct scx_sched_pcpu *pcpu = per_cpu_ptr(pos->pcpu, cpu); + + if (pos->bypass_depth) + pcpu->flags |= SCX_SCHED_PCPU_BYPASSING; + else + pcpu->flags &= ~SCX_SCHED_PCPU_BYPASSING; } + raw_spin_unlock(&scx_sched_lock); + /* * We need to guarantee that no tasks are on the BPF scheduler * while bypassing. Either we see enabled or the enable path - * sees scx_rq_bypassing() before moving tasks to SCX. + * sees scx_bypassing() before moving tasks to SCX. */ if (!scx_enabled()) { raw_spin_rq_unlock(rq); @@ -4922,11 +5522,13 @@ static void scx_ops_bypass(bool bypass) */ list_for_each_entry_safe_reverse(p, n, &rq->scx.runnable_list, scx.runnable_node) { - struct sched_enq_and_set_ctx ctx; + if (!scx_is_descendant(scx_task_sched(p), sch)) + continue; /* cycling deq/enq is enough, see the function comment */ - sched_deq_and_put_task(p, DEQUEUE_SAVE | DEQUEUE_MOVE, &ctx); - sched_enq_and_set_task(&ctx); + scoped_guard (sched_change, p, DEQUEUE_SAVE | DEQUEUE_MOVE) { + /* nothing */ ; + } } /* resched to restore ticks and idle state */ @@ -4936,15 +5538,16 @@ static void scx_ops_bypass(bool bypass) raw_spin_rq_unlock(rq); } - atomic_dec(&scx_ops_breather_depth); + /* disarming must come after moving all tasks out of the bypass DSQs */ + if (!bypass) + disable_bypass_dsp(sch); unlock: - raw_spin_unlock_irqrestore(&bypass_lock, flags); - scx_clear_softlockup(); + raw_spin_unlock_irqrestore(&scx_bypass_lock, flags); } static void free_exit_info(struct scx_exit_info *ei) { - kfree(ei->dump); + kvfree(ei->dump); kfree(ei->msg); kfree(ei->bt); kfree(ei); @@ -4954,13 +5557,13 @@ static struct scx_exit_info *alloc_exit_info(size_t exit_dump_len) { struct scx_exit_info *ei; - ei = kzalloc(sizeof(*ei), GFP_KERNEL); + ei = kzalloc_obj(*ei); if (!ei) return NULL; - ei->bt = kcalloc(SCX_EXIT_BT_LEN, sizeof(ei->bt[0]), GFP_KERNEL); + ei->bt = kzalloc_objs(ei->bt[0], SCX_EXIT_BT_LEN); ei->msg = kzalloc(SCX_EXIT_MSG_LEN, GFP_KERNEL); - ei->dump = kzalloc(exit_dump_len, GFP_KERNEL); + ei->dump = kvzalloc(exit_dump_len, GFP_KERNEL); if (!ei->bt || !ei->msg || !ei->dump) { free_exit_info(ei); @@ -4981,6 +5584,8 @@ static const char *scx_exit_reason(enum scx_exit_kind kind) return "unregistered from the main kernel"; case SCX_EXIT_SYSRQ: return "disabled by sysrq-S"; + case SCX_EXIT_PARENT: + return "parent exiting"; case SCX_EXIT_ERROR: return "runtime error"; case SCX_EXIT_ERROR_BPF: @@ -4992,42 +5597,332 @@ static const char *scx_exit_reason(enum scx_exit_kind kind) } } -static void scx_ops_disable_workfn(struct kthread_work *work) +static void free_kick_syncs(void) +{ + int cpu; + + for_each_possible_cpu(cpu) { + struct scx_kick_syncs **ksyncs = per_cpu_ptr(&scx_kick_syncs, cpu); + struct scx_kick_syncs *to_free; + + to_free = rcu_replace_pointer(*ksyncs, NULL, true); + if (to_free) + kvfree_rcu(to_free, rcu); + } +} + +static void refresh_watchdog(void) +{ + struct scx_sched *sch; + unsigned long intv = ULONG_MAX; + + /* take the shortest timeout and use its half for watchdog interval */ + rcu_read_lock(); + list_for_each_entry_rcu(sch, &scx_sched_all, all) + intv = max(min(intv, sch->watchdog_timeout / 2), 1); + rcu_read_unlock(); + + WRITE_ONCE(scx_watchdog_timestamp, jiffies); + WRITE_ONCE(scx_watchdog_interval, intv); + + if (intv < ULONG_MAX) + mod_delayed_work(system_dfl_wq, &scx_watchdog_work, intv); + else + cancel_delayed_work_sync(&scx_watchdog_work); +} + +static s32 scx_link_sched(struct scx_sched *sch) +{ + const char *err_msg = ""; + s32 ret = 0; + + scoped_guard(raw_spinlock_irq, &scx_sched_lock) { +#ifdef CONFIG_EXT_SUB_SCHED + struct scx_sched *parent = scx_parent(sch); + + if (parent) { + /* + * scx_claim_exit() propagates exit_kind transition to + * its sub-scheds while holding scx_sched_lock - either + * we can see the parent's non-NONE exit_kind or the + * parent can shoot us down. + */ + if (atomic_read(&parent->exit_kind) != SCX_EXIT_NONE) { + err_msg = "parent disabled"; + ret = -ENOENT; + break; + } + + ret = rhashtable_lookup_insert_fast(&scx_sched_hash, + &sch->hash_node, scx_sched_hash_params); + if (ret) { + err_msg = "failed to insert into scx_sched_hash"; + break; + } + + list_add_tail(&sch->sibling, &parent->children); + } +#endif /* CONFIG_EXT_SUB_SCHED */ + + list_add_tail_rcu(&sch->all, &scx_sched_all); + } + + /* + * scx_error() takes scx_sched_lock via scx_claim_exit(), so it must run after + * the guard above is released. + */ + if (ret) { + scx_error(sch, "%s (%d)", err_msg, ret); + return ret; + } + + refresh_watchdog(); + return 0; +} + +static void scx_unlink_sched(struct scx_sched *sch) +{ + scoped_guard(raw_spinlock_irq, &scx_sched_lock) { +#ifdef CONFIG_EXT_SUB_SCHED + if (scx_parent(sch)) { + rhashtable_remove_fast(&scx_sched_hash, &sch->hash_node, + scx_sched_hash_params); + list_del_init(&sch->sibling); + } +#endif /* CONFIG_EXT_SUB_SCHED */ + list_del_rcu(&sch->all); + } + + refresh_watchdog(); +} + +/* + * Called to disable future dumps and wait for in-progress one while disabling + * @sch. Once @sch becomes empty during disable, there's no point in dumping it. + * This prevents calling dump ops on a dead sch. + */ +static void scx_disable_dump(struct scx_sched *sch) +{ + guard(raw_spinlock_irqsave)(&scx_dump_lock); + sch->dump_disabled = true; +} + +#ifdef CONFIG_EXT_SUB_SCHED +static DECLARE_WAIT_QUEUE_HEAD(scx_unlink_waitq); + +static void drain_descendants(struct scx_sched *sch) +{ + /* + * Child scheds that finished the critical part of disabling will take + * themselves off @sch->children. Wait for it to drain. As propagation + * is recursive, empty @sch->children means that all proper descendant + * scheds reached unlinking stage. + */ + wait_event(scx_unlink_waitq, list_empty(&sch->children)); +} + +static void scx_fail_parent(struct scx_sched *sch, + struct task_struct *failed, s32 fail_code) { - struct scx_exit_info *ei = scx_exit_info; + struct scx_sched *parent = scx_parent(sch); struct scx_task_iter sti; struct task_struct *p; - struct rhashtable_iter rht_iter; - struct scx_dispatch_q *dsq; - int i, kind, cpu; - kind = atomic_read(&scx_exit_kind); - while (true) { + scx_error(parent, "ops.init_task() failed (%d) for %s[%d] while disabling a sub-scheduler", + fail_code, failed->comm, failed->pid); + + /* + * Once $parent is bypassed, it's safe to put SCX_TASK_NONE tasks into + * it. This may cause downstream failures on the BPF side but $parent is + * dying anyway. + */ + scx_bypass(parent, true); + + scx_task_iter_start(&sti, sch->cgrp); + while ((p = scx_task_iter_next_locked(&sti))) { + if (scx_task_on_sched(parent, p)) + continue; + + scoped_guard (sched_change, p, DEQUEUE_SAVE | DEQUEUE_MOVE) { + scx_disable_and_exit_task(sch, p); + scx_set_task_sched(p, parent); + } + } + scx_task_iter_stop(&sti); +} + +static void scx_sub_disable(struct scx_sched *sch) +{ + struct scx_sched *parent = scx_parent(sch); + struct scx_task_iter sti; + struct task_struct *p; + int ret; + + /* + * Guarantee forward progress and wait for descendants to be disabled. + * To limit disruptions, $parent is not bypassed. Tasks are fully + * prepped and then inserted back into $parent. + */ + scx_bypass(sch, true); + drain_descendants(sch); + + /* + * Here, every runnable task is guaranteed to make forward progress and + * we can safely use blocking synchronization constructs. Actually + * disable ops. + */ + mutex_lock(&scx_enable_mutex); + percpu_down_write(&scx_fork_rwsem); + scx_cgroup_lock(); + + set_cgroup_sched(sch_cgroup(sch), parent); + + scx_task_iter_start(&sti, sch->cgrp); + while ((p = scx_task_iter_next_locked(&sti))) { + struct rq *rq; + struct rq_flags rf; + + /* filter out duplicate visits */ + if (scx_task_on_sched(parent, p)) + continue; + /* - * NONE indicates that a new scx_ops has been registered since - * disable was scheduled - don't kill the new ops. DONE - * indicates that the ops has already been disabled. + * By the time control reaches here, all descendant schedulers + * should already have been disabled. */ - if (kind == SCX_EXIT_NONE || kind == SCX_EXIT_DONE) - return; - if (atomic_try_cmpxchg(&scx_exit_kind, &kind, SCX_EXIT_DONE)) + WARN_ON_ONCE(!scx_task_on_sched(sch, p)); + + /* + * If $p is about to be freed, nothing prevents $sch from + * unloading before $p reaches sched_ext_free(). Disable and + * exit $p right away. + */ + if (!tryget_task_struct(p)) { + scx_disable_and_exit_task(sch, p); + continue; + } + + scx_task_iter_unlock(&sti); + + /* + * $p is READY or ENABLED on @sch. Initialize for $parent, + * disable and exit from @sch, and then switch over to $parent. + * + * If a task fails to initialize for $parent, the only available + * action is disabling $parent too. While this allows disabling + * of a child sched to cause the parent scheduler to fail, the + * failure can only originate from ops.init_task() of the + * parent. A child can't directly affect the parent through its + * own failures. + */ + ret = __scx_init_task(parent, p, false); + if (ret) { + scx_fail_parent(sch, p, ret); + put_task_struct(p); break; + } + + rq = task_rq_lock(p, &rf); + + if (scx_get_task_state(p) == SCX_TASK_DEAD) { + /* + * sched_ext_dead() raced us between __scx_init_task() + * and this rq lock and ran exit_task() on @sch (the + * sched @p was on at that point), not on $parent. + * $parent's just-completed init is owed an exit_task() + * and we issue it here. + */ + scx_sub_init_cancel_task(parent, p); + task_rq_unlock(rq, p, &rf); + put_task_struct(p); + continue; + } + + scoped_guard (sched_change, p, DEQUEUE_SAVE | DEQUEUE_MOVE) { + /* + * $p is initialized for $parent and still attached to + * @sch. Disable and exit for @sch, switch over to + * $parent, override the state to READY to account for + * $p having already been initialized, and then enable. + */ + scx_disable_and_exit_task(sch, p); + scx_set_task_state(p, SCX_TASK_INIT_BEGIN); + scx_set_task_state(p, SCX_TASK_INIT); + scx_set_task_sched(p, parent); + scx_set_task_state(p, SCX_TASK_READY); + scx_enable_task(parent, p); + } + + task_rq_unlock(rq, p, &rf); + put_task_struct(p); } - ei->kind = kind; - ei->reason = scx_exit_reason(ei->kind); + scx_task_iter_stop(&sti); + + scx_disable_dump(sch); + + scx_cgroup_unlock(); + percpu_up_write(&scx_fork_rwsem); + + /* + * All tasks are moved off of @sch but there may still be on-going + * operations (e.g. ops.select_cpu()). Drain them by flushing RCU. Use + * the expedited version as ancestors may be waiting in bypass mode. + * Also, tell the parent that there is no need to keep running bypass + * DSQs for us. + */ + synchronize_rcu_expedited(); + disable_bypass_dsp(sch); + + scx_unlink_sched(sch); + + mutex_unlock(&scx_enable_mutex); - /* guarantee forward progress by bypassing scx_ops */ - scx_ops_bypass(true); + /* + * @sch is now unlinked from the parent's children list. Notify and call + * ops.sub_detach/exit(). Note that ops.sub_detach/exit() must be called + * after unlinking and releasing all locks. See scx_claim_exit(). + */ + wake_up_all(&scx_unlink_waitq); + + if (parent->ops.sub_detach && sch->sub_attached) { + struct scx_sub_detach_args sub_detach_args = { + .ops = &sch->ops, + .cgroup_path = sch->cgrp_path, + }; + SCX_CALL_OP(parent, sub_detach, NULL, + &sub_detach_args); + } + + if (sch->ops.exit) + SCX_CALL_OP(sch, exit, NULL, sch->exit_info); + if (sch->sub_kset) + kobject_del(&sch->sub_kset->kobj); + kobject_del(&sch->kobj); +} +#else /* CONFIG_EXT_SUB_SCHED */ +static void drain_descendants(struct scx_sched *sch) { } +static void scx_sub_disable(struct scx_sched *sch) { } +#endif /* CONFIG_EXT_SUB_SCHED */ + +static void scx_root_disable(struct scx_sched *sch) +{ + struct scx_exit_info *ei = sch->exit_info; + struct scx_task_iter sti; + struct task_struct *p; + int cpu; - switch (scx_ops_set_enable_state(SCX_OPS_DISABLING)) { - case SCX_OPS_DISABLING: + /* guarantee forward progress and wait for descendants to be disabled */ + scx_bypass(sch, true); + drain_descendants(sch); + + switch (scx_set_enable_state(SCX_DISABLING)) { + case SCX_DISABLING: WARN_ONCE(true, "sched_ext: duplicate disabling instance?"); break; - case SCX_OPS_DISABLED: + case SCX_DISABLED: pr_warn("sched_ext: ops error detected without ops (%s)\n", - scx_exit_info->msg); - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_DISABLED) != - SCX_OPS_DISABLING); + sch->exit_info->msg); + WARN_ON_ONCE(scx_set_enable_state(SCX_DISABLED) != SCX_DISABLING); goto done; default: break; @@ -5038,17 +5933,17 @@ static void scx_ops_disable_workfn(struct kthread_work *work) * we can safely use blocking synchronization constructs. Actually * disable ops. */ - mutex_lock(&scx_ops_enable_mutex); + mutex_lock(&scx_enable_mutex); static_branch_disable(&__scx_switched_all); WRITE_ONCE(scx_switching_all, false); /* * Shut down cgroup support before tasks so that the cgroup attach path - * doesn't race against scx_ops_exit_task(). + * doesn't race against scx_disable_and_exit_task(). */ scx_cgroup_lock(); - scx_cgroup_exit(); + scx_cgroup_exit(sch); scx_cgroup_unlock(); /* @@ -5057,29 +5952,33 @@ static void scx_ops_disable_workfn(struct kthread_work *work) */ percpu_down_write(&scx_fork_rwsem); - scx_ops_init_task_enabled = false; + scx_init_task_enabled = false; - scx_task_iter_start(&sti); + scx_task_iter_start(&sti, NULL); while ((p = scx_task_iter_next_locked(&sti))) { + unsigned int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; const struct sched_class *old_class = p->sched_class; - const struct sched_class *new_class = - __setscheduler_class(p->policy, p->prio); - struct sched_enq_and_set_ctx ctx; - - if (old_class != new_class && p->se.sched_delayed) - dequeue_task(task_rq(p), p, DEQUEUE_SLEEP | DEQUEUE_DELAYED); + const struct sched_class *new_class = scx_setscheduler_class(p); - sched_deq_and_put_task(p, DEQUEUE_SAVE | DEQUEUE_MOVE, &ctx); + update_rq_clock(task_rq(p)); - p->sched_class = new_class; - check_class_changing(task_rq(p), p, old_class); + if (old_class != new_class) + queue_flags |= DEQUEUE_CLASS; - sched_enq_and_set_task(&ctx); + scoped_guard (sched_change, p, queue_flags) { + p->sched_class = new_class; + } - check_class_changed(task_rq(p), p, old_class, p->prio); - scx_ops_exit_task(p); + scx_disable_and_exit_task(scx_task_sched(p), p); } scx_task_iter_stop(&sti); + + scx_disable_dump(sch); + + scx_cgroup_lock(); + set_cgroup_sched(sch_cgroup(sch), NULL); + scx_cgroup_unlock(); + percpu_up_write(&scx_fork_rwsem); /* @@ -5092,97 +5991,156 @@ static void scx_ops_disable_workfn(struct kthread_work *work) } /* no task is on scx, turn off all the switches and flush in-progress calls */ - static_branch_disable(&__scx_ops_enabled); - for (i = SCX_OPI_BEGIN; i < SCX_OPI_END; i++) - static_branch_disable(&scx_has_op[i]); - static_branch_disable(&scx_ops_enq_last); - static_branch_disable(&scx_ops_enq_exiting); - static_branch_disable(&scx_ops_enq_migration_disabled); - static_branch_disable(&scx_ops_cpu_preempt); - static_branch_disable(&scx_builtin_idle_enabled); + static_branch_disable(&__scx_enabled); + bitmap_zero(sch->has_op, SCX_OPI_END); + scx_idle_disable(); synchronize_rcu(); if (ei->kind >= SCX_EXIT_ERROR) { pr_err("sched_ext: BPF scheduler \"%s\" disabled (%s)\n", - scx_ops.name, ei->reason); + sch->ops.name, ei->reason); if (ei->msg[0] != '\0') - pr_err("sched_ext: %s: %s\n", scx_ops.name, ei->msg); + pr_err("sched_ext: %s: %s\n", sch->ops.name, ei->msg); #ifdef CONFIG_STACKTRACE stack_trace_print(ei->bt, ei->bt_len, 2); #endif } else { pr_info("sched_ext: BPF scheduler \"%s\" disabled (%s)\n", - scx_ops.name, ei->reason); + sch->ops.name, ei->reason); } - if (scx_ops.exit) - SCX_CALL_OP(SCX_KF_UNLOCKED, exit, ei); + if (sch->ops.exit) + SCX_CALL_OP(sch, exit, NULL, ei); - cancel_delayed_work_sync(&scx_watchdog_work); + scx_unlink_sched(sch); /* - * Delete the kobject from the hierarchy eagerly in addition to just - * dropping a reference. Otherwise, if the object is deleted - * asynchronously, sysfs could observe an object of the same name still - * in the hierarchy when another scheduler is loaded. + * scx_root clearing must be inside cpus_read_lock(). See + * handle_hotplug(). */ - kobject_del(scx_root_kobj); - kobject_put(scx_root_kobj); - scx_root_kobj = NULL; - - memset(&scx_ops, 0, sizeof(scx_ops)); - - rhashtable_walk_enter(&dsq_hash, &rht_iter); - do { - rhashtable_walk_start(&rht_iter); - - while ((dsq = rhashtable_walk_next(&rht_iter)) && !IS_ERR(dsq)) - destroy_dsq(dsq->id); - - rhashtable_walk_stop(&rht_iter); - } while (dsq == ERR_PTR(-EAGAIN)); - rhashtable_walk_exit(&rht_iter); + cpus_read_lock(); + RCU_INIT_POINTER(scx_root, NULL); + cpus_read_unlock(); - free_percpu(scx_dsp_ctx); - scx_dsp_ctx = NULL; - scx_dsp_max_batch = 0; + /* + * Delete the kobject from the hierarchy synchronously. Otherwise, sysfs + * could observe an object of the same name still in the hierarchy when + * the next scheduler is loaded. + */ +#ifdef CONFIG_EXT_SUB_SCHED + if (sch->sub_kset) + kobject_del(&sch->sub_kset->kobj); +#endif + kobject_del(&sch->kobj); - free_exit_info(scx_exit_info); - scx_exit_info = NULL; + free_kick_syncs(); - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_DISABLED) != - SCX_OPS_DISABLING); + WARN_ON_ONCE(scx_set_enable_state(SCX_DISABLED) != SCX_DISABLING); done: - scx_ops_bypass(false); + scx_bypass(sch, false); } -static DEFINE_KTHREAD_WORK(scx_ops_disable_work, scx_ops_disable_workfn); - -static void schedule_scx_ops_disable_work(void) +/* + * Claim the exit on @sch. The caller must ensure that the helper kthread work + * is kicked before the current task can be preempted. Once exit_kind is + * claimed, scx_error() can no longer trigger, so if the current task gets + * preempted and the BPF scheduler fails to schedule it back, the helper work + * will never be kicked and the whole system can wedge. + */ +static bool scx_claim_exit(struct scx_sched *sch, enum scx_exit_kind kind) { - struct kthread_worker *helper = READ_ONCE(scx_ops_helper); + int none = SCX_EXIT_NONE; + + lockdep_assert_preemption_disabled(); + + if (WARN_ON_ONCE(kind == SCX_EXIT_NONE || kind == SCX_EXIT_DONE)) + kind = SCX_EXIT_ERROR; + + if (!atomic_try_cmpxchg(&sch->exit_kind, &none, kind)) + return false; /* - * We may be called spuriously before the first bpf_sched_ext_reg(). If - * scx_ops_helper isn't set up yet, there's nothing to do. + * Some CPUs may be trapped in the dispatch paths. Set the aborting + * flag to break potential live-lock scenarios, ensuring we can + * successfully reach scx_bypass(). */ - if (helper) - kthread_queue_work(helper, &scx_ops_disable_work); + WRITE_ONCE(sch->aborting, true); + + /* + * Propagate exits to descendants immediately. Each has a dedicated + * helper kthread and can run in parallel. While most of disabling is + * serialized, running them in separate threads allows parallelizing + * ops.exit(), which can take arbitrarily long prolonging bypass mode. + * + * To guarantee forward progress, this propagation must be in-line so + * that ->aborting is synchronously asserted for all sub-scheds. The + * propagation is also the interlocking point against sub-sched + * attachment. See scx_link_sched(). + * + * This doesn't cause recursions as propagation only takes place for + * non-propagation exits. + */ + if (kind != SCX_EXIT_PARENT) { + scoped_guard (raw_spinlock_irqsave, &scx_sched_lock) { + struct scx_sched *pos; + scx_for_each_descendant_pre(pos, sch) + scx_disable(pos, SCX_EXIT_PARENT); + } + } + + return true; } -static void scx_ops_disable(enum scx_exit_kind kind) +static void scx_disable_workfn(struct kthread_work *work) { - int none = SCX_EXIT_NONE; + struct scx_sched *sch = container_of(work, struct scx_sched, disable_work); + struct scx_exit_info *ei = sch->exit_info; + int kind; - if (WARN_ON_ONCE(kind == SCX_EXIT_NONE || kind == SCX_EXIT_DONE)) - kind = SCX_EXIT_ERROR; + kind = atomic_read(&sch->exit_kind); + while (true) { + if (kind == SCX_EXIT_DONE) /* already disabled? */ + return; + WARN_ON_ONCE(kind == SCX_EXIT_NONE); + if (atomic_try_cmpxchg(&sch->exit_kind, &kind, SCX_EXIT_DONE)) + break; + } + ei->kind = kind; + ei->reason = scx_exit_reason(ei->kind); - atomic_try_cmpxchg(&scx_exit_kind, &none, kind); + if (scx_parent(sch)) + scx_sub_disable(sch); + else + scx_root_disable(sch); +} - schedule_scx_ops_disable_work(); +static void scx_disable(struct scx_sched *sch, enum scx_exit_kind kind) +{ + guard(preempt)(); + if (scx_claim_exit(sch, kind)) + irq_work_queue(&sch->disable_irq_work); +} + +/** + * scx_flush_disable_work - flush the disable work and wait for it to finish + * @sch: the scheduler + * + * sch->disable_work might still not queued, causing kthread_flush_work() + * as a noop. Syncing the irq_work first is required to guarantee the + * kthread work has been queued before waiting for it. + */ +static void scx_flush_disable_work(struct scx_sched *sch) +{ + int kind; + + do { + irq_work_sync(&sch->disable_irq_work); + kthread_flush_work(&sch->disable_work); + kind = atomic_read(&sch->exit_kind); + } while (kind != SCX_EXIT_NONE && kind != SCX_EXIT_DONE); } static void dump_newline(struct seq_buf *s) @@ -5200,14 +6158,14 @@ static __printf(2, 3) void dump_line(struct seq_buf *s, const char *fmt, ...) #ifdef CONFIG_TRACEPOINTS if (trace_sched_ext_dump_enabled()) { - /* protected by scx_dump_state()::dump_lock */ + /* protected by scx_dump_lock */ static char line_buf[SCX_EXIT_MSG_LEN]; va_start(args, fmt); vscnprintf(line_buf, sizeof(line_buf), fmt, args); va_end(args); - trace_sched_ext_dump(line_buf); + trace_call__sched_ext_dump(line_buf); } #endif /* @s may be zero sized and seq_buf triggers WARN if so */ @@ -5296,35 +6254,49 @@ static void ops_dump_exit(void) scx_dump_data.cpu = -1; } -static void scx_dump_task(struct seq_buf *s, struct scx_dump_ctx *dctx, - struct task_struct *p, char marker) +static void scx_dump_task(struct scx_sched *sch, struct seq_buf *s, struct scx_dump_ctx *dctx, + struct rq *rq, struct task_struct *p, char marker) { static unsigned long bt[SCX_EXIT_BT_LEN]; + struct scx_sched *task_sch = scx_task_sched(p); + const char *own_marker; + char sch_id_buf[32]; char dsq_id_buf[19] = "(n/a)"; unsigned long ops_state = atomic_long_read(&p->scx.ops_state); unsigned int bt_len = 0; + own_marker = task_sch == sch ? "*" : ""; + + if (task_sch->level == 0) + scnprintf(sch_id_buf, sizeof(sch_id_buf), "root"); + else + scnprintf(sch_id_buf, sizeof(sch_id_buf), "sub%d-%llu", + task_sch->level, task_sch->ops.sub_cgroup_id); + if (p->scx.dsq) scnprintf(dsq_id_buf, sizeof(dsq_id_buf), "0x%llx", (unsigned long long)p->scx.dsq->id); dump_newline(s); - dump_line(s, " %c%c %s[%d] %+ldms", + dump_line(s, " %c%c %s[%d] %s%s %+ldms", marker, task_state_to_char(p), p->comm, p->pid, + own_marker, sch_id_buf, jiffies_delta_msecs(p->scx.runnable_at, dctx->at_jiffies)); dump_line(s, " scx_state/flags=%u/0x%x dsq_flags=0x%x ops_state/qseq=%lu/%lu", - scx_get_task_state(p), p->scx.flags & ~SCX_TASK_STATE_MASK, + scx_get_task_state(p) >> SCX_TASK_STATE_SHIFT, + p->scx.flags & ~SCX_TASK_STATE_MASK, p->scx.dsq_flags, ops_state & SCX_OPSS_STATE_MASK, ops_state >> SCX_OPSS_QSEQ_SHIFT); dump_line(s, " sticky/holding_cpu=%d/%d dsq_id=%s", p->scx.sticky_cpu, p->scx.holding_cpu, dsq_id_buf); dump_line(s, " dsq_vtime=%llu slice=%llu weight=%u", p->scx.dsq_vtime, p->scx.slice, p->scx.weight); - dump_line(s, " cpus=%*pb", cpumask_pr_args(p->cpus_ptr)); + dump_line(s, " cpus=%*pb no_mig=%u", cpumask_pr_args(p->cpus_ptr), + p->migration_disabled); - if (SCX_HAS_OP(dump_task)) { + if (SCX_HAS_OP(sch, dump_task)) { ops_dump_init(s, " "); - SCX_CALL_OP(SCX_KF_REST, dump_task, dctx, p); + SCX_CALL_OP(sch, dump_task, rq, dctx, p); ops_dump_exit(); } @@ -5337,9 +6309,16 @@ static void scx_dump_task(struct seq_buf *s, struct scx_dump_ctx *dctx, } } -static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) +/* + * Dump scheduler state. If @dump_all_tasks is true, dump all tasks regardless + * of which scheduler they belong to. If false, only dump tasks owned by @sch. + * For SysRq-D dumps, @dump_all_tasks=false since all schedulers are dumped + * separately. For error dumps, @dump_all_tasks=true since only the failing + * scheduler is dumped. + */ +static void scx_dump_state(struct scx_sched *sch, struct scx_exit_info *ei, + size_t dump_len, bool dump_all_tasks) { - static DEFINE_SPINLOCK(dump_lock); static const char trunc_marker[] = "\n\n~~~~ TRUNCATED ~~~~\n"; struct scx_dump_ctx dctx = { .kind = ei->kind, @@ -5349,14 +6328,25 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) .at_jiffies = jiffies, }; struct seq_buf s; - unsigned long flags; + struct scx_event_stats events; char *buf; int cpu; - spin_lock_irqsave(&dump_lock, flags); + guard(raw_spinlock_irqsave)(&scx_dump_lock); + + if (sch->dump_disabled) + return; seq_buf_init(&s, ei->dump, dump_len); +#ifdef CONFIG_EXT_SUB_SCHED + if (sch->level == 0) + dump_line(&s, "%s: root", sch->ops.name); + else + dump_line(&s, "%s: sub%d-%llu %s", + sch->ops.name, sch->level, sch->ops.sub_cgroup_id, + sch->cgrp_path); +#endif if (ei->kind == SCX_EXIT_NONE) { dump_line(&s, "Debug dump triggered by %s", ei->reason); } else { @@ -5368,9 +6358,9 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) dump_stack_trace(&s, " ", ei->bt, ei->bt_len); } - if (SCX_HAS_OP(dump)) { + if (SCX_HAS_OP(sch, dump)) { ops_dump_init(&s, ""); - SCX_CALL_OP(SCX_KF_UNLOCKED, dump, &dctx); + SCX_CALL_OP(sch, dump, NULL, &dctx); ops_dump_exit(); } @@ -5386,12 +6376,12 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) size_t avail, used; bool idle; - rq_lock(rq, &rf); + rq_lock_irqsave(rq, &rf); idle = list_empty(&rq->scx.runnable_list) && rq->curr->sched_class == &idle_sched_class; - if (idle && !SCX_HAS_OP(dump_cpu)) + if (idle && !SCX_HAS_OP(sch, dump_cpu)) goto next; /* @@ -5404,10 +6394,10 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) seq_buf_init(&ns, buf, avail); dump_newline(&ns); - dump_line(&ns, "CPU %-4d: nr_run=%u flags=0x%x cpu_rel=%d ops_qseq=%lu pnt_seq=%lu", + dump_line(&ns, "CPU %-4d: nr_run=%u flags=0x%x cpu_rel=%d ops_qseq=%lu ksync=%lu", cpu, rq->scx.nr_running, rq->scx.flags, rq->scx.cpu_released, rq->scx.ops_qseq, - rq->scx.pnt_seq); + rq->scx.kick_sync); dump_line(&ns, " curr=%s[%d] class=%ps", rq->curr->comm, rq->curr->pid, rq->curr->sched_class); @@ -5423,11 +6413,14 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) if (!cpumask_empty(rq->scx.cpus_to_wait)) dump_line(&ns, " cpus_to_wait : %*pb", cpumask_pr_args(rq->scx.cpus_to_wait)); + if (!cpumask_empty(rq->scx.cpus_to_sync)) + dump_line(&ns, " cpus_to_sync : %*pb", + cpumask_pr_args(rq->scx.cpus_to_sync)); used = seq_buf_used(&ns); - if (SCX_HAS_OP(dump_cpu)) { + if (SCX_HAS_OP(sch, dump_cpu)) { ops_dump_init(&ns, " "); - SCX_CALL_OP(SCX_KF_REST, dump_cpu, &dctx, cpu, idle); + SCX_CALL_OP(sch, dump_cpu, rq, &dctx, cpu, idle); ops_dump_exit(); } @@ -5448,75 +6441,321 @@ static void scx_dump_state(struct scx_exit_info *ei, size_t dump_len) seq_buf_set_overflow(&s); } - if (rq->curr->sched_class == &ext_sched_class) - scx_dump_task(&s, &dctx, rq->curr, '*'); + if (rq->curr->sched_class == &ext_sched_class && + (dump_all_tasks || scx_task_on_sched(sch, rq->curr))) + scx_dump_task(sch, &s, &dctx, rq, rq->curr, '*'); list_for_each_entry(p, &rq->scx.runnable_list, scx.runnable_node) - scx_dump_task(&s, &dctx, p, ' '); + if (dump_all_tasks || scx_task_on_sched(sch, p)) + scx_dump_task(sch, &s, &dctx, rq, p, ' '); next: - rq_unlock(rq, &rf); + rq_unlock_irqrestore(rq, &rf); } + dump_newline(&s); + dump_line(&s, "Event counters"); + dump_line(&s, "--------------"); + + scx_read_events(sch, &events); + scx_dump_event(s, &events, SCX_EV_SELECT_CPU_FALLBACK); + scx_dump_event(s, &events, SCX_EV_DISPATCH_LOCAL_DSQ_OFFLINE); + scx_dump_event(s, &events, SCX_EV_DISPATCH_KEEP_LAST); + scx_dump_event(s, &events, SCX_EV_ENQ_SKIP_EXITING); + scx_dump_event(s, &events, SCX_EV_ENQ_SKIP_MIGRATION_DISABLED); + scx_dump_event(s, &events, SCX_EV_REENQ_IMMED); + scx_dump_event(s, &events, SCX_EV_REENQ_LOCAL_REPEAT); + scx_dump_event(s, &events, SCX_EV_REFILL_SLICE_DFL); + scx_dump_event(s, &events, SCX_EV_BYPASS_DURATION); + scx_dump_event(s, &events, SCX_EV_BYPASS_DISPATCH); + scx_dump_event(s, &events, SCX_EV_BYPASS_ACTIVATE); + scx_dump_event(s, &events, SCX_EV_INSERT_NOT_OWNED); + scx_dump_event(s, &events, SCX_EV_SUB_BYPASS_DISPATCH); + if (seq_buf_has_overflowed(&s) && dump_len >= sizeof(trunc_marker)) memcpy(ei->dump + dump_len - sizeof(trunc_marker), trunc_marker, sizeof(trunc_marker)); - - spin_unlock_irqrestore(&dump_lock, flags); } -static void scx_ops_error_irq_workfn(struct irq_work *irq_work) +static void scx_disable_irq_workfn(struct irq_work *irq_work) { - struct scx_exit_info *ei = scx_exit_info; + struct scx_sched *sch = container_of(irq_work, struct scx_sched, disable_irq_work); + struct scx_exit_info *ei = sch->exit_info; if (ei->kind >= SCX_EXIT_ERROR) - scx_dump_state(ei, scx_ops.exit_dump_len); + scx_dump_state(sch, ei, sch->ops.exit_dump_len, true); - schedule_scx_ops_disable_work(); + kthread_queue_work(sch->helper, &sch->disable_work); } -static DEFINE_IRQ_WORK(scx_ops_error_irq_work, scx_ops_error_irq_workfn); - -static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind, - s64 exit_code, - const char *fmt, ...) +static bool scx_vexit(struct scx_sched *sch, + enum scx_exit_kind kind, s64 exit_code, + const char *fmt, va_list args) { - struct scx_exit_info *ei = scx_exit_info; - int none = SCX_EXIT_NONE; - va_list args; + struct scx_exit_info *ei = sch->exit_info; - if (!atomic_try_cmpxchg(&scx_exit_kind, &none, kind)) - return; + guard(preempt)(); + + if (!scx_claim_exit(sch, kind)) + return false; ei->exit_code = exit_code; #ifdef CONFIG_STACKTRACE if (kind >= SCX_EXIT_ERROR) ei->bt_len = stack_trace_save(ei->bt, SCX_EXIT_BT_LEN, 1); #endif - va_start(args, fmt); vscnprintf(ei->msg, SCX_EXIT_MSG_LEN, fmt, args); - va_end(args); /* * Set ei->kind and ->reason for scx_dump_state(). They'll be set again - * in scx_ops_disable_workfn(). + * in scx_disable_workfn(). */ ei->kind = kind; ei->reason = scx_exit_reason(ei->kind); - irq_work_queue(&scx_ops_error_irq_work); + irq_work_queue(&sch->disable_irq_work); + return true; +} + +static int alloc_kick_syncs(void) +{ + int cpu; + + /* + * Allocate per-CPU arrays sized by nr_cpu_ids. Use kvzalloc as size + * can exceed percpu allocator limits on large machines. + */ + for_each_possible_cpu(cpu) { + struct scx_kick_syncs **ksyncs = per_cpu_ptr(&scx_kick_syncs, cpu); + struct scx_kick_syncs *new_ksyncs; + + WARN_ON_ONCE(rcu_access_pointer(*ksyncs)); + + new_ksyncs = kvzalloc_node(struct_size(new_ksyncs, syncs, nr_cpu_ids), + GFP_KERNEL, cpu_to_node(cpu)); + if (!new_ksyncs) { + free_kick_syncs(); + return -ENOMEM; + } + + rcu_assign_pointer(*ksyncs, new_ksyncs); + } + + return 0; +} + +static void free_pnode(struct scx_sched_pnode *pnode) +{ + if (!pnode) + return; + exit_dsq(&pnode->global_dsq); + kfree(pnode); } -static struct kthread_worker *scx_create_rt_helper(const char *name) +static struct scx_sched_pnode *alloc_pnode(struct scx_sched *sch, int node) { - struct kthread_worker *helper; + struct scx_sched_pnode *pnode; + + pnode = kzalloc_node(sizeof(*pnode), GFP_KERNEL, node); + if (!pnode) + return NULL; + + if (init_dsq(&pnode->global_dsq, SCX_DSQ_GLOBAL, sch)) { + kfree(pnode); + return NULL; + } - helper = kthread_run_worker(0, name); - if (helper) - sched_set_fifo(helper->task); - return helper; + return pnode; } -static void check_hotplug_seq(const struct sched_ext_ops *ops) +/* + * Allocate and initialize a new scx_sched. @cgrp's reference is always + * consumed whether the function succeeds or fails. + */ +static struct scx_sched *scx_alloc_and_add_sched(struct sched_ext_ops *ops, + struct cgroup *cgrp, + struct scx_sched *parent) +{ + struct scx_sched *sch; + s32 level = parent ? parent->level + 1 : 0; + s32 node, cpu, ret, bypass_fail_cpu = nr_cpu_ids; + + sch = kzalloc_flex(*sch, ancestors, level + 1); + if (!sch) { + ret = -ENOMEM; + goto err_put_cgrp; + } + + sch->exit_info = alloc_exit_info(ops->exit_dump_len); + if (!sch->exit_info) { + ret = -ENOMEM; + goto err_free_sch; + } + + ret = rhashtable_init(&sch->dsq_hash, &dsq_hash_params); + if (ret < 0) + goto err_free_ei; + + sch->pnode = kzalloc_objs(sch->pnode[0], nr_node_ids); + if (!sch->pnode) { + ret = -ENOMEM; + goto err_free_hash; + } + + for_each_node_state(node, N_POSSIBLE) { + sch->pnode[node] = alloc_pnode(sch, node); + if (!sch->pnode[node]) { + ret = -ENOMEM; + goto err_free_pnode; + } + } + + sch->dsp_max_batch = ops->dispatch_max_batch ?: SCX_DSP_DFL_MAX_BATCH; + sch->pcpu = __alloc_percpu(struct_size_t(struct scx_sched_pcpu, + dsp_ctx.buf, sch->dsp_max_batch), + __alignof__(struct scx_sched_pcpu)); + if (!sch->pcpu) { + ret = -ENOMEM; + goto err_free_pnode; + } + + for_each_possible_cpu(cpu) { + ret = init_dsq(bypass_dsq(sch, cpu), SCX_DSQ_BYPASS, sch); + if (ret) { + bypass_fail_cpu = cpu; + goto err_free_pcpu; + } + } + + for_each_possible_cpu(cpu) { + struct scx_sched_pcpu *pcpu = per_cpu_ptr(sch->pcpu, cpu); + + pcpu->sch = sch; + INIT_LIST_HEAD(&pcpu->deferred_reenq_local.node); + } + + sch->helper = kthread_run_worker(0, "sched_ext_helper"); + if (IS_ERR(sch->helper)) { + ret = PTR_ERR(sch->helper); + goto err_free_pcpu; + } + + sched_set_fifo(sch->helper->task); + + if (parent) + memcpy(sch->ancestors, parent->ancestors, + level * sizeof(parent->ancestors[0])); + sch->ancestors[level] = sch; + sch->level = level; + + if (ops->timeout_ms) + sch->watchdog_timeout = msecs_to_jiffies(ops->timeout_ms); + else + sch->watchdog_timeout = SCX_WATCHDOG_MAX_TIMEOUT; + + sch->slice_dfl = SCX_SLICE_DFL; + atomic_set(&sch->exit_kind, SCX_EXIT_NONE); + sch->disable_irq_work = IRQ_WORK_INIT_HARD(scx_disable_irq_workfn); + kthread_init_work(&sch->disable_work, scx_disable_workfn); + timer_setup(&sch->bypass_lb_timer, scx_bypass_lb_timerfn, 0); + + if (!alloc_cpumask_var(&sch->bypass_lb_donee_cpumask, GFP_KERNEL)) { + ret = -ENOMEM; + goto err_stop_helper; + } + if (!alloc_cpumask_var(&sch->bypass_lb_resched_cpumask, GFP_KERNEL)) { + ret = -ENOMEM; + goto err_free_lb_cpumask; + } + sch->ops = *ops; + rcu_assign_pointer(ops->priv, sch); + + sch->kobj.kset = scx_kset; + INIT_LIST_HEAD(&sch->all); + +#ifdef CONFIG_EXT_SUB_SCHED + char *buf = kzalloc(PATH_MAX, GFP_KERNEL); + if (!buf) { + ret = -ENOMEM; + goto err_free_lb_resched; + } + cgroup_path(cgrp, buf, PATH_MAX); + sch->cgrp_path = kstrdup(buf, GFP_KERNEL); + kfree(buf); + if (!sch->cgrp_path) { + ret = -ENOMEM; + goto err_free_lb_resched; + } + + sch->cgrp = cgrp; + INIT_LIST_HEAD(&sch->children); + INIT_LIST_HEAD(&sch->sibling); + + if (parent) + ret = kobject_init_and_add(&sch->kobj, &scx_ktype, + &parent->sub_kset->kobj, + "sub-%llu", cgroup_id(cgrp)); + else + ret = kobject_init_and_add(&sch->kobj, &scx_ktype, NULL, "root"); + + if (ret < 0) { + RCU_INIT_POINTER(ops->priv, NULL); + kobject_put(&sch->kobj); + return ERR_PTR(ret); + } + + if (ops->sub_attach) { + sch->sub_kset = kset_create_and_add("sub", NULL, &sch->kobj); + if (!sch->sub_kset) { + RCU_INIT_POINTER(ops->priv, NULL); + kobject_put(&sch->kobj); + return ERR_PTR(-ENOMEM); + } + } +#else /* CONFIG_EXT_SUB_SCHED */ + ret = kobject_init_and_add(&sch->kobj, &scx_ktype, NULL, "root"); + if (ret < 0) { + RCU_INIT_POINTER(ops->priv, NULL); + kobject_put(&sch->kobj); + return ERR_PTR(ret); + } +#endif /* CONFIG_EXT_SUB_SCHED */ + return sch; + +#ifdef CONFIG_EXT_SUB_SCHED +err_free_lb_resched: + RCU_INIT_POINTER(ops->priv, NULL); + free_cpumask_var(sch->bypass_lb_resched_cpumask); +#endif +err_free_lb_cpumask: + free_cpumask_var(sch->bypass_lb_donee_cpumask); +err_stop_helper: + kthread_destroy_worker(sch->helper); +err_free_pcpu: + for_each_possible_cpu(cpu) { + if (cpu == bypass_fail_cpu) + break; + exit_dsq(bypass_dsq(sch, cpu)); + } + free_percpu(sch->pcpu); +err_free_pnode: + for_each_node_state(node, N_POSSIBLE) + free_pnode(sch->pnode[node]); + kfree(sch->pnode); +err_free_hash: + rhashtable_free_and_destroy(&sch->dsq_hash, NULL, NULL); +err_free_ei: + free_exit_info(sch->exit_info); +err_free_sch: + kfree(sch); +err_put_cgrp: +#ifdef CONFIG_EXT_SUB_SCHED + cgroup_put(cgrp); +#endif + return ERR_PTR(ret); +} + +static int check_hotplug_seq(struct scx_sched *sch, + const struct sched_ext_ops *ops) { unsigned long long global_hotplug_seq; @@ -5528,117 +6767,115 @@ static void check_hotplug_seq(const struct sched_ext_ops *ops) if (ops->hotplug_seq) { global_hotplug_seq = atomic_long_read(&scx_hotplug_seq); if (ops->hotplug_seq != global_hotplug_seq) { - scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, - "expected hotplug seq %llu did not match actual %llu", - ops->hotplug_seq, global_hotplug_seq); + scx_exit(sch, SCX_EXIT_UNREG_KERN, + SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG, + "expected hotplug seq %llu did not match actual %llu", + ops->hotplug_seq, global_hotplug_seq); + return -EBUSY; } } + + return 0; } -static int validate_ops(const struct sched_ext_ops *ops) +static int validate_ops(struct scx_sched *sch, const struct sched_ext_ops *ops) { /* * It doesn't make sense to specify the SCX_OPS_ENQ_LAST flag if the * ops.enqueue() callback isn't implemented. */ if ((ops->flags & SCX_OPS_ENQ_LAST) && !ops->enqueue) { - scx_ops_error("SCX_OPS_ENQ_LAST requires ops.enqueue() to be implemented"); + scx_error(sch, "SCX_OPS_ENQ_LAST requires ops.enqueue() to be implemented"); return -EINVAL; } - return 0; -} - -static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) -{ - struct scx_task_iter sti; - struct task_struct *p; - unsigned long timeout; - int i, cpu, node, ret; - - if (!cpumask_equal(housekeeping_cpumask(HK_TYPE_DOMAIN), - cpu_possible_mask)) { - pr_err("sched_ext: Not compatible with \"isolcpus=\" domain isolation\n"); + /* + * SCX_OPS_BUILTIN_IDLE_PER_NODE requires built-in CPU idle + * selection policy to be enabled. + */ + if ((ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE) && + (ops->update_idle && !(ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE))) { + scx_error(sch, "SCX_OPS_BUILTIN_IDLE_PER_NODE requires CPU idle selection enabled"); return -EINVAL; } - mutex_lock(&scx_ops_enable_mutex); - - if (!scx_ops_helper) { - WRITE_ONCE(scx_ops_helper, - scx_create_rt_helper("sched_ext_ops_helper")); - if (!scx_ops_helper) { - ret = -ENOMEM; - goto err_unlock; - } - } - - if (!global_dsqs) { - struct scx_dispatch_q **dsqs; - - dsqs = kcalloc(nr_node_ids, sizeof(dsqs[0]), GFP_KERNEL); - if (!dsqs) { - ret = -ENOMEM; - goto err_unlock; - } + if (ops->cpu_acquire || ops->cpu_release) + pr_warn("ops->cpu_acquire/release() are deprecated, use sched_switch TP instead\n"); - for_each_node_state(node, N_POSSIBLE) { - struct scx_dispatch_q *dsq; + return 0; +} - dsq = kzalloc_node(sizeof(*dsq), GFP_KERNEL, node); - if (!dsq) { - for_each_node_state(node, N_POSSIBLE) - kfree(dsqs[node]); - kfree(dsqs); - ret = -ENOMEM; - goto err_unlock; - } +/* + * scx_enable() is offloaded to a dedicated system-wide RT kthread to avoid + * starvation. During the READY -> ENABLED task switching loop, the calling + * thread's sched_class gets switched from fair to ext. As fair has higher + * priority than ext, the calling thread can be indefinitely starved under + * fair-class saturation, leading to a system hang. + */ +struct scx_enable_cmd { + struct kthread_work work; + struct sched_ext_ops *ops; + int ret; +}; - init_dsq(dsq, SCX_DSQ_GLOBAL); - dsqs[node] = dsq; - } +static void scx_root_enable_workfn(struct kthread_work *work) +{ + struct scx_enable_cmd *cmd = container_of(work, struct scx_enable_cmd, work); + struct sched_ext_ops *ops = cmd->ops; + struct cgroup *cgrp = root_cgroup(); + struct scx_sched *sch; + struct scx_task_iter sti; + struct task_struct *p; + int i, cpu, ret; - global_dsqs = dsqs; - } + mutex_lock(&scx_enable_mutex); - if (scx_ops_enable_state() != SCX_OPS_DISABLED) { + if (scx_enable_state() != SCX_DISABLED) { ret = -EBUSY; goto err_unlock; } - scx_root_kobj = kzalloc(sizeof(*scx_root_kobj), GFP_KERNEL); - if (!scx_root_kobj) { - ret = -ENOMEM; + /* + * @ops->priv binds @ops to its scx_sched instance. It is set here by + * scx_alloc_and_add_sched() and cleared at the tail of bpf_scx_unreg(), + * which runs after scx_root_disable() has dropped scx_enable_mutex. If + * it's still non-NULL here, a previous attachment on @ops has not + * finished tearing down; proceeding would let the in-flight unreg's + * RCU_INIT_POINTER(NULL) clobber the @ops->priv we are about to assign. + */ + if (rcu_access_pointer(ops->priv)) { + ret = -EBUSY; goto err_unlock; } - scx_root_kobj->kset = scx_kset; - ret = kobject_init_and_add(scx_root_kobj, &scx_ktype, NULL, "root"); - if (ret < 0) - goto err; + ret = alloc_kick_syncs(); + if (ret) + goto err_unlock; - scx_exit_info = alloc_exit_info(ops->exit_dump_len); - if (!scx_exit_info) { - ret = -ENOMEM; - goto err_del; +#ifdef CONFIG_EXT_SUB_SCHED + cgroup_get(cgrp); +#endif + sch = scx_alloc_and_add_sched(ops, cgrp, NULL); + if (IS_ERR(sch)) { + ret = PTR_ERR(sch); + goto err_free_ksyncs; } /* - * Set scx_ops, transition to ENABLING and clear exit info to arm the - * disable path. Failure triggers full disabling from here on. + * Transition to ENABLING and clear exit info to arm the disable path. + * Failure triggers full disabling from here on. */ - scx_ops = *ops; - - WARN_ON_ONCE(scx_ops_set_enable_state(SCX_OPS_ENABLING) != - SCX_OPS_DISABLED); - - atomic_set(&scx_exit_kind, SCX_EXIT_NONE); - scx_warned_zero_slice = false; + WARN_ON_ONCE(scx_set_enable_state(SCX_ENABLING) != SCX_DISABLED); + WARN_ON_ONCE(scx_root); atomic_long_set(&scx_nr_rejected, 0); - for_each_possible_cpu(cpu) - cpu_rq(cpu)->scx.cpuperf_target = SCX_CPUPERF_ONE; + for_each_possible_cpu(cpu) { + struct rq *rq = cpu_rq(cpu); + + rq->scx.local_dsq.sched = sch; + rq->scx.cpuperf_target = SCX_CPUPERF_ONE; + } /* * Keep CPUs stable during enable so that the BPF scheduler can track @@ -5646,78 +6883,62 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) */ cpus_read_lock(); - if (scx_ops.init) { - ret = SCX_CALL_OP_RET(SCX_KF_UNLOCKED, init); + /* + * Make the scheduler instance visible. Must be inside cpus_read_lock(). + * See handle_hotplug(). + */ + rcu_assign_pointer(scx_root, sch); + + ret = scx_link_sched(sch); + if (ret) { + cpus_read_unlock(); + goto err_disable; + } + + scx_idle_enable(ops); + + if (sch->ops.init) { + ret = SCX_CALL_OP_RET(sch, init, NULL); if (ret) { - ret = ops_sanitize_err("init", ret); + ret = ops_sanitize_err(sch, "init", ret); cpus_read_unlock(); - scx_ops_error("ops.init() failed (%d)", ret); + scx_error(sch, "ops.init() failed (%d)", ret); goto err_disable; } + sch->exit_info->flags |= SCX_EFLAG_INITIALIZED; } for (i = SCX_OPI_CPU_HOTPLUG_BEGIN; i < SCX_OPI_CPU_HOTPLUG_END; i++) if (((void (**)(void))ops)[i]) - static_branch_enable_cpuslocked(&scx_has_op[i]); - - check_hotplug_seq(ops); -#ifdef CONFIG_SMP - update_selcpu_topology(); -#endif - cpus_read_unlock(); - - ret = validate_ops(ops); - if (ret) - goto err_disable; + set_bit(i, sch->has_op); - WARN_ON_ONCE(scx_dsp_ctx); - scx_dsp_max_batch = ops->dispatch_max_batch ?: SCX_DSP_DFL_MAX_BATCH; - scx_dsp_ctx = __alloc_percpu(struct_size_t(struct scx_dsp_ctx, buf, - scx_dsp_max_batch), - __alignof__(struct scx_dsp_ctx)); - if (!scx_dsp_ctx) { - ret = -ENOMEM; + ret = check_hotplug_seq(sch, ops); + if (ret) { + cpus_read_unlock(); goto err_disable; } + scx_idle_update_selcpu_topology(ops); - if (ops->timeout_ms) - timeout = msecs_to_jiffies(ops->timeout_ms); - else - timeout = SCX_WATCHDOG_MAX_TIMEOUT; + cpus_read_unlock(); - WRITE_ONCE(scx_watchdog_timeout, timeout); - WRITE_ONCE(scx_watchdog_timestamp, jiffies); - queue_delayed_work(system_unbound_wq, &scx_watchdog_work, - scx_watchdog_timeout / 2); + ret = validate_ops(sch, ops); + if (ret) + goto err_disable; /* - * Once __scx_ops_enabled is set, %current can be switched to SCX - * anytime. This can lead to stalls as some BPF schedulers (e.g. - * userspace scheduling) may not function correctly before all tasks are - * switched. Init in bypass mode to guarantee forward progress. + * Once __scx_enabled is set, %current can be switched to SCX anytime. + * This can lead to stalls as some BPF schedulers (e.g. userspace + * scheduling) may not function correctly before all tasks are switched. + * Init in bypass mode to guarantee forward progress. */ - scx_ops_bypass(true); + scx_bypass(sch, true); for (i = SCX_OPI_NORMAL_BEGIN; i < SCX_OPI_NORMAL_END; i++) if (((void (**)(void))ops)[i]) - static_branch_enable(&scx_has_op[i]); - - if (ops->flags & SCX_OPS_ENQ_LAST) - static_branch_enable(&scx_ops_enq_last); - - if (ops->flags & SCX_OPS_ENQ_EXITING) - static_branch_enable(&scx_ops_enq_exiting); - if (ops->flags & SCX_OPS_ENQ_MIGRATION_DISABLED) - static_branch_enable(&scx_ops_enq_migration_disabled); - if (scx_ops.cpu_acquire || scx_ops.cpu_release) - static_branch_enable(&scx_ops_cpu_preempt); + set_bit(i, sch->has_op); - if (!ops->update_idle || (ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE)) { - reset_idle_masks(); - static_branch_enable(&scx_builtin_idle_enabled); - } else { - static_branch_disable(&scx_builtin_idle_enabled); - } + if (sch->ops.cpu_acquire || sch->ops.cpu_release) + sch->ops.flags |= SCX_OPS_HAS_CPU_PREEMPT; /* * Lock out forks, cgroup on/offlining and moves before opening the @@ -5725,8 +6946,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) */ percpu_down_write(&scx_fork_rwsem); - WARN_ON_ONCE(scx_ops_init_task_enabled); - scx_ops_init_task_enabled = true; + WARN_ON_ONCE(scx_init_task_enabled); + scx_init_task_enabled = true; /* * Enable ops for every task. Fork is excluded by scx_fork_rwsem @@ -5735,19 +6956,23 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) * tasks. Prep all tasks first and then enable them with preemption * disabled. * - * All cgroups should be initialized before scx_ops_init_task() so that - * the BPF scheduler can reliably track each task's cgroup membership - * from scx_ops_init_task(). Lock out cgroup on/offlining and task - * migrations while tasks are being initialized so that - * scx_cgroup_can_attach() never sees uninitialized tasks. + * All cgroups should be initialized before scx_init_task() so that the + * BPF scheduler can reliably track each task's cgroup membership from + * scx_init_task(). Lock out cgroup on/offlining and task migrations + * while tasks are being initialized so that scx_cgroup_can_attach() + * never sees uninitialized tasks. */ scx_cgroup_lock(); - ret = scx_cgroup_init(); + set_cgroup_sched(sch_cgroup(sch), sch); + ret = scx_cgroup_init(sch); if (ret) goto err_disable_unlock_all; - scx_task_iter_start(&sti); + scx_task_iter_start(&sti, NULL); while ((p = scx_task_iter_next_locked(&sti))) { + struct rq_flags rf; + struct rq *rq; + /* * @p may already be dead, have lost all its usages counts and * be waiting for RCU grace period before being freed. @p can't @@ -5756,22 +6981,48 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) if (!tryget_task_struct(p)) continue; + /* + * Set %INIT_BEGIN under the iter's rq lock so that a concurrent + * sched_ext_dead() does not call ops.exit_task() on @p while + * ops.init_task() is running. If sched_ext_dead() runs before + * this store, it has already removed @p from scx_tasks and the + * iter won't visit @p; if it runs after, it observes + * %INIT_BEGIN and transitions to %DEAD without calling ops, + * leaving the post-init recheck below to unwind. + */ + scx_set_task_state(p, SCX_TASK_INIT_BEGIN); scx_task_iter_unlock(&sti); - ret = scx_ops_init_task(p, task_group(p), false); - if (ret) { - put_task_struct(p); - scx_task_iter_relock(&sti); + ret = __scx_init_task(sch, p, false); + + rq = task_rq_lock(p, &rf); + + if (unlikely(ret)) { + if (scx_get_task_state(p) != SCX_TASK_DEAD) + scx_set_task_state(p, SCX_TASK_NONE); + task_rq_unlock(rq, p, &rf); scx_task_iter_stop(&sti); - scx_ops_error("ops.init_task() failed (%d) for %s[%d]", - ret, p->comm, p->pid); + scx_error(sch, "ops.init_task() failed (%d) for %s[%d]", + ret, p->comm, p->pid); + put_task_struct(p); goto err_disable_unlock_all; } - scx_set_task_state(p, SCX_TASK_READY); + if (scx_get_task_state(p) == SCX_TASK_DEAD) { + /* + * sched_ext_dead() observed %INIT_BEGIN and set %DEAD. + * ops.exit_task() is owed to the sched __scx_init_task() + * ran against; call it now. + */ + scx_sub_init_cancel_task(sch, p); + } else { + scx_set_task_state(p, SCX_TASK_INIT); + scx_set_task_sched(p, sch); + scx_set_task_state(p, SCX_TASK_READY); + } + task_rq_unlock(rq, p, &rf); put_task_struct(p); - scx_task_iter_relock(&sti); } scx_task_iter_stop(&sti); scx_cgroup_unlock(); @@ -5782,7 +7033,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) * all eligible tasks. */ WRITE_ONCE(scx_switching_all, !(ops->flags & SCX_OPS_SWITCH_PARTIAL)); - static_branch_enable(&__scx_ops_enabled); + static_branch_enable(&__scx_enabled); /* * We're fully committed and can't fail. The task READY -> ENABLED @@ -5790,33 +7041,30 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) * scx_tasks_lock. */ percpu_down_write(&scx_fork_rwsem); - scx_task_iter_start(&sti); + scx_task_iter_start(&sti, NULL); while ((p = scx_task_iter_next_locked(&sti))) { + unsigned int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE; const struct sched_class *old_class = p->sched_class; - const struct sched_class *new_class = - __setscheduler_class(p->policy, p->prio); - struct sched_enq_and_set_ctx ctx; + const struct sched_class *new_class = scx_setscheduler_class(p); - if (old_class != new_class && p->se.sched_delayed) - dequeue_task(task_rq(p), p, DEQUEUE_SLEEP | DEQUEUE_DELAYED); - - sched_deq_and_put_task(p, DEQUEUE_SAVE | DEQUEUE_MOVE, &ctx); - - p->scx.slice = SCX_SLICE_DFL; - p->sched_class = new_class; - check_class_changing(task_rq(p), p, old_class); + if (scx_get_task_state(p) != SCX_TASK_READY) + continue; - sched_enq_and_set_task(&ctx); + if (old_class != new_class) + queue_flags |= DEQUEUE_CLASS; - check_class_changed(task_rq(p), p, old_class, p->prio); + scoped_guard (sched_change, p, queue_flags) { + p->scx.slice = READ_ONCE(sch->slice_dfl); + p->sched_class = new_class; + } } scx_task_iter_stop(&sti); percpu_up_write(&scx_fork_rwsem); - scx_ops_bypass(false); + scx_bypass(sch, false); - if (!scx_ops_tryset_enable_state(SCX_OPS_ENABLED, SCX_OPS_ENABLING)) { - WARN_ON_ONCE(atomic_read(&scx_exit_kind) == SCX_EXIT_NONE); + if (!scx_tryset_enable_state(SCX_ENABLED, SCX_ENABLING)) { + WARN_ON_ONCE(atomic_read(&sch->exit_kind) == SCX_EXIT_NONE); goto err_disable; } @@ -5824,45 +7072,413 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link) static_branch_enable(&__scx_switched_all); pr_info("sched_ext: BPF scheduler \"%s\" enabled%s\n", - scx_ops.name, scx_switched_all() ? "" : " (partial)"); - kobject_uevent(scx_root_kobj, KOBJ_ADD); - mutex_unlock(&scx_ops_enable_mutex); + sch->ops.name, scx_switched_all() ? "" : " (partial)"); + kobject_uevent(&sch->kobj, KOBJ_ADD); + mutex_unlock(&scx_enable_mutex); atomic_long_inc(&scx_enable_seq); - return 0; + cmd->ret = 0; + return; -err_del: - kobject_del(scx_root_kobj); -err: - kobject_put(scx_root_kobj); - scx_root_kobj = NULL; - if (scx_exit_info) { - free_exit_info(scx_exit_info); - scx_exit_info = NULL; - } +err_free_ksyncs: + free_kick_syncs(); err_unlock: - mutex_unlock(&scx_ops_enable_mutex); - return ret; + mutex_unlock(&scx_enable_mutex); + cmd->ret = ret; + return; err_disable_unlock_all: scx_cgroup_unlock(); percpu_up_write(&scx_fork_rwsem); - scx_ops_bypass(false); + /* we'll soon enter disable path, keep bypass on */ err_disable: - mutex_unlock(&scx_ops_enable_mutex); + mutex_unlock(&scx_enable_mutex); /* * Returning an error code here would not pass all the error information - * to userspace. Record errno using scx_ops_error() for cases - * scx_ops_error() wasn't already invoked and exit indicating success so - * that the error is notified through ops.exit() with all the details. + * to userspace. Record errno using scx_error() for cases scx_error() + * wasn't already invoked and exit indicating success so that the error + * is notified through ops.exit() with all the details. * - * Flush scx_ops_disable_work to ensure that error is reported before - * init completion. + * Flush scx_disable_work to ensure that error is reported before init + * completion. sch's base reference will be put by bpf_scx_unreg(). */ - scx_ops_error("scx_ops_enable() failed (%d)", ret); - kthread_flush_work(&scx_ops_disable_work); - return 0; + scx_error(sch, "scx_root_enable() failed (%d)", ret); + scx_flush_disable_work(sch); + cmd->ret = 0; +} + +#ifdef CONFIG_EXT_SUB_SCHED +/* verify that a scheduler can be attached to @cgrp and return the parent */ +static struct scx_sched *find_parent_sched(struct cgroup *cgrp) +{ + struct scx_sched *parent = cgrp->scx_sched; + struct scx_sched *pos; + + lockdep_assert_held(&scx_sched_lock); + + /* can't attach twice to the same cgroup */ + if (parent->cgrp == cgrp) + return ERR_PTR(-EBUSY); + + /* does $parent allow sub-scheds? */ + if (!parent->ops.sub_attach) + return ERR_PTR(-EOPNOTSUPP); + + /* can't insert between $parent and its exiting children */ + list_for_each_entry(pos, &parent->children, sibling) + if (cgroup_is_descendant(pos->cgrp, cgrp)) + return ERR_PTR(-EBUSY); + + return parent; +} + +static bool assert_task_ready_or_enabled(struct task_struct *p) +{ + u32 state = scx_get_task_state(p); + + switch (state) { + case SCX_TASK_READY: + case SCX_TASK_ENABLED: + return true; + default: + WARN_ONCE(true, "sched_ext: Invalid task state %d for %s[%d] during enabling sub sched", + state, p->comm, p->pid); + return false; + } +} + +static void scx_sub_enable_workfn(struct kthread_work *work) +{ + struct scx_enable_cmd *cmd = container_of(work, struct scx_enable_cmd, work); + struct sched_ext_ops *ops = cmd->ops; + struct cgroup *cgrp; + struct scx_sched *parent, *sch; + struct scx_task_iter sti; + struct task_struct *p; + s32 i, ret; + + mutex_lock(&scx_enable_mutex); + + if (!scx_enabled()) { + ret = -ENODEV; + goto out_unlock; + } + + /* See scx_root_enable_workfn() for the @ops->priv check. */ + if (rcu_access_pointer(ops->priv)) { + ret = -EBUSY; + goto out_unlock; + } + + cgrp = cgroup_get_from_id(ops->sub_cgroup_id); + if (IS_ERR(cgrp)) { + ret = PTR_ERR(cgrp); + goto out_unlock; + } + + raw_spin_lock_irq(&scx_sched_lock); + parent = find_parent_sched(cgrp); + if (IS_ERR(parent)) { + raw_spin_unlock_irq(&scx_sched_lock); + ret = PTR_ERR(parent); + goto out_put_cgrp; + } + kobject_get(&parent->kobj); + raw_spin_unlock_irq(&scx_sched_lock); + + /* scx_alloc_and_add_sched() consumes @cgrp whether it succeeds or not */ + sch = scx_alloc_and_add_sched(ops, cgrp, parent); + kobject_put(&parent->kobj); + if (IS_ERR(sch)) { + ret = PTR_ERR(sch); + goto out_unlock; + } + + ret = scx_link_sched(sch); + if (ret) + goto err_disable; + + if (sch->level >= SCX_SUB_MAX_DEPTH) { + scx_error(sch, "max nesting depth %d violated", + SCX_SUB_MAX_DEPTH); + goto err_disable; + } + + if (sch->ops.init) { + ret = SCX_CALL_OP_RET(sch, init, NULL); + if (ret) { + ret = ops_sanitize_err(sch, "init", ret); + scx_error(sch, "ops.init() failed (%d)", ret); + goto err_disable; + } + sch->exit_info->flags |= SCX_EFLAG_INITIALIZED; + } + + if (validate_ops(sch, ops)) + goto err_disable; + + struct scx_sub_attach_args sub_attach_args = { + .ops = &sch->ops, + .cgroup_path = sch->cgrp_path, + }; + + ret = SCX_CALL_OP_RET(parent, sub_attach, NULL, + &sub_attach_args); + if (ret) { + ret = ops_sanitize_err(sch, "sub_attach", ret); + scx_error(sch, "parent rejected (%d)", ret); + goto err_disable; + } + sch->sub_attached = true; + + scx_bypass(sch, true); + + for (i = SCX_OPI_BEGIN; i < SCX_OPI_END; i++) + if (((void (**)(void))ops)[i]) + set_bit(i, sch->has_op); + + percpu_down_write(&scx_fork_rwsem); + scx_cgroup_lock(); + + /* + * Set cgroup->scx_sched's and check CSS_ONLINE. Either we see + * !CSS_ONLINE or scx_cgroup_lifetime_notify() sees and shoots us down. + */ + set_cgroup_sched(sch_cgroup(sch), sch); + if (!(cgrp->self.flags & CSS_ONLINE)) { + scx_error(sch, "cgroup is not online"); + goto err_unlock_and_disable; + } + + /* + * Initialize tasks for the new child $sch without exiting them for + * $parent so that the tasks can always be reverted back to $parent + * sched on child init failure. + */ + WARN_ON_ONCE(scx_enabling_sub_sched); + scx_enabling_sub_sched = sch; + + scx_task_iter_start(&sti, sch->cgrp); + while ((p = scx_task_iter_next_locked(&sti))) { + struct rq *rq; + struct rq_flags rf; + + /* + * Task iteration may visit the same task twice when racing + * against exiting. Use %SCX_TASK_SUB_INIT to mark tasks which + * finished __scx_init_task() and skip if set. + * + * A task may exit and get freed between __scx_init_task() + * completion and scx_enable_task(). In such cases, + * scx_disable_and_exit_task() must exit the task for both the + * parent and child scheds. + */ + if (p->scx.flags & SCX_TASK_SUB_INIT) + continue; + + /* see scx_root_enable() */ + if (!tryget_task_struct(p)) + continue; + + if (!assert_task_ready_or_enabled(p)) { + ret = -EINVAL; + goto abort; + } + + scx_task_iter_unlock(&sti); + + /* + * As $p is still on $parent, it can't be transitioned to INIT. + * Let's worry about task state later. Use __scx_init_task(). + */ + ret = __scx_init_task(sch, p, false); + if (ret) + goto abort; + + rq = task_rq_lock(p, &rf); + + if (scx_get_task_state(p) == SCX_TASK_DEAD) { + /* + * sched_ext_dead() raced us between __scx_init_task() + * and this rq lock and ran exit_task() on $parent (the + * sched @p was on at that point), not on @sch. @sch's + * just-completed init is owed an exit_task() and we + * issue it here. + */ + scx_sub_init_cancel_task(sch, p); + task_rq_unlock(rq, p, &rf); + put_task_struct(p); + continue; + } + + p->scx.flags |= SCX_TASK_SUB_INIT; + task_rq_unlock(rq, p, &rf); + + put_task_struct(p); + } + scx_task_iter_stop(&sti); + + /* + * All tasks are prepped. Disable/exit tasks for $parent and enable for + * the new @sch. + */ + scx_task_iter_start(&sti, sch->cgrp); + while ((p = scx_task_iter_next_locked(&sti))) { + /* + * Use clearing of %SCX_TASK_SUB_INIT to detect and skip + * duplicate iterations. + */ + if (!(p->scx.flags & SCX_TASK_SUB_INIT)) + continue; + + scoped_guard (sched_change, p, DEQUEUE_SAVE | DEQUEUE_MOVE) { + /* + * $p must be either READY or ENABLED. If ENABLED, + * __scx_disabled_and_exit_task() first disables and + * makes it READY. However, after exiting $p, it will + * leave $p as READY. + */ + assert_task_ready_or_enabled(p); + __scx_disable_and_exit_task(parent, p); + + /* + * $p is now only initialized for @sch and READY, which + * is what we want. Assign it to @sch and enable. + */ + scx_set_task_sched(p, sch); + scx_enable_task(sch, p); + + p->scx.flags &= ~SCX_TASK_SUB_INIT; + } + } + scx_task_iter_stop(&sti); + + scx_enabling_sub_sched = NULL; + + scx_cgroup_unlock(); + percpu_up_write(&scx_fork_rwsem); + + scx_bypass(sch, false); + + pr_info("sched_ext: BPF sub-scheduler \"%s\" enabled\n", sch->ops.name); + kobject_uevent(&sch->kobj, KOBJ_ADD); + ret = 0; + goto out_unlock; + +out_put_cgrp: + cgroup_put(cgrp); +out_unlock: + mutex_unlock(&scx_enable_mutex); + cmd->ret = ret; + return; + +abort: + put_task_struct(p); + scx_task_iter_stop(&sti); + + /* + * Undo __scx_init_task() for tasks we marked. scx_enable_task() never + * ran for @sch on them, so calling scx_disable_task() here would invoke + * ops.disable() without a matching ops.enable(). scx_enabling_sub_sched + * must stay set until SUB_INIT is cleared from every marked task - + * scx_disable_and_exit_task() reads it when a task exits concurrently. + */ + scx_task_iter_start(&sti, sch->cgrp); + while ((p = scx_task_iter_next_locked(&sti))) { + if (p->scx.flags & SCX_TASK_SUB_INIT) { + scx_sub_init_cancel_task(sch, p); + p->scx.flags &= ~SCX_TASK_SUB_INIT; + } + } + scx_task_iter_stop(&sti); + scx_enabling_sub_sched = NULL; +err_unlock_and_disable: + /* we'll soon enter disable path, keep bypass on */ + scx_cgroup_unlock(); + percpu_up_write(&scx_fork_rwsem); +err_disable: + mutex_unlock(&scx_enable_mutex); + scx_flush_disable_work(sch); + cmd->ret = 0; +} + +static s32 scx_cgroup_lifetime_notify(struct notifier_block *nb, + unsigned long action, void *data) +{ + struct cgroup *cgrp = data; + struct cgroup *parent = cgroup_parent(cgrp); + + if (!cgroup_on_dfl(cgrp)) + return NOTIFY_OK; + + switch (action) { + case CGROUP_LIFETIME_ONLINE: + /* inherit ->scx_sched from $parent */ + if (parent) + rcu_assign_pointer(cgrp->scx_sched, parent->scx_sched); + break; + case CGROUP_LIFETIME_OFFLINE: + /* if there is a sched attached, shoot it down */ + if (cgrp->scx_sched && cgrp->scx_sched->cgrp == cgrp) + scx_exit(cgrp->scx_sched, SCX_EXIT_UNREG_KERN, + SCX_ECODE_RSN_CGROUP_OFFLINE, + "cgroup %llu going offline", cgroup_id(cgrp)); + break; + } + + return NOTIFY_OK; +} + +static struct notifier_block scx_cgroup_lifetime_nb = { + .notifier_call = scx_cgroup_lifetime_notify, +}; + +static s32 __init scx_cgroup_lifetime_notifier_init(void) +{ + return blocking_notifier_chain_register(&cgroup_lifetime_notifier, + &scx_cgroup_lifetime_nb); +} +core_initcall(scx_cgroup_lifetime_notifier_init); +#endif /* CONFIG_EXT_SUB_SCHED */ + +static s32 scx_enable(struct sched_ext_ops *ops, struct bpf_link *link) +{ + static struct kthread_worker *helper; + static DEFINE_MUTEX(helper_mutex); + struct scx_enable_cmd cmd; + + if (housekeeping_enabled(HK_TYPE_DOMAIN_BOOT)) { + pr_err("sched_ext: Not compatible with \"isolcpus=\" domain isolation\n"); + return -EINVAL; + } + + if (!READ_ONCE(helper)) { + mutex_lock(&helper_mutex); + if (!helper) { + struct kthread_worker *w = + kthread_run_worker(0, "scx_enable_helper"); + if (IS_ERR_OR_NULL(w)) { + mutex_unlock(&helper_mutex); + return -ENOMEM; + } + sched_set_fifo(w->task); + WRITE_ONCE(helper, w); + } + mutex_unlock(&helper_mutex); + } + +#ifdef CONFIG_EXT_SUB_SCHED + if (ops->sub_cgroup_id > 1) + kthread_init_work(&cmd.work, scx_sub_enable_workfn); + else +#endif /* CONFIG_EXT_SUB_SCHED */ + kthread_init_work(&cmd.work, scx_root_enable_workfn); + cmd.ops = ops; + + kthread_queue_work(READ_ONCE(helper), &cmd.work); + kthread_flush_work(&cmd.work); + return cmd.ret; } @@ -5898,12 +7514,17 @@ static int bpf_scx_btf_struct_access(struct bpf_verifier_log *log, t = btf_type_by_id(reg->btf, reg->btf_id); if (t == task_struct_type) { - if (off >= offsetof(struct task_struct, scx.slice) && - off + size <= offsetofend(struct task_struct, scx.slice)) - return SCALAR_VALUE; - if (off >= offsetof(struct task_struct, scx.dsq_vtime) && - off + size <= offsetofend(struct task_struct, scx.dsq_vtime)) + /* + * COMPAT: Will be removed in v6.23. + */ + if ((off >= offsetof(struct task_struct, scx.slice) && + off + size <= offsetofend(struct task_struct, scx.slice)) || + (off >= offsetof(struct task_struct, scx.dsq_vtime) && + off + size <= offsetofend(struct task_struct, scx.dsq_vtime))) { + pr_warn("sched_ext: Writing directly to p->scx.slice/dsq_vtime is deprecated, use scx_bpf_task_set_slice/dsq_vtime()"); return SCALAR_VALUE; + } + if (off >= offsetof(struct task_struct, scx.disallow) && off + size <= offsetofend(struct task_struct, scx.disallow)) return SCALAR_VALUE; @@ -5912,21 +7533,8 @@ static int bpf_scx_btf_struct_access(struct bpf_verifier_log *log, return -EACCES; } -static const struct bpf_func_proto * -bpf_scx_get_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog) -{ - switch (func_id) { - case BPF_FUNC_task_storage_get: - return &bpf_task_storage_get_proto; - case BPF_FUNC_task_storage_delete: - return &bpf_task_storage_delete_proto; - default: - return bpf_base_func_proto(func_id, prog); - } -} - static const struct bpf_verifier_ops bpf_scx_verifier_ops = { - .get_func_proto = bpf_scx_get_func_proto, + .get_func_proto = bpf_base_func_proto, .is_valid_access = bpf_scx_is_valid_access, .btf_struct_access = bpf_scx_btf_struct_access, }; @@ -5972,11 +7580,30 @@ static int bpf_scx_init_member(const struct btf_type *t, case offsetof(struct sched_ext_ops, hotplug_seq): ops->hotplug_seq = *(u64 *)(udata + moff); return 1; +#ifdef CONFIG_EXT_SUB_SCHED + case offsetof(struct sched_ext_ops, sub_cgroup_id): + ops->sub_cgroup_id = *(u64 *)(udata + moff); + return 1; +#endif /* CONFIG_EXT_SUB_SCHED */ } return 0; } +#ifdef CONFIG_EXT_SUB_SCHED +static void scx_pstack_recursion_on_dispatch(struct bpf_prog *prog) +{ + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(prog->aux); + if (unlikely(!sch)) + return; + + scx_error(sch, "dispatch recursion detected"); +} +#endif /* CONFIG_EXT_SUB_SCHED */ + static int bpf_scx_check_member(const struct btf_type *t, const struct btf_member *member, const struct bpf_prog *prog) @@ -5994,24 +7621,47 @@ static int bpf_scx_check_member(const struct btf_type *t, case offsetof(struct sched_ext_ops, cpu_offline): case offsetof(struct sched_ext_ops, init): case offsetof(struct sched_ext_ops, exit): + case offsetof(struct sched_ext_ops, sub_attach): + case offsetof(struct sched_ext_ops, sub_detach): break; default: if (prog->sleepable) return -EINVAL; } +#ifdef CONFIG_EXT_SUB_SCHED + /* + * Enable private stack for operations that can nest along the + * hierarchy. + * + * XXX - Ideally, we should only do this for scheds that allow + * sub-scheds and sub-scheds themselves but I don't know how to access + * struct_ops from here. + */ + switch (moff) { + case offsetof(struct sched_ext_ops, dispatch): + prog->aux->priv_stack_requested = true; + prog->aux->recursion_detected = scx_pstack_recursion_on_dispatch; + } +#endif /* CONFIG_EXT_SUB_SCHED */ + return 0; } static int bpf_scx_reg(void *kdata, struct bpf_link *link) { - return scx_ops_enable(kdata, link); + return scx_enable(kdata, link); } static void bpf_scx_unreg(void *kdata, struct bpf_link *link) { - scx_ops_disable(SCX_EXIT_UNREG); - kthread_flush_work(&scx_ops_disable_work); + struct sched_ext_ops *ops = kdata; + struct scx_sched *sch = rcu_dereference_protected(ops->priv, true); + + scx_disable(sch, SCX_EXIT_UNREG); + scx_flush_disable_work(sch); + RCU_INIT_POINTER(ops->priv, NULL); + kobject_put(&sch->kobj); } static int bpf_scx_init(struct btf *btf) @@ -6065,7 +7715,11 @@ static s32 sched_ext_ops__cgroup_prep_move(struct task_struct *p, struct cgroup static void sched_ext_ops__cgroup_move(struct task_struct *p, struct cgroup *from, struct cgroup *to) {} static void sched_ext_ops__cgroup_cancel_move(struct task_struct *p, struct cgroup *from, struct cgroup *to) {} static void sched_ext_ops__cgroup_set_weight(struct cgroup *cgrp, u32 weight) {} -#endif +static void sched_ext_ops__cgroup_set_bandwidth(struct cgroup *cgrp, u64 period_us, u64 quota_us, u64 burst_us) {} +static void sched_ext_ops__cgroup_set_idle(struct cgroup *cgrp, bool idle) {} +#endif /* CONFIG_EXT_GROUP_SCHED */ +static s32 sched_ext_ops__sub_attach(struct scx_sub_attach_args *args) { return -EINVAL; } +static void sched_ext_ops__sub_detach(struct scx_sub_detach_args *args) {} static void sched_ext_ops__cpu_online(s32 cpu) {} static void sched_ext_ops__cpu_offline(s32 cpu) {} static s32 sched_ext_ops__init(void) { return -EINVAL; } @@ -6102,7 +7756,11 @@ static struct sched_ext_ops __bpf_ops_sched_ext_ops = { .cgroup_move = sched_ext_ops__cgroup_move, .cgroup_cancel_move = sched_ext_ops__cgroup_cancel_move, .cgroup_set_weight = sched_ext_ops__cgroup_set_weight, + .cgroup_set_bandwidth = sched_ext_ops__cgroup_set_bandwidth, + .cgroup_set_idle = sched_ext_ops__cgroup_set_idle, #endif + .sub_attach = sched_ext_ops__sub_attach, + .sub_detach = sched_ext_ops__sub_detach, .cpu_online = sched_ext_ops__cpu_online, .cpu_offline = sched_ext_ops__cpu_offline, .init = sched_ext_ops__init, @@ -6133,10 +7791,15 @@ static struct bpf_struct_ops bpf_sched_ext_ops = { static void sysrq_handle_sched_ext_reset(u8 key) { - if (scx_ops_helper) - scx_ops_disable(SCX_EXIT_SYSRQ); + struct scx_sched *sch; + + rcu_read_lock(); + sch = rcu_dereference(scx_root); + if (likely(sch)) + scx_disable(sch, SCX_EXIT_SYSRQ); else - pr_info("sched_ext: BPF scheduler not yet used\n"); + pr_info("sched_ext: BPF schedulers not loaded\n"); + rcu_read_unlock(); } static const struct sysrq_key_op sysrq_sched_ext_reset_op = { @@ -6149,9 +7812,10 @@ static const struct sysrq_key_op sysrq_sched_ext_reset_op = { static void sysrq_handle_sched_ext_dump(u8 key) { struct scx_exit_info ei = { .kind = SCX_EXIT_NONE, .reason = "SysRq-D" }; + struct scx_sched *sch; - if (scx_enabled()) - scx_dump_state(&ei, 0); + list_for_each_entry_rcu(sch, &scx_sched_all, all) + scx_dump_state(sch, &ei, 0, false); } static const struct sysrq_key_op sysrq_sched_ext_dump_op = { @@ -6178,29 +7842,38 @@ static bool can_skip_idle_kick(struct rq *rq) return !is_idle_task(rq->curr) && !(rq->scx.flags & SCX_RQ_IN_BALANCE); } -static bool kick_one_cpu(s32 cpu, struct rq *this_rq, unsigned long *pseqs) +static bool kick_one_cpu(s32 cpu, struct rq *this_rq, unsigned long *ksyncs) { struct rq *rq = cpu_rq(cpu); struct scx_rq *this_scx = &this_rq->scx; + const struct sched_class *cur_class; bool should_wait = false; unsigned long flags; raw_spin_rq_lock_irqsave(rq, flags); + cur_class = rq->curr->sched_class; /* * During CPU hotplug, a CPU may depend on kicking itself to make - * forward progress. Allow kicking self regardless of online state. + * forward progress. Allow kicking self regardless of online state. If + * @cpu is running a higher class task, we have no control over @cpu. + * Skip kicking. */ - if (cpu_online(cpu) || cpu == cpu_of(this_rq)) { + if ((cpu_online(cpu) || cpu == cpu_of(this_rq)) && + !sched_class_above(cur_class, &ext_sched_class)) { if (cpumask_test_cpu(cpu, this_scx->cpus_to_preempt)) { - if (rq->curr->sched_class == &ext_sched_class) + if (cur_class == &ext_sched_class) rq->curr->scx.slice = 0; cpumask_clear_cpu(cpu, this_scx->cpus_to_preempt); } if (cpumask_test_cpu(cpu, this_scx->cpus_to_wait)) { - pseqs[cpu] = rq->scx.pnt_seq; - should_wait = true; + if (cur_class == &ext_sched_class) { + cpumask_set_cpu(cpu, this_scx->cpus_to_sync); + ksyncs[cpu] = rq->scx.kick_sync; + should_wait = true; + } + cpumask_clear_cpu(cpu, this_scx->cpus_to_wait); } resched_curr(rq); @@ -6232,12 +7905,19 @@ static void kick_cpus_irq_workfn(struct irq_work *irq_work) { struct rq *this_rq = this_rq(); struct scx_rq *this_scx = &this_rq->scx; - unsigned long *pseqs = this_cpu_ptr(scx_kick_cpus_pnt_seqs); + struct scx_kick_syncs __rcu *ksyncs_pcpu = __this_cpu_read(scx_kick_syncs); bool should_wait = false; + unsigned long *ksyncs; s32 cpu; + /* can race with free_kick_syncs() during scheduler disable */ + if (unlikely(!ksyncs_pcpu)) + return; + + ksyncs = rcu_dereference_bh(ksyncs_pcpu)->syncs; + for_each_cpu(cpu, this_scx->cpus_to_kick) { - should_wait |= kick_one_cpu(cpu, this_rq, pseqs); + should_wait |= kick_one_cpu(cpu, this_rq, ksyncs); cpumask_clear_cpu(cpu, this_scx->cpus_to_kick); cpumask_clear_cpu(cpu, this_scx->cpus_to_kick_if_idle); } @@ -6247,26 +7927,15 @@ static void kick_cpus_irq_workfn(struct irq_work *irq_work) cpumask_clear_cpu(cpu, this_scx->cpus_to_kick_if_idle); } - if (!should_wait) - return; - - for_each_cpu(cpu, this_scx->cpus_to_wait) { - unsigned long *wait_pnt_seq = &cpu_rq(cpu)->scx.pnt_seq; - - if (cpu != cpu_of(this_rq)) { - /* - * Pairs with smp_store_release() issued by this CPU in - * switch_class() on the resched path. - * - * We busy-wait here to guarantee that no other task can - * be scheduled on our core before the target CPU has - * entered the resched path. - */ - while (smp_load_acquire(wait_pnt_seq) == pseqs[cpu]) - cpu_relax(); - } - - cpumask_clear_cpu(cpu, this_scx->cpus_to_wait); + /* + * Can't wait in hardirq — kick_sync can't advance, deadlocking if + * CPUs wait for each other. Defer to kick_sync_wait_bal_cb(). + */ + if (should_wait) { + raw_spin_rq_lock(this_rq); + this_scx->kick_sync_pending = true; + resched_curr(this_rq); + raw_spin_rq_unlock(this_rq); } } @@ -6284,13 +7953,18 @@ static void kick_cpus_irq_workfn(struct irq_work *irq_work) */ void print_scx_info(const char *log_lvl, struct task_struct *p) { - enum scx_ops_enable_state state = scx_ops_enable_state(); + struct scx_sched *sch; + enum scx_enable_state state = scx_enable_state(); const char *all = READ_ONCE(scx_switching_all) ? "+all" : ""; char runnable_at_buf[22] = "?"; struct sched_class *class; unsigned long runnable_at; - if (state == SCX_OPS_DISABLED) + guard(rcu)(); + + sch = scx_task_sched_rcu(p); + + if (!sch) return; /* @@ -6299,8 +7973,8 @@ void print_scx_info(const char *log_lvl, struct task_struct *p) */ if (copy_from_kernel_nofault(&class, &p->sched_class, sizeof(class)) || class != &ext_sched_class) { - printk("%sSched_ext: %s (%s%s)", log_lvl, scx_ops.name, - scx_ops_enable_state_str[state], all); + printk("%sSched_ext: %s (%s%s)", log_lvl, sch->ops.name, + scx_enable_state_str[state], all); return; } @@ -6311,12 +7985,20 @@ void print_scx_info(const char *log_lvl, struct task_struct *p) /* print everything onto one line to conserve console space */ printk("%sSched_ext: %s (%s%s), task: runnable_at=%s", - log_lvl, scx_ops.name, scx_ops_enable_state_str[state], all, + log_lvl, sch->ops.name, scx_enable_state_str[state], all, runnable_at_buf); } static int scx_pm_handler(struct notifier_block *nb, unsigned long event, void *ptr) { + struct scx_sched *sch; + + guard(rcu)(); + + sch = rcu_dereference(scx_root); + if (!sch) + return NOTIFY_OK; + /* * SCX schedulers often have userspace components which are sometimes * involved in critial scheduling paths. PM operations involve freezing @@ -6327,12 +8009,12 @@ static int scx_pm_handler(struct notifier_block *nb, unsigned long event, void * case PM_HIBERNATION_PREPARE: case PM_SUSPEND_PREPARE: case PM_RESTORE_PREPARE: - scx_ops_bypass(true); + scx_bypass(sch, true); break; case PM_POST_HIBERNATION: case PM_POST_SUSPEND: case PM_POST_RESTORE: - scx_ops_bypass(false); + scx_bypass(sch, false); break; } @@ -6355,29 +8037,28 @@ void __init init_sched_ext_class(void) WRITE_ONCE(v, SCX_ENQ_WAKEUP | SCX_DEQ_SLEEP | SCX_KICK_PREEMPT | SCX_TG_ONLINE); - BUG_ON(rhashtable_init(&dsq_hash, &dsq_hash_params)); -#ifdef CONFIG_SMP - BUG_ON(!alloc_cpumask_var(&idle_masks.cpu, GFP_KERNEL)); - BUG_ON(!alloc_cpumask_var(&idle_masks.smt, GFP_KERNEL)); -#endif - scx_kick_cpus_pnt_seqs = - __alloc_percpu(sizeof(scx_kick_cpus_pnt_seqs[0]) * nr_cpu_ids, - __alignof__(scx_kick_cpus_pnt_seqs[0])); - BUG_ON(!scx_kick_cpus_pnt_seqs); + scx_idle_init_masks(); for_each_possible_cpu(cpu) { struct rq *rq = cpu_rq(cpu); + int n = cpu_to_node(cpu); + + /* local_dsq's sch will be set during scx_root_enable() */ + BUG_ON(init_dsq(&rq->scx.local_dsq, SCX_DSQ_LOCAL, NULL)); - init_dsq(&rq->scx.local_dsq, SCX_DSQ_LOCAL); INIT_LIST_HEAD(&rq->scx.runnable_list); INIT_LIST_HEAD(&rq->scx.ddsp_deferred_locals); - BUG_ON(!zalloc_cpumask_var(&rq->scx.cpus_to_kick, GFP_KERNEL)); - BUG_ON(!zalloc_cpumask_var(&rq->scx.cpus_to_kick_if_idle, GFP_KERNEL)); - BUG_ON(!zalloc_cpumask_var(&rq->scx.cpus_to_preempt, GFP_KERNEL)); - BUG_ON(!zalloc_cpumask_var(&rq->scx.cpus_to_wait, GFP_KERNEL)); - init_irq_work(&rq->scx.deferred_irq_work, deferred_irq_workfn); - init_irq_work(&rq->scx.kick_cpus_irq_work, kick_cpus_irq_workfn); + BUG_ON(!zalloc_cpumask_var_node(&rq->scx.cpus_to_kick, GFP_KERNEL, n)); + BUG_ON(!zalloc_cpumask_var_node(&rq->scx.cpus_to_kick_if_idle, GFP_KERNEL, n)); + BUG_ON(!zalloc_cpumask_var_node(&rq->scx.cpus_to_preempt, GFP_KERNEL, n)); + BUG_ON(!zalloc_cpumask_var_node(&rq->scx.cpus_to_wait, GFP_KERNEL, n)); + BUG_ON(!zalloc_cpumask_var_node(&rq->scx.cpus_to_sync, GFP_KERNEL, n)); + raw_spin_lock_init(&rq->scx.deferred_reenq_lock); + INIT_LIST_HEAD(&rq->scx.deferred_reenq_locals); + INIT_LIST_HEAD(&rq->scx.deferred_reenq_users); + rq->scx.deferred_irq_work = IRQ_WORK_INIT_HARD(deferred_irq_workfn); + rq->scx.kick_cpus_irq_work = IRQ_WORK_INIT_HARD(kick_cpus_irq_workfn); if (cpu_online(cpu)) cpu_rq(cpu)->scx.flags |= SCX_RQ_ONLINE; @@ -6386,105 +8067,74 @@ void __init init_sched_ext_class(void) register_sysrq_key('S', &sysrq_sched_ext_reset_op); register_sysrq_key('D', &sysrq_sched_ext_dump_op); INIT_DELAYED_WORK(&scx_watchdog_work, scx_watchdog_workfn); + +#ifdef CONFIG_EXT_SUB_SCHED + BUG_ON(rhashtable_init(&scx_sched_hash, &scx_sched_hash_params)); +#endif /* CONFIG_EXT_SUB_SCHED */ } /******************************************************************************** * Helpers that can be called from the BPF scheduler. */ -#include <linux/btf_ids.h> - -__bpf_kfunc_start_defs(); - -static bool check_builtin_idle_enabled(void) +static bool scx_vet_enq_flags(struct scx_sched *sch, u64 dsq_id, u64 *enq_flags) { - if (static_branch_likely(&scx_builtin_idle_enabled)) - return true; - - scx_ops_error("built-in idle tracking is disabled"); - return false; -} - -/** - * scx_bpf_select_cpu_dfl - The default implementation of ops.select_cpu() - * @p: task_struct to select a CPU for - * @prev_cpu: CPU @p was on previously - * @wake_flags: %SCX_WAKE_* flags - * @is_idle: out parameter indicating whether the returned CPU is idle - * - * Can only be called from ops.select_cpu() if the built-in CPU selection is - * enabled - ops.update_idle() is missing or %SCX_OPS_KEEP_BUILTIN_IDLE is set. - * @p, @prev_cpu and @wake_flags match ops.select_cpu(). - * - * Returns the picked CPU with *@is_idle indicating whether the picked CPU is - * currently idle and thus a good candidate for direct dispatching. - */ -__bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, - u64 wake_flags, bool *is_idle) -{ - if (!ops_cpu_valid(prev_cpu, NULL)) - goto prev_cpu; - - if (!check_builtin_idle_enabled()) - goto prev_cpu; + bool is_local = dsq_id == SCX_DSQ_LOCAL || + (dsq_id & SCX_DSQ_LOCAL_ON) == SCX_DSQ_LOCAL_ON; - if (!scx_kf_allowed(SCX_KF_SELECT_CPU)) - goto prev_cpu; - -#ifdef CONFIG_SMP - return scx_select_cpu_dfl(p, prev_cpu, wake_flags, is_idle); -#endif + if (*enq_flags & SCX_ENQ_IMMED) { + if (unlikely(!is_local)) { + scx_error(sch, "SCX_ENQ_IMMED on a non-local DSQ 0x%llx", dsq_id); + return false; + } + } else if ((sch->ops.flags & SCX_OPS_ALWAYS_ENQ_IMMED) && is_local) { + *enq_flags |= SCX_ENQ_IMMED; + } -prev_cpu: - *is_idle = false; - return prev_cpu; + return true; } -__bpf_kfunc_end_defs(); - -BTF_KFUNCS_START(scx_kfunc_ids_select_cpu) -BTF_ID_FLAGS(func, scx_bpf_select_cpu_dfl, KF_RCU) -BTF_KFUNCS_END(scx_kfunc_ids_select_cpu) - -static const struct btf_kfunc_id_set scx_kfunc_set_select_cpu = { - .owner = THIS_MODULE, - .set = &scx_kfunc_ids_select_cpu, -}; - -static bool scx_dsq_insert_preamble(struct task_struct *p, u64 enq_flags) +static bool scx_dsq_insert_preamble(struct scx_sched *sch, struct task_struct *p, + u64 dsq_id, u64 *enq_flags) { - if (!scx_kf_allowed(SCX_KF_ENQUEUE | SCX_KF_DISPATCH)) - return false; - lockdep_assert_irqs_disabled(); if (unlikely(!p)) { - scx_ops_error("called with NULL task"); + scx_error(sch, "called with NULL task"); return false; } - if (unlikely(enq_flags & __SCX_ENQ_INTERNAL_MASK)) { - scx_ops_error("invalid enq_flags 0x%llx", enq_flags); + if (unlikely(*enq_flags & __SCX_ENQ_INTERNAL_MASK)) { + scx_error(sch, "invalid enq_flags 0x%llx", *enq_flags); return false; } + /* see SCX_EV_INSERT_NOT_OWNED definition */ + if (unlikely(!scx_task_on_sched(sch, p))) { + __scx_add_event(sch, SCX_EV_INSERT_NOT_OWNED, 1); + return false; + } + + if (!scx_vet_enq_flags(sch, dsq_id, enq_flags)) + return false; + return true; } -static void scx_dsq_insert_commit(struct task_struct *p, u64 dsq_id, - u64 enq_flags) +static void scx_dsq_insert_commit(struct scx_sched *sch, struct task_struct *p, + u64 dsq_id, u64 enq_flags) { - struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); + struct scx_dsp_ctx *dspc = &this_cpu_ptr(sch->pcpu)->dsp_ctx; struct task_struct *ddsp_task; ddsp_task = __this_cpu_read(direct_dispatch_task); if (ddsp_task) { - mark_direct_dispatch(ddsp_task, p, dsq_id, enq_flags); + mark_direct_dispatch(sch, ddsp_task, p, dsq_id, enq_flags); return; } - if (unlikely(dspc->cursor >= scx_dsp_max_batch)) { - scx_ops_error("dispatch buffer overflow"); + if (unlikely(dspc->cursor >= sch->dsp_max_batch)) { + scx_error(sch, "dispatch buffer overflow"); return; } @@ -6504,6 +8154,7 @@ __bpf_kfunc_start_defs(); * @dsq_id: DSQ to insert into * @slice: duration @p can run for in nsecs, 0 to keep the current value * @enq_flags: SCX_ENQ_* + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Insert @p into the FIFO queue of the DSQ identified by @dsq_id. It is safe to * call this function spuriously. Can be called from ops.enqueue(), @@ -6522,7 +8173,8 @@ __bpf_kfunc_start_defs(); * When called from ops.dispatch(), there are no restrictions on @p or @dsq_id * and this function can be called upto ops.dispatch_max_batch times to insert * multiple tasks. scx_bpf_dispatch_nr_slots() returns the number of the - * remaining slots. scx_bpf_consume() flushes the batch and resets the counter. + * remaining slots. scx_bpf_dsq_move_to_local() flushes the batch and resets the + * counter. * * This function doesn't have any locking restrictions and may be called under * BPF locks (in the future when BPF introduces more flexible locking). @@ -6531,101 +8183,199 @@ __bpf_kfunc_start_defs(); * exhaustion. If zero, the current residual slice is maintained. If * %SCX_SLICE_INF, @p never expires and the BPF scheduler must kick the CPU with * scx_bpf_kick_cpu() to trigger scheduling. + * + * Returns %true on successful insertion, %false on failure. On the root + * scheduler, %false return triggers scheduler abort and the caller doesn't need + * to check the return value. */ -__bpf_kfunc void scx_bpf_dsq_insert(struct task_struct *p, u64 dsq_id, u64 slice, - u64 enq_flags) +__bpf_kfunc bool scx_bpf_dsq_insert___v2(struct task_struct *p, u64 dsq_id, + u64 slice, u64 enq_flags, + const struct bpf_prog_aux *aux) { - if (!scx_dsq_insert_preamble(p, enq_flags)) - return; + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return false; + + if (!scx_dsq_insert_preamble(sch, p, dsq_id, &enq_flags)) + return false; if (slice) p->scx.slice = slice; else p->scx.slice = p->scx.slice ?: 1; - scx_dsq_insert_commit(p, dsq_id, enq_flags); + scx_dsq_insert_commit(sch, p, dsq_id, enq_flags); + + return true; } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc void scx_bpf_dispatch(struct task_struct *p, u64 dsq_id, u64 slice, - u64 enq_flags) +/* + * COMPAT: Will be removed in v6.23 along with the ___v2 suffix. + */ +__bpf_kfunc void scx_bpf_dsq_insert(struct task_struct *p, u64 dsq_id, + u64 slice, u64 enq_flags, + const struct bpf_prog_aux *aux) +{ + scx_bpf_dsq_insert___v2(p, dsq_id, slice, enq_flags, aux); +} + +static bool scx_dsq_insert_vtime(struct scx_sched *sch, struct task_struct *p, + u64 dsq_id, u64 slice, u64 vtime, u64 enq_flags) { - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch() renamed to scx_bpf_dsq_insert()"); - scx_bpf_dsq_insert(p, dsq_id, slice, enq_flags); + if (!scx_dsq_insert_preamble(sch, p, dsq_id, &enq_flags)) + return false; + + if (slice) + p->scx.slice = slice; + else + p->scx.slice = p->scx.slice ?: 1; + + p->scx.dsq_vtime = vtime; + + scx_dsq_insert_commit(sch, p, dsq_id, enq_flags | SCX_ENQ_DSQ_PRIQ); + + return true; } +struct scx_bpf_dsq_insert_vtime_args { + /* @p can't be packed together as KF_RCU is not transitive */ + u64 dsq_id; + u64 slice; + u64 vtime; + u64 enq_flags; +}; + /** - * scx_bpf_dsq_insert_vtime - Insert a task into the vtime priority queue of a DSQ + * __scx_bpf_dsq_insert_vtime - Arg-wrapped vtime DSQ insertion * @p: task_struct to insert - * @dsq_id: DSQ to insert into - * @slice: duration @p can run for in nsecs, 0 to keep the current value - * @vtime: @p's ordering inside the vtime-sorted queue of the target DSQ - * @enq_flags: SCX_ENQ_* + * @args: struct containing the rest of the arguments + * @args->dsq_id: DSQ to insert into + * @args->slice: duration @p can run for in nsecs, 0 to keep the current value + * @args->vtime: @p's ordering inside the vtime-sorted queue of the target DSQ + * @args->enq_flags: SCX_ENQ_* + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Wrapper kfunc that takes arguments via struct to work around BPF's 5 argument + * limit. BPF programs should use scx_bpf_dsq_insert_vtime() which is provided + * as an inline wrapper in common.bpf.h. * - * Insert @p into the vtime priority queue of the DSQ identified by @dsq_id. - * Tasks queued into the priority queue are ordered by @vtime. All other aspects - * are identical to scx_bpf_dsq_insert(). + * Insert @p into the vtime priority queue of the DSQ identified by + * @args->dsq_id. Tasks queued into the priority queue are ordered by + * @args->vtime. All other aspects are identical to scx_bpf_dsq_insert(). * - * @vtime ordering is according to time_before64() which considers wrapping. A - * numerically larger vtime may indicate an earlier position in the ordering and - * vice-versa. + * @args->vtime ordering is according to time_before64() which considers + * wrapping. A numerically larger vtime may indicate an earlier position in the + * ordering and vice-versa. * * A DSQ can only be used as a FIFO or priority queue at any given time and this * function must not be called on a DSQ which already has one or more FIFO tasks * queued and vice-versa. Also, the built-in DSQs (SCX_DSQ_LOCAL and * SCX_DSQ_GLOBAL) cannot be used as priority queues. + * + * Returns %true on successful insertion, %false on failure. On the root + * scheduler, %false return triggers scheduler abort and the caller doesn't need + * to check the return value. */ -__bpf_kfunc void scx_bpf_dsq_insert_vtime(struct task_struct *p, u64 dsq_id, - u64 slice, u64 vtime, u64 enq_flags) +__bpf_kfunc bool +__scx_bpf_dsq_insert_vtime(struct task_struct *p, + struct scx_bpf_dsq_insert_vtime_args *args, + const struct bpf_prog_aux *aux) { - if (!scx_dsq_insert_preamble(p, enq_flags)) - return; + struct scx_sched *sch; - if (slice) - p->scx.slice = slice; - else - p->scx.slice = p->scx.slice ?: 1; + guard(rcu)(); - p->scx.dsq_vtime = vtime; + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return false; - scx_dsq_insert_commit(p, dsq_id, enq_flags | SCX_ENQ_DSQ_PRIQ); + return scx_dsq_insert_vtime(sch, p, args->dsq_id, args->slice, + args->vtime, args->enq_flags); } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc void scx_bpf_dispatch_vtime(struct task_struct *p, u64 dsq_id, - u64 slice, u64 vtime, u64 enq_flags) +/* + * COMPAT: Will be removed in v6.23. + */ +__bpf_kfunc void scx_bpf_dsq_insert_vtime(struct task_struct *p, u64 dsq_id, + u64 slice, u64 vtime, u64 enq_flags) { - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch_vtime() renamed to scx_bpf_dsq_insert_vtime()"); - scx_bpf_dsq_insert_vtime(p, dsq_id, slice, vtime, enq_flags); + struct scx_sched *sch; + + guard(rcu)(); + + sch = rcu_dereference(scx_root); + if (unlikely(!sch)) + return; + +#ifdef CONFIG_EXT_SUB_SCHED + /* + * Disallow if any sub-scheds are attached. There is no way to tell + * which scheduler called us, just error out @p's scheduler. + */ + if (unlikely(!list_empty(&sch->children))) { + scx_error(scx_task_sched(p), "__scx_bpf_dsq_insert_vtime() must be used"); + return; + } +#endif + + scx_dsq_insert_vtime(sch, p, dsq_id, slice, vtime, enq_flags); } __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_enqueue_dispatch) -BTF_ID_FLAGS(func, scx_bpf_dsq_insert, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_dsq_insert, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_dsq_insert___v2, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, __scx_bpf_dsq_insert_vtime, KF_IMPLICIT_ARGS | KF_RCU) BTF_ID_FLAGS(func, scx_bpf_dsq_insert_vtime, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch_vtime, KF_RCU) BTF_KFUNCS_END(scx_kfunc_ids_enqueue_dispatch) static const struct btf_kfunc_id_set scx_kfunc_set_enqueue_dispatch = { .owner = THIS_MODULE, .set = &scx_kfunc_ids_enqueue_dispatch, + .filter = scx_kfunc_context_filter, }; static bool scx_dsq_move(struct bpf_iter_scx_dsq_kern *kit, struct task_struct *p, u64 dsq_id, u64 enq_flags) { struct scx_dispatch_q *src_dsq = kit->dsq, *dst_dsq; + struct scx_sched *sch; struct rq *this_rq, *src_rq, *locked_rq; bool dispatched = false; bool in_balance; unsigned long flags; - if (!scx_kf_allowed_if_unlocked() && !scx_kf_allowed(SCX_KF_DISPATCH)) + /* + * The verifier considers an iterator slot initialized on any + * KF_ITER_NEW return, so a BPF program may legally reach here after + * bpf_iter_scx_dsq_new() failed and left @kit->dsq NULL. + */ + if (unlikely(!src_dsq)) + return false; + + sch = src_dsq->sched; + + if (!scx_vet_enq_flags(sch, dsq_id, &enq_flags)) return false; /* + * If the BPF scheduler keeps calling this function repeatedly, it can + * cause similar live-lock conditions as consume_dispatch_q(). + */ + if (unlikely(READ_ONCE(sch->aborting))) + return false; + + if (unlikely(!scx_task_on_sched(sch, p))) { + scx_error(sch, "scx_bpf_dsq_move[_vtime]() on %s[%d] but the task belongs to a different scheduler", + p->comm, p->pid); + return false; + } + + /* * Can be called from either ops.dispatch() locking this_rq() or any * context where no rq lock is held. If latter, lock @p's task_rq which * we'll likely need anyway. @@ -6645,30 +8395,17 @@ static bool scx_dsq_move(struct bpf_iter_scx_dsq_kern *kit, raw_spin_rq_lock(src_rq); } - /* - * If the BPF scheduler keeps calling this function repeatedly, it can - * cause similar live-lock conditions as consume_dispatch_q(). Insert a - * breather if necessary. - */ - scx_ops_breather(src_rq); - locked_rq = src_rq; raw_spin_lock(&src_dsq->lock); - /* - * Did someone else get to it? @p could have already left $src_dsq, got - * re-enqueud, or be in the process of being consumed by someone else. - */ - if (unlikely(p->scx.dsq != src_dsq || - u32_before(kit->cursor.priv, p->scx.dsq_seq) || - p->scx.holding_cpu >= 0) || - WARN_ON_ONCE(src_rq != task_rq(p))) { + /* did someone else get to it while we dropped the locks? */ + if (nldsq_cursor_lost_task(&kit->cursor, src_rq, src_dsq, p)) { raw_spin_unlock(&src_dsq->lock); goto out; } /* @p is still on $src_dsq and stable, determine the destination */ - dst_dsq = find_dsq_for_dispatch(this_rq, dsq_id, p); + dst_dsq = find_dsq_for_dispatch(sch, this_rq, dsq_id, task_cpu(p)); /* * Apply vtime and slice updates before moving so that the new time is @@ -6681,7 +8418,7 @@ static bool scx_dsq_move(struct bpf_iter_scx_dsq_kern *kit, p->scx.slice = kit->slice; /* execute move */ - locked_rq = move_task_between_dsqs(p, enq_flags, src_dsq, dst_dsq); + locked_rq = move_task_between_dsqs(sch, p, enq_flags, src_dsq, dst_dsq); dispatched = true; out: if (in_balance) { @@ -6702,42 +8439,66 @@ __bpf_kfunc_start_defs(); /** * scx_bpf_dispatch_nr_slots - Return the number of remaining dispatch slots + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Can only be called from ops.dispatch(). */ -__bpf_kfunc u32 scx_bpf_dispatch_nr_slots(void) +__bpf_kfunc u32 scx_bpf_dispatch_nr_slots(const struct bpf_prog_aux *aux) { - if (!scx_kf_allowed(SCX_KF_DISPATCH)) + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) return 0; - return scx_dsp_max_batch - __this_cpu_read(scx_dsp_ctx->cursor); + return sch->dsp_max_batch - __this_cpu_read(sch->pcpu->dsp_ctx.cursor); } /** * scx_bpf_dispatch_cancel - Cancel the latest dispatch + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Cancel the latest dispatch. Can be called multiple times to cancel further * dispatches. Can only be called from ops.dispatch(). */ -__bpf_kfunc void scx_bpf_dispatch_cancel(void) +__bpf_kfunc void scx_bpf_dispatch_cancel(const struct bpf_prog_aux *aux) { - struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); + struct scx_sched *sch; + struct scx_dsp_ctx *dspc; + + guard(rcu)(); - if (!scx_kf_allowed(SCX_KF_DISPATCH)) + sch = scx_prog_sched(aux); + if (unlikely(!sch)) return; + dspc = &this_cpu_ptr(sch->pcpu)->dsp_ctx; + if (dspc->cursor > 0) dspc->cursor--; else - scx_ops_error("dispatch buffer underflow"); + scx_error(sch, "dispatch buffer underflow"); } /** * scx_bpf_dsq_move_to_local - move a task from a DSQ to the current CPU's local DSQ - * @dsq_id: DSQ to move task from + * @dsq_id: DSQ to move task from. Must be a user-created DSQ + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * @enq_flags: %SCX_ENQ_* * * Move a task from the non-local DSQ identified by @dsq_id to the current CPU's - * local DSQ for execution. Can only be called from ops.dispatch(). + * local DSQ for execution with @enq_flags applied. Can only be called from + * ops.dispatch(). + * + * Built-in DSQs (%SCX_DSQ_GLOBAL and %SCX_DSQ_LOCAL*) are not supported as + * sources. Local DSQs support reenqueueing (a task can be picked up for + * execution, dequeued for property changes, or reenqueued), but the BPF + * scheduler cannot directly iterate or move tasks from them. %SCX_DSQ_GLOBAL + * is similar but also doesn't support reenqueueing, as it maps to multiple + * per-node DSQs making the scope difficult to define; this may change in the + * future. * * This function flushes the in-flight dispatches from scx_bpf_dsq_insert() * before trying to move from the specified DSQ. It may also grab rq locks and @@ -6746,27 +8507,37 @@ __bpf_kfunc void scx_bpf_dispatch_cancel(void) * Returns %true if a task has been moved, %false if there isn't any task to * move. */ -__bpf_kfunc bool scx_bpf_dsq_move_to_local(u64 dsq_id) +__bpf_kfunc bool scx_bpf_dsq_move_to_local___v2(u64 dsq_id, u64 enq_flags, + const struct bpf_prog_aux *aux) { - struct scx_dsp_ctx *dspc = this_cpu_ptr(scx_dsp_ctx); struct scx_dispatch_q *dsq; + struct scx_sched *sch; + struct scx_dsp_ctx *dspc; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return false; - if (!scx_kf_allowed(SCX_KF_DISPATCH)) + if (!scx_vet_enq_flags(sch, SCX_DSQ_LOCAL, &enq_flags)) return false; - flush_dispatch_buf(dspc->rq); + dspc = &this_cpu_ptr(sch->pcpu)->dsp_ctx; - dsq = find_user_dsq(dsq_id); + flush_dispatch_buf(sch, dspc->rq); + + dsq = find_user_dsq(sch, dsq_id); if (unlikely(!dsq)) { - scx_ops_error("invalid DSQ ID 0x%016llx", dsq_id); + scx_error(sch, "invalid DSQ ID 0x%016llx", dsq_id); return false; } - if (consume_dispatch_q(dspc->rq, dsq)) { + if (consume_dispatch_q(sch, dspc->rq, dsq, enq_flags)) { /* * A successfully consumed task can be dequeued before it starts * running while the CPU is trying to migrate other dispatched - * tasks. Bump nr_tasks to tell balance_scx() to retry on empty + * tasks. Bump nr_tasks to tell balance_one() to retry on empty * local DSQ. */ dspc->nr_tasks++; @@ -6776,11 +8547,12 @@ __bpf_kfunc bool scx_bpf_dsq_move_to_local(u64 dsq_id) } } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc bool scx_bpf_consume(u64 dsq_id) +/* + * COMPAT: ___v2 was introduced in v7.1. Remove this and ___v2 tag in the future. + */ +__bpf_kfunc bool scx_bpf_dsq_move_to_local(u64 dsq_id, const struct bpf_prog_aux *aux) { - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_consume() renamed to scx_bpf_dsq_move_to_local()"); - return scx_bpf_dsq_move_to_local(dsq_id); + return scx_bpf_dsq_move_to_local___v2(dsq_id, 0, aux); } /** @@ -6801,14 +8573,6 @@ __bpf_kfunc void scx_bpf_dsq_move_set_slice(struct bpf_iter_scx_dsq *it__iter, kit->cursor.flags |= __SCX_DSQ_ITER_HAS_SLICE; } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc void scx_bpf_dispatch_from_dsq_set_slice( - struct bpf_iter_scx_dsq *it__iter, u64 slice) -{ - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch_from_dsq_set_slice() renamed to scx_bpf_dsq_move_set_slice()"); - scx_bpf_dsq_move_set_slice(it__iter, slice); -} - /** * scx_bpf_dsq_move_set_vtime - Override vtime when moving between DSQs * @it__iter: DSQ iterator in progress @@ -6828,14 +8592,6 @@ __bpf_kfunc void scx_bpf_dsq_move_set_vtime(struct bpf_iter_scx_dsq *it__iter, kit->cursor.flags |= __SCX_DSQ_ITER_HAS_VTIME; } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc void scx_bpf_dispatch_from_dsq_set_vtime( - struct bpf_iter_scx_dsq *it__iter, u64 vtime) -{ - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch_from_dsq_set_vtime() renamed to scx_bpf_dsq_move_set_vtime()"); - scx_bpf_dsq_move_set_vtime(it__iter, vtime); -} - /** * scx_bpf_dsq_move - Move a task from DSQ iteration to a DSQ * @it__iter: DSQ iterator in progress @@ -6857,8 +8613,9 @@ __bpf_kfunc void scx_bpf_dispatch_from_dsq_set_vtime( * Can be called from ops.dispatch() or any BPF context which doesn't hold a rq * lock (e.g. BPF timers or SYSCALL programs). * - * Returns %true if @p has been consumed, %false if @p had already been consumed - * or dequeued. + * Returns %true if @p has been consumed, %false if @p had already been + * consumed, dequeued, or, for sub-scheds, @dsq_id points to a disallowed local + * DSQ. */ __bpf_kfunc bool scx_bpf_dsq_move(struct bpf_iter_scx_dsq *it__iter, struct task_struct *p, u64 dsq_id, @@ -6868,15 +8625,6 @@ __bpf_kfunc bool scx_bpf_dsq_move(struct bpf_iter_scx_dsq *it__iter, p, dsq_id, enq_flags); } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc bool scx_bpf_dispatch_from_dsq(struct bpf_iter_scx_dsq *it__iter, - struct task_struct *p, u64 dsq_id, - u64 enq_flags) -{ - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch_from_dsq() renamed to scx_bpf_dsq_move()"); - return scx_bpf_dsq_move(it__iter, p, dsq_id, enq_flags); -} - /** * scx_bpf_dsq_move_vtime - Move a task from DSQ iteration to a PRIQ DSQ * @it__iter: DSQ iterator in progress @@ -6902,102 +8650,104 @@ __bpf_kfunc bool scx_bpf_dsq_move_vtime(struct bpf_iter_scx_dsq *it__iter, p, dsq_id, enq_flags | SCX_ENQ_DSQ_PRIQ); } -/* for backward compatibility, will be removed in v6.15 */ -__bpf_kfunc bool scx_bpf_dispatch_vtime_from_dsq(struct bpf_iter_scx_dsq *it__iter, - struct task_struct *p, u64 dsq_id, - u64 enq_flags) +#ifdef CONFIG_EXT_SUB_SCHED +/** + * scx_bpf_sub_dispatch - Trigger dispatching on a child scheduler + * @cgroup_id: cgroup ID of the child scheduler to dispatch + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Allows a parent scheduler to trigger dispatching on one of its direct + * child schedulers. The child scheduler runs its dispatch operation to + * move tasks from dispatch queues to the local runqueue. + * + * Returns: true on success, false if cgroup_id is invalid, not a direct + * child, or caller lacks dispatch permission. + */ +__bpf_kfunc bool scx_bpf_sub_dispatch(u64 cgroup_id, const struct bpf_prog_aux *aux) { - printk_deferred_once(KERN_WARNING "sched_ext: scx_bpf_dispatch_from_dsq_vtime() renamed to scx_bpf_dsq_move_vtime()"); - return scx_bpf_dsq_move_vtime(it__iter, p, dsq_id, enq_flags); + struct rq *this_rq = this_rq(); + struct scx_sched *parent, *child; + + guard(rcu)(); + parent = scx_prog_sched(aux); + if (unlikely(!parent)) + return false; + + child = scx_find_sub_sched(cgroup_id); + + if (unlikely(!child)) + return false; + + if (unlikely(scx_parent(child) != parent)) { + scx_error(parent, "trying to dispatch a distant sub-sched on cgroup %llu", + cgroup_id); + return false; + } + + return scx_dispatch_sched(child, this_rq, this_rq->scx.sub_dispatch_prev, + true); } +#endif /* CONFIG_EXT_SUB_SCHED */ __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_dispatch) -BTF_ID_FLAGS(func, scx_bpf_dispatch_nr_slots) -BTF_ID_FLAGS(func, scx_bpf_dispatch_cancel) -BTF_ID_FLAGS(func, scx_bpf_dsq_move_to_local) -BTF_ID_FLAGS(func, scx_bpf_consume) -BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_slice) -BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_vtime) +BTF_ID_FLAGS(func, scx_bpf_dispatch_nr_slots, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dispatch_cancel, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dsq_move_to_local, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dsq_move_to_local___v2, KF_IMPLICIT_ARGS) +/* scx_bpf_dsq_move*() also in scx_kfunc_ids_unlocked: callable from unlocked contexts */ +BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_slice, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_vtime, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_dsq_move, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_dsq_move_vtime, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq_set_slice) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq_set_vtime) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch_vtime_from_dsq, KF_RCU) +#ifdef CONFIG_EXT_SUB_SCHED +BTF_ID_FLAGS(func, scx_bpf_sub_dispatch, KF_IMPLICIT_ARGS) +#endif BTF_KFUNCS_END(scx_kfunc_ids_dispatch) static const struct btf_kfunc_id_set scx_kfunc_set_dispatch = { .owner = THIS_MODULE, .set = &scx_kfunc_ids_dispatch, + .filter = scx_kfunc_context_filter, }; __bpf_kfunc_start_defs(); /** * scx_bpf_reenqueue_local - Re-enqueue tasks on a local DSQ + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Iterate over all of the tasks currently enqueued on the local DSQ of the * caller's CPU, and re-enqueue them in the BPF scheduler. Returns the number of * processed tasks. Can only be called from ops.cpu_release(). */ -__bpf_kfunc u32 scx_bpf_reenqueue_local(void) +__bpf_kfunc u32 scx_bpf_reenqueue_local(const struct bpf_prog_aux *aux) { - LIST_HEAD(tasks); - u32 nr_enqueued = 0; + struct scx_sched *sch; struct rq *rq; - struct task_struct *p, *n; - if (!scx_kf_allowed(SCX_KF_CPU_RELEASE)) + guard(rcu)(); + sch = scx_prog_sched(aux); + if (unlikely(!sch)) return 0; rq = cpu_rq(smp_processor_id()); lockdep_assert_rq_held(rq); - /* - * The BPF scheduler may choose to dispatch tasks back to - * @rq->scx.local_dsq. Move all candidate tasks off to a private list - * first to avoid processing the same tasks repeatedly. - */ - list_for_each_entry_safe(p, n, &rq->scx.local_dsq.list, - scx.dsq_list.node) { - /* - * If @p is being migrated, @p's current CPU may not agree with - * its allowed CPUs and the migration_cpu_stop is about to - * deactivate and re-activate @p anyway. Skip re-enqueueing. - * - * While racing sched property changes may also dequeue and - * re-enqueue a migrating task while its current CPU and allowed - * CPUs disagree, they use %ENQUEUE_RESTORE which is bypassed to - * the current local DSQ for running tasks and thus are not - * visible to the BPF scheduler. - */ - if (p->migration_pending) - continue; - - dispatch_dequeue(rq, p); - list_add_tail(&p->scx.dsq_list.node, &tasks); - } - - list_for_each_entry_safe(p, n, &tasks, scx.dsq_list.node) { - list_del_init(&p->scx.dsq_list.node); - do_enqueue_task(rq, p, SCX_ENQ_REENQ, -1); - nr_enqueued++; - } - - return nr_enqueued; + return reenq_local(sch, rq, SCX_REENQ_ANY); } __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_cpu_release) -BTF_ID_FLAGS(func, scx_bpf_reenqueue_local) +BTF_ID_FLAGS(func, scx_bpf_reenqueue_local, KF_IMPLICIT_ARGS) BTF_KFUNCS_END(scx_kfunc_ids_cpu_release) static const struct btf_kfunc_id_set scx_kfunc_set_cpu_release = { .owner = THIS_MODULE, .set = &scx_kfunc_ids_cpu_release, + .filter = scx_kfunc_context_filter, }; __bpf_kfunc_start_defs(); @@ -7006,55 +8756,132 @@ __bpf_kfunc_start_defs(); * scx_bpf_create_dsq - Create a custom DSQ * @dsq_id: DSQ to create * @node: NUMA node to allocate from + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Create a custom DSQ identified by @dsq_id. Can be called from any sleepable * scx callback, and any BPF_PROG_TYPE_SYSCALL prog. */ -__bpf_kfunc s32 scx_bpf_create_dsq(u64 dsq_id, s32 node) +__bpf_kfunc s32 scx_bpf_create_dsq(u64 dsq_id, s32 node, const struct bpf_prog_aux *aux) { + struct scx_dispatch_q *dsq; + struct scx_sched *sch; + s32 ret; + if (unlikely(node >= (int)nr_node_ids || (node < 0 && node != NUMA_NO_NODE))) return -EINVAL; - return PTR_ERR_OR_ZERO(create_dsq(dsq_id, node)); + + if (unlikely(dsq_id & SCX_DSQ_FLAG_BUILTIN)) + return -EINVAL; + + dsq = kmalloc_node(sizeof(*dsq), GFP_KERNEL, node); + if (!dsq) + return -ENOMEM; + + /* + * init_dsq() must be called in GFP_KERNEL context. Init it with NULL + * @sch and update afterwards. + */ + ret = init_dsq(dsq, dsq_id, NULL); + if (ret) { + kfree(dsq); + return ret; + } + + rcu_read_lock(); + + sch = scx_prog_sched(aux); + if (sch) { + dsq->sched = sch; + ret = rhashtable_lookup_insert_fast(&sch->dsq_hash, &dsq->hash_node, + dsq_hash_params); + } else { + ret = -ENODEV; + } + + rcu_read_unlock(); + if (ret) { + exit_dsq(dsq); + kfree(dsq); + } + return ret; } __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_unlocked) -BTF_ID_FLAGS(func, scx_bpf_create_dsq, KF_SLEEPABLE) -BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_slice) -BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_vtime) +BTF_ID_FLAGS(func, scx_bpf_create_dsq, KF_IMPLICIT_ARGS | KF_SLEEPABLE) +/* also in scx_kfunc_ids_dispatch: also callable from ops.dispatch() */ +BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_slice, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_dsq_move_set_vtime, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_dsq_move, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_dsq_move_vtime, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq_set_slice) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq_set_vtime) -BTF_ID_FLAGS(func, scx_bpf_dispatch_from_dsq, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_dispatch_vtime_from_dsq, KF_RCU) +/* also in scx_kfunc_ids_select_cpu: also callable from ops.select_cpu()/ops.enqueue() */ +BTF_ID_FLAGS(func, __scx_bpf_select_cpu_and, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_select_cpu_and, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_select_cpu_dfl, KF_IMPLICIT_ARGS | KF_RCU) BTF_KFUNCS_END(scx_kfunc_ids_unlocked) static const struct btf_kfunc_id_set scx_kfunc_set_unlocked = { .owner = THIS_MODULE, .set = &scx_kfunc_ids_unlocked, + .filter = scx_kfunc_context_filter, }; __bpf_kfunc_start_defs(); /** - * scx_bpf_kick_cpu - Trigger reschedule on a CPU - * @cpu: cpu to kick - * @flags: %SCX_KICK_* flags + * scx_bpf_task_set_slice - Set task's time slice + * @p: task of interest + * @slice: time slice to set in nsecs + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * - * Kick @cpu into rescheduling. This can be used to wake up an idle CPU or - * trigger rescheduling on a busy CPU. This can be called from any online - * scx_ops operation and the actual kicking is performed asynchronously through - * an irq work. + * Set @p's time slice to @slice. Returns %true on success, %false if the + * calling scheduler doesn't have authority over @p. */ -__bpf_kfunc void scx_bpf_kick_cpu(s32 cpu, u64 flags) +__bpf_kfunc bool scx_bpf_task_set_slice(struct task_struct *p, u64 slice, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(aux); + if (unlikely(!sch || !scx_task_on_sched(sch, p))) + return false; + + p->scx.slice = slice; + return true; +} + +/** + * scx_bpf_task_set_dsq_vtime - Set task's virtual time for DSQ ordering + * @p: task of interest + * @vtime: virtual time to set + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Set @p's virtual time to @vtime. Returns %true on success, %false if the + * calling scheduler doesn't have authority over @p. + */ +__bpf_kfunc bool scx_bpf_task_set_dsq_vtime(struct task_struct *p, u64 vtime, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(aux); + if (unlikely(!sch || !scx_task_on_sched(sch, p))) + return false; + + p->scx.dsq_vtime = vtime; + return true; +} + +static void scx_kick_cpu(struct scx_sched *sch, s32 cpu, u64 flags) { struct rq *this_rq; unsigned long irq_flags; - if (!ops_cpu_valid(cpu, NULL)) + if (!ops_cpu_valid(sch, cpu, NULL)) return; local_irq_save(irq_flags); @@ -7066,7 +8893,7 @@ __bpf_kfunc void scx_bpf_kick_cpu(s32 cpu, u64 flags) * lead to irq_work_queue() malfunction such as infinite busy wait for * IRQ status update. Suppress kicking. */ - if (scx_rq_bypassing(this_rq)) + if (scx_bypassing(sch, cpu_of(this_rq))) goto out; /* @@ -7078,7 +8905,7 @@ __bpf_kfunc void scx_bpf_kick_cpu(s32 cpu, u64 flags) struct rq *target_rq = cpu_rq(cpu); if (unlikely(flags & (SCX_KICK_PREEMPT | SCX_KICK_WAIT))) - scx_ops_error("PREEMPT/WAIT cannot be used with SCX_KICK_IDLE"); + scx_error(sch, "PREEMPT/WAIT cannot be used with SCX_KICK_IDLE"); if (raw_spin_rq_trylock(target_rq)) { if (can_skip_idle_kick(target_rq)) { @@ -7103,31 +8930,60 @@ out: } /** + * scx_bpf_kick_cpu - Trigger reschedule on a CPU + * @cpu: cpu to kick + * @flags: %SCX_KICK_* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Kick @cpu into rescheduling. This can be used to wake up an idle CPU or + * trigger rescheduling on a busy CPU. This can be called from any online + * scx_ops operation and the actual kicking is performed asynchronously through + * an irq work. + */ +__bpf_kfunc void scx_bpf_kick_cpu(s32 cpu, u64 flags, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(aux); + if (likely(sch)) + scx_kick_cpu(sch, cpu, flags); +} + +/** * scx_bpf_dsq_nr_queued - Return the number of queued tasks * @dsq_id: id of the DSQ + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Return the number of tasks in the DSQ matching @dsq_id. If not found, * -%ENOENT is returned. */ -__bpf_kfunc s32 scx_bpf_dsq_nr_queued(u64 dsq_id) +__bpf_kfunc s32 scx_bpf_dsq_nr_queued(u64 dsq_id, const struct bpf_prog_aux *aux) { + struct scx_sched *sch; struct scx_dispatch_q *dsq; s32 ret; preempt_disable(); + sch = scx_prog_sched(aux); + if (unlikely(!sch)) { + ret = -ENODEV; + goto out; + } + if (dsq_id == SCX_DSQ_LOCAL) { ret = READ_ONCE(this_rq()->scx.local_dsq.nr); goto out; } else if ((dsq_id & SCX_DSQ_LOCAL_ON) == SCX_DSQ_LOCAL_ON) { s32 cpu = dsq_id & SCX_DSQ_LOCAL_CPU_MASK; - if (ops_cpu_valid(cpu, NULL)) { + if (ops_cpu_valid(sch, cpu, NULL)) { ret = READ_ONCE(cpu_rq(cpu)->scx.local_dsq.nr); goto out; } } else { - dsq = find_user_dsq(dsq_id); + dsq = find_user_dsq(sch, dsq_id); if (dsq) { ret = READ_ONCE(dsq->nr); goto out; @@ -7142,15 +8998,21 @@ out: /** * scx_bpf_destroy_dsq - Destroy a custom DSQ * @dsq_id: DSQ to destroy + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Destroy the custom DSQ identified by @dsq_id. Only DSQs created with * scx_bpf_create_dsq() can be destroyed. The caller must ensure that the DSQ is * empty and no further tasks are dispatched to it. Ignored if called on a DSQ * which doesn't exist. Can be called from any online scx_ops operations. */ -__bpf_kfunc void scx_bpf_destroy_dsq(u64 dsq_id) +__bpf_kfunc void scx_bpf_destroy_dsq(u64 dsq_id, const struct bpf_prog_aux *aux) { - destroy_dsq(dsq_id); + struct scx_sched *sch; + + guard(rcu)(); + sch = scx_prog_sched(aux); + if (sch) + destroy_dsq(sch, dsq_id); } /** @@ -7158,31 +9020,43 @@ __bpf_kfunc void scx_bpf_destroy_dsq(u64 dsq_id) * @it: iterator to initialize * @dsq_id: DSQ to iterate * @flags: %SCX_DSQ_ITER_* + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Initialize BPF iterator @it which can be used with bpf_for_each() to walk * tasks in the DSQ specified by @dsq_id. Iteration using @it only includes * tasks which are already queued when this function is invoked. */ __bpf_kfunc int bpf_iter_scx_dsq_new(struct bpf_iter_scx_dsq *it, u64 dsq_id, - u64 flags) + u64 flags, const struct bpf_prog_aux *aux) { struct bpf_iter_scx_dsq_kern *kit = (void *)it; + struct scx_sched *sch; BUILD_BUG_ON(sizeof(struct bpf_iter_scx_dsq_kern) > sizeof(struct bpf_iter_scx_dsq)); BUILD_BUG_ON(__alignof__(struct bpf_iter_scx_dsq_kern) != __alignof__(struct bpf_iter_scx_dsq)); + BUILD_BUG_ON(__SCX_DSQ_ITER_ALL_FLAGS & + ((1U << __SCX_DSQ_LNODE_PRIV_SHIFT) - 1)); + + /* + * next() and destroy() will be called regardless of the return value. + * Always clear $kit->dsq. + */ + kit->dsq = NULL; + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; if (flags & ~__SCX_DSQ_ITER_USER_FLAGS) return -EINVAL; - kit->dsq = find_user_dsq(dsq_id); + kit->dsq = find_user_dsq(sch, dsq_id); if (!kit->dsq) return -ENOENT; - INIT_LIST_HEAD(&kit->cursor.node); - kit->cursor.flags = SCX_DSQ_LNODE_ITER_CURSOR | flags; - kit->cursor.priv = READ_ONCE(kit->dsq->seq); + kit->cursor = INIT_DSQ_LIST_CURSOR(kit->cursor, kit->dsq, flags); return 0; } @@ -7196,41 +9070,13 @@ __bpf_kfunc int bpf_iter_scx_dsq_new(struct bpf_iter_scx_dsq *it, u64 dsq_id, __bpf_kfunc struct task_struct *bpf_iter_scx_dsq_next(struct bpf_iter_scx_dsq *it) { struct bpf_iter_scx_dsq_kern *kit = (void *)it; - bool rev = kit->cursor.flags & SCX_DSQ_ITER_REV; - struct task_struct *p; - unsigned long flags; if (!kit->dsq) return NULL; - raw_spin_lock_irqsave(&kit->dsq->lock, flags); - - if (list_empty(&kit->cursor.node)) - p = NULL; - else - p = container_of(&kit->cursor, struct task_struct, scx.dsq_list); - - /* - * Only tasks which were queued before the iteration started are - * visible. This bounds BPF iterations and guarantees that vtime never - * jumps in the other direction while iterating. - */ - do { - p = nldsq_next_task(kit->dsq, p, rev); - } while (p && unlikely(u32_before(kit->cursor.priv, p->scx.dsq_seq))); - - if (p) { - if (rev) - list_move_tail(&kit->cursor.node, &p->scx.dsq_list.node); - else - list_move(&kit->cursor.node, &p->scx.dsq_list.node); - } else { - list_del_init(&kit->cursor.node); - } - - raw_spin_unlock_irqrestore(&kit->dsq->lock, flags); + guard(raw_spinlock_irqsave)(&kit->dsq->lock); - return p; + return nldsq_cursor_next_task(&kit->cursor, kit->dsq); } /** @@ -7256,31 +9102,123 @@ __bpf_kfunc void bpf_iter_scx_dsq_destroy(struct bpf_iter_scx_dsq *it) kit->dsq = NULL; } +/** + * scx_bpf_dsq_peek - Lockless peek at the first element. + * @dsq_id: DSQ to examine. + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Read the first element in the DSQ. This is semantically equivalent to using + * the DSQ iterator, but is lockfree. Of course, like any lockless operation, + * this provides only a point-in-time snapshot, and the contents may change + * by the time any subsequent locking operation reads the queue. + * + * Returns the pointer, or NULL indicates an empty queue OR internal error. + */ +__bpf_kfunc struct task_struct *scx_bpf_dsq_peek(u64 dsq_id, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + struct scx_dispatch_q *dsq; + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return NULL; + + if (unlikely(dsq_id & SCX_DSQ_FLAG_BUILTIN)) { + scx_error(sch, "peek disallowed on builtin DSQ 0x%llx", dsq_id); + return NULL; + } + + dsq = find_user_dsq(sch, dsq_id); + if (unlikely(!dsq)) { + scx_error(sch, "peek on non-existent DSQ 0x%llx", dsq_id); + return NULL; + } + + return rcu_dereference(dsq->first_task); +} + +/** + * scx_bpf_dsq_reenq - Re-enqueue tasks on a DSQ + * @dsq_id: DSQ to re-enqueue + * @reenq_flags: %SCX_RENQ_* + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Iterate over all of the tasks currently enqueued on the DSQ identified by + * @dsq_id, and re-enqueue them in the BPF scheduler. The following DSQs are + * supported: + * + * - Local DSQs (%SCX_DSQ_LOCAL or %SCX_DSQ_LOCAL_ON | $cpu) + * - User DSQs + * + * Re-enqueues are performed asynchronously. Can be called from anywhere. + */ +__bpf_kfunc void scx_bpf_dsq_reenq(u64 dsq_id, u64 reenq_flags, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + struct scx_dispatch_q *dsq; + + guard(preempt)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return; + + if (unlikely(reenq_flags & ~__SCX_REENQ_USER_MASK)) { + scx_error(sch, "invalid SCX_REENQ flags 0x%llx", reenq_flags); + return; + } + + /* not specifying any filter bits is the same as %SCX_REENQ_ANY */ + if (!(reenq_flags & __SCX_REENQ_FILTER_MASK)) + reenq_flags |= SCX_REENQ_ANY; + + dsq = find_dsq_for_dispatch(sch, this_rq(), dsq_id, smp_processor_id()); + schedule_dsq_reenq(sch, dsq, reenq_flags, scx_locked_rq()); +} + +/** + * scx_bpf_reenqueue_local - Re-enqueue tasks on a local DSQ + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Iterate over all of the tasks currently enqueued on the local DSQ of the + * caller's CPU, and re-enqueue them in the BPF scheduler. Can be called from + * anywhere. + * + * This is now a special case of scx_bpf_dsq_reenq() and may be removed in the + * future. + */ +__bpf_kfunc void scx_bpf_reenqueue_local___v2(const struct bpf_prog_aux *aux) +{ + scx_bpf_dsq_reenq(SCX_DSQ_LOCAL, 0, aux); +} + __bpf_kfunc_end_defs(); -static s32 __bstr_format(u64 *data_buf, char *line_buf, size_t line_size, - char *fmt, unsigned long long *data, u32 data__sz) +static s32 __bstr_format(struct scx_sched *sch, u64 *data_buf, char *line_buf, + size_t line_size, char *fmt, unsigned long long *data, + u32 data__sz) { struct bpf_bprintf_data bprintf_data = { .get_bin_args = true }; s32 ret; if (data__sz % 8 || data__sz > MAX_BPRINTF_VARARGS * 8 || (data__sz && !data)) { - scx_ops_error("invalid data=%p and data__sz=%u", - (void *)data, data__sz); + scx_error(sch, "invalid data=%p and data__sz=%u", (void *)data, data__sz); return -EINVAL; } ret = copy_from_kernel_nofault(data_buf, data, data__sz); if (ret < 0) { - scx_ops_error("failed to read data fields (%d)", ret); + scx_error(sch, "failed to read data fields (%d)", ret); return ret; } ret = bpf_bprintf_prepare(fmt, UINT_MAX, data_buf, data__sz / 8, &bprintf_data); if (ret < 0) { - scx_ops_error("format preparation failed (%d)", ret); + scx_error(sch, "format preparation failed (%d)", ret); return ret; } @@ -7288,18 +9226,17 @@ static s32 __bstr_format(u64 *data_buf, char *line_buf, size_t line_size, bprintf_data.bin_args); bpf_bprintf_cleanup(&bprintf_data); if (ret < 0) { - scx_ops_error("(\"%s\", %p, %u) failed to format", - fmt, data, data__sz); + scx_error(sch, "(\"%s\", %p, %u) failed to format", fmt, data, data__sz); return ret; } return ret; } -static s32 bstr_format(struct scx_bstr_buf *buf, +static s32 bstr_format(struct scx_sched *sch, struct scx_bstr_buf *buf, char *fmt, unsigned long long *data, u32 data__sz) { - return __bstr_format(buf->data, buf->line, sizeof(buf->line), + return __bstr_format(sch, buf->data, buf->line, sizeof(buf->line), fmt, data, data__sz); } @@ -7311,19 +9248,23 @@ __bpf_kfunc_start_defs(); * @fmt: error message format string * @data: format string parameters packaged using ___bpf_fill() macro * @data__sz: @data len, must end in '__sz' for the verifier + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Indicate that the BPF scheduler wants to exit gracefully, and initiate ops * disabling. */ __bpf_kfunc void scx_bpf_exit_bstr(s64 exit_code, char *fmt, - unsigned long long *data, u32 data__sz) + unsigned long long *data, u32 data__sz, + const struct bpf_prog_aux *aux) { + struct scx_sched *sch; unsigned long flags; raw_spin_lock_irqsave(&scx_exit_bstr_buf_lock, flags); - if (bstr_format(&scx_exit_bstr_buf, fmt, data, data__sz) >= 0) - scx_ops_exit_kind(SCX_EXIT_UNREG_BPF, exit_code, "%s", - scx_exit_bstr_buf.line); + sch = scx_prog_sched(aux); + if (likely(sch) && + bstr_format(sch, &scx_exit_bstr_buf, fmt, data, data__sz) >= 0) + scx_exit(sch, SCX_EXIT_UNREG_BPF, exit_code, "%s", scx_exit_bstr_buf.line); raw_spin_unlock_irqrestore(&scx_exit_bstr_buf_lock, flags); } @@ -7332,19 +9273,22 @@ __bpf_kfunc void scx_bpf_exit_bstr(s64 exit_code, char *fmt, * @fmt: error message format string * @data: format string parameters packaged using ___bpf_fill() macro * @data__sz: @data len, must end in '__sz' for the verifier + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Indicate that the BPF scheduler encountered a fatal error and initiate ops * disabling. */ __bpf_kfunc void scx_bpf_error_bstr(char *fmt, unsigned long long *data, - u32 data__sz) + u32 data__sz, const struct bpf_prog_aux *aux) { + struct scx_sched *sch; unsigned long flags; raw_spin_lock_irqsave(&scx_exit_bstr_buf_lock, flags); - if (bstr_format(&scx_exit_bstr_buf, fmt, data, data__sz) >= 0) - scx_ops_exit_kind(SCX_EXIT_ERROR_BPF, 0, "%s", - scx_exit_bstr_buf.line); + sch = scx_prog_sched(aux); + if (likely(sch) && + bstr_format(sch, &scx_exit_bstr_buf, fmt, data, data__sz) >= 0) + scx_exit(sch, SCX_EXIT_ERROR_BPF, 0, "%s", scx_exit_bstr_buf.line); raw_spin_unlock_irqrestore(&scx_exit_bstr_buf_lock, flags); } @@ -7353,6 +9297,7 @@ __bpf_kfunc void scx_bpf_error_bstr(char *fmt, unsigned long long *data, * @fmt: format string * @data: format string parameters packaged using ___bpf_fill() macro * @data__sz: @data len, must end in '__sz' for the verifier + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * To be called through scx_bpf_dump() helper from ops.dump(), dump_cpu() and * dump_task() to generate extra debug dump specific to the BPF scheduler. @@ -7361,19 +9306,26 @@ __bpf_kfunc void scx_bpf_error_bstr(char *fmt, unsigned long long *data, * multiple calls. The last line is automatically terminated. */ __bpf_kfunc void scx_bpf_dump_bstr(char *fmt, unsigned long long *data, - u32 data__sz) + u32 data__sz, const struct bpf_prog_aux *aux) { + struct scx_sched *sch; struct scx_dump_data *dd = &scx_dump_data; struct scx_bstr_buf *buf = &dd->buf; s32 ret; + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return; + if (raw_smp_processor_id() != dd->cpu) { - scx_ops_error("scx_bpf_dump() must only be called from ops.dump() and friends"); + scx_error(sch, "scx_bpf_dump() must only be called from ops.dump() and friends"); return; } /* append the formatted string to the line buf */ - ret = __bstr_format(buf->data, buf->line + dd->cursor, + ret = __bstr_format(sch, buf->data, buf->line + dd->cursor, sizeof(buf->line) - dd->cursor, fmt, data, data__sz); if (ret < 0) { dump_line(dd->s, "%s[!] (\"%s\", %p, %u) failed to format (%d)", @@ -7402,14 +9354,20 @@ __bpf_kfunc void scx_bpf_dump_bstr(char *fmt, unsigned long long *data, /** * scx_bpf_cpuperf_cap - Query the maximum relative capacity of a CPU * @cpu: CPU of interest + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Return the maximum relative capacity of @cpu in relation to the most * performant CPU in the system. The return value is in the range [1, * %SCX_CPUPERF_ONE]. See scx_bpf_cpuperf_cur(). */ -__bpf_kfunc u32 scx_bpf_cpuperf_cap(s32 cpu) +__bpf_kfunc u32 scx_bpf_cpuperf_cap(s32 cpu, const struct bpf_prog_aux *aux) { - if (ops_cpu_valid(cpu, NULL)) + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (likely(sch) && ops_cpu_valid(sch, cpu, NULL)) return arch_scale_cpu_capacity(cpu); else return SCX_CPUPERF_ONE; @@ -7418,6 +9376,7 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cap(s32 cpu) /** * scx_bpf_cpuperf_cur - Query the current relative performance of a CPU * @cpu: CPU of interest + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Return the current relative performance of @cpu in relation to its maximum. * The return value is in the range [1, %SCX_CPUPERF_ONE]. @@ -7429,9 +9388,14 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cap(s32 cpu) * * The result is in the range [1, %SCX_CPUPERF_ONE]. */ -__bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu) +__bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu, const struct bpf_prog_aux *aux) { - if (ops_cpu_valid(cpu, NULL)) + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (likely(sch) && ops_cpu_valid(sch, cpu, NULL)) return arch_scale_freq_capacity(cpu); else return SCX_CPUPERF_ONE; @@ -7441,6 +9405,7 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu) * scx_bpf_cpuperf_set - Set the relative performance target of a CPU * @cpu: CPU of interest * @perf: target performance level [0, %SCX_CPUPERF_ONE] + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * * Set the target performance level of @cpu to @perf. @perf is in linear * relative scale between 0 and %SCX_CPUPERF_ONE. This determines how the @@ -7451,25 +9416,62 @@ __bpf_kfunc u32 scx_bpf_cpuperf_cur(s32 cpu) * use. Consult hardware and cpufreq documentation for more information. The * current performance level can be monitored using scx_bpf_cpuperf_cur(). */ -__bpf_kfunc void scx_bpf_cpuperf_set(s32 cpu, u32 perf) +__bpf_kfunc void scx_bpf_cpuperf_set(s32 cpu, u32 perf, const struct bpf_prog_aux *aux) { + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return; + if (unlikely(perf > SCX_CPUPERF_ONE)) { - scx_ops_error("Invalid cpuperf target %u for CPU %d", perf, cpu); + scx_error(sch, "Invalid cpuperf target %u for CPU %d", perf, cpu); return; } - if (ops_cpu_valid(cpu, NULL)) { - struct rq *rq = cpu_rq(cpu); + if (ops_cpu_valid(sch, cpu, NULL)) { + struct rq *rq = cpu_rq(cpu), *locked_rq = scx_locked_rq(); + struct rq_flags rf; + + /* + * When called with an rq lock held, restrict the operation + * to the corresponding CPU to prevent ABBA deadlocks. + */ + if (locked_rq && rq != locked_rq) { + scx_error(sch, "Invalid target CPU %d", cpu); + return; + } + + /* + * If no rq lock is held, allow to operate on any CPU by + * acquiring the corresponding rq lock. + */ + if (!locked_rq) { + rq_lock_irqsave(rq, &rf); + update_rq_clock(rq); + } rq->scx.cpuperf_target = perf; + cpufreq_update_util(rq, 0); - rcu_read_lock_sched_notrace(); - cpufreq_update_util(cpu_rq(cpu), 0); - rcu_read_unlock_sched_notrace(); + if (!locked_rq) + rq_unlock_irqrestore(rq, &rf); } } /** + * scx_bpf_nr_node_ids - Return the number of possible node IDs + * + * All valid node IDs in the system are smaller than the returned value. + */ +__bpf_kfunc u32 scx_bpf_nr_node_ids(void) +{ + return nr_node_ids; +} + +/** * scx_bpf_nr_cpu_ids - Return the number of possible CPU IDs * * All valid CPU IDs in the system are smaller than the returned value. @@ -7510,198 +9512,100 @@ __bpf_kfunc void scx_bpf_put_cpumask(const struct cpumask *cpumask) } /** - * scx_bpf_get_idle_cpumask - Get a referenced kptr to the idle-tracking - * per-CPU cpumask. - * - * Returns NULL if idle tracking is not enabled, or running on a UP kernel. - */ -__bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void) -{ - if (!check_builtin_idle_enabled()) - return cpu_none_mask; - -#ifdef CONFIG_SMP - return idle_masks.cpu; -#else - return cpu_none_mask; -#endif -} - -/** - * scx_bpf_get_idle_smtmask - Get a referenced kptr to the idle-tracking, - * per-physical-core cpumask. Can be used to determine if an entire physical - * core is free. - * - * Returns NULL if idle tracking is not enabled, or running on a UP kernel. + * scx_bpf_task_running - Is task currently running? + * @p: task of interest */ -__bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void) +__bpf_kfunc bool scx_bpf_task_running(const struct task_struct *p) { - if (!check_builtin_idle_enabled()) - return cpu_none_mask; - -#ifdef CONFIG_SMP - if (sched_smt_active()) - return idle_masks.smt; - else - return idle_masks.cpu; -#else - return cpu_none_mask; -#endif + return task_rq(p)->curr == p; } /** - * scx_bpf_put_idle_cpumask - Release a previously acquired referenced kptr to - * either the percpu, or SMT idle-tracking cpumask. - * @idle_mask: &cpumask to use + * scx_bpf_task_cpu - CPU a task is currently associated with + * @p: task of interest */ -__bpf_kfunc void scx_bpf_put_idle_cpumask(const struct cpumask *idle_mask) +__bpf_kfunc s32 scx_bpf_task_cpu(const struct task_struct *p) { - /* - * Empty function body because we aren't actually acquiring or releasing - * a reference to a global idle cpumask, which is read-only in the - * caller and is never released. The acquire / release semantics here - * are just used to make the cpumask a trusted pointer in the caller. - */ + return task_cpu(p); } /** - * scx_bpf_test_and_clear_cpu_idle - Test and clear @cpu's idle state - * @cpu: cpu to test and clear idle for - * - * Returns %true if @cpu was idle and its idle state was successfully cleared. - * %false otherwise. - * - * Unavailable if ops.update_idle() is implemented and - * %SCX_OPS_KEEP_BUILTIN_IDLE is not set. + * scx_bpf_cpu_rq - Fetch the rq of a CPU + * @cpu: CPU of the rq + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs */ -__bpf_kfunc bool scx_bpf_test_and_clear_cpu_idle(s32 cpu) +__bpf_kfunc struct rq *scx_bpf_cpu_rq(s32 cpu, const struct bpf_prog_aux *aux) { - if (!check_builtin_idle_enabled()) - return false; + struct scx_sched *sch; - if (ops_cpu_valid(cpu, NULL)) - return test_and_clear_cpu_idle(cpu); - else - return false; -} + guard(rcu)(); -/** - * scx_bpf_pick_idle_cpu - Pick and claim an idle cpu - * @cpus_allowed: Allowed cpumask - * @flags: %SCX_PICK_IDLE_CPU_* flags - * - * Pick and claim an idle cpu in @cpus_allowed. Returns the picked idle cpu - * number on success. -%EBUSY if no matching cpu was found. - * - * Idle CPU tracking may race against CPU scheduling state transitions. For - * example, this function may return -%EBUSY as CPUs are transitioning into the - * idle state. If the caller then assumes that there will be dispatch events on - * the CPUs as they were all busy, the scheduler may end up stalling with CPUs - * idling while there are pending tasks. Use scx_bpf_pick_any_cpu() and - * scx_bpf_kick_cpu() to guarantee that there will be at least one dispatch - * event in the near future. - * - * Unavailable if ops.update_idle() is implemented and - * %SCX_OPS_KEEP_BUILTIN_IDLE is not set. - */ -__bpf_kfunc s32 scx_bpf_pick_idle_cpu(const struct cpumask *cpus_allowed, - u64 flags) -{ - if (!check_builtin_idle_enabled()) - return -EBUSY; - - return scx_pick_idle_cpu(cpus_allowed, flags); -} + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return NULL; -/** - * scx_bpf_pick_any_cpu - Pick and claim an idle cpu if available or pick any CPU - * @cpus_allowed: Allowed cpumask - * @flags: %SCX_PICK_IDLE_CPU_* flags - * - * Pick and claim an idle cpu in @cpus_allowed. If none is available, pick any - * CPU in @cpus_allowed. Guaranteed to succeed and returns the picked idle cpu - * number if @cpus_allowed is not empty. -%EBUSY is returned if @cpus_allowed is - * empty. - * - * If ops.update_idle() is implemented and %SCX_OPS_KEEP_BUILTIN_IDLE is not - * set, this function can't tell which CPUs are idle and will always pick any - * CPU. - */ -__bpf_kfunc s32 scx_bpf_pick_any_cpu(const struct cpumask *cpus_allowed, - u64 flags) -{ - s32 cpu; + if (!ops_cpu_valid(sch, cpu, NULL)) + return NULL; - if (static_branch_likely(&scx_builtin_idle_enabled)) { - cpu = scx_pick_idle_cpu(cpus_allowed, flags); - if (cpu >= 0) - return cpu; + if (!sch->warned_deprecated_rq) { + printk_deferred(KERN_WARNING "sched_ext: %s() is deprecated; " + "use scx_bpf_locked_rq() when holding rq lock " + "or scx_bpf_cpu_curr() to read remote curr safely.\n", __func__); + sch->warned_deprecated_rq = true; } - cpu = cpumask_any_distribute(cpus_allowed); - if (cpu < nr_cpu_ids) - return cpu; - else - return -EBUSY; + return cpu_rq(cpu); } /** - * scx_bpf_task_running - Is task currently running? - * @p: task of interest + * scx_bpf_locked_rq - Return the rq currently locked by SCX + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns the rq if a rq lock is currently held by SCX. + * Otherwise emits an error and returns NULL. */ -__bpf_kfunc bool scx_bpf_task_running(const struct task_struct *p) +__bpf_kfunc struct rq *scx_bpf_locked_rq(const struct bpf_prog_aux *aux) { - return task_rq(p)->curr == p; -} + struct scx_sched *sch; + struct rq *rq; -/** - * scx_bpf_task_cpu - CPU a task is currently associated with - * @p: task of interest - */ -__bpf_kfunc s32 scx_bpf_task_cpu(const struct task_struct *p) -{ - return task_cpu(p); -} + guard(preempt)(); -/** - * scx_bpf_cpu_rq - Fetch the rq of a CPU - * @cpu: CPU of the rq - */ -__bpf_kfunc struct rq *scx_bpf_cpu_rq(s32 cpu) -{ - if (!ops_cpu_valid(cpu, NULL)) + sch = scx_prog_sched(aux); + if (unlikely(!sch)) return NULL; - return cpu_rq(cpu); + rq = scx_locked_rq(); + if (!rq) { + scx_error(sch, "accessing rq without holding rq lock"); + return NULL; + } + + return rq; } /** - * scx_bpf_task_cgroup - Return the sched cgroup of a task - * @p: task of interest + * scx_bpf_cpu_curr - Return remote CPU's curr task + * @cpu: CPU of interest + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs * - * @p->sched_task_group->css.cgroup represents the cgroup @p is associated with - * from the scheduler's POV. SCX operations should use this function to - * determine @p's current cgroup as, unlike following @p->cgroups, - * @p->sched_task_group is protected by @p's rq lock and thus atomic w.r.t. all - * rq-locked operations. Can be called on the parameter tasks of rq-locked - * operations. The restriction guarantees that @p's rq is locked by the caller. + * Callers must hold RCU read lock (KF_RCU). */ -#ifdef CONFIG_CGROUP_SCHED -__bpf_kfunc struct cgroup *scx_bpf_task_cgroup(struct task_struct *p) +__bpf_kfunc struct task_struct *scx_bpf_cpu_curr(s32 cpu, const struct bpf_prog_aux *aux) { - struct task_group *tg = p->sched_task_group; - struct cgroup *cgrp = &cgrp_dfl_root.cgrp; + struct scx_sched *sch; - if (!scx_kf_allowed_on_arg_tasks(__SCX_KF_RQ_LOCKED, p)) - goto out; + guard(rcu)(); - cgrp = tg_cgrp(tg); + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return NULL; -out: - cgroup_get(cgrp); - return cgrp; + if (!ops_cpu_valid(sch, cpu, NULL)) + return NULL; + + return rcu_dereference(cpu_rq(cpu)->curr); } -#endif /** * scx_bpf_now - Returns a high-performance monotonically non-decreasing @@ -7765,45 +9669,252 @@ __bpf_kfunc u64 scx_bpf_now(void) return clock; } +static void scx_read_events(struct scx_sched *sch, struct scx_event_stats *events) +{ + struct scx_event_stats *e_cpu; + int cpu; + + /* Aggregate per-CPU event counters into @events. */ + memset(events, 0, sizeof(*events)); + for_each_possible_cpu(cpu) { + e_cpu = &per_cpu_ptr(sch->pcpu, cpu)->event_stats; + scx_agg_event(events, e_cpu, SCX_EV_SELECT_CPU_FALLBACK); + scx_agg_event(events, e_cpu, SCX_EV_DISPATCH_LOCAL_DSQ_OFFLINE); + scx_agg_event(events, e_cpu, SCX_EV_DISPATCH_KEEP_LAST); + scx_agg_event(events, e_cpu, SCX_EV_ENQ_SKIP_EXITING); + scx_agg_event(events, e_cpu, SCX_EV_ENQ_SKIP_MIGRATION_DISABLED); + scx_agg_event(events, e_cpu, SCX_EV_REENQ_IMMED); + scx_agg_event(events, e_cpu, SCX_EV_REENQ_LOCAL_REPEAT); + scx_agg_event(events, e_cpu, SCX_EV_REFILL_SLICE_DFL); + scx_agg_event(events, e_cpu, SCX_EV_BYPASS_DURATION); + scx_agg_event(events, e_cpu, SCX_EV_BYPASS_DISPATCH); + scx_agg_event(events, e_cpu, SCX_EV_BYPASS_ACTIVATE); + scx_agg_event(events, e_cpu, SCX_EV_INSERT_NOT_OWNED); + scx_agg_event(events, e_cpu, SCX_EV_SUB_BYPASS_DISPATCH); + } +} + +/* + * scx_bpf_events - Get a system-wide event counter to + * @events: output buffer from a BPF program + * @events__sz: @events len, must end in '__sz'' for the verifier + */ +__bpf_kfunc void scx_bpf_events(struct scx_event_stats *events, + size_t events__sz) +{ + struct scx_sched *sch; + struct scx_event_stats e_sys; + + rcu_read_lock(); + sch = rcu_dereference(scx_root); + if (sch) + scx_read_events(sch, &e_sys); + else + memset(&e_sys, 0, sizeof(e_sys)); + rcu_read_unlock(); + + /* + * We cannot entirely trust a BPF-provided size since a BPF program + * might be compiled against a different vmlinux.h, of which + * scx_event_stats would be larger (a newer vmlinux.h) or smaller + * (an older vmlinux.h). Hence, we use the smaller size to avoid + * memory corruption. + */ + events__sz = min(events__sz, sizeof(*events)); + memcpy(events, &e_sys, events__sz); +} + +#ifdef CONFIG_CGROUP_SCHED +/** + * scx_bpf_task_cgroup - Return the sched cgroup of a task + * @p: task of interest + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * @p->sched_task_group->css.cgroup represents the cgroup @p is associated with + * from the scheduler's POV. SCX operations should use this function to + * determine @p's current cgroup as, unlike following @p->cgroups, + * @p->sched_task_group is stable for the duration of the SCX op. See + * SCX_CALL_OP_TASK() for details. + */ +__bpf_kfunc struct cgroup *scx_bpf_task_cgroup(struct task_struct *p, + const struct bpf_prog_aux *aux) +{ + struct task_group *tg = p->sched_task_group; + struct cgroup *cgrp = &cgrp_dfl_root.cgrp; + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + goto out; + + if (!scx_kf_arg_task_ok(sch, p)) + goto out; + + cgrp = tg_cgrp(tg); + +out: + cgroup_get(cgrp); + return cgrp; +} +#endif /* CONFIG_CGROUP_SCHED */ + __bpf_kfunc_end_defs(); BTF_KFUNCS_START(scx_kfunc_ids_any) -BTF_ID_FLAGS(func, scx_bpf_kick_cpu) -BTF_ID_FLAGS(func, scx_bpf_dsq_nr_queued) -BTF_ID_FLAGS(func, scx_bpf_destroy_dsq) -BTF_ID_FLAGS(func, bpf_iter_scx_dsq_new, KF_ITER_NEW | KF_RCU_PROTECTED) +BTF_ID_FLAGS(func, scx_bpf_task_set_slice, KF_IMPLICIT_ARGS | KF_RCU); +BTF_ID_FLAGS(func, scx_bpf_task_set_dsq_vtime, KF_IMPLICIT_ARGS | KF_RCU); +BTF_ID_FLAGS(func, scx_bpf_kick_cpu, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dsq_nr_queued, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_destroy_dsq, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dsq_peek, KF_IMPLICIT_ARGS | KF_RCU_PROTECTED | KF_RET_NULL) +BTF_ID_FLAGS(func, scx_bpf_dsq_reenq, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_reenqueue_local___v2, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, bpf_iter_scx_dsq_new, KF_IMPLICIT_ARGS | KF_ITER_NEW | KF_RCU_PROTECTED) BTF_ID_FLAGS(func, bpf_iter_scx_dsq_next, KF_ITER_NEXT | KF_RET_NULL) BTF_ID_FLAGS(func, bpf_iter_scx_dsq_destroy, KF_ITER_DESTROY) -BTF_ID_FLAGS(func, scx_bpf_exit_bstr, KF_TRUSTED_ARGS) -BTF_ID_FLAGS(func, scx_bpf_error_bstr, KF_TRUSTED_ARGS) -BTF_ID_FLAGS(func, scx_bpf_dump_bstr, KF_TRUSTED_ARGS) -BTF_ID_FLAGS(func, scx_bpf_cpuperf_cap) -BTF_ID_FLAGS(func, scx_bpf_cpuperf_cur) -BTF_ID_FLAGS(func, scx_bpf_cpuperf_set) +BTF_ID_FLAGS(func, scx_bpf_exit_bstr, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_error_bstr, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_dump_bstr, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_cpuperf_cap, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_cpuperf_cur, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_cpuperf_set, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_nr_node_ids) BTF_ID_FLAGS(func, scx_bpf_nr_cpu_ids) BTF_ID_FLAGS(func, scx_bpf_get_possible_cpumask, KF_ACQUIRE) BTF_ID_FLAGS(func, scx_bpf_get_online_cpumask, KF_ACQUIRE) BTF_ID_FLAGS(func, scx_bpf_put_cpumask, KF_RELEASE) -BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask, KF_ACQUIRE) -BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask, KF_ACQUIRE) -BTF_ID_FLAGS(func, scx_bpf_put_idle_cpumask, KF_RELEASE) -BTF_ID_FLAGS(func, scx_bpf_test_and_clear_cpu_idle) -BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_running, KF_RCU) BTF_ID_FLAGS(func, scx_bpf_task_cpu, KF_RCU) -BTF_ID_FLAGS(func, scx_bpf_cpu_rq) +BTF_ID_FLAGS(func, scx_bpf_cpu_rq, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_locked_rq, KF_IMPLICIT_ARGS | KF_RET_NULL) +BTF_ID_FLAGS(func, scx_bpf_cpu_curr, KF_IMPLICIT_ARGS | KF_RET_NULL | KF_RCU_PROTECTED) +BTF_ID_FLAGS(func, scx_bpf_now) +BTF_ID_FLAGS(func, scx_bpf_events) #ifdef CONFIG_CGROUP_SCHED -BTF_ID_FLAGS(func, scx_bpf_task_cgroup, KF_RCU | KF_ACQUIRE) +BTF_ID_FLAGS(func, scx_bpf_task_cgroup, KF_IMPLICIT_ARGS | KF_RCU | KF_ACQUIRE) #endif -BTF_ID_FLAGS(func, scx_bpf_now) BTF_KFUNCS_END(scx_kfunc_ids_any) static const struct btf_kfunc_id_set scx_kfunc_set_any = { .owner = THIS_MODULE, .set = &scx_kfunc_ids_any, + .filter = scx_kfunc_context_filter, +}; + +/* + * Per-op kfunc allow flags. Each bit corresponds to a context-sensitive kfunc + * group; an op may permit zero or more groups, with the union expressed in + * scx_kf_allow_flags[]. The verifier-time filter (scx_kfunc_context_filter()) + * consults this table to decide whether a context-sensitive kfunc is callable + * from a given SCX op. + */ +enum scx_kf_allow_flags { + SCX_KF_ALLOW_UNLOCKED = 1 << 0, + SCX_KF_ALLOW_CPU_RELEASE = 1 << 1, + SCX_KF_ALLOW_DISPATCH = 1 << 2, + SCX_KF_ALLOW_ENQUEUE = 1 << 3, + SCX_KF_ALLOW_SELECT_CPU = 1 << 4, +}; + +/* + * Map each SCX op to the union of kfunc groups it permits, indexed by + * SCX_OP_IDX(op). Ops not listed only permit kfuncs that are not + * context-sensitive. + */ +static const u32 scx_kf_allow_flags[] = { + [SCX_OP_IDX(select_cpu)] = SCX_KF_ALLOW_SELECT_CPU | SCX_KF_ALLOW_ENQUEUE, + [SCX_OP_IDX(enqueue)] = SCX_KF_ALLOW_SELECT_CPU | SCX_KF_ALLOW_ENQUEUE, + [SCX_OP_IDX(dispatch)] = SCX_KF_ALLOW_ENQUEUE | SCX_KF_ALLOW_DISPATCH, + [SCX_OP_IDX(cpu_release)] = SCX_KF_ALLOW_CPU_RELEASE, + [SCX_OP_IDX(init_task)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(dump)] = SCX_KF_ALLOW_UNLOCKED, +#ifdef CONFIG_EXT_GROUP_SCHED + [SCX_OP_IDX(cgroup_init)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_exit)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_prep_move)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_cancel_move)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_set_weight)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_set_bandwidth)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cgroup_set_idle)] = SCX_KF_ALLOW_UNLOCKED, +#endif /* CONFIG_EXT_GROUP_SCHED */ + [SCX_OP_IDX(sub_attach)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(sub_detach)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cpu_online)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(cpu_offline)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(init)] = SCX_KF_ALLOW_UNLOCKED, + [SCX_OP_IDX(exit)] = SCX_KF_ALLOW_UNLOCKED, }; +/* + * Verifier-time filter for SCX kfuncs. Registered via the .filter field on + * each per-group btf_kfunc_id_set. The BPF core invokes this for every kfunc + * call in the registered hook (BPF_PROG_TYPE_STRUCT_OPS or + * BPF_PROG_TYPE_SYSCALL), regardless of which set originally introduced the + * kfunc - so the filter must short-circuit on kfuncs it doesn't govern by + * falling through to "allow" when none of the SCX sets contain the kfunc. + */ +int scx_kfunc_context_filter(const struct bpf_prog *prog, u32 kfunc_id) +{ + bool in_unlocked = btf_id_set8_contains(&scx_kfunc_ids_unlocked, kfunc_id); + bool in_select_cpu = btf_id_set8_contains(&scx_kfunc_ids_select_cpu, kfunc_id); + bool in_enqueue = btf_id_set8_contains(&scx_kfunc_ids_enqueue_dispatch, kfunc_id); + bool in_dispatch = btf_id_set8_contains(&scx_kfunc_ids_dispatch, kfunc_id); + bool in_cpu_release = btf_id_set8_contains(&scx_kfunc_ids_cpu_release, kfunc_id); + bool in_idle = btf_id_set8_contains(&scx_kfunc_ids_idle, kfunc_id); + bool in_any = btf_id_set8_contains(&scx_kfunc_ids_any, kfunc_id); + u32 moff, flags; + + /* Not an SCX kfunc - allow. */ + if (!(in_unlocked || in_select_cpu || in_enqueue || in_dispatch || + in_cpu_release || in_idle || in_any)) + return 0; + + /* SYSCALL progs (e.g. BPF test_run()) may call unlocked and select_cpu kfuncs. */ + if (prog->type == BPF_PROG_TYPE_SYSCALL) + return (in_unlocked || in_select_cpu || in_idle || in_any) ? 0 : -EACCES; + + if (prog->type != BPF_PROG_TYPE_STRUCT_OPS) + return (in_any || in_idle) ? 0 : -EACCES; + + /* + * add_subprog_and_kfunc() collects all kfunc calls, including dead code + * guarded by bpf_ksym_exists(), before check_attach_btf_id() sets + * prog->aux->st_ops. Allow all kfuncs when st_ops is not yet set; + * do_check_main() re-runs the filter with st_ops set and enforces the + * actual restrictions. + */ + if (!prog->aux->st_ops) + return 0; + + /* + * Non-SCX struct_ops: SCX kfuncs are not permitted. + */ + if (prog->aux->st_ops != &bpf_sched_ext_ops) + return -EACCES; + + /* SCX struct_ops: check the per-op allow list. */ + if (in_any || in_idle) + return 0; + + moff = prog->aux->attach_st_ops_member_off; + flags = scx_kf_allow_flags[SCX_MOFF_IDX(moff)]; + + if ((flags & SCX_KF_ALLOW_UNLOCKED) && in_unlocked) + return 0; + if ((flags & SCX_KF_ALLOW_CPU_RELEASE) && in_cpu_release) + return 0; + if ((flags & SCX_KF_ALLOW_DISPATCH) && in_dispatch) + return 0; + if ((flags & SCX_KF_ALLOW_ENQUEUE) && in_enqueue) + return 0; + if ((flags & SCX_KF_ALLOW_SELECT_CPU) && in_select_cpu) + return 0; + + return -EACCES; +} + static int __init scx_init(void) { int ret; @@ -7813,15 +9924,14 @@ static int __init scx_init(void) * register_btf_kfunc_id_set() needs most of the system to be up. * * Some kfuncs are context-sensitive and can only be called from - * specific SCX ops. They are grouped into BTF sets accordingly. - * Unfortunately, BPF currently doesn't have a way of enforcing such - * restrictions. Eventually, the verifier should be able to enforce - * them. For now, register them the same and make each kfunc explicitly - * check using scx_kf_allowed(). + * specific SCX ops. They are grouped into per-context BTF sets, each + * registered with scx_kfunc_context_filter as its .filter callback. The + * BPF core dedups identical filter pointers per hook + * (btf_populate_kfunc_set()), so the filter is invoked exactly once per + * kfunc lookup; it consults scx_kf_allow_flags[] to enforce per-op + * restrictions at verify time. */ if ((ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, - &scx_kfunc_set_select_cpu)) || - (ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &scx_kfunc_set_enqueue_dispatch)) || (ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &scx_kfunc_set_dispatch)) || @@ -7841,6 +9951,12 @@ static int __init scx_init(void) return ret; } + ret = scx_idle_init(); + if (ret) { + pr_err("sched_ext: Failed to initialize idle tracking (%d)\n", ret); + return ret; + } + ret = register_bpf_struct_ops(&bpf_sched_ext_ops, sched_ext_ops); if (ret) { pr_err("sched_ext: Failed to register struct_ops (%d)\n", ret); diff --git a/kernel/sched/ext.h b/kernel/sched/ext.h index 1079b56b0f7a..0b7fc46aee08 100644 --- a/kernel/sched/ext.h +++ b/kernel/sched/ext.h @@ -11,7 +11,7 @@ void scx_tick(struct rq *rq); void init_scx_entity(struct sched_ext_entity *scx); void scx_pre_fork(struct task_struct *p); -int scx_fork(struct task_struct *p); +int scx_fork(struct task_struct *p, struct kernel_clone_args *kargs); void scx_post_fork(struct task_struct *p); void scx_cancel_fork(struct task_struct *p); bool scx_can_stop_tick(struct rq *rq); @@ -19,6 +19,7 @@ void scx_rq_activate(struct rq *rq); void scx_rq_deactivate(struct rq *rq); int scx_check_setscheduler(struct task_struct *p, int policy); bool task_should_scx(int policy); +bool scx_allow_ttwu_queue(const struct task_struct *p); void init_sched_ext_class(void); static inline u32 scx_cpuperf_target(s32 cpu) @@ -43,7 +44,7 @@ bool scx_prio_less(const struct task_struct *a, const struct task_struct *b, static inline void scx_tick(struct rq *rq) {} static inline void scx_pre_fork(struct task_struct *p) {} -static inline int scx_fork(struct task_struct *p) { return 0; } +static inline int scx_fork(struct task_struct *p, struct kernel_clone_args *kargs) { return 0; } static inline void scx_post_fork(struct task_struct *p) {} static inline void scx_cancel_fork(struct task_struct *p) {} static inline u32 scx_cpuperf_target(s32 cpu) { return 0; } @@ -52,11 +53,12 @@ static inline void scx_rq_activate(struct rq *rq) {} static inline void scx_rq_deactivate(struct rq *rq) {} static inline int scx_check_setscheduler(struct task_struct *p, int policy) { return 0; } static inline bool task_on_scx(const struct task_struct *p) { return false; } +static inline bool scx_allow_ttwu_queue(const struct task_struct *p) { return true; } static inline void init_sched_ext_class(void) {} #endif /* CONFIG_SCHED_CLASS_EXT */ -#if defined(CONFIG_SCHED_CLASS_EXT) && defined(CONFIG_SMP) +#ifdef CONFIG_SCHED_CLASS_EXT void __scx_update_idle(struct rq *rq, bool idle, bool do_notify); static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) @@ -70,22 +72,24 @@ static inline void scx_update_idle(struct rq *rq, bool idle, bool do_notify) {} #ifdef CONFIG_CGROUP_SCHED #ifdef CONFIG_EXT_GROUP_SCHED +void scx_tg_init(struct task_group *tg); int scx_tg_online(struct task_group *tg); void scx_tg_offline(struct task_group *tg); int scx_cgroup_can_attach(struct cgroup_taskset *tset); void scx_cgroup_move_task(struct task_struct *p); -void scx_cgroup_finish_attach(void); void scx_cgroup_cancel_attach(struct cgroup_taskset *tset); void scx_group_set_weight(struct task_group *tg, unsigned long cgrp_weight); void scx_group_set_idle(struct task_group *tg, bool idle); +void scx_group_set_bandwidth(struct task_group *tg, u64 period_us, u64 quota_us, u64 burst_us); #else /* CONFIG_EXT_GROUP_SCHED */ +static inline void scx_tg_init(struct task_group *tg) {} static inline int scx_tg_online(struct task_group *tg) { return 0; } static inline void scx_tg_offline(struct task_group *tg) {} static inline int scx_cgroup_can_attach(struct cgroup_taskset *tset) { return 0; } static inline void scx_cgroup_move_task(struct task_struct *p) {} -static inline void scx_cgroup_finish_attach(void) {} static inline void scx_cgroup_cancel_attach(struct cgroup_taskset *tset) {} static inline void scx_group_set_weight(struct task_group *tg, unsigned long cgrp_weight) {} static inline void scx_group_set_idle(struct task_group *tg, bool idle) {} +static inline void scx_group_set_bandwidth(struct task_group *tg, u64 period_us, u64 quota_us, u64 burst_us) {} #endif /* CONFIG_EXT_GROUP_SCHED */ #endif /* CONFIG_CGROUP_SCHED */ diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c new file mode 100644 index 000000000000..6e1980763270 --- /dev/null +++ b/kernel/sched/ext_idle.c @@ -0,0 +1,1522 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * BPF extensible scheduler class: Documentation/scheduler/sched-ext.rst + * + * Built-in idle CPU tracking policy. + * + * Copyright (c) 2022 Meta Platforms, Inc. and affiliates. + * Copyright (c) 2022 Tejun Heo <tj@kernel.org> + * Copyright (c) 2022 David Vernet <dvernet@meta.com> + * Copyright (c) 2024 Andrea Righi <arighi@nvidia.com> + */ +#include "ext_idle.h" + +/* Enable/disable built-in idle CPU selection policy */ +static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_enabled); + +/* Enable/disable per-node idle cpumasks */ +static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_per_node); + +/* Enable/disable LLC aware optimizations */ +static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_llc); + +/* Enable/disable NUMA aware optimizations */ +static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_numa); + +/* + * cpumasks to track idle CPUs within each NUMA node. + * + * If SCX_OPS_BUILTIN_IDLE_PER_NODE is not enabled, a single global cpumask + * from is used to track all the idle CPUs in the system. + */ +struct scx_idle_cpus { + cpumask_var_t cpu; + cpumask_var_t smt; +}; + +/* + * Global host-wide idle cpumasks (used when SCX_OPS_BUILTIN_IDLE_PER_NODE + * is not enabled). + */ +static struct scx_idle_cpus scx_idle_global_masks; + +/* + * Per-node idle cpumasks. + */ +static struct scx_idle_cpus **scx_idle_node_masks; + +/* + * Local per-CPU cpumasks (used to generate temporary idle cpumasks). + */ +static DEFINE_PER_CPU(cpumask_var_t, local_idle_cpumask); +static DEFINE_PER_CPU(cpumask_var_t, local_llc_idle_cpumask); +static DEFINE_PER_CPU(cpumask_var_t, local_numa_idle_cpumask); + +/* + * Return the idle masks associated to a target @node. + * + * NUMA_NO_NODE identifies the global idle cpumask. + */ +static struct scx_idle_cpus *idle_cpumask(int node) +{ + return node == NUMA_NO_NODE ? &scx_idle_global_masks : scx_idle_node_masks[node]; +} + +/* + * Returns the NUMA node ID associated with a @cpu, or NUMA_NO_NODE if + * per-node idle cpumasks are disabled. + */ +static int scx_cpu_node_if_enabled(int cpu) +{ + if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) + return NUMA_NO_NODE; + + return cpu_to_node(cpu); +} + +static bool scx_idle_test_and_clear_cpu(int cpu) +{ + int node = scx_cpu_node_if_enabled(cpu); + struct cpumask *idle_cpus = idle_cpumask(node)->cpu; + +#ifdef CONFIG_SCHED_SMT + /* + * SMT mask should be cleared whether we can claim @cpu or not. The SMT + * cluster is not wholly idle either way. This also prevents + * scx_pick_idle_cpu() from getting caught in an infinite loop. + */ + if (sched_smt_active()) { + const struct cpumask *smt = cpu_smt_mask(cpu); + struct cpumask *idle_smts = idle_cpumask(node)->smt; + + /* + * If offline, @cpu is not its own sibling and + * scx_pick_idle_cpu() can get caught in an infinite loop as + * @cpu is never cleared from the idle SMT mask. Ensure that + * @cpu is eventually cleared. + * + * NOTE: Use cpumask_intersects() and cpumask_test_cpu() to + * reduce memory writes, which may help alleviate cache + * coherence pressure. + */ + if (cpumask_intersects(smt, idle_smts)) + cpumask_andnot(idle_smts, idle_smts, smt); + else if (cpumask_test_cpu(cpu, idle_smts)) + __cpumask_clear_cpu(cpu, idle_smts); + } +#endif + + return cpumask_test_and_clear_cpu(cpu, idle_cpus); +} + +/* + * Pick an idle CPU in a specific NUMA node. + */ +static s32 pick_idle_cpu_in_node(const struct cpumask *cpus_allowed, int node, u64 flags) +{ + int cpu; + +retry: + if (sched_smt_active()) { + cpu = cpumask_any_and_distribute(idle_cpumask(node)->smt, cpus_allowed); + if (cpu < nr_cpu_ids) + goto found; + + if (flags & SCX_PICK_IDLE_CORE) + return -EBUSY; + } + + cpu = cpumask_any_and_distribute(idle_cpumask(node)->cpu, cpus_allowed); + if (cpu >= nr_cpu_ids) + return -EBUSY; + +found: + if (scx_idle_test_and_clear_cpu(cpu)) + return cpu; + else + goto retry; +} + +#ifdef CONFIG_NUMA +/* + * Tracks nodes that have not yet been visited when searching for an idle + * CPU across all available nodes. + */ +static DEFINE_PER_CPU(nodemask_t, per_cpu_unvisited); + +/* + * Search for an idle CPU across all nodes, excluding @node. + */ +static s32 pick_idle_cpu_from_online_nodes(const struct cpumask *cpus_allowed, int node, u64 flags) +{ + nodemask_t *unvisited; + s32 cpu = -EBUSY; + + preempt_disable(); + unvisited = this_cpu_ptr(&per_cpu_unvisited); + + /* + * Restrict the search to the online nodes (excluding the current + * node that has been visited already). + */ + nodes_copy(*unvisited, node_states[N_ONLINE]); + node_clear(node, *unvisited); + + /* + * Traverse all nodes in order of increasing distance, starting + * from @node. + * + * This loop is O(N^2), with N being the amount of NUMA nodes, + * which might be quite expensive in large NUMA systems. However, + * this complexity comes into play only when a scheduler enables + * SCX_OPS_BUILTIN_IDLE_PER_NODE and it's requesting an idle CPU + * without specifying a target NUMA node, so it shouldn't be a + * bottleneck is most cases. + * + * As a future optimization we may want to cache the list of nodes + * in a per-node array, instead of actually traversing them every + * time. + */ + for_each_node_numadist(node, *unvisited) { + cpu = pick_idle_cpu_in_node(cpus_allowed, node, flags); + if (cpu >= 0) + break; + } + preempt_enable(); + + return cpu; +} +#else +static inline s32 +pick_idle_cpu_from_online_nodes(const struct cpumask *cpus_allowed, int node, u64 flags) +{ + return -EBUSY; +} +#endif + +/* + * Find an idle CPU in the system, starting from @node. + */ +static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, int node, u64 flags) +{ + s32 cpu; + + /* + * Always search in the starting node first (this is an + * optimization that can save some cycles even when the search is + * not limited to a single node). + */ + cpu = pick_idle_cpu_in_node(cpus_allowed, node, flags); + if (cpu >= 0) + return cpu; + + /* + * Stop the search if we are using only a single global cpumask + * (NUMA_NO_NODE) or if the search is restricted to the first node + * only. + */ + if (node == NUMA_NO_NODE || flags & SCX_PICK_IDLE_IN_NODE) + return -EBUSY; + + /* + * Extend the search to the other online nodes. + */ + return pick_idle_cpu_from_online_nodes(cpus_allowed, node, flags); +} + +/* + * Return the amount of CPUs in the same LLC domain of @cpu (or zero if the LLC + * domain is not defined). + */ +static unsigned int llc_weight(s32 cpu) +{ + struct sched_domain *sd; + + sd = rcu_dereference(per_cpu(sd_llc, cpu)); + if (!sd) + return 0; + + return sd->span_weight; +} + +/* + * Return the cpumask representing the LLC domain of @cpu (or NULL if the LLC + * domain is not defined). + */ +static struct cpumask *llc_span(s32 cpu) +{ + struct sched_domain *sd; + + sd = rcu_dereference(per_cpu(sd_llc, cpu)); + if (!sd) + return NULL; + + return sched_domain_span(sd); +} + +/* + * Return the amount of CPUs in the same NUMA domain of @cpu (or zero if the + * NUMA domain is not defined). + */ +static unsigned int numa_weight(s32 cpu) +{ + struct sched_domain *sd; + struct sched_group *sg; + + sd = rcu_dereference(per_cpu(sd_numa, cpu)); + if (!sd) + return 0; + sg = sd->groups; + if (!sg) + return 0; + + return sg->group_weight; +} + +/* + * Return the cpumask representing the NUMA domain of @cpu (or NULL if the NUMA + * domain is not defined). + */ +static struct cpumask *numa_span(s32 cpu) +{ + struct sched_domain *sd; + struct sched_group *sg; + + sd = rcu_dereference(per_cpu(sd_numa, cpu)); + if (!sd) + return NULL; + sg = sd->groups; + if (!sg) + return NULL; + + return sched_group_span(sg); +} + +/* + * Return true if the LLC domains do not perfectly overlap with the NUMA + * domains, false otherwise. + */ +static bool llc_numa_mismatch(void) +{ + int cpu; + + /* + * We need to scan all online CPUs to verify whether their scheduling + * domains overlap. + * + * While it is rare to encounter architectures with asymmetric NUMA + * topologies, CPU hotplugging or virtualized environments can result + * in asymmetric configurations. + * + * For example: + * + * NUMA 0: + * - LLC 0: cpu0..cpu7 + * - LLC 1: cpu8..cpu15 [offline] + * + * NUMA 1: + * - LLC 0: cpu16..cpu23 + * - LLC 1: cpu24..cpu31 + * + * In this case, if we only check the first online CPU (cpu0), we might + * incorrectly assume that the LLC and NUMA domains are fully + * overlapping, which is incorrect (as NUMA 1 has two distinct LLC + * domains). + */ + for_each_online_cpu(cpu) + if (llc_weight(cpu) != numa_weight(cpu)) + return true; + + return false; +} + +/* + * Initialize topology-aware scheduling. + * + * Detect if the system has multiple LLC or multiple NUMA domains and enable + * cache-aware / NUMA-aware scheduling optimizations in the default CPU idle + * selection policy. + * + * Assumption: the kernel's internal topology representation assumes that each + * CPU belongs to a single LLC domain, and that each LLC domain is entirely + * contained within a single NUMA node. + */ +void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops) +{ + bool enable_llc = false, enable_numa = false; + unsigned int nr_cpus; + s32 cpu = cpumask_first(cpu_online_mask); + + /* + * Enable LLC domain optimization only when there are multiple LLC + * domains among the online CPUs. If all online CPUs are part of a + * single LLC domain, the idle CPU selection logic can choose any + * online CPU without bias. + * + * Note that it is sufficient to check the LLC domain of the first + * online CPU to determine whether a single LLC domain includes all + * CPUs. + */ + rcu_read_lock(); + nr_cpus = llc_weight(cpu); + if (nr_cpus > 0) { + if (nr_cpus < num_online_cpus()) + enable_llc = true; + pr_debug("sched_ext: LLC=%*pb weight=%u\n", + cpumask_pr_args(llc_span(cpu)), llc_weight(cpu)); + } + + /* + * Enable NUMA optimization only when there are multiple NUMA domains + * among the online CPUs and the NUMA domains don't perfectly overlap + * with the LLC domains. + * + * If all CPUs belong to the same NUMA node and the same LLC domain, + * enabling both NUMA and LLC optimizations is unnecessary, as checking + * for an idle CPU in the same domain twice is redundant. + * + * If SCX_OPS_BUILTIN_IDLE_PER_NODE is enabled ignore the NUMA + * optimization, as we would naturally select idle CPUs within + * specific NUMA nodes querying the corresponding per-node cpumask. + */ + if (!(ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE)) { + nr_cpus = numa_weight(cpu); + if (nr_cpus > 0) { + if (nr_cpus < num_online_cpus() && llc_numa_mismatch()) + enable_numa = true; + pr_debug("sched_ext: NUMA=%*pb weight=%u\n", + cpumask_pr_args(numa_span(cpu)), nr_cpus); + } + } + rcu_read_unlock(); + + pr_debug("sched_ext: LLC idle selection %s\n", + str_enabled_disabled(enable_llc)); + pr_debug("sched_ext: NUMA idle selection %s\n", + str_enabled_disabled(enable_numa)); + + if (enable_llc) + static_branch_enable_cpuslocked(&scx_selcpu_topo_llc); + else + static_branch_disable_cpuslocked(&scx_selcpu_topo_llc); + if (enable_numa) + static_branch_enable_cpuslocked(&scx_selcpu_topo_numa); + else + static_branch_disable_cpuslocked(&scx_selcpu_topo_numa); +} + +/* + * Return true if @p can run on all possible CPUs, false otherwise. + */ +static inline bool task_affinity_all(const struct task_struct *p) +{ + return p->nr_cpus_allowed >= num_possible_cpus(); +} + +/* + * Built-in CPU idle selection policy: + * + * 1. Prioritize full-idle cores: + * - always prioritize CPUs from fully idle cores (both logical CPUs are + * idle) to avoid interference caused by SMT. + * + * 2. Reuse the same CPU: + * - prefer the last used CPU to take advantage of cached data (L1, L2) and + * branch prediction optimizations. + * + * 3. Prefer @prev_cpu's SMT sibling: + * - if @prev_cpu is busy and no fully idle core is available, try to + * place the task on an idle SMT sibling of @prev_cpu; keeping the + * task on the same core makes migration cheaper, preserves L1 cache + * locality and reduces wakeup latency. + * + * 4. Pick a CPU within the same LLC (Last-Level Cache): + * - if the above conditions aren't met, pick a CPU that shares the same + * LLC, if the LLC domain is a subset of @cpus_allowed, to maintain + * cache locality. + * + * 5. Pick a CPU within the same NUMA node, if enabled: + * - choose a CPU from the same NUMA node, if the node cpumask is a + * subset of @cpus_allowed, to reduce memory access latency. + * + * 6. Pick any idle CPU within the @cpus_allowed domain. + * + * Step 4 and 5 are performed only if the system has, respectively, + * multiple LLCs / multiple NUMA nodes (see scx_selcpu_topo_llc and + * scx_selcpu_topo_numa) and they don't contain the same subset of CPUs. + * + * If %SCX_OPS_BUILTIN_IDLE_PER_NODE is enabled, the search will always + * begin in @prev_cpu's node and proceed to other nodes in order of + * increasing distance. + * + * Return the picked CPU if idle, or a negative value otherwise. + * + * NOTE: tasks that can only run on 1 CPU are excluded by this logic, because + * we never call ops.select_cpu() for them, see select_task_rq(). + */ +s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags) +{ + const struct cpumask *llc_cpus = NULL, *numa_cpus = NULL; + const struct cpumask *allowed = cpus_allowed ?: p->cpus_ptr; + int node = scx_cpu_node_if_enabled(prev_cpu); + bool is_prev_allowed; + s32 cpu; + + preempt_disable(); + + /* + * Determine the subset of CPUs usable by @p within @cpus_allowed. + */ + if (allowed != p->cpus_ptr) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_idle_cpumask); + + if (task_affinity_all(p)) { + allowed = cpus_allowed; + } else if (cpumask_and(local_cpus, cpus_allowed, p->cpus_ptr)) { + allowed = local_cpus; + } else { + cpu = -EBUSY; + goto out_enable; + } + } + + /* + * Check whether @prev_cpu is still within the allowed set. If not, + * we can still try selecting a nearby CPU. + */ + is_prev_allowed = cpumask_test_cpu(prev_cpu, allowed); + + /* + * This is necessary to protect llc_cpus. + */ + rcu_read_lock(); + + /* + * Determine the subset of CPUs that the task can use in its + * current LLC and node. + * + * If the task can run on all CPUs, use the node and LLC cpumasks + * directly. + */ + if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_numa_idle_cpumask); + const struct cpumask *cpus = numa_span(prev_cpu); + + if (allowed == p->cpus_ptr && task_affinity_all(p)) + numa_cpus = cpus; + else if (cpus && cpumask_and(local_cpus, allowed, cpus)) + numa_cpus = local_cpus; + } + + if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) { + struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_llc_idle_cpumask); + const struct cpumask *cpus = llc_span(prev_cpu); + + if (allowed == p->cpus_ptr && task_affinity_all(p)) + llc_cpus = cpus; + else if (cpus && cpumask_and(local_cpus, allowed, cpus)) + llc_cpus = local_cpus; + } + + /* + * If WAKE_SYNC, try to migrate the wakee to the waker's CPU. + */ + if (wake_flags & SCX_WAKE_SYNC) { + int waker_node; + + /* + * If the waker's CPU is cache affine and prev_cpu is idle, + * then avoid a migration. + */ + cpu = smp_processor_id(); + if (is_prev_allowed && cpus_share_cache(cpu, prev_cpu) && + scx_idle_test_and_clear_cpu(prev_cpu)) { + cpu = prev_cpu; + goto out_unlock; + } + + /* + * If the waker's local DSQ is empty, and the system is under + * utilized, try to wake up @p to the local DSQ of the waker. + * + * Checking only for an empty local DSQ is insufficient as it + * could give the wakee an unfair advantage when the system is + * oversaturated. + * + * Checking only for the presence of idle CPUs is also + * insufficient as the local DSQ of the waker could have tasks + * piled up on it even if there is an idle core elsewhere on + * the system. + */ + waker_node = scx_cpu_node_if_enabled(cpu); + if (!(current->flags & PF_EXITING) && + cpu_rq(cpu)->scx.local_dsq.nr == 0 && + (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) && + !cpumask_empty(idle_cpumask(waker_node)->cpu)) { + if (cpumask_test_cpu(cpu, allowed)) + goto out_unlock; + } + } + + /* + * If CPU has SMT, any wholly idle CPU is likely a better pick than + * partially idle @prev_cpu. + */ + if (sched_smt_active()) { + /* + * Keep using @prev_cpu if it's part of a fully idle core. + */ + if (is_prev_allowed && + cpumask_test_cpu(prev_cpu, idle_cpumask(node)->smt) && + scx_idle_test_and_clear_cpu(prev_cpu)) { + cpu = prev_cpu; + goto out_unlock; + } + + /* + * Search for any fully idle core in the same LLC domain. + */ + if (llc_cpus) { + cpu = pick_idle_cpu_in_node(llc_cpus, node, SCX_PICK_IDLE_CORE); + if (cpu >= 0) + goto out_unlock; + } + + /* + * Search for any fully idle core in the same NUMA node. + */ + if (numa_cpus) { + cpu = pick_idle_cpu_in_node(numa_cpus, node, SCX_PICK_IDLE_CORE); + if (cpu >= 0) + goto out_unlock; + } + + /* + * Search for any full-idle core usable by the task. + * + * If the node-aware idle CPU selection policy is enabled + * (%SCX_OPS_BUILTIN_IDLE_PER_NODE), the search will always + * begin in prev_cpu's node and proceed to other nodes in + * order of increasing distance. + */ + cpu = scx_pick_idle_cpu(allowed, node, flags | SCX_PICK_IDLE_CORE); + if (cpu >= 0) + goto out_unlock; + + /* + * Give up if we're strictly looking for a full-idle SMT + * core. + */ + if (flags & SCX_PICK_IDLE_CORE) { + cpu = -EBUSY; + goto out_unlock; + } + } + + /* + * Use @prev_cpu if it's idle. + */ + if (is_prev_allowed && scx_idle_test_and_clear_cpu(prev_cpu)) { + cpu = prev_cpu; + goto out_unlock; + } + +#ifdef CONFIG_SCHED_SMT + /* + * Use @prev_cpu's sibling if it's idle. + */ + if (sched_smt_active()) { + for_each_cpu_and(cpu, cpu_smt_mask(prev_cpu), allowed) { + if (cpu == prev_cpu) + continue; + if (scx_idle_test_and_clear_cpu(cpu)) + goto out_unlock; + } + } +#endif + + /* + * Search for any idle CPU in the same LLC domain. + */ + if (llc_cpus) { + cpu = pick_idle_cpu_in_node(llc_cpus, node, 0); + if (cpu >= 0) + goto out_unlock; + } + + /* + * Search for any idle CPU in the same NUMA node. + */ + if (numa_cpus) { + cpu = pick_idle_cpu_in_node(numa_cpus, node, 0); + if (cpu >= 0) + goto out_unlock; + } + + /* + * Search for any idle CPU usable by the task. + * + * If the node-aware idle CPU selection policy is enabled + * (%SCX_OPS_BUILTIN_IDLE_PER_NODE), the search will always begin + * in prev_cpu's node and proceed to other nodes in order of + * increasing distance. + */ + cpu = scx_pick_idle_cpu(allowed, node, flags); + +out_unlock: + rcu_read_unlock(); +out_enable: + preempt_enable(); + + return cpu; +} + +/* + * Initialize global and per-node idle cpumasks. + */ +void scx_idle_init_masks(void) +{ + int i; + + /* Allocate global idle cpumasks */ + BUG_ON(!alloc_cpumask_var(&scx_idle_global_masks.cpu, GFP_KERNEL)); + BUG_ON(!alloc_cpumask_var(&scx_idle_global_masks.smt, GFP_KERNEL)); + + /* Allocate per-node idle cpumasks (use nr_node_ids for non-contiguous NUMA nodes) */ + scx_idle_node_masks = kzalloc_objs(*scx_idle_node_masks, nr_node_ids); + BUG_ON(!scx_idle_node_masks); + + for_each_node(i) { + scx_idle_node_masks[i] = kzalloc_node(sizeof(**scx_idle_node_masks), + GFP_KERNEL, i); + BUG_ON(!scx_idle_node_masks[i]); + + BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[i]->cpu, GFP_KERNEL, i)); + BUG_ON(!alloc_cpumask_var_node(&scx_idle_node_masks[i]->smt, GFP_KERNEL, i)); + } + + /* Allocate local per-cpu idle cpumasks */ + for_each_possible_cpu(i) { + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_llc_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); + BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_numa_idle_cpumask, i), + GFP_KERNEL, cpu_to_node(i))); + } +} + +static void update_builtin_idle(int cpu, bool idle) +{ + int node = scx_cpu_node_if_enabled(cpu); + struct cpumask *idle_cpus = idle_cpumask(node)->cpu; + + assign_cpu(cpu, idle_cpus, idle); + +#ifdef CONFIG_SCHED_SMT + if (sched_smt_active()) { + const struct cpumask *smt = cpu_smt_mask(cpu); + struct cpumask *idle_smts = idle_cpumask(node)->smt; + + if (idle) { + /* + * idle_smt handling is racy but that's fine as it's + * only for optimization and self-correcting. + */ + if (!cpumask_subset(smt, idle_cpus)) + return; + cpumask_or(idle_smts, idle_smts, smt); + } else { + cpumask_andnot(idle_smts, idle_smts, smt); + } + } +#endif +} + +/* + * Update the idle state of a CPU to @idle. + * + * If @do_notify is true, ops.update_idle() is invoked to notify the scx + * scheduler of an actual idle state transition (idle to busy or vice + * versa). If @do_notify is false, only the idle state in the idle masks is + * refreshed without invoking ops.update_idle(). + * + * This distinction is necessary, because an idle CPU can be "reserved" and + * awakened via scx_bpf_pick_idle_cpu() + scx_bpf_kick_cpu(), marking it as + * busy even if no tasks are dispatched. In this case, the CPU may return + * to idle without a true state transition. Refreshing the idle masks + * without invoking ops.update_idle() ensures accurate idle state tracking + * while avoiding unnecessary updates and maintaining balanced state + * transitions. + */ +void __scx_update_idle(struct rq *rq, bool idle, bool do_notify) +{ + struct scx_sched *sch = scx_root; + int cpu = cpu_of(rq); + + lockdep_assert_rq_held(rq); + + /* + * Update the idle masks: + * - for real idle transitions (do_notify == true) + * - for idle-to-idle transitions (indicated by the previous task + * being the idle thread, managed by pick_task_idle()) + * + * Skip updating idle masks if the previous task is not the idle + * thread, since set_next_task_idle() has already handled it when + * transitioning from a task to the idle thread (calling this + * function with do_notify == true). + * + * In this way we can avoid updating the idle masks twice, + * unnecessarily. + */ + if (static_branch_likely(&scx_builtin_idle_enabled)) + if (do_notify || is_idle_task(rq->curr)) + update_builtin_idle(cpu, idle); + + /* + * Trigger ops.update_idle() only when transitioning from a task to + * the idle thread and vice versa. + * + * Idle transitions are indicated by do_notify being set to true, + * managed by put_prev_task_idle()/set_next_task_idle(). + * + * This must come after builtin idle update so that BPF schedulers can + * create interlocking between ops.update_idle() and ops.enqueue() - + * either enqueue() sees the idle bit or update_idle() sees the task + * that enqueue() queued. + */ + if (SCX_HAS_OP(sch, update_idle) && do_notify && + !scx_bypassing(sch, cpu_of(rq))) + SCX_CALL_OP(sch, update_idle, rq, cpu_of(rq), idle); +} + +static void reset_idle_masks(struct sched_ext_ops *ops) +{ + int node; + + /* + * Consider all online cpus idle. Should converge to the actual state + * quickly. + */ + if (!(ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE)) { + cpumask_copy(idle_cpumask(NUMA_NO_NODE)->cpu, cpu_online_mask); + cpumask_copy(idle_cpumask(NUMA_NO_NODE)->smt, cpu_online_mask); + return; + } + + for_each_node(node) { + const struct cpumask *node_mask = cpumask_of_node(node); + + cpumask_and(idle_cpumask(node)->cpu, cpu_online_mask, node_mask); + cpumask_and(idle_cpumask(node)->smt, cpu_online_mask, node_mask); + } +} + +void scx_idle_enable(struct sched_ext_ops *ops) +{ + if (!ops->update_idle || (ops->flags & SCX_OPS_KEEP_BUILTIN_IDLE)) + static_branch_enable_cpuslocked(&scx_builtin_idle_enabled); + else + static_branch_disable_cpuslocked(&scx_builtin_idle_enabled); + + if (ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE) + static_branch_enable_cpuslocked(&scx_builtin_idle_per_node); + else + static_branch_disable_cpuslocked(&scx_builtin_idle_per_node); + + reset_idle_masks(ops); +} + +void scx_idle_disable(void) +{ + static_branch_disable(&scx_builtin_idle_enabled); + static_branch_disable(&scx_builtin_idle_per_node); +} + +/******************************************************************************** + * Helpers that can be called from the BPF scheduler. + */ + +static int validate_node(struct scx_sched *sch, int node) +{ + if (!static_branch_likely(&scx_builtin_idle_per_node)) { + scx_error(sch, "per-node idle tracking is disabled"); + return -EOPNOTSUPP; + } + + /* Return no entry for NUMA_NO_NODE (not a critical scx error) */ + if (node == NUMA_NO_NODE) + return -ENOENT; + + /* Make sure node is in a valid range */ + if (node < 0 || node >= nr_node_ids) { + scx_error(sch, "invalid node %d", node); + return -EINVAL; + } + + /* Make sure the node is part of the set of possible nodes */ + if (!node_possible(node)) { + scx_error(sch, "unavailable node %d", node); + return -EINVAL; + } + + return node; +} + +__bpf_kfunc_start_defs(); + +static bool check_builtin_idle_enabled(struct scx_sched *sch) +{ + if (static_branch_likely(&scx_builtin_idle_enabled)) + return true; + + scx_error(sch, "built-in idle tracking is disabled"); + return false; +} + +/* + * Determine whether @p is a migration-disabled task in the context of BPF + * code. + * + * We can't simply check whether @p->migration_disabled is set in a + * sched_ext callback, because the BPF prolog (__bpf_prog_enter) may disable + * migration for the current task while running BPF code. + * + * Since the BPF prolog calls migrate_disable() only when CONFIG_PREEMPT_RCU + * is enabled (via rcu_read_lock_dont_migrate()), migration_disabled == 1 for + * the current task is ambiguous only in that case: it could be from the BPF + * prolog rather than a real migrate_disable() call. + * + * Without CONFIG_PREEMPT_RCU, the BPF prolog never calls migrate_disable(), + * so migration_disabled == 1 always means the task is truly + * migration-disabled. + * + * Therefore, when migration_disabled == 1 and CONFIG_PREEMPT_RCU is enabled, + * check whether @p is the current task or not: if it is, then migration was + * not disabled before entering the callback, otherwise migration was disabled. + * + * Returns true if @p is migration-disabled, false otherwise. + */ +static bool is_bpf_migration_disabled(const struct task_struct *p) +{ + if (p->migration_disabled == 1) { + if (IS_ENABLED(CONFIG_PREEMPT_RCU)) + return p != current; + return true; + } + return p->migration_disabled; +} + +static s32 select_cpu_from_kfunc(struct scx_sched *sch, struct task_struct *p, + s32 prev_cpu, u64 wake_flags, + const struct cpumask *allowed, u64 flags) +{ + unsigned long irq_flags; + bool we_locked = false; + s32 cpu; + + if (!ops_cpu_valid(sch, prev_cpu, NULL)) + return -EINVAL; + + if (!check_builtin_idle_enabled(sch)) + return -EBUSY; + + /* + * Accessing p->cpus_ptr / p->nr_cpus_allowed needs either @p's rq + * lock or @p's pi_lock. Three cases: + * + * - inside ops.select_cpu(): try_to_wake_up() holds the wake-up + * task's pi_lock; the wake-up task is recorded in kf_tasks[0] + * by SCX_CALL_OP_TASK_RET(). + * - other rq-locked SCX op: scx_locked_rq() points at the held rq. + * - truly unlocked (UNLOCKED ops, SYSCALL, non-SCX struct_ops): + * nothing held, take pi_lock ourselves. + * + * In the first two cases, BPF schedulers may pass an arbitrary task + * that the held lock doesn't cover. Refuse those. + */ + if (this_rq()->scx.in_select_cpu) { + if (!scx_kf_arg_task_ok(sch, p)) + return -EINVAL; + lockdep_assert_held(&p->pi_lock); + } else if (scx_locked_rq()) { + if (task_rq(p) != scx_locked_rq()) + goto cross_task; + } else { + raw_spin_lock_irqsave(&p->pi_lock, irq_flags); + we_locked = true; + } + + /* + * This may also be called from ops.enqueue(), so we need to handle + * per-CPU tasks as well. For these tasks, we can skip all idle CPU + * selection optimizations and simply check whether the previously + * used CPU is idle and within the allowed cpumask. + */ + if (p->nr_cpus_allowed == 1 || is_bpf_migration_disabled(p)) { + if (cpumask_test_cpu(prev_cpu, allowed ?: p->cpus_ptr) && + scx_idle_test_and_clear_cpu(prev_cpu)) + cpu = prev_cpu; + else + cpu = -EBUSY; + } else { + cpu = scx_select_cpu_dfl(p, prev_cpu, wake_flags, + allowed ?: p->cpus_ptr, flags); + } + + if (we_locked) + raw_spin_unlock_irqrestore(&p->pi_lock, irq_flags); + + return cpu; + +cross_task: + scx_error(sch, "select_cpu kfunc called cross-task on %s[%d]", + p->comm, p->pid); + return -EINVAL; +} + +/** + * scx_bpf_cpu_node - Return the NUMA node the given @cpu belongs to, or + * trigger an error if @cpu is invalid + * @cpu: target CPU + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + */ +__bpf_kfunc s32 scx_bpf_cpu_node(s32 cpu, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch) || !ops_cpu_valid(sch, cpu, NULL)) + return NUMA_NO_NODE; + return cpu_to_node(cpu); +} + +/** + * scx_bpf_select_cpu_dfl - The default implementation of ops.select_cpu() + * @p: task_struct to select a CPU for + * @prev_cpu: CPU @p was on previously + * @wake_flags: %SCX_WAKE_* flags + * @is_idle: out parameter indicating whether the returned CPU is idle + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Can be called from ops.select_cpu(), ops.enqueue(), or from an unlocked + * context such as a BPF test_run() call, as long as built-in CPU selection + * is enabled: ops.update_idle() is missing or %SCX_OPS_KEEP_BUILTIN_IDLE + * is set. + * + * Returns the picked CPU with *@is_idle indicating whether the picked CPU is + * currently idle and thus a good candidate for direct dispatching. + */ +__bpf_kfunc s32 scx_bpf_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, + u64 wake_flags, bool *is_idle, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + s32 cpu; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + cpu = select_cpu_from_kfunc(sch, p, prev_cpu, wake_flags, NULL, 0); + if (cpu >= 0) { + *is_idle = true; + return cpu; + } + *is_idle = false; + return prev_cpu; +} + +struct scx_bpf_select_cpu_and_args { + /* @p and @cpus_allowed can't be packed together as KF_RCU is not transitive */ + s32 prev_cpu; + u64 wake_flags; + u64 flags; +}; + +/** + * __scx_bpf_select_cpu_and - Arg-wrapped CPU selection with cpumask + * @p: task_struct to select a CPU for + * @cpus_allowed: cpumask of allowed CPUs + * @args: struct containing the rest of the arguments + * @args->prev_cpu: CPU @p was on previously + * @args->wake_flags: %SCX_WAKE_* flags + * @args->flags: %SCX_PICK_IDLE* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Wrapper kfunc that takes arguments via struct to work around BPF's 5 argument + * limit. BPF programs should use scx_bpf_select_cpu_and() which is provided + * as an inline wrapper in common.bpf.h. + * + * Can be called from ops.select_cpu(), ops.enqueue(), or from an unlocked + * context such as a BPF test_run() call, as long as built-in CPU selection + * is enabled: ops.update_idle() is missing or %SCX_OPS_KEEP_BUILTIN_IDLE + * is set. + * + * @p, @args->prev_cpu and @args->wake_flags match ops.select_cpu(). + * + * Returns the selected idle CPU, which will be automatically awakened upon + * returning from ops.select_cpu() and can be used for direct dispatch, or + * a negative value if no idle CPU is available. + */ +__bpf_kfunc s32 +__scx_bpf_select_cpu_and(struct task_struct *p, const struct cpumask *cpus_allowed, + struct scx_bpf_select_cpu_and_args *args, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + return select_cpu_from_kfunc(sch, p, args->prev_cpu, args->wake_flags, + cpus_allowed, args->flags); +} + +/* + * COMPAT: Will be removed in v6.22. + */ +__bpf_kfunc s32 scx_bpf_select_cpu_and(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = rcu_dereference(scx_root); + if (unlikely(!sch)) + return -ENODEV; + +#ifdef CONFIG_EXT_SUB_SCHED + /* + * Disallow if any sub-scheds are attached. There is no way to tell + * which scheduler called us, just error out @p's scheduler. + */ + if (unlikely(!list_empty(&sch->children))) { + scx_error(scx_task_sched(p), "__scx_bpf_select_cpu_and() must be used"); + return -EINVAL; + } +#endif + + return select_cpu_from_kfunc(sch, p, prev_cpu, wake_flags, + cpus_allowed, flags); +} + +/** + * scx_bpf_get_idle_cpumask_node - Get a referenced kptr to the + * idle-tracking per-CPU cpumask of a target NUMA node. + * @node: target NUMA node + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns an empty cpumask if idle tracking is not enabled, if @node is + * not valid, or running on a UP kernel. In this case the actual error will + * be reported to the BPF scheduler via scx_error(). + */ +__bpf_kfunc const struct cpumask * +scx_bpf_get_idle_cpumask_node(s32 node, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return cpu_none_mask; + + node = validate_node(sch, node); + if (node < 0) + return cpu_none_mask; + + return idle_cpumask(node)->cpu; +} + +/** + * scx_bpf_get_idle_cpumask - Get a referenced kptr to the idle-tracking + * per-CPU cpumask. + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns an empty mask if idle tracking is not enabled, or running on a + * UP kernel. + */ +__bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return cpu_none_mask; + + if (static_branch_unlikely(&scx_builtin_idle_per_node)) { + scx_error(sch, "SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); + return cpu_none_mask; + } + + if (!check_builtin_idle_enabled(sch)) + return cpu_none_mask; + + return idle_cpumask(NUMA_NO_NODE)->cpu; +} + +/** + * scx_bpf_get_idle_smtmask_node - Get a referenced kptr to the + * idle-tracking, per-physical-core cpumask of a target NUMA node. Can be + * used to determine if an entire physical core is free. + * @node: target NUMA node + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns an empty cpumask if idle tracking is not enabled, if @node is + * not valid, or running on a UP kernel. In this case the actual error will + * be reported to the BPF scheduler via scx_error(). + */ +__bpf_kfunc const struct cpumask * +scx_bpf_get_idle_smtmask_node(s32 node, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return cpu_none_mask; + + node = validate_node(sch, node); + if (node < 0) + return cpu_none_mask; + + if (sched_smt_active()) + return idle_cpumask(node)->smt; + else + return idle_cpumask(node)->cpu; +} + +/** + * scx_bpf_get_idle_smtmask - Get a referenced kptr to the idle-tracking, + * per-physical-core cpumask. Can be used to determine if an entire physical + * core is free. + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns an empty mask if idle tracking is not enabled, or running on a + * UP kernel. + */ +__bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return cpu_none_mask; + + if (static_branch_unlikely(&scx_builtin_idle_per_node)) { + scx_error(sch, "SCX_OPS_BUILTIN_IDLE_PER_NODE enabled"); + return cpu_none_mask; + } + + if (!check_builtin_idle_enabled(sch)) + return cpu_none_mask; + + if (sched_smt_active()) + return idle_cpumask(NUMA_NO_NODE)->smt; + else + return idle_cpumask(NUMA_NO_NODE)->cpu; +} + +/** + * scx_bpf_put_idle_cpumask - Release a previously acquired referenced kptr to + * either the percpu, or SMT idle-tracking cpumask. + * @idle_mask: &cpumask to use + */ +__bpf_kfunc void scx_bpf_put_idle_cpumask(const struct cpumask *idle_mask) +{ + /* + * Empty function body because we aren't actually acquiring or releasing + * a reference to a global idle cpumask, which is read-only in the + * caller and is never released. The acquire / release semantics here + * are just used to make the cpumask a trusted pointer in the caller. + */ +} + +/** + * scx_bpf_test_and_clear_cpu_idle - Test and clear @cpu's idle state + * @cpu: cpu to test and clear idle for + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Returns %true if @cpu was idle and its idle state was successfully cleared. + * %false otherwise. + * + * Unavailable if ops.update_idle() is implemented and + * %SCX_OPS_KEEP_BUILTIN_IDLE is not set. + */ +__bpf_kfunc bool scx_bpf_test_and_clear_cpu_idle(s32 cpu, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return false; + + if (!check_builtin_idle_enabled(sch)) + return false; + + if (!ops_cpu_valid(sch, cpu, NULL)) + return false; + + return scx_idle_test_and_clear_cpu(cpu); +} + +/** + * scx_bpf_pick_idle_cpu_node - Pick and claim an idle cpu from @node + * @cpus_allowed: Allowed cpumask + * @node: target NUMA node + * @flags: %SCX_PICK_IDLE_* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Pick and claim an idle cpu in @cpus_allowed from the NUMA node @node. + * + * Returns the picked idle cpu number on success, or -%EBUSY if no matching + * cpu was found. + * + * The search starts from @node and proceeds to other online NUMA nodes in + * order of increasing distance (unless SCX_PICK_IDLE_IN_NODE is specified, + * in which case the search is limited to the target @node). + * + * Always returns an error if ops.update_idle() is implemented and + * %SCX_OPS_KEEP_BUILTIN_IDLE is not set, or if + * %SCX_OPS_BUILTIN_IDLE_PER_NODE is not set. + */ +__bpf_kfunc s32 scx_bpf_pick_idle_cpu_node(const struct cpumask *cpus_allowed, + s32 node, u64 flags, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + node = validate_node(sch, node); + if (node < 0) + return node; + + return scx_pick_idle_cpu(cpus_allowed, node, flags); +} + +/** + * scx_bpf_pick_idle_cpu - Pick and claim an idle cpu + * @cpus_allowed: Allowed cpumask + * @flags: %SCX_PICK_IDLE_CPU_* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Pick and claim an idle cpu in @cpus_allowed. Returns the picked idle cpu + * number on success. -%EBUSY if no matching cpu was found. + * + * Idle CPU tracking may race against CPU scheduling state transitions. For + * example, this function may return -%EBUSY as CPUs are transitioning into the + * idle state. If the caller then assumes that there will be dispatch events on + * the CPUs as they were all busy, the scheduler may end up stalling with CPUs + * idling while there are pending tasks. Use scx_bpf_pick_any_cpu() and + * scx_bpf_kick_cpu() to guarantee that there will be at least one dispatch + * event in the near future. + * + * Unavailable if ops.update_idle() is implemented and + * %SCX_OPS_KEEP_BUILTIN_IDLE is not set. + * + * Always returns an error if %SCX_OPS_BUILTIN_IDLE_PER_NODE is set, use + * scx_bpf_pick_idle_cpu_node() instead. + */ +__bpf_kfunc s32 scx_bpf_pick_idle_cpu(const struct cpumask *cpus_allowed, + u64 flags, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + if (static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) { + scx_error(sch, "per-node idle tracking is enabled"); + return -EBUSY; + } + + if (!check_builtin_idle_enabled(sch)) + return -EBUSY; + + return scx_pick_idle_cpu(cpus_allowed, NUMA_NO_NODE, flags); +} + +/** + * scx_bpf_pick_any_cpu_node - Pick and claim an idle cpu if available + * or pick any CPU from @node + * @cpus_allowed: Allowed cpumask + * @node: target NUMA node + * @flags: %SCX_PICK_IDLE_CPU_* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Pick and claim an idle cpu in @cpus_allowed. If none is available, pick any + * CPU in @cpus_allowed. Guaranteed to succeed and returns the picked idle cpu + * number if @cpus_allowed is not empty. -%EBUSY is returned if @cpus_allowed is + * empty. + * + * The search starts from @node and proceeds to other online NUMA nodes in + * order of increasing distance (unless %SCX_PICK_IDLE_IN_NODE is specified, + * in which case the search is limited to the target @node, regardless of + * the CPU idle state). + * + * If ops.update_idle() is implemented and %SCX_OPS_KEEP_BUILTIN_IDLE is not + * set, this function can't tell which CPUs are idle and will always pick any + * CPU. + */ +__bpf_kfunc s32 scx_bpf_pick_any_cpu_node(const struct cpumask *cpus_allowed, + s32 node, u64 flags, + const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + s32 cpu; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + node = validate_node(sch, node); + if (node < 0) + return node; + + cpu = scx_pick_idle_cpu(cpus_allowed, node, flags); + if (cpu >= 0) + return cpu; + + if (flags & SCX_PICK_IDLE_IN_NODE) + cpu = cpumask_any_and_distribute(cpumask_of_node(node), cpus_allowed); + else + cpu = cpumask_any_distribute(cpus_allowed); + if (cpu < nr_cpu_ids) + return cpu; + else + return -EBUSY; +} + +/** + * scx_bpf_pick_any_cpu - Pick and claim an idle cpu if available or pick any CPU + * @cpus_allowed: Allowed cpumask + * @flags: %SCX_PICK_IDLE_CPU_* flags + * @aux: implicit BPF argument to access bpf_prog_aux hidden from BPF progs + * + * Pick and claim an idle cpu in @cpus_allowed. If none is available, pick any + * CPU in @cpus_allowed. Guaranteed to succeed and returns the picked idle cpu + * number if @cpus_allowed is not empty. -%EBUSY is returned if @cpus_allowed is + * empty. + * + * If ops.update_idle() is implemented and %SCX_OPS_KEEP_BUILTIN_IDLE is not + * set, this function can't tell which CPUs are idle and will always pick any + * CPU. + * + * Always returns an error if %SCX_OPS_BUILTIN_IDLE_PER_NODE is set, use + * scx_bpf_pick_any_cpu_node() instead. + */ +__bpf_kfunc s32 scx_bpf_pick_any_cpu(const struct cpumask *cpus_allowed, + u64 flags, const struct bpf_prog_aux *aux) +{ + struct scx_sched *sch; + s32 cpu; + + guard(rcu)(); + + sch = scx_prog_sched(aux); + if (unlikely(!sch)) + return -ENODEV; + + if (static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) { + scx_error(sch, "per-node idle tracking is enabled"); + return -EBUSY; + } + + if (static_branch_likely(&scx_builtin_idle_enabled)) { + cpu = scx_pick_idle_cpu(cpus_allowed, NUMA_NO_NODE, flags); + if (cpu >= 0) + return cpu; + } + + cpu = cpumask_any_distribute(cpus_allowed); + if (cpu < nr_cpu_ids) + return cpu; + else + return -EBUSY; +} + +__bpf_kfunc_end_defs(); + +BTF_KFUNCS_START(scx_kfunc_ids_idle) +BTF_ID_FLAGS(func, scx_bpf_cpu_node, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask_node, KF_IMPLICIT_ARGS | KF_ACQUIRE) +BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask, KF_IMPLICIT_ARGS | KF_ACQUIRE) +BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask_node, KF_IMPLICIT_ARGS | KF_ACQUIRE) +BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask, KF_IMPLICIT_ARGS | KF_ACQUIRE) +BTF_ID_FLAGS(func, scx_bpf_put_idle_cpumask, KF_RELEASE) +BTF_ID_FLAGS(func, scx_bpf_test_and_clear_cpu_idle, KF_IMPLICIT_ARGS) +BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu_node, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu_node, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu, KF_IMPLICIT_ARGS | KF_RCU) +BTF_KFUNCS_END(scx_kfunc_ids_idle) + +static const struct btf_kfunc_id_set scx_kfunc_set_idle = { + .owner = THIS_MODULE, + .set = &scx_kfunc_ids_idle, + .filter = scx_kfunc_context_filter, +}; + +/* + * The select_cpu kfuncs internally call task_rq_lock() when invoked from an + * rq-unlocked context, and thus cannot be safely called from arbitrary tracing + * contexts where @p's pi_lock state is unknown. Keep them out of + * BPF_PROG_TYPE_TRACING by registering them in their own set which is exposed + * only to STRUCT_OPS and SYSCALL programs. + * + * These kfuncs are also members of scx_kfunc_ids_unlocked (see ext.c) because + * they're callable from unlocked contexts in addition to ops.select_cpu() and + * ops.enqueue(). + */ +BTF_KFUNCS_START(scx_kfunc_ids_select_cpu) +BTF_ID_FLAGS(func, __scx_bpf_select_cpu_and, KF_IMPLICIT_ARGS | KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_select_cpu_and, KF_RCU) +BTF_ID_FLAGS(func, scx_bpf_select_cpu_dfl, KF_IMPLICIT_ARGS | KF_RCU) +BTF_KFUNCS_END(scx_kfunc_ids_select_cpu) + +static const struct btf_kfunc_id_set scx_kfunc_set_select_cpu = { + .owner = THIS_MODULE, + .set = &scx_kfunc_ids_select_cpu, + .filter = scx_kfunc_context_filter, +}; + +int scx_idle_init(void) +{ + int ret; + + ret = register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &scx_kfunc_set_idle) || + register_btf_kfunc_id_set(BPF_PROG_TYPE_TRACING, &scx_kfunc_set_idle) || + register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &scx_kfunc_set_idle) || + register_btf_kfunc_id_set(BPF_PROG_TYPE_STRUCT_OPS, &scx_kfunc_set_select_cpu) || + register_btf_kfunc_id_set(BPF_PROG_TYPE_SYSCALL, &scx_kfunc_set_select_cpu); + + return ret; +} diff --git a/kernel/sched/ext_idle.h b/kernel/sched/ext_idle.h new file mode 100644 index 000000000000..8d169d3bbdf9 --- /dev/null +++ b/kernel/sched/ext_idle.h @@ -0,0 +1,27 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * BPF extensible scheduler class: Documentation/scheduler/sched-ext.rst + * + * Copyright (c) 2022 Meta Platforms, Inc. and affiliates. + * Copyright (c) 2022 Tejun Heo <tj@kernel.org> + * Copyright (c) 2022 David Vernet <dvernet@meta.com> + * Copyright (c) 2024 Andrea Righi <arighi@nvidia.com> + */ +#ifndef _KERNEL_SCHED_EXT_IDLE_H +#define _KERNEL_SCHED_EXT_IDLE_H + +struct sched_ext_ops; + +extern struct btf_id_set8 scx_kfunc_ids_idle; +extern struct btf_id_set8 scx_kfunc_ids_select_cpu; + +void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops); +void scx_idle_init_masks(void); + +s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags, + const struct cpumask *cpus_allowed, u64 flags); +void scx_idle_enable(struct sched_ext_ops *ops); +void scx_idle_disable(void); +int scx_idle_init(void); + +#endif /* _KERNEL_SCHED_EXT_IDLE_H */ diff --git a/kernel/sched/ext_internal.h b/kernel/sched/ext_internal.h new file mode 100644 index 000000000000..a075732d4430 --- /dev/null +++ b/kernel/sched/ext_internal.h @@ -0,0 +1,1483 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * BPF extensible scheduler class: Documentation/scheduler/sched-ext.rst + * + * Copyright (c) 2025 Meta Platforms, Inc. and affiliates. + * Copyright (c) 2025 Tejun Heo <tj@kernel.org> + */ +#define SCX_OP_IDX(op) (offsetof(struct sched_ext_ops, op) / sizeof(void (*)(void))) +#define SCX_MOFF_IDX(moff) ((moff) / sizeof(void (*)(void))) + +enum scx_consts { + SCX_DSP_DFL_MAX_BATCH = 32, + SCX_DSP_MAX_LOOPS = 32, + SCX_WATCHDOG_MAX_TIMEOUT = 30 * HZ, + + SCX_EXIT_BT_LEN = 64, + SCX_EXIT_MSG_LEN = 1024, + SCX_EXIT_DUMP_DFL_LEN = 32768, + + SCX_CPUPERF_ONE = SCHED_CAPACITY_SCALE, + + /* + * Iterating all tasks may take a while. Periodically drop + * scx_tasks_lock to avoid causing e.g. CSD and RCU stalls. + */ + SCX_TASK_ITER_BATCH = 32, + + SCX_BYPASS_HOST_NTH = 2, + + SCX_BYPASS_LB_DFL_INTV_US = 500 * USEC_PER_MSEC, + SCX_BYPASS_LB_DONOR_PCT = 125, + SCX_BYPASS_LB_MIN_DELTA_DIV = 4, + SCX_BYPASS_LB_BATCH = 256, + + SCX_REENQ_LOCAL_MAX_REPEAT = 256, + + SCX_SUB_MAX_DEPTH = 4, +}; + +enum scx_exit_kind { + SCX_EXIT_NONE, + SCX_EXIT_DONE, + + SCX_EXIT_UNREG = 64, /* user-space initiated unregistration */ + SCX_EXIT_UNREG_BPF, /* BPF-initiated unregistration */ + SCX_EXIT_UNREG_KERN, /* kernel-initiated unregistration */ + SCX_EXIT_SYSRQ, /* requested by 'S' sysrq */ + SCX_EXIT_PARENT, /* parent exiting */ + + SCX_EXIT_ERROR = 1024, /* runtime error, error msg contains details */ + SCX_EXIT_ERROR_BPF, /* ERROR but triggered through scx_bpf_error() */ + SCX_EXIT_ERROR_STALL, /* watchdog detected stalled runnable tasks */ +}; + +/* + * An exit code can be specified when exiting with scx_bpf_exit() or scx_exit(), + * corresponding to exit_kind UNREG_BPF and UNREG_KERN respectively. The codes + * are 64bit of the format: + * + * Bits: [63 .. 48 47 .. 32 31 .. 0] + * [ SYS ACT ] [ SYS RSN ] [ USR ] + * + * SYS ACT: System-defined exit actions + * SYS RSN: System-defined exit reasons + * USR : User-defined exit codes and reasons + * + * Using the above, users may communicate intention and context by ORing system + * actions and/or system reasons with a user-defined exit code. + */ +enum scx_exit_code { + /* Reasons */ + SCX_ECODE_RSN_HOTPLUG = 1LLU << 32, + SCX_ECODE_RSN_CGROUP_OFFLINE = 2LLU << 32, + + /* Actions */ + SCX_ECODE_ACT_RESTART = 1LLU << 48, +}; + +enum scx_exit_flags { + /* + * ops.exit() may be called even if the loading failed before ops.init() + * finishes successfully. This is because ops.exit() allows rich exit + * info communication. The following flag indicates whether ops.init() + * finished successfully. + */ + SCX_EFLAG_INITIALIZED = 1LLU << 0, +}; + +/* + * scx_exit_info is passed to ops.exit() to describe why the BPF scheduler is + * being disabled. + */ +struct scx_exit_info { + /* %SCX_EXIT_* - broad category of the exit reason */ + enum scx_exit_kind kind; + + /* exit code if gracefully exiting */ + s64 exit_code; + + /* %SCX_EFLAG_* */ + u64 flags; + + /* textual representation of the above */ + const char *reason; + + /* backtrace if exiting due to an error */ + unsigned long *bt; + u32 bt_len; + + /* informational message */ + char *msg; + + /* debug dump */ + char *dump; +}; + +/* sched_ext_ops.flags */ +enum scx_ops_flags { + /* + * Keep built-in idle tracking even if ops.update_idle() is implemented. + */ + SCX_OPS_KEEP_BUILTIN_IDLE = 1LLU << 0, + + /* + * By default, if there are no other task to run on the CPU, ext core + * keeps running the current task even after its slice expires. If this + * flag is specified, such tasks are passed to ops.enqueue() with + * %SCX_ENQ_LAST. See the comment above %SCX_ENQ_LAST for more info. + */ + SCX_OPS_ENQ_LAST = 1LLU << 1, + + /* + * An exiting task may schedule after PF_EXITING is set. In such cases, + * bpf_task_from_pid() may not be able to find the task and if the BPF + * scheduler depends on pid lookup for dispatching, the task will be + * lost leading to various issues including RCU grace period stalls. + * + * To mask this problem, by default, unhashed tasks are automatically + * dispatched to the local DSQ on enqueue. If the BPF scheduler doesn't + * depend on pid lookups and wants to handle these tasks directly, the + * following flag can be used. + */ + SCX_OPS_ENQ_EXITING = 1LLU << 2, + + /* + * If set, only tasks with policy set to SCHED_EXT are attached to + * sched_ext. If clear, SCHED_NORMAL tasks are also included. + */ + SCX_OPS_SWITCH_PARTIAL = 1LLU << 3, + + /* + * A migration disabled task can only execute on its current CPU. By + * default, such tasks are automatically put on the CPU's local DSQ with + * the default slice on enqueue. If this ops flag is set, they also go + * through ops.enqueue(). + * + * A migration disabled task never invokes ops.select_cpu() as it can + * only select the current CPU. Also, p->cpus_ptr will only contain its + * current CPU while p->nr_cpus_allowed keeps tracking p->user_cpus_ptr + * and thus may disagree with cpumask_weight(p->cpus_ptr). + */ + SCX_OPS_ENQ_MIGRATION_DISABLED = 1LLU << 4, + + /* + * Queued wakeup (ttwu_queue) is a wakeup optimization that invokes + * ops.enqueue() on the ops.select_cpu() selected or the wakee's + * previous CPU via IPI (inter-processor interrupt) to reduce cacheline + * transfers. When this optimization is enabled, ops.select_cpu() is + * skipped in some cases (when racing against the wakee switching out). + * As the BPF scheduler may depend on ops.select_cpu() being invoked + * during wakeups, queued wakeup is disabled by default. + * + * If this ops flag is set, queued wakeup optimization is enabled and + * the BPF scheduler must be able to handle ops.enqueue() invoked on the + * wakee's CPU without preceding ops.select_cpu() even for tasks which + * may be executed on multiple CPUs. + */ + SCX_OPS_ALLOW_QUEUED_WAKEUP = 1LLU << 5, + + /* + * If set, enable per-node idle cpumasks. If clear, use a single global + * flat idle cpumask. + */ + SCX_OPS_BUILTIN_IDLE_PER_NODE = 1LLU << 6, + + /* + * If set, %SCX_ENQ_IMMED is assumed to be set on all local DSQ + * enqueues. + */ + SCX_OPS_ALWAYS_ENQ_IMMED = 1LLU << 7, + + SCX_OPS_ALL_FLAGS = SCX_OPS_KEEP_BUILTIN_IDLE | + SCX_OPS_ENQ_LAST | + SCX_OPS_ENQ_EXITING | + SCX_OPS_ENQ_MIGRATION_DISABLED | + SCX_OPS_ALLOW_QUEUED_WAKEUP | + SCX_OPS_SWITCH_PARTIAL | + SCX_OPS_BUILTIN_IDLE_PER_NODE | + SCX_OPS_ALWAYS_ENQ_IMMED, + + /* high 8 bits are internal, don't include in SCX_OPS_ALL_FLAGS */ + __SCX_OPS_INTERNAL_MASK = 0xffLLU << 56, + + SCX_OPS_HAS_CPU_PREEMPT = 1LLU << 56, +}; + +/* argument container for ops.init_task() */ +struct scx_init_task_args { + /* + * Set if ops.init_task() is being invoked on the fork path, as opposed + * to the scheduler transition path. + */ + bool fork; +#ifdef CONFIG_EXT_GROUP_SCHED + /* the cgroup the task is joining */ + struct cgroup *cgroup; +#endif +}; + +/* argument container for ops.exit_task() */ +struct scx_exit_task_args { + /* Whether the task exited before running on sched_ext. */ + bool cancelled; +}; + +/* argument container for ops.cgroup_init() */ +struct scx_cgroup_init_args { + /* the weight of the cgroup [1..10000] */ + u32 weight; + + /* bandwidth control parameters from cpu.max and cpu.max.burst */ + u64 bw_period_us; + u64 bw_quota_us; + u64 bw_burst_us; +}; + +enum scx_cpu_preempt_reason { + /* next task is being scheduled by &sched_class_rt */ + SCX_CPU_PREEMPT_RT, + /* next task is being scheduled by &sched_class_dl */ + SCX_CPU_PREEMPT_DL, + /* next task is being scheduled by &sched_class_stop */ + SCX_CPU_PREEMPT_STOP, + /* unknown reason for SCX being preempted */ + SCX_CPU_PREEMPT_UNKNOWN, +}; + +/* + * Argument container for ops.cpu_acquire(). Currently empty, but may be + * expanded in the future. + */ +struct scx_cpu_acquire_args {}; + +/* argument container for ops.cpu_release() */ +struct scx_cpu_release_args { + /* the reason the CPU was preempted */ + enum scx_cpu_preempt_reason reason; + + /* the task that's going to be scheduled on the CPU */ + struct task_struct *task; +}; + +/* informational context provided to dump operations */ +struct scx_dump_ctx { + enum scx_exit_kind kind; + s64 exit_code; + const char *reason; + u64 at_ns; + u64 at_jiffies; +}; + +/* argument container for ops.sub_attach() */ +struct scx_sub_attach_args { + struct sched_ext_ops *ops; + char *cgroup_path; +}; + +/* argument container for ops.sub_detach() */ +struct scx_sub_detach_args { + struct sched_ext_ops *ops; + char *cgroup_path; +}; + +/** + * struct sched_ext_ops - Operation table for BPF scheduler implementation + * + * A BPF scheduler can implement an arbitrary scheduling policy by + * implementing and loading operations in this table. Note that a userland + * scheduling policy can also be implemented using the BPF scheduler + * as a shim layer. + */ +struct sched_ext_ops { + /** + * @select_cpu: Pick the target CPU for a task which is being woken up + * @p: task being woken up + * @prev_cpu: the cpu @p was on before sleeping + * @wake_flags: SCX_WAKE_* + * + * Decision made here isn't final. @p may be moved to any CPU while it + * is getting dispatched for execution later. However, as @p is not on + * the rq at this point, getting the eventual execution CPU right here + * saves a small bit of overhead down the line. + * + * If an idle CPU is returned, the CPU is kicked and will try to + * dispatch. While an explicit custom mechanism can be added, + * select_cpu() serves as the default way to wake up idle CPUs. + * + * @p may be inserted into a DSQ directly by calling + * scx_bpf_dsq_insert(). If so, the ops.enqueue() will be skipped. + * Directly inserting into %SCX_DSQ_LOCAL will put @p in the local DSQ + * of the CPU returned by this operation. + * + * Note that select_cpu() is never called for tasks that can only run + * on a single CPU or tasks with migration disabled, as they don't have + * the option to select a different CPU. See select_task_rq() for + * details. + */ + s32 (*select_cpu)(struct task_struct *p, s32 prev_cpu, u64 wake_flags); + + /** + * @enqueue: Enqueue a task on the BPF scheduler + * @p: task being enqueued + * @enq_flags: %SCX_ENQ_* + * + * @p is ready to run. Insert directly into a DSQ by calling + * scx_bpf_dsq_insert() or enqueue on the BPF scheduler. If not directly + * inserted, the bpf scheduler owns @p and if it fails to dispatch @p, + * the task will stall. + * + * If @p was inserted into a DSQ from ops.select_cpu(), this callback is + * skipped. + */ + void (*enqueue)(struct task_struct *p, u64 enq_flags); + + /** + * @dequeue: Remove a task from the BPF scheduler + * @p: task being dequeued + * @deq_flags: %SCX_DEQ_* + * + * Remove @p from the BPF scheduler. This is usually called to isolate + * the task while updating its scheduling properties (e.g. priority). + * + * The ext core keeps track of whether the BPF side owns a given task or + * not and can gracefully ignore spurious dispatches from BPF side, + * which makes it safe to not implement this method. However, depending + * on the scheduling logic, this can lead to confusing behaviors - e.g. + * scheduling position not being updated across a priority change. + */ + void (*dequeue)(struct task_struct *p, u64 deq_flags); + + /** + * @dispatch: Dispatch tasks from the BPF scheduler and/or user DSQs + * @cpu: CPU to dispatch tasks for + * @prev: previous task being switched out + * + * Called when a CPU's local dsq is empty. The operation should dispatch + * one or more tasks from the BPF scheduler into the DSQs using + * scx_bpf_dsq_insert() and/or move from user DSQs into the local DSQ + * using scx_bpf_dsq_move_to_local(). + * + * The maximum number of times scx_bpf_dsq_insert() can be called + * without an intervening scx_bpf_dsq_move_to_local() is specified by + * ops.dispatch_max_batch. See the comments on top of the two functions + * for more details. + * + * When not %NULL, @prev is an SCX task with its slice depleted. If + * @prev is still runnable as indicated by set %SCX_TASK_QUEUED in + * @prev->scx.flags, it is not enqueued yet and will be enqueued after + * ops.dispatch() returns. To keep executing @prev, return without + * dispatching or moving any tasks. Also see %SCX_OPS_ENQ_LAST. + */ + void (*dispatch)(s32 cpu, struct task_struct *prev); + + /** + * @tick: Periodic tick + * @p: task running currently + * + * This operation is called every 1/HZ seconds on CPUs which are + * executing an SCX task. Setting @p->scx.slice to 0 will trigger an + * immediate dispatch cycle on the CPU. + */ + void (*tick)(struct task_struct *p); + + /** + * @runnable: A task is becoming runnable on its associated CPU + * @p: task becoming runnable + * @enq_flags: %SCX_ENQ_* + * + * This and the following three functions can be used to track a task's + * execution state transitions. A task becomes ->runnable() on a CPU, + * and then goes through one or more ->running() and ->stopping() pairs + * as it runs on the CPU, and eventually becomes ->quiescent() when it's + * done running on the CPU. + * + * @p is becoming runnable on the CPU because it's + * + * - waking up (%SCX_ENQ_WAKEUP) + * - being moved from another CPU + * - being restored after temporarily taken off the queue for an + * attribute change. + * + * This and ->enqueue() are related but not coupled. This operation + * notifies @p's state transition and may not be followed by ->enqueue() + * e.g. when @p is being dispatched to a remote CPU, or when @p is + * being enqueued on a CPU experiencing a hotplug event. Likewise, a + * task may be ->enqueue()'d without being preceded by this operation + * e.g. after exhausting its slice. + */ + void (*runnable)(struct task_struct *p, u64 enq_flags); + + /** + * @running: A task is starting to run on its associated CPU + * @p: task starting to run + * + * Note that this callback may be called from a CPU other than the + * one the task is going to run on. This can happen when a task + * property is changed (i.e., affinity), since scx_next_task_scx(), + * which triggers this callback, may run on a CPU different from + * the task's assigned CPU. + * + * Therefore, always use scx_bpf_task_cpu(@p) to determine the + * target CPU the task is going to use. + * + * See ->runnable() for explanation on the task state notifiers. + */ + void (*running)(struct task_struct *p); + + /** + * @stopping: A task is stopping execution + * @p: task stopping to run + * @runnable: is task @p still runnable? + * + * Note that this callback may be called from a CPU other than the + * one the task was running on. This can happen when a task + * property is changed (i.e., affinity), since dequeue_task_scx(), + * which triggers this callback, may run on a CPU different from + * the task's assigned CPU. + * + * Therefore, always use scx_bpf_task_cpu(@p) to retrieve the CPU + * the task was running on. + * + * See ->runnable() for explanation on the task state notifiers. If + * !@runnable, ->quiescent() will be invoked after this operation + * returns. + */ + void (*stopping)(struct task_struct *p, bool runnable); + + /** + * @quiescent: A task is becoming not runnable on its associated CPU + * @p: task becoming not runnable + * @deq_flags: %SCX_DEQ_* + * + * See ->runnable() for explanation on the task state notifiers. + * + * @p is becoming quiescent on the CPU because it's + * + * - sleeping (%SCX_DEQ_SLEEP) + * - being moved to another CPU + * - being temporarily taken off the queue for an attribute change + * (%SCX_DEQ_SAVE) + * + * This and ->dequeue() are related but not coupled. This operation + * notifies @p's state transition and may not be preceded by ->dequeue() + * e.g. when @p is being dispatched to a remote CPU. + */ + void (*quiescent)(struct task_struct *p, u64 deq_flags); + + /** + * @yield: Yield CPU + * @from: yielding task + * @to: optional yield target task + * + * If @to is NULL, @from is yielding the CPU to other runnable tasks. + * The BPF scheduler should ensure that other available tasks are + * dispatched before the yielding task. Return value is ignored in this + * case. + * + * If @to is not-NULL, @from wants to yield the CPU to @to. If the bpf + * scheduler can implement the request, return %true; otherwise, %false. + */ + bool (*yield)(struct task_struct *from, struct task_struct *to); + + /** + * @core_sched_before: Task ordering for core-sched + * @a: task A + * @b: task B + * + * Used by core-sched to determine the ordering between two tasks. See + * Documentation/admin-guide/hw-vuln/core-scheduling.rst for details on + * core-sched. + * + * Both @a and @b are runnable and may or may not currently be queued on + * the BPF scheduler. Should return %true if @a should run before @b. + * %false if there's no required ordering or @b should run before @a. + * + * If not specified, the default is ordering them according to when they + * became runnable. + */ + bool (*core_sched_before)(struct task_struct *a, struct task_struct *b); + + /** + * @set_weight: Set task weight + * @p: task to set weight for + * @weight: new weight [1..10000] + * + * Update @p's weight to @weight. + */ + void (*set_weight)(struct task_struct *p, u32 weight); + + /** + * @set_cpumask: Set CPU affinity + * @p: task to set CPU affinity for + * @cpumask: cpumask of cpus that @p can run on + * + * Update @p's CPU affinity to @cpumask. + */ + void (*set_cpumask)(struct task_struct *p, + const struct cpumask *cpumask); + + /** + * @update_idle: Update the idle state of a CPU + * @cpu: CPU to update the idle state for + * @idle: whether entering or exiting the idle state + * + * This operation is called when @rq's CPU goes or leaves the idle + * state. By default, implementing this operation disables the built-in + * idle CPU tracking and the following helpers become unavailable: + * + * - scx_bpf_select_cpu_dfl() + * - scx_bpf_select_cpu_and() + * - scx_bpf_test_and_clear_cpu_idle() + * - scx_bpf_pick_idle_cpu() + * + * The user also must implement ops.select_cpu() as the default + * implementation relies on scx_bpf_select_cpu_dfl(). + * + * Specify the %SCX_OPS_KEEP_BUILTIN_IDLE flag to keep the built-in idle + * tracking. + */ + void (*update_idle)(s32 cpu, bool idle); + + /** + * @cpu_acquire: A CPU is becoming available to the BPF scheduler + * @cpu: The CPU being acquired by the BPF scheduler. + * @args: Acquire arguments, see the struct definition. + * + * A CPU that was previously released from the BPF scheduler is now once + * again under its control. + */ + void (*cpu_acquire)(s32 cpu, struct scx_cpu_acquire_args *args); + + /** + * @cpu_release: A CPU is taken away from the BPF scheduler + * @cpu: The CPU being released by the BPF scheduler. + * @args: Release arguments, see the struct definition. + * + * The specified CPU is no longer under the control of the BPF + * scheduler. This could be because it was preempted by a higher + * priority sched_class, though there may be other reasons as well. The + * caller should consult @args->reason to determine the cause. + */ + void (*cpu_release)(s32 cpu, struct scx_cpu_release_args *args); + + /** + * @init_task: Initialize a task to run in a BPF scheduler + * @p: task to initialize for BPF scheduling + * @args: init arguments, see the struct definition + * + * Either we're loading a BPF scheduler or a new task is being forked. + * Initialize @p for BPF scheduling. This operation may block and can + * be used for allocations, and is called exactly once for a task. + * + * Return 0 for success, -errno for failure. An error return while + * loading will abort loading of the BPF scheduler. During a fork, it + * will abort that specific fork. + */ + s32 (*init_task)(struct task_struct *p, struct scx_init_task_args *args); + + /** + * @exit_task: Exit a previously-running task from the system + * @p: task to exit + * @args: exit arguments, see the struct definition + * + * @p is exiting or the BPF scheduler is being unloaded. Perform any + * necessary cleanup for @p. + */ + void (*exit_task)(struct task_struct *p, struct scx_exit_task_args *args); + + /** + * @enable: Enable BPF scheduling for a task + * @p: task to enable BPF scheduling for + * + * Enable @p for BPF scheduling. enable() is called on @p any time it + * enters SCX, and is always paired with a matching disable(). + */ + void (*enable)(struct task_struct *p); + + /** + * @disable: Disable BPF scheduling for a task + * @p: task to disable BPF scheduling for + * + * @p is exiting, leaving SCX or the BPF scheduler is being unloaded. + * Disable BPF scheduling for @p. A disable() call is always matched + * with a prior enable() call. + */ + void (*disable)(struct task_struct *p); + + /** + * @dump: Dump BPF scheduler state on error + * @ctx: debug dump context + * + * Use scx_bpf_dump() to generate BPF scheduler specific debug dump. + */ + void (*dump)(struct scx_dump_ctx *ctx); + + /** + * @dump_cpu: Dump BPF scheduler state for a CPU on error + * @ctx: debug dump context + * @cpu: CPU to generate debug dump for + * @idle: @cpu is currently idle without any runnable tasks + * + * Use scx_bpf_dump() to generate BPF scheduler specific debug dump for + * @cpu. If @idle is %true and this operation doesn't produce any + * output, @cpu is skipped for dump. + */ + void (*dump_cpu)(struct scx_dump_ctx *ctx, s32 cpu, bool idle); + + /** + * @dump_task: Dump BPF scheduler state for a runnable task on error + * @ctx: debug dump context + * @p: runnable task to generate debug dump for + * + * Use scx_bpf_dump() to generate BPF scheduler specific debug dump for + * @p. + */ + void (*dump_task)(struct scx_dump_ctx *ctx, struct task_struct *p); + +#ifdef CONFIG_EXT_GROUP_SCHED + /** + * @cgroup_init: Initialize a cgroup + * @cgrp: cgroup being initialized + * @args: init arguments, see the struct definition + * + * Either the BPF scheduler is being loaded or @cgrp created, initialize + * @cgrp for sched_ext. This operation may block. + * + * Return 0 for success, -errno for failure. An error return while + * loading will abort loading of the BPF scheduler. During cgroup + * creation, it will abort the specific cgroup creation. + */ + s32 (*cgroup_init)(struct cgroup *cgrp, + struct scx_cgroup_init_args *args); + + /** + * @cgroup_exit: Exit a cgroup + * @cgrp: cgroup being exited + * + * Either the BPF scheduler is being unloaded or @cgrp destroyed, exit + * @cgrp for sched_ext. This operation my block. + */ + void (*cgroup_exit)(struct cgroup *cgrp); + + /** + * @cgroup_prep_move: Prepare a task to be moved to a different cgroup + * @p: task being moved + * @from: cgroup @p is being moved from + * @to: cgroup @p is being moved to + * + * Prepare @p for move from cgroup @from to @to. This operation may + * block and can be used for allocations. + * + * Return 0 for success, -errno for failure. An error return aborts the + * migration. + */ + s32 (*cgroup_prep_move)(struct task_struct *p, + struct cgroup *from, struct cgroup *to); + + /** + * @cgroup_move: Commit cgroup move + * @p: task being moved + * @from: cgroup @p is being moved from + * @to: cgroup @p is being moved to + * + * Commit the move. @p is dequeued during this operation. + */ + void (*cgroup_move)(struct task_struct *p, + struct cgroup *from, struct cgroup *to); + + /** + * @cgroup_cancel_move: Cancel cgroup move + * @p: task whose cgroup move is being canceled + * @from: cgroup @p was being moved from + * @to: cgroup @p was being moved to + * + * @p was cgroup_prep_move()'d but failed before reaching cgroup_move(). + * Undo the preparation. + */ + void (*cgroup_cancel_move)(struct task_struct *p, + struct cgroup *from, struct cgroup *to); + + /** + * @cgroup_set_weight: A cgroup's weight is being changed + * @cgrp: cgroup whose weight is being updated + * @weight: new weight [1..10000] + * + * Update @cgrp's weight to @weight. + */ + void (*cgroup_set_weight)(struct cgroup *cgrp, u32 weight); + + /** + * @cgroup_set_bandwidth: A cgroup's bandwidth is being changed + * @cgrp: cgroup whose bandwidth is being updated + * @period_us: bandwidth control period + * @quota_us: bandwidth control quota + * @burst_us: bandwidth control burst + * + * Update @cgrp's bandwidth control parameters. This is from the cpu.max + * cgroup interface. + * + * @quota_us / @period_us determines the CPU bandwidth @cgrp is entitled + * to. For example, if @period_us is 1_000_000 and @quota_us is + * 2_500_000. @cgrp is entitled to 2.5 CPUs. @burst_us can be + * interpreted in the same fashion and specifies how much @cgrp can + * burst temporarily. The specific control mechanism and thus the + * interpretation of @period_us and burstiness is up to the BPF + * scheduler. + */ + void (*cgroup_set_bandwidth)(struct cgroup *cgrp, + u64 period_us, u64 quota_us, u64 burst_us); + + /** + * @cgroup_set_idle: A cgroup's idle state is being changed + * @cgrp: cgroup whose idle state is being updated + * @idle: whether the cgroup is entering or exiting idle state + * + * Update @cgrp's idle state to @idle. This callback is invoked when + * a cgroup transitions between idle and non-idle states, allowing the + * BPF scheduler to adjust its behavior accordingly. + */ + void (*cgroup_set_idle)(struct cgroup *cgrp, bool idle); + +#endif /* CONFIG_EXT_GROUP_SCHED */ + + /** + * @sub_attach: Attach a sub-scheduler + * @args: argument container, see the struct definition + * + * Return 0 to accept the sub-scheduler. -errno to reject. + */ + s32 (*sub_attach)(struct scx_sub_attach_args *args); + + /** + * @sub_detach: Detach a sub-scheduler + * @args: argument container, see the struct definition + */ + void (*sub_detach)(struct scx_sub_detach_args *args); + + /* + * All online ops must come before ops.cpu_online(). + */ + + /** + * @cpu_online: A CPU became online + * @cpu: CPU which just came up + * + * @cpu just came online. @cpu will not call ops.enqueue() or + * ops.dispatch(), nor run tasks associated with other CPUs beforehand. + */ + void (*cpu_online)(s32 cpu); + + /** + * @cpu_offline: A CPU is going offline + * @cpu: CPU which is going offline + * + * @cpu is going offline. @cpu will not call ops.enqueue() or + * ops.dispatch(), nor run tasks associated with other CPUs afterwards. + */ + void (*cpu_offline)(s32 cpu); + + /* + * All CPU hotplug ops must come before ops.init(). + */ + + /** + * @init: Initialize the BPF scheduler + */ + s32 (*init)(void); + + /** + * @exit: Clean up after the BPF scheduler + * @info: Exit info + * + * ops.exit() is also called on ops.init() failure, which is a bit + * unusual. This is to allow rich reporting through @info on how + * ops.init() failed. + */ + void (*exit)(struct scx_exit_info *info); + + /* + * Data fields must comes after all ops fields. + */ + + /** + * @dispatch_max_batch: Max nr of tasks that dispatch() can dispatch + */ + u32 dispatch_max_batch; + + /** + * @flags: %SCX_OPS_* flags + */ + u64 flags; + + /** + * @timeout_ms: The maximum amount of time, in milliseconds, that a + * runnable task should be able to wait before being scheduled. The + * maximum timeout may not exceed the default timeout of 30 seconds. + * + * Defaults to the maximum allowed timeout value of 30 seconds. + */ + u32 timeout_ms; + + /** + * @exit_dump_len: scx_exit_info.dump buffer length. If 0, the default + * value of 32768 is used. + */ + u32 exit_dump_len; + + /** + * @hotplug_seq: A sequence number that may be set by the scheduler to + * detect when a hotplug event has occurred during the loading process. + * If 0, no detection occurs. Otherwise, the scheduler will fail to + * load if the sequence number does not match @scx_hotplug_seq on the + * enable path. + */ + u64 hotplug_seq; + + /** + * @cgroup_id: When >1, attach the scheduler as a sub-scheduler on the + * specified cgroup. + */ + u64 sub_cgroup_id; + + /** + * @name: BPF scheduler's name + * + * Must be a non-zero valid BPF object name including only isalnum(), + * '_' and '.' chars. Shows up in kernel.sched_ext_ops sysctl while the + * BPF scheduler is enabled. + */ + char name[SCX_OPS_NAME_LEN]; + + /* internal use only, must be NULL */ + void __rcu *priv; +}; + +enum scx_opi { + SCX_OPI_BEGIN = 0, + SCX_OPI_NORMAL_BEGIN = 0, + SCX_OPI_NORMAL_END = SCX_OP_IDX(cpu_online), + SCX_OPI_CPU_HOTPLUG_BEGIN = SCX_OP_IDX(cpu_online), + SCX_OPI_CPU_HOTPLUG_END = SCX_OP_IDX(init), + SCX_OPI_END = SCX_OP_IDX(init), +}; + +/* + * Collection of event counters. Event types are placed in descending order. + */ +struct scx_event_stats { + /* + * If ops.select_cpu() returns a CPU which can't be used by the task, + * the core scheduler code silently picks a fallback CPU. + */ + s64 SCX_EV_SELECT_CPU_FALLBACK; + + /* + * When dispatching to a local DSQ, the CPU may have gone offline in + * the meantime. In this case, the task is bounced to the global DSQ. + */ + s64 SCX_EV_DISPATCH_LOCAL_DSQ_OFFLINE; + + /* + * If SCX_OPS_ENQ_LAST is not set, the number of times that a task + * continued to run because there were no other tasks on the CPU. + */ + s64 SCX_EV_DISPATCH_KEEP_LAST; + + /* + * If SCX_OPS_ENQ_EXITING is not set, the number of times that a task + * is dispatched to a local DSQ when exiting. + */ + s64 SCX_EV_ENQ_SKIP_EXITING; + + /* + * If SCX_OPS_ENQ_MIGRATION_DISABLED is not set, the number of times a + * migration disabled task skips ops.enqueue() and is dispatched to its + * local DSQ. + */ + s64 SCX_EV_ENQ_SKIP_MIGRATION_DISABLED; + + /* + * The number of times a task, enqueued on a local DSQ with + * SCX_ENQ_IMMED, was re-enqueued because the CPU was not available for + * immediate execution. + */ + s64 SCX_EV_REENQ_IMMED; + + /* + * The number of times a reenq of local DSQ caused another reenq of + * local DSQ. This can happen when %SCX_ENQ_IMMED races against a higher + * priority class task even if the BPF scheduler always satisfies the + * prerequisites for %SCX_ENQ_IMMED at the time of enqueue. However, + * that scenario is very unlikely and this count going up regularly + * indicates that the BPF scheduler is handling %SCX_ENQ_REENQ + * incorrectly causing recursive reenqueues. + */ + s64 SCX_EV_REENQ_LOCAL_REPEAT; + + /* + * Total number of times a task's time slice was refilled with the + * default value (SCX_SLICE_DFL). + */ + s64 SCX_EV_REFILL_SLICE_DFL; + + /* + * The total duration of bypass modes in nanoseconds. + */ + s64 SCX_EV_BYPASS_DURATION; + + /* + * The number of tasks dispatched in the bypassing mode. + */ + s64 SCX_EV_BYPASS_DISPATCH; + + /* + * The number of times the bypassing mode has been activated. + */ + s64 SCX_EV_BYPASS_ACTIVATE; + + /* + * The number of times the scheduler attempted to insert a task that it + * doesn't own into a DSQ. Such attempts are ignored. + * + * As BPF schedulers are allowed to ignore dequeues, it's difficult to + * tell whether such an attempt is from a scheduler malfunction or an + * ignored dequeue around sub-sched enabling. If this count keeps going + * up regardless of sub-sched enabling, it likely indicates a bug in the + * scheduler. + */ + s64 SCX_EV_INSERT_NOT_OWNED; + + /* + * The number of times tasks from bypassing descendants are scheduled + * from sub_bypass_dsq's. + */ + s64 SCX_EV_SUB_BYPASS_DISPATCH; +}; + +struct scx_sched; + +enum scx_sched_pcpu_flags { + SCX_SCHED_PCPU_BYPASSING = 1LLU << 0, +}; + +/* dispatch buf */ +struct scx_dsp_buf_ent { + struct task_struct *task; + unsigned long qseq; + u64 dsq_id; + u64 enq_flags; +}; + +struct scx_dsp_ctx { + struct rq *rq; + u32 cursor; + u32 nr_tasks; + struct scx_dsp_buf_ent buf[]; +}; + +struct scx_deferred_reenq_local { + struct list_head node; + u64 flags; + u64 seq; + u32 cnt; +}; + +struct scx_sched_pcpu { + struct scx_sched *sch; + u64 flags; /* protected by rq lock */ + + /* + * The event counters are in a per-CPU variable to minimize the + * accounting overhead. A system-wide view on the event counter is + * constructed when requested by scx_bpf_events(). + */ + struct scx_event_stats event_stats; + + struct scx_deferred_reenq_local deferred_reenq_local; + struct scx_dispatch_q bypass_dsq; +#ifdef CONFIG_EXT_SUB_SCHED + u32 bypass_host_seq; +#endif + + /* must be the last entry - contains flex array */ + struct scx_dsp_ctx dsp_ctx; +}; + +struct scx_sched_pnode { + struct scx_dispatch_q global_dsq; +}; + +struct scx_sched { + struct sched_ext_ops ops; + DECLARE_BITMAP(has_op, SCX_OPI_END); + + /* + * Dispatch queues. + * + * The global DSQ (%SCX_DSQ_GLOBAL) is split per-node for scalability. + * This is to avoid live-locking in bypass mode where all tasks are + * dispatched to %SCX_DSQ_GLOBAL and all CPUs consume from it. If + * per-node split isn't sufficient, it can be further split. + */ + struct rhashtable dsq_hash; + struct scx_sched_pnode **pnode; + struct scx_sched_pcpu __percpu *pcpu; + + u64 slice_dfl; + u64 bypass_timestamp; + s32 bypass_depth; + + /* bypass dispatch path enable state, see bypass_dsp_enabled() */ + unsigned long bypass_dsp_claim; + atomic_t bypass_dsp_enable_depth; + + bool aborting; + bool dump_disabled; /* protected by scx_dump_lock */ + u32 dsp_max_batch; + s32 level; + + /* + * Updates to the following warned bitfields can race causing RMW issues + * but it doesn't really matter. + */ + bool warned_zero_slice:1; + bool warned_deprecated_rq:1; + bool warned_unassoc_progs:1; + + struct list_head all; + +#ifdef CONFIG_EXT_SUB_SCHED + struct rhash_head hash_node; + + struct list_head children; + struct list_head sibling; + struct cgroup *cgrp; + char *cgrp_path; + struct kset *sub_kset; + + bool sub_attached; +#endif /* CONFIG_EXT_SUB_SCHED */ + + /* + * The maximum amount of time in jiffies that a task may be runnable + * without being scheduled on a CPU. If this timeout is exceeded, it + * will trigger scx_error(). + */ + unsigned long watchdog_timeout; + + atomic_t exit_kind; + struct scx_exit_info *exit_info; + + struct kobject kobj; + + struct kthread_worker *helper; + struct irq_work disable_irq_work; + struct kthread_work disable_work; + struct timer_list bypass_lb_timer; + cpumask_var_t bypass_lb_donee_cpumask; + cpumask_var_t bypass_lb_resched_cpumask; + struct rcu_work rcu_work; + + /* all ancestors including self */ + struct scx_sched *ancestors[]; +}; + +enum scx_wake_flags { + /* expose select WF_* flags as enums */ + SCX_WAKE_FORK = WF_FORK, + SCX_WAKE_TTWU = WF_TTWU, + SCX_WAKE_SYNC = WF_SYNC, +}; + +enum scx_enq_flags { + /* expose select ENQUEUE_* flags as enums */ + SCX_ENQ_WAKEUP = ENQUEUE_WAKEUP, + SCX_ENQ_HEAD = ENQUEUE_HEAD, + SCX_ENQ_CPU_SELECTED = ENQUEUE_RQ_SELECTED, + + /* high 32bits are SCX specific */ + + /* + * Set the following to trigger preemption when calling + * scx_bpf_dsq_insert() with a local dsq as the target. The slice of the + * current task is cleared to zero and the CPU is kicked into the + * scheduling path. Implies %SCX_ENQ_HEAD. + */ + SCX_ENQ_PREEMPT = 1LLU << 32, + + /* + * Only allowed on local DSQs. Guarantees that the task either gets + * on the CPU immediately and stays on it, or gets reenqueued back + * to the BPF scheduler. It will never linger on a local DSQ or be + * silently put back after preemption. + * + * The protection persists until the next fresh enqueue - it + * survives SAVE/RESTORE cycles, slice extensions and preemption. + * If the task can't stay on the CPU for any reason, it gets + * reenqueued back to the BPF scheduler. + * + * Exiting and migration-disabled tasks bypass ops.enqueue() and + * are placed directly on a local DSQ without IMMED protection + * unless %SCX_OPS_ENQ_EXITING and %SCX_OPS_ENQ_MIGRATION_DISABLED + * are set respectively. + */ + SCX_ENQ_IMMED = 1LLU << 33, + + /* + * The task being enqueued was previously enqueued on a DSQ, but was + * removed and is being re-enqueued. See SCX_TASK_REENQ_* flags to find + * out why a given task is being reenqueued. + */ + SCX_ENQ_REENQ = 1LLU << 40, + + /* + * The task being enqueued is the only task available for the cpu. By + * default, ext core keeps executing such tasks but when + * %SCX_OPS_ENQ_LAST is specified, they're ops.enqueue()'d with the + * %SCX_ENQ_LAST flag set. + * + * The BPF scheduler is responsible for triggering a follow-up + * scheduling event. Otherwise, Execution may stall. + */ + SCX_ENQ_LAST = 1LLU << 41, + + /* high 8 bits are internal */ + __SCX_ENQ_INTERNAL_MASK = 0xffLLU << 56, + + SCX_ENQ_CLEAR_OPSS = 1LLU << 56, + SCX_ENQ_DSQ_PRIQ = 1LLU << 57, + SCX_ENQ_NESTED = 1LLU << 58, + SCX_ENQ_GDSQ_FALLBACK = 1LLU << 59, /* fell back to global DSQ */ +}; + +enum scx_deq_flags { + /* expose select DEQUEUE_* flags as enums */ + SCX_DEQ_SLEEP = DEQUEUE_SLEEP, + + /* high 32bits are SCX specific */ + + /* + * The generic core-sched layer decided to execute the task even though + * it hasn't been dispatched yet. Dequeue from the BPF side. + */ + SCX_DEQ_CORE_SCHED_EXEC = 1LLU << 32, + + /* + * The task is being dequeued due to a property change (e.g., + * sched_setaffinity(), sched_setscheduler(), set_user_nice(), + * etc.). + */ + SCX_DEQ_SCHED_CHANGE = 1LLU << 33, +}; + +enum scx_reenq_flags { + /* low 16bits determine which tasks should be reenqueued */ + SCX_REENQ_ANY = 1LLU << 0, /* all tasks */ + + __SCX_REENQ_FILTER_MASK = 0xffffLLU, + + __SCX_REENQ_USER_MASK = SCX_REENQ_ANY, + + /* bits 32-35 used by task_should_reenq() */ + SCX_REENQ_TSR_RQ_OPEN = 1LLU << 32, + SCX_REENQ_TSR_NOT_FIRST = 1LLU << 33, + + __SCX_REENQ_TSR_MASK = 0xfLLU << 32, +}; + +enum scx_pick_idle_cpu_flags { + SCX_PICK_IDLE_CORE = 1LLU << 0, /* pick a CPU whose SMT siblings are also idle */ + SCX_PICK_IDLE_IN_NODE = 1LLU << 1, /* pick a CPU in the same target NUMA node */ +}; + +enum scx_kick_flags { + /* + * Kick the target CPU if idle. Guarantees that the target CPU goes + * through at least one full scheduling cycle before going idle. If the + * target CPU can be determined to be currently not idle and going to go + * through a scheduling cycle before going idle, noop. + */ + SCX_KICK_IDLE = 1LLU << 0, + + /* + * Preempt the current task and execute the dispatch path. If the + * current task of the target CPU is an SCX task, its ->scx.slice is + * cleared to zero before the scheduling path is invoked so that the + * task expires and the dispatch path is invoked. + */ + SCX_KICK_PREEMPT = 1LLU << 1, + + /* + * The scx_bpf_kick_cpu() call will return after the current SCX task of + * the target CPU switches out. This can be used to implement e.g. core + * scheduling. This has no effect if the current task on the target CPU + * is not on SCX. + */ + SCX_KICK_WAIT = 1LLU << 2, +}; + +enum scx_tg_flags { + SCX_TG_ONLINE = 1U << 0, + SCX_TG_INITED = 1U << 1, +}; + +enum scx_enable_state { + SCX_ENABLING, + SCX_ENABLED, + SCX_DISABLING, + SCX_DISABLED, +}; + +static const char *scx_enable_state_str[] = { + [SCX_ENABLING] = "enabling", + [SCX_ENABLED] = "enabled", + [SCX_DISABLING] = "disabling", + [SCX_DISABLED] = "disabled", +}; + +/* + * Task Ownership State Machine (sched_ext_entity->ops_state) + * + * The sched_ext core uses this state machine to track task ownership + * between the SCX core and the BPF scheduler. This allows the BPF + * scheduler to dispatch tasks without strict ordering requirements, while + * the SCX core safely rejects invalid dispatches. + * + * State Transitions + * + * .------------> NONE (owned by SCX core) + * | | ^ + * | enqueue | | direct dispatch + * | v | + * | QUEUEING -------' + * | | + * | enqueue | + * | completes | + * | v + * | QUEUED (owned by BPF scheduler) + * | | + * | dispatch | + * | | + * | v + * | DISPATCHING + * | | + * | dispatch | + * | completes | + * `---------------' + * + * State Descriptions + * + * - %SCX_OPSS_NONE: + * Task is owned by the SCX core. It's either on a run queue, running, + * or being manipulated by the core scheduler. The BPF scheduler has no + * claim on this task. + * + * - %SCX_OPSS_QUEUEING: + * Transitional state while transferring a task from the SCX core to + * the BPF scheduler. The task's rq lock is held during this state. + * Since QUEUEING is both entered and exited under the rq lock, dequeue + * can never observe this state (it would be a BUG). When finishing a + * dispatch, if the task is still in %SCX_OPSS_QUEUEING the completion + * path busy-waits for it to leave this state (via wait_ops_state()) + * before retrying. + * + * - %SCX_OPSS_QUEUED: + * Task is owned by the BPF scheduler. It's on a DSQ (dispatch queue) + * and the BPF scheduler is responsible for dispatching it. A QSEQ + * (queue sequence number) is embedded in this state to detect + * dispatch/dequeue races: if a task is dequeued and re-enqueued, the + * QSEQ changes and any in-flight dispatch operations targeting the old + * QSEQ are safely ignored. + * + * - %SCX_OPSS_DISPATCHING: + * Transitional state while transferring a task from the BPF scheduler + * back to the SCX core. This state indicates the BPF scheduler has + * selected the task for execution. When dequeue needs to take the task + * off a DSQ and it is still in %SCX_OPSS_DISPATCHING, the dequeue path + * busy-waits for it to leave this state (via wait_ops_state()) before + * proceeding. Exits to %SCX_OPSS_NONE when dispatch completes. + * + * Memory Ordering + * + * Transitions out of %SCX_OPSS_QUEUEING and %SCX_OPSS_DISPATCHING into + * %SCX_OPSS_NONE or %SCX_OPSS_QUEUED must use atomic_long_set_release() + * and waiters must use atomic_long_read_acquire(). This ensures proper + * synchronization between concurrent operations. + * + * Cross-CPU Task Migration + * + * When moving a task in the %SCX_OPSS_DISPATCHING state, we can't simply + * grab the target CPU's rq lock because a concurrent dequeue might be + * waiting on %SCX_OPSS_DISPATCHING while holding the source rq lock + * (deadlock). + * + * The sched_ext core uses a "lock dancing" protocol coordinated by + * p->scx.holding_cpu. When moving a task to a different rq: + * + * 1. Verify task can be moved (CPU affinity, migration_disabled, etc.) + * 2. Set p->scx.holding_cpu to the current CPU + * 3. Set task state to %SCX_OPSS_NONE; dequeue waits while DISPATCHING + * is set, so clearing DISPATCHING first prevents the circular wait + * (safe to lock the rq we need) + * 4. Unlock the current CPU's rq + * 5. Lock src_rq (where the task currently lives) + * 6. Verify p->scx.holding_cpu == current CPU, if not, dequeue won the + * race (dequeue clears holding_cpu to -1 when it takes the task), in + * this case migration is aborted + * 7. If src_rq == dst_rq: clear holding_cpu and enqueue directly + * into dst_rq's local DSQ (no lock swap needed) + * 8. Otherwise: call move_remote_task_to_local_dsq(), which releases + * src_rq, locks dst_rq, and performs the deactivate/activate + * migration cycle (dst_rq is held on return) + * 9. Unlock dst_rq and re-lock the current CPU's rq to restore + * the lock state expected by the caller + * + * If any verification fails, abort the migration. + * + * This state tracking allows the BPF scheduler to try to dispatch any task + * at any time regardless of its state. The SCX core can safely + * reject/ignore invalid dispatches, simplifying the BPF scheduler + * implementation. + */ +enum scx_ops_state { + SCX_OPSS_NONE, /* owned by the SCX core */ + SCX_OPSS_QUEUEING, /* in transit to the BPF scheduler */ + SCX_OPSS_QUEUED, /* owned by the BPF scheduler */ + SCX_OPSS_DISPATCHING, /* in transit back to the SCX core */ + + /* + * QSEQ brands each QUEUED instance so that, when dispatch races + * dequeue/requeue, the dispatcher can tell whether it still has a claim + * on the task being dispatched. + * + * As some 32bit archs can't do 64bit store_release/load_acquire, + * p->scx.ops_state is atomic_long_t which leaves 30 bits for QSEQ on + * 32bit machines. The dispatch race window QSEQ protects is very narrow + * and runs with IRQ disabled. 30 bits should be sufficient. + */ + SCX_OPSS_QSEQ_SHIFT = 2, +}; + +/* Use macros to ensure that the type is unsigned long for the masks */ +#define SCX_OPSS_STATE_MASK ((1LU << SCX_OPSS_QSEQ_SHIFT) - 1) +#define SCX_OPSS_QSEQ_MASK (~SCX_OPSS_STATE_MASK) + +extern struct scx_sched __rcu *scx_root; +DECLARE_PER_CPU(struct rq *, scx_locked_rq_state); + +int scx_kfunc_context_filter(const struct bpf_prog *prog, u32 kfunc_id); + +/* + * Return the rq currently locked from an scx callback, or NULL if no rq is + * locked. + */ +static inline struct rq *scx_locked_rq(void) +{ + return __this_cpu_read(scx_locked_rq_state); +} + +static inline bool scx_bypassing(struct scx_sched *sch, s32 cpu) +{ + return unlikely(per_cpu_ptr(sch->pcpu, cpu)->flags & + SCX_SCHED_PCPU_BYPASSING); +} + +#ifdef CONFIG_EXT_SUB_SCHED +/** + * scx_task_sched - Find scx_sched scheduling a task + * @p: task of interest + * + * Return @p's scheduler instance. Must be called with @p's pi_lock or rq lock + * held. + */ +static inline struct scx_sched *scx_task_sched(const struct task_struct *p) +{ + return rcu_dereference_protected(p->scx.sched, + lockdep_is_held(&p->pi_lock) || + lockdep_is_held(__rq_lockp(task_rq(p)))); +} + +/** + * scx_task_sched_rcu - Find scx_sched scheduling a task + * @p: task of interest + * + * Return @p's scheduler instance. The returned scx_sched is RCU protected. + */ +static inline struct scx_sched *scx_task_sched_rcu(const struct task_struct *p) +{ + return rcu_dereference_all(p->scx.sched); +} + +/** + * scx_task_on_sched - Is a task on the specified sched? + * @sch: sched to test against + * @p: task of interest + * + * Returns %true if @p is on @sch, %false otherwise. + */ +static inline bool scx_task_on_sched(struct scx_sched *sch, + const struct task_struct *p) +{ + return rcu_access_pointer(p->scx.sched) == sch; +} + +/** + * scx_prog_sched - Find scx_sched associated with a BPF prog + * @aux: aux passed in from BPF to a kfunc + * + * To be called from kfuncs. Return the scheduler instance associated with the + * BPF program given the implicit kfunc argument aux. The returned scx_sched is + * RCU protected. + */ +static inline struct scx_sched *scx_prog_sched(const struct bpf_prog_aux *aux) +{ + struct sched_ext_ops *ops; + struct scx_sched *root; + + ops = bpf_prog_get_assoc_struct_ops(aux); + if (likely(ops)) + return rcu_dereference_all(ops->priv); + + root = rcu_dereference_all(scx_root); + if (root) { + /* + * COMPAT-v6.19: Schedulers built before sub-sched support was + * introduced may have unassociated non-struct_ops programs. + */ + if (!root->ops.sub_attach) + return root; + + if (!root->warned_unassoc_progs) { + printk_deferred(KERN_WARNING "sched_ext: Unassociated program %s (id %d)\n", + aux->name, aux->id); + root->warned_unassoc_progs = true; + } + } + + return NULL; +} +#else /* CONFIG_EXT_SUB_SCHED */ +static inline struct scx_sched *scx_task_sched(const struct task_struct *p) +{ + return rcu_dereference_protected(scx_root, + lockdep_is_held(&p->pi_lock) || + lockdep_is_held(__rq_lockp(task_rq(p)))); +} + +static inline struct scx_sched *scx_task_sched_rcu(const struct task_struct *p) +{ + return rcu_dereference_all(scx_root); +} + +static inline bool scx_task_on_sched(struct scx_sched *sch, + const struct task_struct *p) +{ + return true; +} + +static struct scx_sched *scx_prog_sched(const struct bpf_prog_aux *aux) +{ + return rcu_dereference_all(scx_root); +} +#endif /* CONFIG_EXT_SUB_SCHED */ diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index c798d2795243..3ebec186f982 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -15,7 +15,7 @@ * Author: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> * * Scaled math optimizations by Thomas Gleixner - * Copyright (C) 2007, Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2007, Linutronix GmbH, Thomas Gleixner <tglx@kernel.org> * * Adaptive scheduling granularity, math enhancements by Peter Zijlstra * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra @@ -74,12 +74,12 @@ unsigned int sysctl_sched_tunable_scaling = SCHED_TUNABLESCALING_LOG; /* * Minimal preemption granularity for CPU-bound tasks: * - * (default: 0.75 msec * (1 + ilog(ncpus)), units: nanoseconds) + * (default: 0.70 msec * (1 + ilog(ncpus)), units: nanoseconds) */ -unsigned int sysctl_sched_base_slice = 750000ULL; -static unsigned int normalized_sysctl_sched_base_slice = 750000ULL; +unsigned int sysctl_sched_base_slice = 700000ULL; +static unsigned int normalized_sysctl_sched_base_slice = 700000ULL; -const_debug unsigned int sysctl_sched_migration_cost = 500000UL; +__read_mostly unsigned int sysctl_sched_migration_cost = 500000UL; static int __init setup_sched_thermal_decay_shift(char *str) { @@ -88,7 +88,6 @@ static int __init setup_sched_thermal_decay_shift(char *str) } __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. */ @@ -111,7 +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 /* @@ -162,7 +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) { @@ -227,6 +225,7 @@ void __init sched_init_granularity(void) update_sysctl(); } +#ifndef CONFIG_64BIT #define WMULT_CONST (~0U) #define WMULT_SHIFT 32 @@ -285,6 +284,12 @@ static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight return mul_u64_u32_shr(delta_exec, fact, shift); } +#else +static u64 __calc_delta(u64 delta_exec, unsigned long weight, struct load_weight *lw) +{ + return (delta_exec * weight) / lw->weight; +} +#endif /* * delta /= w @@ -399,7 +404,7 @@ 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 through all leaf cfs_rq's on a runqueue */ @@ -471,7 +476,7 @@ 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) @@ -517,7 +522,7 @@ static int se_is_idle(struct sched_entity *se) return task_has_idle_policy(task_of(se)); } -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* !CONFIG_FAIR_GROUP_SCHED */ static __always_inline void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec); @@ -526,10 +531,48 @@ void account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec); * Scheduling class tree data structure manipulation methods: */ +extern void __BUILD_BUG_vruntime_cmp(void); + +/* Use __builtin_strcmp() because of __HAVE_ARCH_STRCMP: */ + +#define vruntime_cmp(A, CMP_STR, B) ({ \ + int __res = 0; \ + \ + if (!__builtin_strcmp(CMP_STR, "<")) { \ + __res = ((s64)((A)-(B)) < 0); \ + } else if (!__builtin_strcmp(CMP_STR, "<=")) { \ + __res = ((s64)((A)-(B)) <= 0); \ + } else if (!__builtin_strcmp(CMP_STR, ">")) { \ + __res = ((s64)((A)-(B)) > 0); \ + } else if (!__builtin_strcmp(CMP_STR, ">=")) { \ + __res = ((s64)((A)-(B)) >= 0); \ + } else { \ + /* Unknown operator throws linker error: */ \ + __BUILD_BUG_vruntime_cmp(); \ + } \ + \ + __res; \ +}) + +extern void __BUILD_BUG_vruntime_op(void); + +#define vruntime_op(A, OP_STR, B) ({ \ + s64 __res = 0; \ + \ + if (!__builtin_strcmp(OP_STR, "-")) { \ + __res = (s64)((A)-(B)); \ + } else { \ + /* Unknown operator throws linker error: */ \ + __BUILD_BUG_vruntime_op(); \ + } \ + \ + __res; \ +}) + + static inline __maybe_unused u64 max_vruntime(u64 max_vruntime, u64 vruntime) { - s64 delta = (s64)(vruntime - max_vruntime); - if (delta > 0) + if (vruntime_cmp(vruntime, ">", max_vruntime)) max_vruntime = vruntime; return max_vruntime; @@ -537,8 +580,7 @@ static inline __maybe_unused u64 max_vruntime(u64 max_vruntime, u64 vruntime) static inline __maybe_unused u64 min_vruntime(u64 min_vruntime, u64 vruntime) { - s64 delta = (s64)(vruntime - min_vruntime); - if (delta < 0) + if (vruntime_cmp(vruntime, "<", min_vruntime)) min_vruntime = vruntime; return min_vruntime; @@ -551,12 +593,27 @@ static inline bool entity_before(const struct sched_entity *a, * Tiebreak on vruntime seems unnecessary since it can * hardly happen. */ - return (s64)(a->deadline - b->deadline) < 0; + return vruntime_cmp(a->deadline, "<", b->deadline); } +/* + * Per avg_vruntime() below, cfs_rq::zero_vruntime is only slightly stale + * and this value should be no more than two lag bounds. Which puts it in the + * general order of: + * + * (slice + TICK_NSEC) << NICE_0_LOAD_SHIFT + * + * which is around 44 bits in size (on 64bit); that is 20 for + * NICE_0_LOAD_SHIFT, another 20 for NSEC_PER_MSEC and then a handful for + * however many msec the actual slice+tick ends up begin. + * + * (disregarding the actual divide-by-weight part makes for the worst case + * weight of 2, which nicely cancels vs the fuzz in zero_vruntime not actually + * being the zero-lag point). + */ static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) { - return (s64)(se->vruntime - cfs_rq->min_vruntime); + return vruntime_op(se->vruntime, "-", cfs_rq->zero_vruntime); } #define __node_2_se(node) \ @@ -578,7 +635,7 @@ static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) * * \Sum lag_i = 0 * \Sum w_i * (V - v_i) = 0 - * \Sum w_i * V - w_i * 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): @@ -608,74 +665,156 @@ static inline s64 entity_key(struct cfs_rq *cfs_rq, struct sched_entity *se) * * Which we track using: * - * v0 := cfs_rq->min_vruntime - * \Sum (v_i - v0) * w_i := cfs_rq->avg_vruntime - * \Sum w_i := cfs_rq->avg_load - * - * Since min_vruntime is a monotonic increasing variable that 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. + * v0 := cfs_rq->zero_vruntime + * \Sum (v_i - v0) * w_i := cfs_rq->sum_w_vruntime + * \Sum w_i := cfs_rq->sum_weight * - * Also, we use scale_load_down() to reduce the size. - * - * As measured, the max (key * weight) value was ~44 bits for a kernel build. + * Since zero_vruntime closely tracks the per-task service, these + * deltas: (v_i - v0), will be in the order of the maximal (virtual) lag + * induced in the system due to quantisation. */ +static inline unsigned long avg_vruntime_weight(struct cfs_rq *cfs_rq, unsigned long w) +{ +#ifdef CONFIG_64BIT + if (cfs_rq->sum_shift) + w = max(2UL, w >> cfs_rq->sum_shift); +#endif + return w; +} + +static inline void +__sum_w_vruntime_add(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + unsigned long weight = avg_vruntime_weight(cfs_rq, se->load.weight); + s64 w_vruntime, key = entity_key(cfs_rq, se); + + w_vruntime = key * weight; + WARN_ON_ONCE((w_vruntime >> 63) != (w_vruntime >> 62)); + + cfs_rq->sum_w_vruntime += w_vruntime; + cfs_rq->sum_weight += weight; +} + static void -avg_vruntime_add(struct cfs_rq *cfs_rq, struct sched_entity *se) +sum_w_vruntime_add_paranoid(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); + unsigned long weight; + s64 key, tmp; + +again: + weight = avg_vruntime_weight(cfs_rq, se->load.weight); + key = entity_key(cfs_rq, se); + + if (check_mul_overflow(key, weight, &key)) + goto overflow; + + if (check_add_overflow(cfs_rq->sum_w_vruntime, key, &tmp)) + goto overflow; + + cfs_rq->sum_w_vruntime = tmp; + cfs_rq->sum_weight += weight; + return; + +overflow: + /* + * There's gotta be a limit -- if we're still failing at this point + * there's really nothing much to be done about things. + */ + BUG_ON(cfs_rq->sum_shift >= 10); + cfs_rq->sum_shift++; + + /* + * Note: \Sum (k_i * (w_i >> 1)) != (\Sum (k_i * w_i)) >> 1 + */ + cfs_rq->sum_w_vruntime = 0; + cfs_rq->sum_weight = 0; + + for (struct rb_node *node = cfs_rq->tasks_timeline.rb_leftmost; + node; node = rb_next(node)) + __sum_w_vruntime_add(cfs_rq, __node_2_se(node)); + + goto again; +} + +static void +sum_w_vruntime_add(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + if (sched_feat(PARANOID_AVG)) + return sum_w_vruntime_add_paranoid(cfs_rq, se); - cfs_rq->avg_vruntime += key * weight; - cfs_rq->avg_load += weight; + __sum_w_vruntime_add(cfs_rq, se); } static void -avg_vruntime_sub(struct cfs_rq *cfs_rq, struct sched_entity *se) +sum_w_vruntime_sub(struct cfs_rq *cfs_rq, struct sched_entity *se) { - unsigned long weight = scale_load_down(se->load.weight); + unsigned long weight = avg_vruntime_weight(cfs_rq, se->load.weight); s64 key = entity_key(cfs_rq, se); - cfs_rq->avg_vruntime -= key * weight; - cfs_rq->avg_load -= weight; + cfs_rq->sum_w_vruntime -= key * weight; + cfs_rq->sum_weight -= weight; } static inline -void avg_vruntime_update(struct cfs_rq *cfs_rq, s64 delta) +void update_zero_vruntime(struct cfs_rq *cfs_rq, s64 delta) { /* - * v' = v + d ==> avg_vruntime' = avg_runtime - d*avg_load + * v' = v + d ==> sum_w_vruntime' = sum_w_vruntime - d*sum_weight */ - cfs_rq->avg_vruntime -= cfs_rq->avg_load * delta; + cfs_rq->sum_w_vruntime -= cfs_rq->sum_weight * delta; + cfs_rq->zero_vruntime += delta; } /* - * Specifically: avg_runtime() + 0 must result in entity_eligible() := true + * Specifically: avg_vruntime() + 0 must result in entity_eligible() := true * For this to be so, the result of this function must have a left bias. + * + * Called in: + * - place_entity() -- before enqueue + * - update_entity_lag() -- before dequeue + * - update_deadline() -- slice expiration + * + * This means it is one entry 'behind' but that puts it close enough to where + * the bound on entity_key() is at most two lag bounds. */ u64 avg_vruntime(struct cfs_rq *cfs_rq) { struct sched_entity *curr = cfs_rq->curr; - s64 avg = cfs_rq->avg_vruntime; - long load = cfs_rq->avg_load; + long weight = cfs_rq->sum_weight; + s64 delta = 0; - if (curr && curr->on_rq) { - unsigned long weight = scale_load_down(curr->load.weight); + if (curr && !curr->on_rq) + curr = NULL; - avg += entity_key(cfs_rq, curr) * weight; - load += weight; - } + if (weight) { + s64 runtime = cfs_rq->sum_w_vruntime; + + if (curr) { + unsigned long w = avg_vruntime_weight(cfs_rq, curr->load.weight); + + runtime += entity_key(cfs_rq, curr) * w; + weight += w; + } - if (load) { /* sign flips effective floor / ceiling */ - if (avg < 0) - avg -= (load - 1); - avg = div_s64(avg, load); + if (runtime < 0) + runtime -= (weight - 1); + + delta = div64_long(runtime, weight); + } else if (curr) { + /* + * When there is but one element, it is the average. + */ + delta = curr->vruntime - cfs_rq->zero_vruntime; } - return cfs_rq->min_vruntime + avg; + update_zero_vruntime(cfs_rq, delta); + + return cfs_rq->zero_vruntime; } +static inline u64 cfs_rq_max_slice(struct cfs_rq *cfs_rq); + /* * lag_i = S - s_i = w_i * (V - v_i) * @@ -689,19 +828,50 @@ u64 avg_vruntime(struct cfs_rq *cfs_rq) * 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) +static s64 entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se, u64 avruntime) { + u64 max_slice = cfs_rq_max_slice(cfs_rq) + TICK_NSEC; s64 vlag, limit; - SCHED_WARN_ON(!se->on_rq); + vlag = avruntime - se->vruntime; + limit = calc_delta_fair(max_slice, se); + + return clamp(vlag, -limit, limit); +} + +/* + * Delayed dequeue aims to reduce the negative lag of a dequeued task. While + * updating the lag of an entity, check that negative lag didn't increase + * during the delayed dequeue period which would be unfair. + * Similarly, check that the entity didn't gain positive lag when DELAY_ZERO + * is set. + * + * Return true if the vlag has been modified. Specifically: + * + * se->vlag != avg_vruntime() - se->vruntime + * + * This can be due to clamping in entity_lag() or clamping due to + * sched_delayed. Either way, when vlag is modified and the entity is + * retained, the tree needs to be adjusted. + */ +static __always_inline +bool update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + u64 avruntime = avg_vruntime(cfs_rq); + s64 vlag = entity_lag(cfs_rq, se, avruntime); + + 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); + if (se->sched_delayed) { + /* previous vlag < 0 otherwise se would not be delayed */ + vlag = max(vlag, se->vlag); + if (sched_feat(DELAY_ZERO)) + vlag = min(vlag, 0); + } + se->vlag = vlag; - se->vlag = clamp(vlag, -limit, limit); + return avruntime - vlag != se->vruntime; } /* @@ -712,11 +882,11 @@ static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) * * lag_i >= 0 -> V >= v_i * - * \Sum (v_i - v)*w_i - * V = ------------------ + v + * \Sum (v_i - v0)*w_i + * V = ------------------- + v0 * \Sum w_i * - * lag_i >= 0 -> \Sum (v_i - v)*w_i >= (v_i - v)*(\Sum w_i) + * lag_i >= 0 -> \Sum (v_i - v0)*w_i >= (v_i - v0)*(\Sum w_i) * * Note: using 'avg_vruntime() > se->vruntime' is inaccurate due * to the loss in precision caused by the division. @@ -724,17 +894,46 @@ static void update_entity_lag(struct cfs_rq *cfs_rq, struct sched_entity *se) static int vruntime_eligible(struct cfs_rq *cfs_rq, u64 vruntime) { struct sched_entity *curr = cfs_rq->curr; - s64 avg = cfs_rq->avg_vruntime; - long load = cfs_rq->avg_load; + s64 key, avg = cfs_rq->sum_w_vruntime; + long load = cfs_rq->sum_weight; if (curr && curr->on_rq) { - unsigned long weight = scale_load_down(curr->load.weight); + unsigned long weight = avg_vruntime_weight(cfs_rq, curr->load.weight); avg += entity_key(cfs_rq, curr) * weight; load += weight; } - return avg >= (s64)(vruntime - cfs_rq->min_vruntime) * load; + key = vruntime_op(vruntime, "-", cfs_rq->zero_vruntime); + + /* + * The worst case term for @key includes 'NSEC_TICK * NICE_0_LOAD' + * and @load obviously includes NICE_0_LOAD. NSEC_TICK is around 24 + * bits, while NICE_0_LOAD is 20 on 64bit and 10 otherwise. + * + * This gives that on 64bit the product will be at least 64bit which + * overflows s64, while on 32bit it will only be 44bits and should fit + * comfortably. + */ +#ifdef CONFIG_64BIT +#ifdef CONFIG_ARCH_SUPPORTS_INT128 + /* This often results in simpler code than __builtin_mul_overflow(). */ + return avg >= (__int128)key * load; +#else + s64 rhs; + /* + * On overflow, the sign of key tells us the correct answer: a large + * positive key means vruntime >> V, so not eligible; a large negative + * key means vruntime << V, so eligible. + */ + if (check_mul_overflow(key, load, &rhs)) + return key <= 0; + + return avg >= rhs; +#endif +#else /* 32bit */ + return avg >= key * load; +#endif } int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se) @@ -742,57 +941,34 @@ int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se) return vruntime_eligible(cfs_rq, se->vruntime); } -static u64 __update_min_vruntime(struct cfs_rq *cfs_rq, u64 vruntime) -{ - u64 min_vruntime = cfs_rq->min_vruntime; - /* - * open coded max_vruntime() to allow updating avg_vruntime - */ - s64 delta = (s64)(vruntime - min_vruntime); - if (delta > 0) { - avg_vruntime_update(cfs_rq, delta); - min_vruntime = vruntime; - } - return min_vruntime; -} - -static void update_min_vruntime(struct cfs_rq *cfs_rq) +static inline u64 cfs_rq_min_slice(struct cfs_rq *cfs_rq) { - struct sched_entity *se = __pick_root_entity(cfs_rq); + struct sched_entity *root = __pick_root_entity(cfs_rq); struct sched_entity *curr = cfs_rq->curr; - u64 vruntime = cfs_rq->min_vruntime; + u64 min_slice = ~0ULL; - if (curr) { - if (curr->on_rq) - vruntime = curr->vruntime; - else - curr = NULL; - } + if (curr && curr->on_rq) + min_slice = curr->slice; - if (se) { - if (!curr) - vruntime = se->min_vruntime; - else - vruntime = min_vruntime(vruntime, se->min_vruntime); - } + if (root) + min_slice = min(min_slice, root->min_slice); - /* ensure we never gain time by being placed backwards. */ - cfs_rq->min_vruntime = __update_min_vruntime(cfs_rq, vruntime); + return min_slice; } -static inline u64 cfs_rq_min_slice(struct cfs_rq *cfs_rq) +static inline u64 cfs_rq_max_slice(struct cfs_rq *cfs_rq) { struct sched_entity *root = __pick_root_entity(cfs_rq); struct sched_entity *curr = cfs_rq->curr; - u64 min_slice = ~0ULL; + u64 max_slice = 0ULL; if (curr && curr->on_rq) - min_slice = curr->slice; + max_slice = curr->slice; if (root) - min_slice = min(min_slice, root->min_slice); + max_slice = max(max_slice, root->max_slice); - return min_slice; + return max_slice; } static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) @@ -800,13 +976,12 @@ static inline bool __entity_less(struct rb_node *a, const struct rb_node *b) return entity_before(__node_2_se(a), __node_2_se(b)); } -#define vruntime_gt(field, lse, rse) ({ (s64)((lse)->field - (rse)->field) > 0; }) - 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)) + + if (vruntime_cmp(se->min_vruntime, ">", rse->min_vruntime)) se->min_vruntime = rse->min_vruntime; } } @@ -820,6 +995,15 @@ static inline void __min_slice_update(struct sched_entity *se, struct rb_node *n } } +static inline void __max_slice_update(struct sched_entity *se, struct rb_node *node) +{ + if (node) { + struct sched_entity *rse = __node_2_se(node); + if (rse->max_slice > se->max_slice) + se->max_slice = rse->max_slice; + } +} + /* * se->min_vruntime = min(se->vruntime, {left,right}->min_vruntime) */ @@ -827,6 +1011,7 @@ static inline bool min_vruntime_update(struct sched_entity *se, bool exit) { u64 old_min_vruntime = se->min_vruntime; u64 old_min_slice = se->min_slice; + u64 old_max_slice = se->max_slice; struct rb_node *node = &se->run_node; se->min_vruntime = se->vruntime; @@ -837,8 +1022,13 @@ static inline bool min_vruntime_update(struct sched_entity *se, bool exit) __min_slice_update(se, node->rb_right); __min_slice_update(se, node->rb_left); + se->max_slice = se->slice; + __max_slice_update(se, node->rb_right); + __max_slice_update(se, node->rb_left); + return se->min_vruntime == old_min_vruntime && - se->min_slice == old_min_slice; + se->min_slice == old_min_slice && + se->max_slice == old_max_slice; } RB_DECLARE_CALLBACKS(static, min_vruntime_cb, struct sched_entity, @@ -849,7 +1039,7 @@ RB_DECLARE_CALLBACKS(static, min_vruntime_cb, struct sched_entity, */ static void __enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) { - avg_vruntime_add(cfs_rq, se); + sum_w_vruntime_add(cfs_rq, se); se->min_vruntime = se->vruntime; se->min_slice = se->slice; rb_add_augmented_cached(&se->run_node, &cfs_rq->tasks_timeline, @@ -860,7 +1050,7 @@ 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); + sum_w_vruntime_sub(cfs_rq, se); } struct sched_entity *__pick_root_entity(struct cfs_rq *cfs_rq) @@ -884,6 +1074,47 @@ struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) } /* + * 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 inline void set_protect_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + u64 slice = normalized_sysctl_sched_base_slice; + u64 vprot = se->deadline; + + if (sched_feat(RUN_TO_PARITY)) + slice = cfs_rq_min_slice(cfs_rq); + + slice = min(slice, se->slice); + if (slice != se->slice) + vprot = min_vruntime(vprot, se->vruntime + calc_delta_fair(slice, se)); + + se->vprot = vprot; +} + +static inline void update_protect_slice(struct cfs_rq *cfs_rq, struct sched_entity *se) +{ + u64 slice = cfs_rq_min_slice(cfs_rq); + + se->vprot = min_vruntime(se->vprot, se->vruntime + calc_delta_fair(slice, se)); +} + +static inline bool protect_slice(struct sched_entity *se) +{ + return vruntime_cmp(se->vruntime, "<", se->vprot); +} + +static inline void cancel_protect_slice(struct sched_entity *se) +{ + if (protect_slice(se)) + se->vprot = se->vruntime; +} + +/* * Earliest Eligible Virtual Deadline First * * In order to provide latency guarantees for different request sizes @@ -902,7 +1133,7 @@ struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq) * * Which allows tree pruning through eligibility. */ -static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) +static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq, bool protect) { struct rb_node *node = cfs_rq->tasks_timeline.rb_root.rb_node; struct sched_entity *se = __pick_first_entity(cfs_rq); @@ -916,14 +1147,20 @@ static struct sched_entity *pick_eevdf(struct cfs_rq *cfs_rq) 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) && protect && + cfs_rq->next && entity_eligible(cfs_rq, cfs_rq->next)) { + /* ->next will never be delayed */ + WARN_ON_ONCE(cfs_rq->next->sched_delayed); + return cfs_rq->next; + } + if (curr && (!curr->on_rq || !entity_eligible(cfs_rq, curr))) curr = NULL; - /* - * Once selected, run a task until it either becomes non-eligible or - * until it gets a new slice. See the HACK in set_next_entity(). - */ - if (sched_feat(RUN_TO_PARITY) && curr && curr->vlag == curr->deadline) + if (curr && protect && protect_slice(curr)) return curr; /* Pick the leftmost entity if it's eligible */ @@ -967,7 +1204,6 @@ found: return best; } -#ifdef CONFIG_SCHED_DEBUG struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) { struct rb_node *last = rb_last(&cfs_rq->tasks_timeline.rb_root); @@ -981,7 +1217,6 @@ struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) /************************************************************** * Scheduling class statistics methods: */ -#ifdef CONFIG_SMP int sched_update_scaling(void) { unsigned int factor = get_update_sysctl_factor(); @@ -993,8 +1228,6 @@ int sched_update_scaling(void) return 0; } -#endif -#endif static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se); @@ -1004,7 +1237,7 @@ static void clear_buddies(struct cfs_rq *cfs_rq, struct sched_entity *se); */ static bool update_deadline(struct cfs_rq *cfs_rq, struct sched_entity *se) { - if ((s64)(se->vruntime - se->deadline) < 0) + if (vruntime_cmp(se->vruntime, "<", se->deadline)) return false; /* @@ -1019,6 +1252,7 @@ static bool update_deadline(struct cfs_rq *cfs_rq, struct sched_entity *se) * EEVDF: vd_i = ve_i + r_i / w_i */ se->deadline = se->vruntime + calc_delta_fair(se->slice, se); + avg_vruntime(cfs_rq); /* * The task has consumed its request, reschedule. @@ -1027,7 +1261,6 @@ static bool update_deadline(struct cfs_rq *cfs_rq, struct sched_entity *se) } #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); @@ -1117,34 +1350,40 @@ 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 void update_tg_load_avg(struct cfs_rq *cfs_rq) -{ -} -#endif /* CONFIG_SMP */ - -static s64 update_curr_se(struct rq *rq, struct sched_entity *curr) +static s64 update_se(struct rq *rq, struct sched_entity *se) { u64 now = rq_clock_task(rq); s64 delta_exec; - delta_exec = now - curr->exec_start; + delta_exec = now - se->exec_start; if (unlikely(delta_exec <= 0)) return delta_exec; - curr->exec_start = now; - curr->sum_exec_runtime += 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(curr); + stats = __schedstats_from_se(se); __schedstat_set(stats->exec_max, max(delta_exec, stats->exec_max)); } @@ -1152,58 +1391,14 @@ static s64 update_curr_se(struct rq *rq, struct sched_entity *curr) return delta_exec; } -static inline void update_curr_task(struct task_struct *p, s64 delta_exec) -{ - trace_sched_stat_runtime(p, delta_exec); - account_group_exec_runtime(p, delta_exec); - cgroup_account_cputime(p, delta_exec); -} - -static inline bool did_preempt_short(struct cfs_rq *cfs_rq, struct sched_entity *curr) -{ - if (!sched_feat(PREEMPT_SHORT)) - return false; - - if (curr->vlag == curr->deadline) - return false; - - return !entity_eligible(cfs_rq, curr); -} - -static inline bool do_preempt_short(struct cfs_rq *cfs_rq, - struct sched_entity *pse, struct sched_entity *se) -{ - if (!sched_feat(PREEMPT_SHORT)) - return false; - - if (pse->slice >= se->slice) - return false; - - if (!entity_eligible(cfs_rq, pse)) - return false; - - if (entity_before(pse, se)) - return true; - - if (!entity_eligible(cfs_rq, se)) - return true; - - return false; -} +static void set_next_buddy(struct sched_entity *se); /* * Used by other classes to account runtime. */ s64 update_curr_common(struct rq *rq) { - struct task_struct *donor = rq->donor; - s64 delta_exec; - - delta_exec = update_curr_se(rq, &donor->se); - if (likely(delta_exec > 0)) - update_curr_task(donor, delta_exec); - - return delta_exec; + return update_se(rq, &rq->donor->se); } /* @@ -1211,6 +1406,12 @@ s64 update_curr_common(struct rq *rq) */ 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; struct rq *rq = rq_of(cfs_rq); s64 delta_exec; @@ -1219,19 +1420,14 @@ static void update_curr(struct cfs_rq *cfs_rq) if (unlikely(!curr)) return; - delta_exec = update_curr_se(rq, curr); + delta_exec = update_se(rq, curr); if (unlikely(delta_exec <= 0)) return; curr->vruntime += calc_delta_fair(delta_exec, curr); resched = update_deadline(cfs_rq, curr); - update_min_vruntime(cfs_rq); if (entity_is_task(curr)) { - struct task_struct *p = task_of(curr); - - update_curr_task(p, 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 @@ -1242,8 +1438,7 @@ static void update_curr(struct cfs_rq *cfs_rq) * against fair_server such that it can account for this time * and possibly avoid running this period. */ - if (dl_server_active(&rq->fair_server)) - dl_server_update(&rq->fair_server, delta_exec); + dl_server_update(&rq->fair_server, delta_exec); } account_cfs_rq_runtime(cfs_rq, delta_exec); @@ -1251,7 +1446,7 @@ static void update_curr(struct cfs_rq *cfs_rq) if (cfs_rq->nr_queued == 1) return; - if (resched || did_preempt_short(cfs_rq, curr)) { + if (resched || !protect_slice(curr)) { resched_curr_lazy(rq); clear_buddies(cfs_rq, curr); } @@ -1498,7 +1693,7 @@ 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; @@ -1532,7 +1727,7 @@ static unsigned int task_scan_start(struct task_struct *p) /* Scale the maximum scan period with the amount of shared memory. */ rcu_read_lock(); - ng = rcu_dereference(p->numa_group); + ng = rcu_dereference_all(p->numa_group); if (ng) { unsigned long shared = group_faults_shared(ng); unsigned long private = group_faults_priv(ng); @@ -1599,7 +1794,7 @@ pid_t task_numa_group_id(struct task_struct *p) pid_t gid = 0; rcu_read_lock(); - ng = rcu_dereference(p->numa_group); + ng = rcu_dereference_all(p->numa_group); if (ng) gid = ng->gid; rcu_read_unlock(); @@ -1926,17 +2121,18 @@ bool should_numa_migrate_memory(struct task_struct *p, struct folio *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; @@ -1944,8 +2140,7 @@ bool should_numa_migrate_memory(struct task_struct *p, struct folio *folio, if (latency >= th) return false; - return !numa_promotion_rate_limit(pgdat, rate_limit, - folio_nr_pages(folio)); + return !numa_promotion_rate_limit(pgdat, rate_limit, nr); } this_cpupid = cpu_pid_to_cpupid(dst_cpu, current->pid); @@ -2100,12 +2295,12 @@ 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 @@ -2258,8 +2453,9 @@ static bool task_numa_compare(struct task_numa_env *env, return false; rcu_read_lock(); - cur = rcu_dereference(dst_rq->curr); - if (cur && ((cur->flags & PF_EXITING) || is_idle_task(cur))) + cur = rcu_dereference_all(dst_rq->curr); + if (cur && ((cur->flags & (PF_EXITING | PF_KTHREAD)) || + !cur->mm)) cur = NULL; /* @@ -2302,7 +2498,7 @@ static bool task_numa_compare(struct task_numa_env *env, * If dst and source tasks are in the same NUMA group, or not * in any group then look only at task weights. */ - cur_ng = rcu_dereference(cur->numa_group); + cur_ng = rcu_dereference_all(cur->numa_group); if (cur_ng == p_ng) { /* * Do not swap within a group or between tasks that have @@ -2476,11 +2672,8 @@ static void task_numa_find_cpu(struct task_numa_env *env, maymove = !load_too_imbalanced(src_load, dst_load, env); } - for_each_cpu(cpu, cpumask_of_node(env->dst_nid)) { - /* Skip this CPU if the source task cannot migrate */ - if (!cpumask_test_cpu(cpu, env->p->cpus_ptr)) - continue; - + /* Skip CPUs if the source task cannot migrate */ + for_each_cpu_and(cpu, cpumask_of_node(env->dst_nid), env->p->cpus_ptr) { env->dst_cpu = cpu; if (task_numa_compare(env, taskimp, groupimp, maymove)) break; @@ -2517,7 +2710,7 @@ static int task_numa_migrate(struct task_struct *p) * to satisfy here. */ rcu_read_lock(); - sd = rcu_dereference(per_cpu(sd_numa, env.src_cpu)); + sd = rcu_dereference_all(per_cpu(sd_numa, env.src_cpu)); if (sd) { env.imbalance_pct = 100 + (sd->imbalance_pct - 100) / 2; env.imb_numa_nr = sd->imb_numa_nr; @@ -2878,6 +3071,7 @@ static int preferred_group_nid(struct task_struct *p, int nid) } static void task_numa_placement(struct task_struct *p) + __context_unsafe(/* conditional locking */) { int seq, nid, max_nid = NUMA_NO_NODE; unsigned long max_faults = 0; @@ -3040,7 +3234,7 @@ static void task_numa_group(struct task_struct *p, int cpupid, int flags, if (!cpupid_match_pid(tsk, cpupid)) goto no_join; - grp = rcu_dereference(tsk->numa_group); + grp = rcu_dereference_all(tsk->numa_group); if (!grp) goto no_join; @@ -3301,7 +3495,7 @@ static void task_numa_work(struct callback_head *work) 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; /* @@ -3315,6 +3509,15 @@ 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); @@ -3402,7 +3605,7 @@ retry_pids: if (!vma->numab_state) { struct vma_numab_state *ptr; - ptr = kzalloc(sizeof(*ptr), GFP_KERNEL); + ptr = kzalloc_obj(*ptr); if (!ptr) continue; @@ -3535,7 +3738,7 @@ 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; @@ -3649,7 +3852,8 @@ 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) { } @@ -3666,20 +3870,18 @@ 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_queued++; } @@ -3687,12 +3889,10 @@ 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_queued--; } @@ -3705,7 +3905,7 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) */ #define add_positive(_ptr, _val) do { \ typeof(_ptr) ptr = (_ptr); \ - typeof(_val) val = (_val); \ + __signed_scalar_typeof(*ptr) val = (_val); \ typeof(*ptr) res, var = READ_ONCE(*ptr); \ \ res = var + val; \ @@ -3717,23 +3917,6 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) } while (0) /* - * Unsigned subtract and clamp on underflow. - * - * Explicitly do a load-store to ensure the intermediate value never hits - * memory. This allows lockless observations without ever seeing the negative - * values. - */ -#define sub_positive(_ptr, _val) do { \ - typeof(_ptr) ptr = (_ptr); \ - typeof(*ptr) val = (_val); \ - typeof(*ptr) res, var = READ_ONCE(*ptr); \ - res = var - val; \ - if (res > var) \ - res = 0; \ - WRITE_ONCE(*ptr, res); \ -} while (0) - -/* * Remove and clamp on negative, from a local variable. * * A variant of sub_positive(), which does not use explicit load-store @@ -3744,82 +3927,191 @@ account_entity_dequeue(struct cfs_rq *cfs_rq, struct sched_entity *se) *ptr -= min_t(typeof(*ptr), *ptr, _val); \ } while (0) -#ifdef CONFIG_SMP + +/* + * Because of rounding, se->util_sum might ends up being +1 more than + * cfs->util_sum. Although this is not a problem by itself, detaching + * a lot of tasks with the rounding problem between 2 updates of + * util_avg (~1ms) can make cfs->util_sum becoming null whereas + * cfs_util_avg is not. + * + * Check that util_sum is still above its lower bound for the new + * util_avg. Given that period_contrib might have moved since the last + * sync, we are only sure that util_sum must be above or equal to + * util_avg * minimum possible divider + */ +#define __update_sa(sa, name, delta_avg, delta_sum) do { \ + add_positive(&(sa)->name##_avg, delta_avg); \ + add_positive(&(sa)->name##_sum, delta_sum); \ + (sa)->name##_sum = max_t(typeof((sa)->name##_sum), \ + (sa)->name##_sum, \ + (sa)->name##_avg * PELT_MIN_DIVIDER); \ +} while (0) + static inline void enqueue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { - cfs_rq->avg.load_avg += se->avg.load_avg; - cfs_rq->avg.load_sum += se_weight(se) * se->avg.load_sum; + __update_sa(&cfs_rq->avg, load, se->avg.load_avg, + se_weight(se) * se->avg.load_sum); } static inline void dequeue_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { - sub_positive(&cfs_rq->avg.load_avg, se->avg.load_avg); - sub_positive(&cfs_rq->avg.load_sum, se_weight(se) * se->avg.load_sum); - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum, - cfs_rq->avg.load_avg * PELT_MIN_DIVIDER); + __update_sa(&cfs_rq->avg, load, -se->avg.load_avg, + se_weight(se) * -se->avg.load_sum); } -#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 +rescale_entity(struct sched_entity *se, unsigned long weight, bool rel_vprot) +{ + unsigned long old_weight = se->load.weight; + + /* + * VRUNTIME + * -------- + * + * COROLLARY #1: The virtual runtime of the entity needs to be + * adjusted if re-weight at !0-lag point. + * + * Proof: For contradiction assume this is not true, so we can + * re-weight without changing vruntime at !0-lag point. + * + * Weight VRuntime Avg-VRuntime + * before w v V + * after w' v' V' + * + * Since lag needs to be preserved through re-weight: + * + * lag = (V - v)*w = (V'- v')*w', where v = v' + * ==> V' = (V - v)*w/w' + v (1) + * + * Let W be the total weight of the entities before reweight, + * since V' is the new weighted average of entities: + * + * V' = (WV + w'v - wv) / (W + w' - w) (2) + * + * by using (1) & (2) we obtain: + * + * (WV + w'v - wv) / (W + w' - w) = (V - v)*w/w' + v + * ==> (WV-Wv+Wv+w'v-wv)/(W+w'-w) = (V - v)*w/w' + v + * ==> (WV - Wv)/(W + w' - w) + v = (V - v)*w/w' + v + * ==> (V - v)*W/(W + w' - w) = (V - v)*w/w' (3) + * + * Since we are doing at !0-lag point which means V != v, we + * can simplify (3): + * + * ==> W / (W + w' - w) = w / w' + * ==> Ww' = Ww + ww' - ww + * ==> W * (w' - w) = w * (w' - w) + * ==> W = w (re-weight indicates w' != w) + * + * So the cfs_rq contains only one entity, hence vruntime of + * the entity @v should always equal to the cfs_rq's weighted + * average vruntime @V, which means we will always re-weight + * at 0-lag point, thus breach assumption. Proof completed. + * + * + * COROLLARY #2: Re-weight does NOT affect weighted average + * vruntime of all the entities. + * + * Proof: According to corollary #1, Eq. (1) should be: + * + * (V - v)*w = (V' - v')*w' + * ==> v' = V' - (V - v)*w/w' (4) + * + * According to the weighted average formula, we have: + * + * V' = (WV - wv + w'v') / (W - w + w') + * = (WV - wv + w'(V' - (V - v)w/w')) / (W - w + w') + * = (WV - wv + w'V' - Vw + wv) / (W - w + w') + * = (WV + w'V' - Vw) / (W - w + w') + * + * ==> V'*(W - w + w') = WV + w'V' - Vw + * ==> V' * (W - w) = (W - w) * V (5) + * + * If the entity is the only one in the cfs_rq, then reweight + * always occurs at 0-lag point, so V won't change. Or else + * there are other entities, hence W != w, then Eq. (5) turns + * into V' = V. So V won't change in either case, proof done. + * + * + * So according to corollary #1 & #2, the effect of re-weight + * on vruntime should be: + * + * v' = V' - (V - v) * w / w' (4) + * = V - (V - v) * w / w' + * = V - vl * w / w' + * = V - vl' + */ + se->vlag = div64_long(se->vlag * old_weight, weight); + + /* + * DEADLINE + * -------- + * + * When the weight changes, the virtual time slope changes and + * we should adjust the relative virtual deadline accordingly. + * + * d' = v' + (d - v)*w/w' + * = V' - (V - v)*w/w' + (d - v)*w/w' + * = V - (V - v)*w/w' + (d - v)*w/w' + * = V + (d - V)*w/w' + */ + if (se->rel_deadline) + se->deadline = div64_long(se->deadline * old_weight, weight); + + if (rel_vprot) + se->vprot = div64_long(se->vprot * old_weight, weight); +} static void reweight_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, unsigned long weight) { bool curr = cfs_rq->curr == se; + bool rel_vprot = false; + u64 avruntime = 0; if (se->on_rq) { /* commit outstanding execution time */ update_curr(cfs_rq); - update_entity_lag(cfs_rq, se); - se->deadline -= se->vruntime; + avruntime = avg_vruntime(cfs_rq); + se->vlag = entity_lag(cfs_rq, se, avruntime); + se->deadline -= avruntime; se->rel_deadline = 1; + if (curr && protect_slice(se)) { + se->vprot -= avruntime; + rel_vprot = true; + } + + 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); + rescale_entity(se, weight, rel_vprot); 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 (rel_vprot) + se->vprot += avruntime; + se->deadline += avruntime; + se->rel_deadline = 0; + se->vruntime = avruntime - se->vlag; + update_load_add(&cfs_rq->load, se->load.weight); - place_entity(cfs_rq, se, 0); if (!curr) __enqueue_entity(cfs_rq, se); - - /* - * The entity's vruntime has been adjusted, so let's check - * whether the rq-wide min_vruntime needs updated too. Since - * the calculations above require stable min_vruntime rather - * than up-to-date one, we do the update at the end of the - * reweight process. - */ - update_min_vruntime(cfs_rq); + cfs_rq->nr_queued++; } } @@ -3837,7 +4129,6 @@ static void reweight_task_fair(struct rq *rq, struct task_struct *p, 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. @@ -3944,7 +4235,6 @@ 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 @@ -3962,23 +4252,16 @@ static void update_cfs_group(struct sched_entity *se) if (!gcfs_rq || !gcfs_rq->load.weight) return; - if (throttled_hierarchy(gcfs_rq)) - return; - -#ifndef CONFIG_SMP - shares = READ_ONCE(gcfs_rq->tg->shares); -#else shares = calc_group_shares(gcfs_rq); -#endif 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) { @@ -4003,7 +4286,6 @@ 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) @@ -4020,7 +4302,7 @@ 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); @@ -4072,6 +4354,9 @@ 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; } @@ -4277,7 +4562,6 @@ update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq */ divider = get_pelt_divider(&cfs_rq->avg); - /* Set new sched_entity's utilization */ se->avg.util_avg = gcfs_rq->avg.util_avg; new_sum = se->avg.util_avg * divider; @@ -4285,12 +4569,7 @@ update_tg_cfs_util(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq se->avg.util_sum = new_sum; /* Update parent cfs_rq utilization */ - add_positive(&cfs_rq->avg.util_avg, delta_avg); - add_positive(&cfs_rq->avg.util_sum, delta_sum); - - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum, - cfs_rq->avg.util_avg * PELT_MIN_DIVIDER); + __update_sa(&cfs_rq->avg, util, delta_avg, delta_sum); } static inline void @@ -4316,11 +4595,7 @@ update_tg_cfs_runnable(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cf se->avg.runnable_sum = new_sum; /* Update parent cfs_rq runnable */ - add_positive(&cfs_rq->avg.runnable_avg, delta_avg); - add_positive(&cfs_rq->avg.runnable_sum, delta_sum); - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum, - cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER); + __update_sa(&cfs_rq->avg, runnable, delta_avg, delta_sum); } static inline void @@ -4384,11 +4659,7 @@ update_tg_cfs_load(struct cfs_rq *cfs_rq, struct sched_entity *se, struct cfs_rq se->avg.load_sum = runnable_sum; se->avg.load_avg = load_avg; - add_positive(&cfs_rq->avg.load_avg, delta_avg); - add_positive(&cfs_rq->avg.load_sum, delta_sum); - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.load_sum = max_t(u32, cfs_rq->avg.load_sum, - cfs_rq->avg.load_avg * PELT_MIN_DIVIDER); + __update_sa(&cfs_rq->avg, load, delta_avg, delta_sum); } static inline void add_tg_cfs_propagate(struct cfs_rq *cfs_rq, long runnable_sum) @@ -4455,7 +4726,7 @@ 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) {} @@ -4468,7 +4739,7 @@ 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) @@ -4485,7 +4756,7 @@ static inline void migrate_se_pelt_lag(struct sched_entity *se) rq = rq_of(cfs_rq); rcu_read_lock(); - is_idle = is_idle_task(rcu_dereference(rq->curr)); + is_idle = is_idle_task(rcu_dereference_all(rq->curr)); rcu_read_unlock(); /* @@ -4549,9 +4820,9 @@ 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 @@ -4587,33 +4858,13 @@ update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq) raw_spin_unlock(&cfs_rq->removed.lock); r = removed_load; - sub_positive(&sa->load_avg, r); - sub_positive(&sa->load_sum, r * divider); - /* See sa->util_sum below */ - sa->load_sum = max_t(u32, sa->load_sum, sa->load_avg * PELT_MIN_DIVIDER); + __update_sa(sa, load, -r, -r*divider); r = removed_util; - sub_positive(&sa->util_avg, r); - sub_positive(&sa->util_sum, r * divider); - /* - * Because of rounding, se->util_sum might ends up being +1 more than - * cfs->util_sum. Although this is not a problem by itself, detaching - * a lot of tasks with the rounding problem between 2 updates of - * util_avg (~1ms) can make cfs->util_sum becoming null whereas - * cfs_util_avg is not. - * Check that util_sum is still above its lower bound for the new - * util_avg. Given that period_contrib might have moved since the last - * sync, we are only sure that util_sum must be above or equal to - * util_avg * minimum possible divider - */ - sa->util_sum = max_t(u32, sa->util_sum, sa->util_avg * PELT_MIN_DIVIDER); + __update_sa(sa, util, -r, -r*divider); r = removed_runnable; - sub_positive(&sa->runnable_avg, r); - sub_positive(&sa->runnable_sum, r * divider); - /* See sa->util_sum above */ - sa->runnable_sum = max_t(u32, sa->runnable_sum, - sa->runnable_avg * PELT_MIN_DIVIDER); + __update_sa(sa, runnable, -r, -r*divider); /* * removed_runnable is the unweighted version of removed_load so we @@ -4698,17 +4949,8 @@ static void attach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *s static void detach_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se) { dequeue_load_avg(cfs_rq, se); - sub_positive(&cfs_rq->avg.util_avg, se->avg.util_avg); - sub_positive(&cfs_rq->avg.util_sum, se->avg.util_sum); - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.util_sum = max_t(u32, cfs_rq->avg.util_sum, - cfs_rq->avg.util_avg * PELT_MIN_DIVIDER); - - sub_positive(&cfs_rq->avg.runnable_avg, se->avg.runnable_avg); - sub_positive(&cfs_rq->avg.runnable_sum, se->avg.runnable_sum); - /* See update_cfs_rq_load_avg() */ - cfs_rq->avg.runnable_sum = max_t(u32, cfs_rq->avg.runnable_sum, - cfs_rq->avg.runnable_avg * PELT_MIN_DIVIDER); + __update_sa(&cfs_rq->avg, util, -se->avg.util_avg, -se->avg.util_sum); + __update_sa(&cfs_rq->avg, runnable, -se->avg.runnable_avg, -se->avg.runnable_sum); add_tg_cfs_propagate(cfs_rq, -se->avg.load_sum); @@ -4816,7 +5058,8 @@ static inline unsigned long cfs_rq_load_avg(struct cfs_rq *cfs_rq) return cfs_rq->avg.load_avg; } -static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf); +static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) + __must_hold(__rq_lockp(this_rq)); static inline unsigned long task_util(struct task_struct *p) { @@ -4919,13 +5162,6 @@ static inline void util_est_update(struct cfs_rq *cfs_rq, goto done; /* - * To avoid overestimation of actual task utilization, skip updates if - * we cannot grant there is idle time in this CPU. - */ - if (dequeued > arch_scale_cpu_capacity(cpu_of(rq_of(cfs_rq)))) - return; - - /* * To avoid underestimate of task utilization, skip updates of EWMA if * we cannot grant that thread got all CPU time it wanted. */ @@ -5125,48 +5361,6 @@ 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) -{ - return !cfs_rq->nr_queued; -} - -#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 sched_balance_newidle(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 */ - void __setparam_fair(struct task_struct *p, const struct sched_attr *attr) { struct sched_entity *se = &p->se; @@ -5187,6 +5381,7 @@ static void place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { u64 vslice, vruntime = avg_vruntime(cfs_rq); + bool update_zero = false; s64 lag = 0; if (!se->custom_slice) @@ -5203,7 +5398,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) */ if (sched_feat(PLACE_LAG) && cfs_rq->nr_queued && se->vlag) { struct sched_entity *curr = cfs_rq->curr; - unsigned long load; + long load, weight; lag = se->vlag; @@ -5234,7 +5429,7 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * 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*l) / (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: @@ -5259,19 +5454,46 @@ place_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) * * vl_i = (W + w_i)*vl'_i / W */ - load = cfs_rq->avg_load; + load = cfs_rq->sum_weight; if (curr && curr->on_rq) - load += scale_load_down(curr->load.weight); + load += avg_vruntime_weight(cfs_rq, curr->load.weight); - lag *= load + scale_load_down(se->load.weight); + weight = avg_vruntime_weight(cfs_rq, se->load.weight); + lag *= load + weight; if (WARN_ON_ONCE(!load)) load = 1; - lag = div_s64(lag, load); + lag = div64_long(lag, load); + + /* + * A heavy entity (relative to the tree) will pull the + * avg_vruntime close to its vruntime position on enqueue. But + * the zero_vruntime point is only updated at the next + * update_deadline()/place_entity()/update_entity_lag(). + * + * Specifically (see the comment near avg_vruntime_weight()): + * + * sum_w_vruntime = \Sum (v_i - v0) * w_i + * + * Note that if v0 is near a light entity, both terms will be + * small for the light entity, while in that case both terms + * are large for the heavy entity, leading to risk of + * overflow. + * + * OTOH if v0 is near the heavy entity, then the difference is + * larger for the light entity, but the factor is small, while + * for the heavy entity the difference is small but the factor + * is large. Avoiding the multiplication overflow. + */ + if (weight > load) + update_zero = true; } se->vruntime = vruntime - lag; - if (se->rel_deadline) { + if (update_zero) + update_zero_vruntime(cfs_rq, -lag); + + if (sched_feat(PLACE_REL_DEADLINE) && se->rel_deadline) { se->deadline += se->vruntime; se->rel_deadline = 0; return; @@ -5350,18 +5572,16 @@ enqueue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) if (cfs_rq->nr_queued == 1) { check_enqueue_throttle(cfs_rq); - if (!throttled_hierarchy(cfs_rq)) { - list_add_leaf_cfs_rq(cfs_rq); - } else { + list_add_leaf_cfs_rq(cfs_rq); #ifdef CONFIG_CFS_BANDWIDTH + if (cfs_rq->pelt_clock_throttled) { 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); -#endif + cfs_rq->throttled_clock_pelt_time += rq_clock_pelt(rq) - + cfs_rq->throttled_clock_pelt; + cfs_rq->pelt_clock_throttled = 0; } +#endif } } @@ -5400,8 +5620,6 @@ static void set_delayed(struct sched_entity *se) struct cfs_rq *cfs_rq = cfs_rq_of(se); cfs_rq->h_nr_runnable--; - if (cfs_rq_throttled(cfs_rq)) - break; } } @@ -5422,18 +5640,9 @@ static void clear_delayed(struct sched_entity *se) struct cfs_rq *cfs_rq = cfs_rq_of(se); cfs_rq->h_nr_runnable++; - if (cfs_rq_throttled(cfs_rq)) - break; } } -static inline void finish_delayed_dequeue_entity(struct sched_entity *se) -{ - clear_delayed(se); - if (sched_feat(DELAY_ZERO) && se->vlag > 0) - se->vlag = 0; -} - static bool dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) { @@ -5444,21 +5653,22 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) clear_buddies(cfs_rq, se); if (flags & DEQUEUE_DELAYED) { - SCHED_WARN_ON(!se->sched_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) + if (flags & (DEQUEUE_SPECIAL | DEQUEUE_THROTTLE)) delay = false; - SCHED_WARN_ON(delay && se->sched_delayed); + 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); + update_entity_lag(cfs_rq, se); set_delayed(se); return false; } @@ -5497,26 +5707,27 @@ dequeue_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, int flags) 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 -- i.e. 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); + clear_delayed(se); - if (cfs_rq->nr_queued == 0) + if (cfs_rq->nr_queued == 0) { update_idle_cfs_rq_clock_pelt(cfs_rq); +#ifdef CONFIG_CFS_BANDWIDTH + if (throttled_hierarchy(cfs_rq)) { + struct rq *rq = rq_of(cfs_rq); + + list_del_leaf_cfs_rq(cfs_rq); + cfs_rq->throttled_clock_pelt = rq_clock_pelt(rq); + cfs_rq->pelt_clock_throttled = 1; + } +#endif + } return true; } static void -set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se) +set_next_entity(struct cfs_rq *cfs_rq, struct sched_entity *se, bool first) { clear_buddies(cfs_rq, se); @@ -5530,15 +5741,13 @@ 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); - /* - * HACK, stash a copy of deadline at the point of pick in vlag, - * which isn't used until dequeue. - */ - se->vlag = se->deadline; + + if (first) + set_protect_slice(cfs_rq, se); } update_stats_curr_start(cfs_rq, se); - SCHED_WARN_ON(cfs_rq->curr); + WARN_ON_ONCE(cfs_rq->curr); cfs_rq->curr = se; /* @@ -5569,21 +5778,11 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags); * 4) do not run the "skip" process, if something else is available */ static struct sched_entity * -pick_next_entity(struct rq *rq, struct cfs_rq *cfs_rq) +pick_next_entity(struct rq *rq, struct cfs_rq *cfs_rq, bool protect) { struct sched_entity *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 */ - SCHED_WARN_ON(cfs_rq->next->sched_delayed); - return cfs_rq->next; - } - - se = pick_eevdf(cfs_rq); + se = pick_eevdf(cfs_rq, protect); if (se->sched_delayed) { dequeue_entities(rq, se, DEQUEUE_SLEEP | DEQUEUE_DELAYED); /* @@ -5615,7 +5814,7 @@ 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); } - SCHED_WARN_ON(cfs_rq->curr != prev); + WARN_ON_ONCE(cfs_rq->curr != prev); cfs_rq->curr = NULL; } @@ -5639,7 +5838,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued) * validating it and just reschedule. */ if (queued) { - resched_curr_lazy(rq_of(cfs_rq)); + resched_curr(rq_of(cfs_rq)); return; } #endif @@ -5669,7 +5868,7 @@ 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; @@ -5677,16 +5876,7 @@ 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) { @@ -5796,74 +5986,253 @@ 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; - cfs_rq->throttle_count--; - if (!cfs_rq->throttle_count) { + if (--cfs_rq->throttle_count) + return 0; + + if (cfs_rq->pelt_clock_throttled) { cfs_rq->throttled_clock_pelt_time += rq_clock_pelt(rq) - cfs_rq->throttled_clock_pelt; + cfs_rq->pelt_clock_throttled = 0; + } - /* 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; - if (cfs_rq->throttled_clock_self) { - u64 delta = rq_clock(rq) - cfs_rq->throttled_clock_self; + cfs_rq->throttled_clock_self = 0; - cfs_rq->throttled_clock_self = 0; + if (WARN_ON_ONCE((s64)delta < 0)) + delta = 0; - if (SCHED_WARN_ON((s64)delta < 0)) - delta = 0; + cfs_rq->throttled_clock_self_time += delta; + } - 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); - list_del_leaf_cfs_rq(cfs_rq); + if (cfs_rq->throttle_count++) + return 0; - SCHED_WARN_ON(cfs_rq->throttled_clock_self); - if (cfs_rq->nr_queued) - cfs_rq->throttled_clock_self = rq_clock(rq); + /* + * 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; } @@ -5871,9 +6240,7 @@ 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 queued_delta, runnable_delta, idle_delta, dequeue = 1; - long rq_h_nr_queued = rq->cfs.h_nr_queued; + int dequeue = 1; raw_spin_lock(&cfs_b->lock); /* This will start the period timer if necessary */ @@ -5896,80 +6263,17 @@ 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(); - queued_delta = cfs_rq->h_nr_queued; - runnable_delta = cfs_rq->h_nr_runnable; - idle_delta = cfs_rq->h_nr_idle; - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); - int flags; - - /* throttled entity or throttle-on-deactivate */ - if (!se->on_rq) - goto done; - - /* - * Abuse SPECIAL to avoid delayed dequeue in this instance. - * This avoids teaching dequeue_entities() about throttled - * entities and keeps things relatively simple. - */ - flags = DEQUEUE_SLEEP | DEQUEUE_SPECIAL; - if (se->sched_delayed) - flags |= DEQUEUE_DELAYED; - dequeue_entity(qcfs_rq, se, flags); - - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_delta = cfs_rq->h_nr_queued; - - qcfs_rq->h_nr_queued -= queued_delta; - qcfs_rq->h_nr_runnable -= runnable_delta; - qcfs_rq->h_nr_idle -= idle_delta; - - if (qcfs_rq->load.weight) { - /* Avoid re-evaluating load for this entity: */ - 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_delta = cfs_rq->h_nr_queued; - - qcfs_rq->h_nr_queued -= queued_delta; - qcfs_rq->h_nr_runnable -= runnable_delta; - qcfs_rq->h_nr_idle -= idle_delta; - } - - /* At this point se is NULL and we are at root level*/ - sub_nr_running(rq, queued_delta); - - /* Stop the fair server if throttling resulted in no runnable tasks */ - if (rq_h_nr_queued && !rq->cfs.h_nr_queued) - dl_server_stop(&rq->fair_server); -done: /* * Note: distribution will already see us throttled via the * throttled-list. rq->lock protects completion. */ cfs_rq->throttled = 1; - SCHED_WARN_ON(cfs_rq->throttled_clock); - if (cfs_rq->nr_queued) - cfs_rq->throttled_clock = rq_clock(rq); + WARN_ON_ONCE(cfs_rq->throttled_clock); return true; } @@ -5977,11 +6281,19 @@ 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 queued_delta, runnable_delta, idle_delta; - long rq_h_nr_queued = rq->cfs.h_nr_queued; + 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; @@ -6009,62 +6321,8 @@ void unthrottle_cfs_rq(struct cfs_rq *cfs_rq) if (list_add_leaf_cfs_rq(cfs_rq_of(se))) break; } - goto unthrottle_throttle; - } - - queued_delta = cfs_rq->h_nr_queued; - runnable_delta = cfs_rq->h_nr_runnable; - idle_delta = cfs_rq->h_nr_idle; - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); - - /* Handle any unfinished DELAY_DEQUEUE business first. */ - if (se->sched_delayed) { - int flags = DEQUEUE_SLEEP | DEQUEUE_DELAYED; - - dequeue_entity(qcfs_rq, se, flags); - } else if (se->on_rq) - break; - enqueue_entity(qcfs_rq, se, ENQUEUE_WAKEUP); - - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_delta = cfs_rq->h_nr_queued; - - qcfs_rq->h_nr_queued += queued_delta; - qcfs_rq->h_nr_runnable += runnable_delta; - qcfs_rq->h_nr_idle += idle_delta; - - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(qcfs_rq)) - goto unthrottle_throttle; } - for_each_sched_entity(se) { - struct cfs_rq *qcfs_rq = cfs_rq_of(se); - - update_load_avg(qcfs_rq, se, UPDATE_TG); - se_update_runnable(se); - - if (cfs_rq_is_idle(group_cfs_rq(se))) - idle_delta = cfs_rq->h_nr_queued; - - qcfs_rq->h_nr_queued += queued_delta; - qcfs_rq->h_nr_runnable += runnable_delta; - qcfs_rq->h_nr_idle += idle_delta; - - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(qcfs_rq)) - goto unthrottle_throttle; - } - - /* Start the fair server if un-throttling resulted in new runnable tasks */ - 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, queued_delta); - -unthrottle_throttle: assert_list_leaf_cfs_rq(rq); /* Determine whether we need to wake up potentially idle CPU: */ @@ -6072,7 +6330,6 @@ unthrottle_throttle: resched_curr(rq); } -#ifdef CONFIG_SMP static void __cfsb_csd_unthrottle(void *arg) { struct cfs_rq *cursor, *tmp; @@ -6123,7 +6380,7 @@ static inline void __unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) } /* Already enqueued */ - if (SCHED_WARN_ON(!list_empty(&cfs_rq->throttled_csd_list))) + if (WARN_ON_ONCE(!list_empty(&cfs_rq->throttled_csd_list))) return; first = list_empty(&rq->cfsb_csd_list); @@ -6131,18 +6388,12 @@ static inline void __unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) if (first) smp_call_function_single_async(cpu_of(rq), &rq->cfsb_csd); } -#else -static inline void __unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) -{ - unthrottle_cfs_rq(cfs_rq); -} -#endif static void unthrottle_cfs_rq_async(struct cfs_rq *cfs_rq) { lockdep_assert_rq_held(rq_of(cfs_rq)); - if (SCHED_WARN_ON(!cfs_rq_throttled(cfs_rq) || + if (WARN_ON_ONCE(!cfs_rq_throttled(cfs_rq) || cfs_rq->runtime_remaining <= 0)) return; @@ -6178,7 +6429,7 @@ static bool distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) goto next; /* By the above checks, this should never be true */ - SCHED_WARN_ON(cfs_rq->runtime_remaining > 0); + WARN_ON_ONCE(cfs_rq->runtime_remaining > 0); raw_spin_lock(&cfs_b->lock); runtime = -cfs_rq->runtime_remaining + 1; @@ -6199,7 +6450,7 @@ static bool distribute_cfs_runtime(struct cfs_bandwidth *cfs_b) * We currently only expect to be unthrottling * a single cfs_rq locally. */ - SCHED_WARN_ON(!list_empty(&local_unthrottle)); + WARN_ON_ONCE(!list_empty(&local_unthrottle)); list_add_tail(&cfs_rq->throttled_csd_list, &local_unthrottle); } @@ -6224,7 +6475,7 @@ next: rq_unlock_irqrestore(rq, &rf); } - SCHED_WARN_ON(!list_empty(&local_unthrottle)); + WARN_ON_ONCE(!list_empty(&local_unthrottle)); rcu_read_unlock(); @@ -6238,6 +6489,7 @@ next: * used to track this state. */ static int do_sched_cfs_period_timer(struct cfs_bandwidth *cfs_b, int overrun, unsigned long flags) + __must_hold(&cfs_b->lock) { int throttled; @@ -6443,6 +6695,16 @@ 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() */ @@ -6474,8 +6736,6 @@ 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 = @@ -6502,7 +6762,7 @@ 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; @@ -6536,19 +6796,19 @@ void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b, struct cfs_bandwidth *paren 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_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_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - cfs_b->slack_timer.function = sched_cfs_slack_timer; + hrtimer_setup(&cfs_b->slack_timer, sched_cfs_slack_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL); cfs_b->slack_started = false; } @@ -6557,6 +6817,7 @@ 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) @@ -6592,7 +6853,6 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) * guaranteed at this point that no additional cfs_rq of this group can * join a CSD list. */ -#ifdef CONFIG_SMP for_each_possible_cpu(i) { struct rq *rq = cpu_rq(i); unsigned long flags; @@ -6604,7 +6864,6 @@ static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) __cfsb_csd_unthrottle(rq); local_irq_restore(flags); } -#endif } /* @@ -6717,28 +6976,37 @@ static void sched_fair_update_stop_tick(struct rq *rq, struct task_struct *p) if (cfs_task_bw_constrained(p)) tick_nohz_dep_set_cpu(cpu, TICK_DEP_BIT_SCHED); } -#endif +#endif /* CONFIG_NO_HZ_FULL */ -#else /* CONFIG_CFS_BANDWIDTH */ +#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; } @@ -6761,7 +7029,7 @@ 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) {} @@ -6775,27 +7043,41 @@ static inline void sched_fair_update_stop_tick(struct rq *rq, struct task_struct static void hrtick_start_fair(struct rq *rq, struct task_struct *p) { struct sched_entity *se = &p->se; + unsigned long scale = 1024; + unsigned long util = 0; + u64 vdelta; + u64 delta; - SCHED_WARN_ON(task_rq(p) != rq); + WARN_ON_ONCE(task_rq(p) != rq); - 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 (rq->cfs.h_nr_queued <= 1) + return; - if (delta < 0) { - if (task_current_donor(rq, p)) - resched_curr(rq); - return; - } - hrtick_start(rq, delta); + /* + * Compute time until virtual deadline + */ + vdelta = se->deadline - se->vruntime; + if ((s64)vdelta < 0) { + if (task_current_donor(rq, p)) + resched_curr(rq); + return; } + delta = (se->load.weight * vdelta) / NICE_0_LOAD; + + /* + * Correct for instantaneous load of other classes. + */ + util += cpu_util_irq(rq); + if (util && util < 1024) { + scale *= 1024; + scale /= (1024 - util); + } + + hrtick_start(rq, (scale * delta) / 1024); } /* - * called from enqueue/dequeue and updates the hrtick when the - * current task is from our class and nr_running is low enough - * to matter. + * Called on enqueue to start the hrtick when h_nr_queued becomes more than 1. */ static void hrtick_update(struct rq *rq) { @@ -6804,9 +7086,12 @@ static void hrtick_update(struct rq *rq) if (!hrtick_enabled_fair(rq) || donor->sched_class != &fair_sched_class) return; + if (hrtick_active(rq)) + return; + 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) { @@ -6815,21 +7100,19 @@ 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, rq_util_max; + unsigned long 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); + return !util_fits_cpu(cpu_util_cfs(cpu), 0, rq_util_max, cpu); } /* @@ -6859,9 +7142,6 @@ static inline void check_update_overutilized_status(struct rq *rq) if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu)) set_rd_overutilized(rq->rd, 1); } -#else -static inline void check_update_overutilized_status(struct rq *rq) { } -#endif /* Runqueue only has SCHED_IDLE tasks enqueued */ static int sched_idle_rq(struct rq *rq) @@ -6870,12 +7150,16 @@ static int sched_idle_rq(struct rq *rq) rq->nr_running); } -#ifdef CONFIG_SMP -static int sched_idle_cpu(int cpu) +static int choose_sched_idle_rq(struct rq *rq, struct task_struct *p) { - return sched_idle_rq(cpu_rq(cpu)); + return sched_idle_rq(rq) && !task_has_idle_policy(p); +} + +static int choose_idle_cpu(int cpu, struct task_struct *p) +{ + return available_idle_cpu(cpu) || + choose_sched_idle_rq(cpu_rq(cpu), p); } -#endif static void requeue_delayed_entity(struct sched_entity *se) @@ -6887,21 +7171,17 @@ requeue_delayed_entity(struct sched_entity *se) * Because a delayed entity is one that is still on * the runqueue competing until elegibility. */ - SCHED_WARN_ON(!se->sched_delayed); - SCHED_WARN_ON(!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++; - } + WARN_ON_ONCE(!se->sched_delayed); + WARN_ON_ONCE(!se->on_rq); + + if (update_entity_lag(cfs_rq, se)) { + cfs_rq->nr_queued--; + if (se != cfs_rq->curr) + __dequeue_entity(cfs_rq, se); + 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); @@ -6924,13 +7204,16 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) 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 * the cfs_rq utilization to select a frequency. * Let's add the task's estimated utilization to the cfs_rq's * estimated utilization, before we update schedutil. */ - if (!(p->se.sched_delayed && (task_on_rq_migrating(p) || (flags & ENQUEUE_RESTORE)))) + if (!p->se.sched_delayed || (flags & ENQUEUE_DELAYED)) util_est_enqueue(&rq->cfs, p); if (flags & ENQUEUE_DELAYED) { @@ -6976,10 +7259,6 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) if (cfs_rq_is_idle(cfs_rq)) h_nr_idle = 1; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto enqueue_throttle; - flags = ENQUEUE_WAKEUP; } @@ -6991,6 +7270,8 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) update_cfs_group(se); se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); slice = cfs_rq_min_slice(cfs_rq); cfs_rq->h_nr_runnable += h_nr_runnable; @@ -6999,18 +7280,10 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) if (cfs_rq_is_idle(cfs_rq)) h_nr_idle = 1; - - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - goto enqueue_throttle; } - if (!rq_h_nr_queued && rq->cfs.h_nr_queued) { - /* Account for idle runtime */ - if (!rq->nr_running) - dl_server_update_idle_time(rq, rq->curr); + 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); @@ -7032,14 +7305,11 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags) if (!task_new) 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); - /* * Basically dequeue_task_fair(), except it can deal with dequeue_entity() * failing half-way through and resume the dequeue later. @@ -7052,9 +7322,9 @@ static void set_next_buddy(struct sched_entity *se); static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) { bool was_sched_idle = sched_idle_rq(rq); - int rq_h_nr_queued = rq->cfs.h_nr_queued; 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; @@ -7068,9 +7338,6 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) h_nr_idle = task_has_idle_policy(p); if (task_sleep || task_delayed || !se->sched_delayed) h_nr_runnable = 1; - } else { - cfs_rq = group_cfs_rq(se); - slice = cfs_rq_min_slice(cfs_rq); } for_each_sched_entity(se) { @@ -7080,6 +7347,7 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) if (p && &p->se == se) return -1; + slice = cfs_rq_min_slice(cfs_rq); break; } @@ -7090,9 +7358,8 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) if (cfs_rq_is_idle(cfs_rq)) h_nr_idle = h_nr_queued; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - return 0; + 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) { @@ -7104,7 +7371,7 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) * 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; } @@ -7120,6 +7387,8 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) update_cfs_group(se); se->slice = slice; + if (se != cfs_rq->curr) + min_vruntime_cb_propagate(&se->run_node, NULL); slice = cfs_rq_min_slice(cfs_rq); cfs_rq->h_nr_runnable -= h_nr_runnable; @@ -7129,26 +7398,19 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) if (cfs_rq_is_idle(cfs_rq)) h_nr_idle = h_nr_queued; - /* end evaluation on encountering a throttled cfs_rq */ - if (cfs_rq_throttled(cfs_rq)) - return 0; + if (throttled_hierarchy(cfs_rq) && task_throttled) + record_throttle_clock(cfs_rq); } sub_nr_running(rq, h_nr_queued); - if (rq_h_nr_queued && !rq->cfs.h_nr_queued) - dl_server_stop(&rq->fair_server); - /* balance early to pull high priority tasks */ if (unlikely(!was_sched_idle && sched_idle_rq(rq))) rq->next_balance = jiffies; if (p && task_delayed) { - SCHED_WARN_ON(!task_sleep); - SCHED_WARN_ON(p->on_rq != 1); - - /* Fix-up what dequeue_task_fair() skipped */ - hrtick_update(rq); + WARN_ON_ONCE(!task_sleep); + WARN_ON_ONCE(p->on_rq != 1); /* * Fix-up what block_task() skipped. @@ -7168,7 +7430,12 @@ static int dequeue_entities(struct rq *rq, struct sched_entity *se, int flags) */ static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) { - if (!(p->se.sched_delayed && (task_on_rq_migrating(p) || (flags & DEQUEUE_SAVE)))) + 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); @@ -7178,12 +7445,13 @@ static bool dequeue_task_fair(struct rq *rq, struct task_struct *p, int flags) /* * 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: sched_balance_rq(), sched_balance_newidle(). */ static DEFINE_PER_CPU(cpumask_var_t, load_balance_mask); @@ -7194,8 +7462,7 @@ static DEFINE_PER_CPU(cpumask_var_t, should_we_balance_tmpmask); static struct { cpumask_var_t idle_cpus_mask; - atomic_t nr_cpus; - int has_blocked; /* Idle CPUS has blocked load */ + int has_blocked_load; /* Idle CPUS has blocked load */ int needs_update; /* Newly idle CPUs need their next_balance collated */ unsigned long next_balance; /* in jiffy units */ unsigned long next_blocked; /* Next update of blocked load in jiffies */ @@ -7344,8 +7611,12 @@ 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; @@ -7443,7 +7714,7 @@ sched_balance_find_dst_group_cpu(struct sched_group *group, struct task_struct * if (!sched_core_cookie_match(rq, p)) continue; - if (sched_idle_cpu(i)) + if (choose_sched_idle_rq(rq, p)) return i; if (available_idle_cpu(i)) { @@ -7534,8 +7805,7 @@ static inline int sched_balance_find_dst_cpu(struct sched_domain *sd, struct tas static inline int __select_idle_cpu(int cpu, struct task_struct *p) { - if ((available_idle_cpu(cpu) || sched_idle_cpu(cpu)) && - sched_cpu_cookie_match(cpu_rq(cpu), p)) + if (choose_idle_cpu(cpu, p) && sched_cpu_cookie_match(cpu_rq(cpu), p)) return cpu; return -1; @@ -7549,7 +7819,7 @@ static inline void set_idle_cores(int cpu, int val) { struct sched_domain_shared *sds; - sds = rcu_dereference(per_cpu(sd_llc_shared, cpu)); + sds = rcu_dereference_all(per_cpu(sd_llc_shared, cpu)); if (sds) WRITE_ONCE(sds->has_idle_cores, val); } @@ -7558,7 +7828,7 @@ static inline bool test_idle_cores(int cpu) { struct sched_domain_shared *sds; - sds = rcu_dereference(per_cpu(sd_llc_shared, cpu)); + sds = rcu_dereference_all(per_cpu(sd_llc_shared, cpu)); if (sds) return READ_ONCE(sds->has_idle_cores); @@ -7608,7 +7878,8 @@ 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, cpus)) { + if (choose_sched_idle_rq(cpu_rq(cpu), p) && + cpumask_test_cpu(cpu, cpus)) { *idle_cpu = cpu; break; } @@ -7643,14 +7914,14 @@ static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int t */ if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) continue; - if (available_idle_cpu(cpu) || sched_idle_cpu(cpu)) + if (choose_idle_cpu(cpu, p)) return cpu; } return -1; } -#else /* CONFIG_SCHED_SMT */ +#else /* !CONFIG_SCHED_SMT: */ static inline void set_idle_cores(int cpu, int val) { @@ -7671,7 +7942,7 @@ static inline int select_idle_smt(struct task_struct *p, struct sched_domain *sd return -1; } -#endif /* CONFIG_SCHED_SMT */ +#endif /* !CONFIG_SCHED_SMT */ /* * Scan the LLC domain for idle CPUs; this is dynamically regulated by @@ -7682,21 +7953,26 @@ 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; - - cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr); if (sched_feat(SIS_UTIL)) { - sd_share = rcu_dereference(per_cpu(sd_llc_shared, target)); - if (sd_share) { - /* because !--nr is the condition to stop scan */ - nr = READ_ONCE(sd_share->nr_idle_scan) + 1; - /* overloaded LLC is unlikely to have idle cpu/core */ - if (nr == 1) - return -1; - } + /* + * Increment because !--nr is the condition to stop scan. + * + * Since "sd" is "sd_llc" for target CPU dereferenced in the + * caller, it is safe to directly dereference "sd->shared". + * Topology bits always ensure it assigned for "sd_llc" abd it + * cannot disappear as long as we have a RCU protected + * reference to one the associated "sd" here. + */ + nr = READ_ONCE(sd->shared->nr_idle_scan) + 1; + /* overloaded LLC is unlikely to have idle cpu/core */ + if (nr == 1) + return -1; } + if (!cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr)) + return -1; + if (static_branch_unlikely(&sched_cluster_active)) { struct sched_group *sg = sd->groups; @@ -7765,7 +8041,7 @@ select_idle_capacity(struct task_struct *p, struct sched_domain *sd, int target) for_each_cpu_wrap(cpu, cpus, target) { unsigned long cpu_cap = capacity_of(cpu); - if (!available_idle_cpu(cpu) && !sched_idle_cpu(cpu)) + if (!choose_idle_cpu(cpu, p)) continue; fits = util_fits_cpu(task_util, util_min, util_max, cpu); @@ -7836,7 +8112,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) */ lockdep_assert_irqs_disabled(); - if ((available_idle_cpu(target) || sched_idle_cpu(target)) && + if (choose_idle_cpu(target, p) && asym_fits_cpu(task_util, util_min, util_max, target)) return target; @@ -7844,7 +8120,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * If the previous CPU is cache affine and idle, don't be stupid: */ if (prev != target && cpus_share_cache(prev, target) && - (available_idle_cpu(prev) || sched_idle_cpu(prev)) && + choose_idle_cpu(prev, p) && asym_fits_cpu(task_util, util_min, util_max, prev)) { if (!static_branch_unlikely(&sched_cluster_active) || @@ -7876,7 +8152,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) if (recent_used_cpu != prev && recent_used_cpu != target && cpus_share_cache(recent_used_cpu, target) && - (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) && + choose_idle_cpu(recent_used_cpu, p) && cpumask_test_cpu(recent_used_cpu, p->cpus_ptr) && asym_fits_cpu(task_util, util_min, util_max, recent_used_cpu)) { @@ -7893,7 +8169,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * sd_asym_cpucapacity rather than sd_llc. */ if (sched_asym_cpucap_active()) { - sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, target)); + sd = rcu_dereference_all(per_cpu(sd_asym_cpucapacity, target)); /* * On an asymmetric CPU capacity system where an exclusive * cpuset defines a symmetric island (i.e. one unique @@ -7908,7 +8184,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) } } - sd = rcu_dereference(per_cpu(sd_llc, target)); + sd = rcu_dereference_all(per_cpu(sd_llc, target)); if (!sd) return target; @@ -8376,26 +8652,25 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) struct perf_domain *pd; struct energy_env eenv; - rcu_read_lock(); - pd = rcu_dereference(rd->pd); + pd = rcu_dereference_all(rd->pd); if (!pd) - goto unlock; + return target; /* * Energy-aware wake-up happens on the lowest sched_domain starting * from sd_asym_cpucapacity spanning over this_cpu and prev_cpu. */ - sd = rcu_dereference(*this_cpu_ptr(&sd_asym_cpucapacity)); + sd = rcu_dereference_all(*this_cpu_ptr(&sd_asym_cpucapacity)); while (sd && !cpumask_test_cpu(prev_cpu, sched_domain_span(sd))) sd = sd->parent; if (!sd) - goto unlock; + return target; target = prev_cpu; sync_entity_load_avg(&p->se); if (!task_util_est(p) && p_util_min == 0) - goto unlock; + return target; eenv_task_busy_time(&eenv, p, prev_cpu); @@ -8408,9 +8683,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) int max_spare_cap_cpu = -1; int fits, max_fits = -1; - cpumask_and(cpus, perf_domain_span(pd), cpu_online_mask); - - if (cpumask_empty(cpus)) + if (!cpumask_and(cpus, perf_domain_span(pd), cpu_online_mask)) continue; /* Account external pressure for the energy estimation */ @@ -8492,7 +8765,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) prev_cpu); /* CPU utilization has changed */ if (prev_delta < base_energy) - goto unlock; + return target; prev_delta -= base_energy; prev_actual_cap = cpu_actual_cap; best_delta = min(best_delta, prev_delta); @@ -8516,7 +8789,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) max_spare_cap_cpu); /* CPU utilization has changed */ if (cur_delta < base_energy) - goto unlock; + return target; cur_delta -= base_energy; /* @@ -8533,7 +8806,6 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) best_actual_cap = cpu_actual_cap; } } - rcu_read_unlock(); if ((best_fits > prev_fits) || ((best_fits > 0) && (best_delta < prev_delta)) || @@ -8541,11 +8813,6 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) target = best_energy_cpu; return target; - -unlock: - rcu_read_unlock(); - - return target; } /* @@ -8590,7 +8857,6 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, p->cpus_ptr); } - rcu_read_lock(); for_each_domain(cpu, tmp) { /* * If both 'cpu' and 'prev_cpu' are part of this domain, @@ -8616,14 +8882,13 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags) break; } - if (unlikely(sd)) { - /* Slow path */ - 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); - } - rcu_read_unlock(); + /* Slow path */ + if (unlikely(sd)) + return sched_balance_find_dst_cpu(sd, p, cpu, prev_cpu, sd_flag); + + /* Fast path */ + if (wake_flags & WF_TTWU) + return select_idle_sibling(p, prev_cpu, new_cpu); return new_cpu; } @@ -8708,22 +8973,10 @@ static void set_cpus_allowed_fair(struct task_struct *p, struct affinity_context set_task_max_allowed_capacity(p); } -static int -balance_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) -{ - if (sched_fair_runnable(rq)) - return 1; - - return sched_balance_newidle(rq, rf) != 0; -} -#else -static inline void set_task_max_allowed_capacity(struct task_struct *p) {} -#endif /* CONFIG_SMP */ - 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; @@ -8731,16 +8984,88 @@ static void set_next_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) +{ + 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_fair(struct rq *rq, struct task_struct *p, int wake_flags) +static void wakeup_preempt_fair(struct rq *rq, struct task_struct *p, int wake_flags) { + 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 sched_entity *nse, *se = &donor->se, *pse = &p->se; struct cfs_rq *cfs_rq = task_cfs_rq(donor); int cse_is_idle, pse_is_idle; + /* + * XXX Getting preempted by higher class, try and find idle CPU? + */ + if (p->sched_class != &fair_sched_class) + return; + if (unlikely(se == pse)) return; @@ -8750,13 +9075,9 @@ static void check_preempt_wakeup_fair(struct rq *rq, struct task_struct *p, int * 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) && !(wake_flags & WF_FORK) && !pse->sched_delayed) { - set_next_buddy(pse); - } - /* * We can come here with TIF_NEED_RESCHED already set from new task * wake up path. @@ -8783,8 +9104,15 @@ static void check_preempt_wakeup_fair(struct rq *rq, struct task_struct *p, int * 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; @@ -8799,50 +9127,104 @@ static void check_preempt_wakeup_fair(struct rq *rq, struct task_struct *p, int /* * If @p has a shorter slice than current and @p is eligible, override * current's slice protection in order to allow preemption. - * - * Note that even if @p does not turn out to be the most eligible - * task at this moment, current's slice protection will be lost. */ - if (do_preempt_short(cfs_rq, pse, se) && se->vlag == se->deadline) - se->vlag = se->deadline + 1; + if (sched_feat(PREEMPT_SHORT) && (pse->slice < se->slice)) { + preempt_action = PREEMPT_WAKEUP_SHORT; + goto pick; + } /* - * If @p has become the most eligible task, force preemption. + * 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 (pick_eevdf(cfs_rq) == pse) + if ((wake_flags & WF_FORK) || pse->sched_delayed) + return; + + /* Prefer picking wakee soon if appropriate. */ + if (sched_feat(NEXT_BUDDY) && + set_preempt_buddy(cfs_rq, wake_flags, pse, se)) { + + /* + * 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 (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: + nse = pick_next_entity(rq, cfs_rq, preempt_action != PREEMPT_WAKEUP_SHORT); + /* If @p has become the most eligible task, force preemption */ + if (nse == pse) goto preempt; + /* + * Because p is enqueued, nse being null can only mean that we + * dequeued a delayed task. If there are still entities queued in + * cfs, check if the next one will be p. + */ + if (!nse && cfs_rq->nr_queued) + goto pick; + + if (sched_feat(RUN_TO_PARITY)) + update_protect_slice(cfs_rq, se); + return; preempt: + if (preempt_action == PREEMPT_WAKEUP_SHORT) { + cancel_protect_slice(se); + clear_buddies(cfs_rq, se); + } + resched_curr_lazy(rq); } -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_queued) return NULL; + throttled = false; + do { /* Might not have done put_prev_entity() */ if (cfs_rq->curr && cfs_rq->curr->on_rq) update_curr(cfs_rq); - if (unlikely(check_cfs_rq_runtime(cfs_rq))) - goto again; + throttled |= check_cfs_rq_runtime(cfs_rq); - se = pick_next_entity(rq, cfs_rq); + se = pick_next_entity(rq, cfs_rq, true); 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; } static void __set_next_task_fair(struct rq *rq, struct task_struct *p, bool first); @@ -8850,13 +9232,14 @@ 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) + __must_hold(__rq_lockp(rq)) { struct sched_entity *se; struct task_struct *p; int new_tasks; again: - p = pick_task_fair(rq); + p = pick_task_fair(rq, rf); if (!p) goto idle; se = &p->se; @@ -8891,13 +9274,13 @@ again: pse = parent_entity(pse); } if (se_depth >= pse_depth) { - set_next_entity(cfs_rq_of(se), se); + set_next_entity(cfs_rq_of(se), se, true); se = parent_entity(se); } } put_prev_entity(cfs_rq, pse); - set_next_entity(cfs_rq, se); + set_next_entity(cfs_rq, se, true); __set_next_task_fair(rq, p, true); } @@ -8905,49 +9288,34 @@ again: return p; simple: -#endif +#endif /* CONFIG_FAIR_GROUP_SCHED */ put_prev_set_next_task(rq, prev, p); return p; idle: - if (!rf) - return NULL; - - new_tasks = sched_balance_newidle(rq, rf); - - /* - * 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 (rf) { + new_tasks = sched_balance_newidle(rq, rf); - if (new_tasks > 0) - goto again; + /* + * 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; - /* - * rq is about to be idle, check if we need to update the - * lost_idle_time of clock_pelt - */ - update_idle_rq_clock_pelt(rq); + if (new_tasks > 0) + goto again; + } return NULL; } -static struct task_struct *__pick_next_task_fair(struct rq *rq, struct task_struct *prev) +static struct task_struct * +fair_server_pick_task(struct sched_dl_entity *dl_se, struct rq_flags *rf) { - return pick_next_task_fair(rq, prev, NULL); -} - -static bool fair_server_has_tasks(struct sched_dl_entity *dl_se) -{ - return !!dl_se->rq->cfs.nr_queued; -} - -static struct task_struct *fair_server_pick_task(struct sched_dl_entity *dl_se) -{ - return pick_task_fair(dl_se->rq); + return pick_task_fair(dl_se->rq, rf); } void fair_server_init(struct rq *rq) @@ -8956,7 +9324,7 @@ void fair_server_init(struct rq *rq) init_dl_entity(dl_se); - dl_server_init(dl_se, rq, fair_server_has_tasks, fair_server_pick_task); + dl_server_init(dl_se, rq, fair_server_pick_task); } /* @@ -8978,7 +9346,7 @@ static void put_prev_task_fair(struct rq *rq, struct task_struct *prev, struct t */ 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; @@ -9002,15 +9370,26 @@ static void yield_task_fair(struct rq *rq) */ rq_clock_skip_update(rq); - se->deadline += calc_delta_fair(se->slice, 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; + update_deadline(cfs_rq, 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 se to run next. */ @@ -9021,7 +9400,6 @@ static bool yield_to_task_fair(struct rq *rq, struct task_struct *p) return true; } -#ifdef CONFIG_SMP /************************************************** * Fair scheduling class load-balancing methods. * @@ -9279,7 +9657,7 @@ static int task_hot(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); + struct numa_group *numa_group = rcu_dereference_all(p->numa_group); unsigned long src_weight, dst_weight; int src_nid, dst_nid, dist; @@ -9323,13 +9701,13 @@ static long migrate_degrades_locality(struct task_struct *p, struct lb_env *env) return src_weight - dst_weight; } -#else +#else /* !CONFIG_NUMA_BALANCING: */ static inline long migrate_degrades_locality(struct task_struct *p, struct lb_env *env) { return 0; } -#endif +#endif /* !CONFIG_NUMA_BALANCING */ /* * Check whether the task is ineligible on the destination cpu @@ -9370,15 +9748,16 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) /* * We do not migrate tasks that are: * 1) delayed dequeued unless we migrate load, or - * 2) throttled_lb_pair, 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. + * 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 (throttled_lb_pair(task_group(p), env->src_cpu, env->dst_cpu)) + if (lb_throttled_hierarchy(p, env->dst_cpu)) return 0; /* @@ -9395,6 +9774,9 @@ int can_migrate_task(struct task_struct *p, struct lb_env *env) 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; @@ -9417,12 +9799,11 @@ 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; @@ -9431,7 +9812,8 @@ 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; } @@ -9475,6 +9857,9 @@ static void detach_task(struct task_struct *p, struct lb_env *env) 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); } @@ -9652,32 +10037,6 @@ next: } /* - * attach_task() -- attach the task detached by detach_task() to its new rq. - */ -static void attach_task(struct rq *rq, struct task_struct *p) -{ - lockdep_assert_rq_held(rq); - - WARN_ON_ONCE(task_rq(p) != rq); - activate_task(rq, p, ENQUEUE_NOCLOCK); - wakeup_preempt(rq, p, 0); -} - -/* - * attach_one_task() -- attaches the task returned from detach_one_task() to - * its new rq. - */ -static void attach_one_task(struct rq *rq, struct task_struct *p) -{ - struct rq_flags rf; - - rq_lock(rq, &rf); - update_rq_clock(rq); - attach_task(rq, p); - rq_unlock(rq, &rf); -} - -/* * attach_tasks() -- attaches all tasks detached by detach_tasks() to their * new rq. */ @@ -9701,7 +10060,7 @@ static void attach_tasks(struct lb_env *env) } #ifdef CONFIG_NO_HZ_COMMON -static inline bool cfs_rq_has_blocked(struct cfs_rq *cfs_rq) +static inline bool cfs_rq_has_blocked_load(struct cfs_rq *cfs_rq) { if (cfs_rq->avg.load_avg) return true; @@ -9734,17 +10093,17 @@ static inline void update_blocked_load_tick(struct rq *rq) WRITE_ONCE(rq->last_blocked_load_update_tick, jiffies); } -static inline void update_blocked_load_status(struct rq *rq, bool has_blocked) +static inline void update_has_blocked_load_status(struct rq *rq, bool has_blocked_load) { - if (!has_blocked) + if (!has_blocked_load) rq->has_blocked_load = 0; } -#else -static inline bool cfs_rq_has_blocked(struct cfs_rq *cfs_rq) { return false; } +#else /* !CONFIG_NO_HZ_COMMON: */ +static inline bool cfs_rq_has_blocked_load(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 +static inline void update_has_blocked_load_status(struct rq *rq, bool has_blocked_load) {} +#endif /* !CONFIG_NO_HZ_COMMON */ static bool __update_blocked_others(struct rq *rq, bool *done) { @@ -9800,7 +10159,7 @@ static bool __update_blocked_fair(struct rq *rq, bool *done) list_del_leaf_cfs_rq(cfs_rq); /* Don't need periodic decay once load/util_avg are null */ - if (cfs_rq_has_blocked(cfs_rq)) + if (cfs_rq_has_blocked_load(cfs_rq)) *done = false; } @@ -9853,14 +10212,14 @@ 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; bool decayed; decayed = update_cfs_rq_load_avg(cfs_rq_clock_pelt(cfs_rq), cfs_rq); - if (cfs_rq_has_blocked(cfs_rq)) + if (cfs_rq_has_blocked_load(cfs_rq)) *done = false; return decayed; @@ -9870,25 +10229,29 @@ static unsigned long task_h_load(struct task_struct *p) { return p->se.avg.load_avg; } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ -static void sched_balance_update_blocked_averages(int cpu) +static void __sched_balance_update_blocked_averages(struct rq *rq) { bool decayed = false, done = true; - struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; - rq_lock_irqsave(rq, &rf); update_blocked_load_tick(rq); - update_rq_clock(rq); decayed |= __update_blocked_others(rq, &done); decayed |= __update_blocked_fair(rq, &done); - update_blocked_load_status(rq, !done); + update_has_blocked_load_status(rq, !done); if (decayed) cpufreq_update_util(rq, 0); - rq_unlock_irqrestore(rq, &rf); +} + +static void sched_balance_update_blocked_averages(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + + guard(rq_lock_irqsave)(rq); + update_rq_clock(rq); + __sched_balance_update_blocked_averages(rq); } /********** Helpers for sched_balance_find_src_group ************************/ @@ -9910,6 +10273,7 @@ struct sg_lb_stats { 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 */ + unsigned int group_overutilized; /* At least one CPU is overutilized in the group */ #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running; @@ -10015,9 +10379,9 @@ 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. */ @@ -10030,7 +10394,7 @@ 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. */ @@ -10142,6 +10506,13 @@ group_has_capacity(unsigned int imbalance_pct, struct sg_lb_stats *sgs) static inline bool group_is_overloaded(unsigned int imbalance_pct, struct sg_lb_stats *sgs) { + /* + * With EAS and uclamp, 1 CPU in the group must be overutilized to + * consider the group overloaded. + */ + if (sched_energy_enabled() && !sgs->group_overutilized) + return false; + if (sgs->sum_nr_running <= sgs->group_weight) return false; @@ -10237,7 +10608,7 @@ sched_group_asym(struct lb_env *env, struct sg_lb_stats *sgs, struct sched_group (sgs->group_weight - sgs->idle_cpus != 1)) return false; - return sched_asym(env->sd, env->dst_cpu, group->asym_prefer_cpu); + return sched_asym(env->sd, env->dst_cpu, READ_ONCE(group->asym_prefer_cpu)); } /* One group has more than one SMT CPU while the other group does not */ @@ -10325,14 +10696,12 @@ sched_reduced_capacity(struct rq *rq, struct sched_domain *sd) * @group: sched_group whose statistics are to be updated. * @sgs: variable to hold the statistics for this 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, - bool *sg_overloaded, - bool *sg_overutilized) + bool *sg_overloaded) { int i, nr_running, local_group, sd_flags = env->sd->flags; bool balancing_at_rd = !env->sd->parent; @@ -10354,7 +10723,7 @@ static inline void update_sg_lb_stats(struct lb_env *env, sgs->sum_nr_running += nr_running; if (cpu_overutilized(i)) - *sg_overutilized = 1; + sgs->group_overutilized = 1; /* * No need to call idle_cpu() if nr_running is not 0 @@ -10474,7 +10843,8 @@ static bool update_sd_pick_busiest(struct lb_env *env, case group_asym_packing: /* Prefer to move from lowest priority CPU's work */ - return sched_asym_prefer(sds->busiest->asym_prefer_cpu, sg->asym_prefer_cpu); + return sched_asym_prefer(READ_ONCE(sds->busiest->asym_prefer_cpu), + READ_ONCE(sg->asym_prefer_cpu)); case group_misfit_task: /* @@ -10582,7 +10952,7 @@ 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; @@ -10592,7 +10962,7 @@ 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; @@ -10659,7 +11029,7 @@ 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; @@ -10897,10 +11267,9 @@ sched_balance_find_dst_group(struct sched_domain *sd, struct task_struct *p, int * take care of it. */ if (p->nr_cpus_allowed != NR_CPUS) { - struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_rq_mask); - - cpumask_and(cpus, sched_group_span(local), p->cpus_ptr); - imb_numa_nr = min(cpumask_weight(cpus), sd->imb_numa_nr); + unsigned int w = cpumask_weight_and(p->cpus_ptr, + sched_group_span(local)); + imb_numa_nr = min(w, sd->imb_numa_nr); } imbalance = abs(local_sgs.idle_cpus - idlest_sgs.idle_cpus); @@ -10930,6 +11299,7 @@ static void update_idle_cpu_scan(struct lb_env *env, unsigned long sum_util) { struct sched_domain_shared *sd_share; + struct sched_domain *sd = env->sd; int llc_weight, pct; u64 x, y, tmp; /* @@ -10943,11 +11313,7 @@ static void update_idle_cpu_scan(struct lb_env *env, if (!sched_feat(SIS_UTIL) || env->idle == CPU_NEWLY_IDLE) return; - llc_weight = per_cpu(sd_llc_size, env->dst_cpu); - if (env->sd->span_weight != llc_weight) - return; - - sd_share = rcu_dereference(per_cpu(sd_llc_shared, env->dst_cpu)); + sd_share = sd->shared; if (!sd_share) return; @@ -10981,10 +11347,11 @@ static void update_idle_cpu_scan(struct lb_env *env, */ /* equation [3] */ x = sum_util; + llc_weight = sd->span_weight; do_div(x, llc_weight); /* equation [4] */ - pct = env->sd->imbalance_pct; + pct = sd->imbalance_pct; tmp = x * x * pct * pct; do_div(tmp, 10000 * SCHED_CAPACITY_SCALE); tmp = min_t(long, tmp, SCHED_CAPACITY_SCALE); @@ -11025,13 +11392,15 @@ 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_overloaded, &sg_overutilized); + update_sg_lb_stats(env, sds, sg, sgs, &sg_overloaded); if (!local_group && update_sd_pick_busiest(env, sds, sg, sgs)) { sds->busiest = sg; sds->busiest_stat = *sgs; } + sg_overutilized |= sgs->group_overutilized; + /* Now, start updating sd_lb_stats */ sds->total_load += sgs->group_load; sds->total_capacity += sgs->group_capacity; @@ -11297,7 +11666,7 @@ static struct sched_group *sched_balance_find_src_group(struct lb_env *env) goto force_balance; if (!is_rd_overutilized(env->dst_rq->rd) && - rcu_dereference(env->dst_rq->rd->pd)) + rcu_dereference_all(env->dst_rq->rd->pd)) goto out_balanced; /* ASYM feature bypasses nice load balance check */ @@ -11711,6 +12080,21 @@ static void update_lb_imbalance_stat(struct lb_env *env, struct sched_domain *sd } /* + * 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. */ @@ -11735,6 +12119,7 @@ static int sched_balance_rq(int this_cpu, struct rq *this_rq, .fbq_type = all, .tasks = LIST_HEAD_INIT(env.tasks), }; + bool need_unlock = false; cpumask_and(cpus, sched_domain_span(sd), cpu_active_mask); @@ -11746,6 +12131,14 @@ redo: goto out_balanced; } + 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]); @@ -11986,6 +12379,9 @@ 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; } @@ -12111,21 +12507,6 @@ out_unlock: } /* - * 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); - -/* * 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. */ @@ -12134,24 +12515,66 @@ 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) { + u64 now = sched_clock(); + s64 delta = now - sd->newidle_stamp; + sd->newidle_stamp = now; + int ratio = 0; + + if (delta < 0) + delta = 0; + + if (sched_feat(NI_RATE)) { + /* + * ratio delta freq + * + * 1024 - 4 s - 128 Hz + * 512 - 2 s - 256 Hz + * 256 - 1 s - 512 Hz + * 128 - .5 s - 1024 Hz + * 64 - .25 s - 2048 Hz + */ + ratio = delta >> 22; + } + + ratio += sd->newidle_success; + + sd->newidle_ratio = min(1024, ratio); + 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; } @@ -12168,13 +12591,13 @@ static void sched_balance_domains(struct rq *rq, enum cpu_idle_type idle) { int continue_balancing = 1; int cpu = rq->cpu; - int busy = idle != CPU_IDLE && !sched_idle_cpu(cpu); + int busy = idle != CPU_IDLE && !sched_idle_rq(rq); unsigned long interval; struct sched_domain *sd; /* Earliest time when we have to do rebalance again */ unsigned long next_balance = jiffies + 60*HZ; int update_next_balance = 0; - int need_serialize, need_decay = 0; + int need_decay = 0; u64 max_cost = 0; rcu_read_lock(); @@ -12183,7 +12606,7 @@ static void sched_balance_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; /* @@ -12198,13 +12621,6 @@ static void sched_balance_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 (atomic_cmpxchg_acquire(&sched_balance_running, 0, 1)) - goto out; - } - if (time_after_eq(jiffies, sd->last_balance + interval)) { if (sched_balance_rq(cpu, rq, sd, idle, &continue_balancing)) { /* @@ -12213,14 +12629,11 @@ static void sched_balance_domains(struct rq *rq, enum cpu_idle_type idle) * state even if we migrated tasks. Update it. */ idle = idle_cpu(cpu); - busy = !idle && !sched_idle_cpu(cpu); + busy = !idle && !sched_idle_rq(rq); } sd->last_balance = jiffies; interval = get_sd_balance_interval(sd, busy); } - if (need_serialize) - atomic_set_release(&sched_balance_running, 0); -out: if (time_after(next_balance, sd->last_balance + interval)) { next_balance = sd->last_balance + interval; update_next_balance = 1; @@ -12261,14 +12674,14 @@ static inline int on_null_domain(struct rq *rq) */ static inline int find_new_ilb(void) { + int this_cpu = smp_processor_id(); const struct cpumask *hk_mask; int ilb_cpu; hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); for_each_cpu_and(ilb_cpu, nohz.idle_cpus_mask, hk_mask) { - - if (ilb_cpu == smp_processor_id()) + if (ilb_cpu == this_cpu) continue; if (idle_cpu(ilb_cpu)) @@ -12344,20 +12757,29 @@ static void nohz_balancer_kick(struct rq *rq) */ nohz_balance_exit_idle(rq); - /* - * None are in tickless mode and hence no need for NOHZ idle load - * balancing: - */ - if (likely(!atomic_read(&nohz.nr_cpus))) - return; - - if (READ_ONCE(nohz.has_blocked) && + if (READ_ONCE(nohz.has_blocked_load) && time_after(now, READ_ONCE(nohz.next_blocked))) flags = NOHZ_STATS_KICK; + /* + * Most of the time system is not 100% busy. i.e nohz.nr_cpus > 0 + * Skip the read if time is not due. + * + * If none are in tickless mode, there maybe a narrow window + * (28 jiffies, HZ=1000) where flags maybe set and kick_ilb called. + * But idle load balancing is not done as find_new_ilb fails. + * That's very rare. So read nohz.nr_cpus only if time is due. + */ if (time_before(now, nohz.next_balance)) goto out; + /* + * None are in tickless mode and hence no need for NOHZ idle load + * balancing + */ + if (unlikely(cpumask_empty(nohz.idle_cpus_mask))) + return; + if (rq->nr_running >= 2) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto out; @@ -12365,7 +12787,7 @@ static void nohz_balancer_kick(struct rq *rq) rcu_read_lock(); - sd = rcu_dereference(rq->sd); + sd = rcu_dereference_all(rq->sd); if (sd) { /* * If there's a runnable CFS task and the current CPU has reduced @@ -12377,7 +12799,7 @@ static void nohz_balancer_kick(struct rq *rq) } } - sd = rcu_dereference(per_cpu(sd_asym_packing, cpu)); + sd = rcu_dereference_all(per_cpu(sd_asym_packing, cpu)); if (sd) { /* * When ASYM_PACKING; see if there's a more preferred CPU @@ -12395,7 +12817,7 @@ static void nohz_balancer_kick(struct rq *rq) } } - sd = rcu_dereference(per_cpu(sd_asym_cpucapacity, cpu)); + sd = rcu_dereference_all(per_cpu(sd_asym_cpucapacity, cpu)); if (sd) { /* * When ASYM_CPUCAPACITY; see if there's a higher capacity CPU @@ -12416,7 +12838,7 @@ static void nohz_balancer_kick(struct rq *rq) goto unlock; } - sds = rcu_dereference(per_cpu(sd_llc_shared, cpu)); + sds = rcu_dereference_all(per_cpu(sd_llc_shared, cpu)); if (sds) { /* * If there is an imbalance between LLC domains (IOW we could @@ -12448,7 +12870,7 @@ static void set_cpu_sd_state_busy(int cpu) struct sched_domain *sd; rcu_read_lock(); - sd = rcu_dereference(per_cpu(sd_llc, cpu)); + sd = rcu_dereference_all(per_cpu(sd_llc, cpu)); if (!sd || !sd->nohz_idle) goto unlock; @@ -12461,14 +12883,13 @@ 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; rq->nohz_tick_stopped = 0; cpumask_clear_cpu(rq->cpu, nohz.idle_cpus_mask); - atomic_dec(&nohz.nr_cpus); set_cpu_sd_state_busy(rq->cpu); } @@ -12478,7 +12899,7 @@ static void set_cpu_sd_state_idle(int cpu) struct sched_domain *sd; rcu_read_lock(); - sd = rcu_dereference(per_cpu(sd_llc, cpu)); + sd = rcu_dereference_all(per_cpu(sd_llc, cpu)); if (!sd || sd->nohz_idle) goto unlock; @@ -12497,7 +12918,7 @@ 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)) @@ -12512,9 +12933,9 @@ void nohz_balance_enter_idle(int cpu) /* * The tick is still stopped but load could have been added in the - * meantime. We set the nohz.has_blocked flag to trig a check of the + * meantime. We set the nohz.has_blocked_load flag to trig a check of the * *_avg. The CPU is already part of nohz.idle_cpus_mask so the clear - * of nohz.has_blocked can only happen after checking the new load + * of nohz.has_blocked_load can only happen after checking the new load */ if (rq->nohz_tick_stopped) goto out; @@ -12526,11 +12947,10 @@ void nohz_balance_enter_idle(int cpu) rq->nohz_tick_stopped = 1; cpumask_set_cpu(cpu, nohz.idle_cpus_mask); - atomic_inc(&nohz.nr_cpus); /* * Ensures that if nohz_idle_balance() fails to observe our - * @idle_cpus_mask store, it must observe the @has_blocked + * @idle_cpus_mask store, it must observe the @has_blocked_load * and @needs_update stores. */ smp_mb__after_atomic(); @@ -12543,7 +12963,7 @@ 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 */ - WRITE_ONCE(nohz.has_blocked, 1); + WRITE_ONCE(nohz.has_blocked_load, 1); } static bool update_nohz_stats(struct rq *rq) @@ -12580,12 +13000,12 @@ 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 - * the has_blocked flag. If a cpu enters idle in the mean time, it will - * set the has_blocked flag and trigger another update of idle load. + * the has_blocked_load flag. If a cpu enters idle in the mean time, it will + * set the has_blocked_load flag and trigger another update of idle load. * Because a cpu that becomes idle, is added to idle_cpus_mask before * setting the flag, we are sure to not clear the state and not * check the load of an idle cpu. @@ -12593,12 +13013,12 @@ static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) * Same applies to idle_cpus_mask vs needs_update. */ if (flags & NOHZ_STATS_KICK) - WRITE_ONCE(nohz.has_blocked, 0); + WRITE_ONCE(nohz.has_blocked_load, 0); if (flags & NOHZ_NEXT_KICK) WRITE_ONCE(nohz.needs_update, 0); /* - * Ensures that if we miss the CPU, we must see the has_blocked + * Ensures that if we miss the CPU, we must see the has_blocked_load * store from nohz_balance_enter_idle(). */ smp_mb(); @@ -12665,7 +13085,7 @@ static void _nohz_idle_balance(struct rq *this_rq, unsigned int flags) abort: /* There is still blocked load, enable periodic update */ if (has_blocked_load) - WRITE_ONCE(nohz.has_blocked, 1); + WRITE_ONCE(nohz.has_blocked_load, 1); } /* @@ -12727,7 +13147,7 @@ static void nohz_newidle_balance(struct rq *this_rq) return; /* Don't need to update blocked load of idle CPUs*/ - if (!READ_ONCE(nohz.has_blocked) || + if (!READ_ONCE(nohz.has_blocked_load) || time_before(jiffies, READ_ONCE(nohz.next_blocked))) return; @@ -12738,7 +13158,7 @@ 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) @@ -12747,7 +13167,7 @@ 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 */ /* * sched_balance_newidle is called by schedule() if this_cpu is about to become @@ -12759,6 +13179,7 @@ static inline void nohz_newidle_balance(struct rq *this_rq) { } * > 0 - success, new (fair) tasks present */ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) + __must_hold(__rq_lockp(this_rq)) { unsigned long next_balance = jiffies + HZ; int this_cpu = this_rq->cpu; @@ -12797,26 +13218,28 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) */ rq_unpin_lock(this_rq, rf); - rcu_read_lock(); - sd = rcu_dereference_check_sched_domain(this_rq->sd); + sd = rcu_dereference_sched_domain(this_rq->sd); + if (!sd) + goto out; if (!get_rd_overloaded(this_rq->rd) || - (sd && this_rq->avg_idle < sd->max_newidle_lb_cost)) { - - if (sd) - update_next_balance(sd, &next_balance); - rcu_read_unlock(); + this_rq->avg_idle < sd->max_newidle_lb_cost) { + update_next_balance(sd, &next_balance); goto out; } - rcu_read_unlock(); - - raw_spin_rq_unlock(this_rq); + /* + * Include sched_balance_update_blocked_averages() in the cost + * calculation because it can be quite costly -- this ensures we skip + * it when avg_idle gets to be very low. + */ t0 = sched_clock_cpu(this_cpu); - sched_balance_update_blocked_averages(this_cpu); + __sched_balance_update_blocked_averages(this_rq); + + rq_modified_begin(this_rq, &fair_sched_class); + raw_spin_rq_unlock(this_rq); - rcu_read_lock(); for_each_domain(this_cpu, sd) { u64 domain_cost; @@ -12826,6 +13249,22 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) break; if (sd->flags & SD_BALANCE_NEWIDLE) { + unsigned int weight = 1; + + if (sched_feat(NI_RANDOM) && sd->newidle_ratio < 1024) { + /* + * 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 = sched_balance_rq(this_cpu, this_rq, sd, CPU_NEWLY_IDLE, @@ -12833,10 +13272,14 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) 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); } /* @@ -12846,7 +13289,6 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) if (pulled_task || !continue_balancing) break; } - rcu_read_unlock(); raw_spin_rq_lock(this_rq); @@ -12861,8 +13303,8 @@ static int sched_balance_newidle(struct rq *this_rq, struct rq_flags *rf) 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_queued) + /* If a higher prio class was modified, restart the pick */ + if (rq_modified_above(this_rq, &fair_sched_class)) pulled_task = -1; out: @@ -12944,8 +13386,6 @@ static void rq_offline_fair(struct rq *rq) clear_tg_offline_cfs_rqs(rq); } -#endif /* CONFIG_SMP */ - #ifdef CONFIG_SCHED_CORE static inline bool __entity_slice_used(struct sched_entity *se, int min_nr_tasks) @@ -12982,7 +13422,170 @@ static inline void task_tick_core(struct rq *rq, struct task_struct *curr) } /* - * 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(const struct sched_entity *se, unsigned int fi_seq, bool forceidle) @@ -12996,7 +13599,7 @@ static void se_fi_update(const struct sched_entity *se, unsigned int fi_seq, cfs_rq->forceidle_seq = fi_seq; } - cfs_rq->min_vruntime_fi = cfs_rq->min_vruntime; + cfs_rq->zero_vruntime_fi = cfs_rq->zero_vruntime; } } @@ -13020,7 +13623,7 @@ bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, 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 /* @@ -13042,18 +13645,18 @@ bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, 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); + delta = vruntime_op(sea->vruntime, "-", seb->vruntime) + + vruntime_op(cfs_rqb->zero_vruntime_fi, "-", cfs_rqa->zero_vruntime_fi); return delta > 0; } @@ -13069,9 +13672,9 @@ static int task_is_throttled_fair(struct task_struct *p, int cpu) #endif return throttled_hierarchy(cfs_rq); } -#else +#else /* !CONFIG_SCHED_CORE: */ static inline void task_tick_core(struct rq *rq, struct task_struct *curr) {} -#endif +#endif /* !CONFIG_SCHED_CORE */ /* * scheduler tick hitting a task of our scheduling class. @@ -13091,6 +13694,9 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued) entity_tick(cfs_rq, se, queued); } + if (queued) + return; + if (static_branch_unlikely(&sched_numa_balancing)) task_tick_numa(rq, curr); @@ -13115,11 +13721,14 @@ static void task_fork_fair(struct task_struct *p) * the current task. */ static void -prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) +prio_changed_fair(struct rq *rq, struct task_struct *p, u64 oldprio) { if (!task_on_rq_queued(p)) return; + if (p->prio == oldprio) + return; + if (rq->cfs.nr_queued == 1) return; @@ -13131,8 +13740,9 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio) if (task_current_donor(rq, p)) { if (p->prio > oldprio) resched_curr(rq); - } else + } else { wakeup_preempt(rq, p, 0); + } } #ifdef CONFIG_FAIR_GROUP_SCHED @@ -13144,10 +13754,13 @@ 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 */ @@ -13158,22 +13771,20 @@ 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(). @@ -13182,7 +13793,6 @@ 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); @@ -13216,6 +13826,12 @@ static void attach_task_cfs_rq(struct task_struct *p) attach_entity_cfs_rq(se); } +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) { detach_task_cfs_rq(p); @@ -13223,7 +13839,7 @@ static void switched_from_fair(struct rq *rq, struct task_struct *p) static void switched_to_fair(struct rq *rq, struct task_struct *p) { - SCHED_WARN_ON(p->se.sched_delayed); + WARN_ON_ONCE(p->se.sched_delayed); attach_task_cfs_rq(p); @@ -13246,7 +13862,6 @@ static void __set_next_task_fair(struct rq *rq, struct task_struct *p, bool firs { 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 @@ -13254,11 +13869,10 @@ static void __set_next_task_fair(struct rq *rq, struct task_struct *p, bool firs */ list_move(&se->group_node, &rq->cfs_tasks); } -#endif if (!first) return; - SCHED_WARN_ON(se->sched_delayed); + WARN_ON_ONCE(se->sched_delayed); if (hrtick_enabled_fair(rq)) hrtick_start_fair(rq, p); @@ -13280,7 +13894,7 @@ static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) for_each_sched_entity(se) { struct cfs_rq *cfs_rq = cfs_rq_of(se); - set_next_entity(cfs_rq, se); + set_next_entity(cfs_rq, se, first); /* ensure bandwidth has been allocated on our new cfs_rq */ account_cfs_rq_runtime(cfs_rq, 0); } @@ -13291,10 +13905,8 @@ static void set_next_task_fair(struct rq *rq, struct task_struct *p, bool first) void init_cfs_rq(struct cfs_rq *cfs_rq) { cfs_rq->tasks_timeline = RB_ROOT_CACHED; - 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 @@ -13309,10 +13921,8 @@ 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); } @@ -13338,10 +13948,10 @@ int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) struct cfs_rq *cfs_rq; int i; - tg->cfs_rq = kcalloc(nr_cpu_ids, sizeof(cfs_rq), GFP_KERNEL); + tg->cfs_rq = kzalloc_objs(cfs_rq, nr_cpu_ids); if (!tg->cfs_rq) goto err; - tg->se = kcalloc(nr_cpu_ids, sizeof(se), GFP_KERNEL); + tg->se = kzalloc_objs(se, nr_cpu_ids); if (!tg->se) goto err; @@ -13595,21 +14205,18 @@ static unsigned int get_rr_interval_fair(struct rq *rq, struct task_struct *task * All the scheduling class methods: */ DEFINE_SCHED_CLASS(fair) = { - .enqueue_task = enqueue_task_fair, .dequeue_task = dequeue_task_fair, .yield_task = yield_task_fair, .yield_to_task = yield_to_task_fair, - .wakeup_preempt = check_preempt_wakeup_fair, + .wakeup_preempt = wakeup_preempt_fair, .pick_task = pick_task_fair, - .pick_next_task = __pick_next_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, .select_task_rq = select_task_rq_fair, .migrate_task_rq = migrate_task_rq_fair, @@ -13618,13 +14225,13 @@ DEFINE_SCHED_CLASS(fair) = { .task_dead = task_dead_fair, .set_cpus_allowed = set_cpus_allowed_fair, -#endif .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, @@ -13645,7 +14252,6 @@ DEFINE_SCHED_CLASS(fair) = { #endif }; -#ifdef CONFIG_SCHED_DEBUG void print_cfs_stats(struct seq_file *m, int cpu) { struct cfs_rq *cfs_rq, *pos; @@ -13664,14 +14270,14 @@ void show_numa_stats(struct task_struct *p, struct seq_file *m) struct numa_group *ng; rcu_read_lock(); - ng = rcu_dereference(p->numa_group); + ng = rcu_dereference_all(p->numa_group); for_each_online_node(node) { if (p->numa_faults) { tsf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 0)]; tpf = p->numa_faults[task_faults_idx(NUMA_MEM, node, 1)]; } if (ng) { - gsf = ng->faults[task_faults_idx(NUMA_MEM, node, 0)], + gsf = ng->faults[task_faults_idx(NUMA_MEM, node, 0)]; gpf = ng->faults[task_faults_idx(NUMA_MEM, node, 1)]; } print_numa_stats(m, node, tsf, tpf, gsf, gpf); @@ -13679,11 +14285,9 @@ 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) { @@ -13705,6 +14309,4 @@ __init void init_sched_fair_class(void) nohz.next_blocked = jiffies; zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT); #endif -#endif /* SMP */ - } diff --git a/kernel/sched/features.h b/kernel/sched/features.h index 3c12d9f93331..84c4fe3abd74 100644 --- a/kernel/sched/features.h +++ b/kernel/sched/features.h @@ -58,13 +58,20 @@ SCHED_FEAT(CACHE_HOT_BUDDY, true) SCHED_FEAT(DELAY_DEQUEUE, true) SCHED_FEAT(DELAY_ZERO, true) +SCHED_FEAT(PARANOID_AVG, false) + /* * Allow wakeup-time preemption of the current task: */ SCHED_FEAT(WAKEUP_PREEMPTION, true) +#ifdef CONFIG_HRTIMER_REARM_DEFERRED +SCHED_FEAT(HRTICK, true) +SCHED_FEAT(HRTICK_DL, true) +#else SCHED_FEAT(HRTICK, false) SCHED_FEAT(HRTICK_DL, false) +#endif /* * Decrement CPU capacity based on time not spent running tasks @@ -121,3 +128,9 @@ SCHED_FEAT(WA_BIAS, true) SCHED_FEAT(UTIL_EST, true) SCHED_FEAT(LATENCY_WARN, false) + +/* + * Do newidle balancing proportional to its success rate using randomization. + */ +SCHED_FEAT(NI_RANDOM, true) +SCHED_FEAT(NI_RATE, true) diff --git a/kernel/sched/idle.c b/kernel/sched/idle.c index 2c85c86b455f..a83be0c834dd 100644 --- a/kernel/sched/idle.c +++ b/kernel/sched/idle.c @@ -6,6 +6,11 @@ * (NOTE: these are not related to SCHED_IDLE batch scheduled * tasks which are handled in sched/fair.c ) */ +#include <linux/cpuidle.h> +#include <linux/suspend.h> +#include <linux/livepatch.h> +#include "sched.h" +#include "smp.h" /* Linker adds these: start and end of __cpuidle functions */ extern char __cpuidle_text_start[], __cpuidle_text_end[]; @@ -47,7 +52,7 @@ static int __init cpu_idle_nopoll_setup(char *__unused) return 1; } __setup("hlt", cpu_idle_nopoll_setup); -#endif +#endif /* CONFIG_GENERIC_IDLE_POLL_SETUP */ static noinline int __cpuidle cpu_idle_poll(void) { @@ -95,10 +100,10 @@ static inline void cond_tick_broadcast_exit(void) if (static_branch_unlikely(&arch_needs_tick_broadcast)) tick_broadcast_exit(); } -#else +#else /* !CONFIG_GENERIC_CLOCKEVENTS_BROADCAST_IDLE: */ static inline void cond_tick_broadcast_enter(void) { } static inline void cond_tick_broadcast_exit(void) { } -#endif +#endif /* !CONFIG_GENERIC_CLOCKEVENTS_BROADCAST_IDLE */ /** * default_idle_call - Default CPU idle routine. @@ -126,12 +131,13 @@ void __cpuidle default_idle_call(void) } static int call_cpuidle_s2idle(struct cpuidle_driver *drv, - struct cpuidle_device *dev) + struct cpuidle_device *dev, + u64 max_latency_ns) { if (current_clr_polling_and_test()) return -EBUSY; - return cpuidle_enter_s2idle(drv, dev); + return cpuidle_enter_s2idle(drv, dev, max_latency_ns); } static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev, @@ -155,6 +161,14 @@ static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev, return cpuidle_enter(drv, dev, next_state); } +static void idle_call_stop_or_retain_tick(bool stop_tick) +{ + if (stop_tick || tick_nohz_tick_stopped()) + tick_nohz_idle_stop_tick(); + else + tick_nohz_idle_retain_tick(); +} + /** * cpuidle_idle_call - the main idle function * @@ -164,7 +178,7 @@ static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev, * set, and it returns with polling set. If it ever stops polling, it * must clear the polling bit. */ -static void cpuidle_idle_call(void) +static void cpuidle_idle_call(bool stop_tick) { struct cpuidle_device *dev = cpuidle_get_device(); struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev); @@ -180,7 +194,7 @@ static void cpuidle_idle_call(void) } if (cpuidle_not_available(drv, dev)) { - tick_nohz_idle_stop_tick(); + idle_call_stop_or_retain_tick(stop_tick); default_idle_call(); goto exit_idle; @@ -200,12 +214,13 @@ static void cpuidle_idle_call(void) u64 max_latency_ns; if (idle_should_enter_s2idle()) { + max_latency_ns = cpu_wakeup_latency_qos_limit() * + NSEC_PER_USEC; - entered_state = call_cpuidle_s2idle(drv, dev); + entered_state = call_cpuidle_s2idle(drv, dev, + max_latency_ns); if (entered_state > 0) goto exit_idle; - - max_latency_ns = U64_MAX; } else { max_latency_ns = dev->forced_idle_latency_limit_ns; } @@ -214,24 +229,35 @@ static void cpuidle_idle_call(void) next_state = cpuidle_find_deepest_state(drv, dev, max_latency_ns); call_cpuidle(drv, dev, next_state); - } else { - bool stop_tick = true; + } else if (drv->state_count > 1) { + /* + * stop_tick is expected to be true by default by cpuidle + * governors, which allows them to select idle states with + * target residency above the tick period length. + */ + stop_tick = true; /* * Ask the cpuidle framework to choose a convenient idle state. */ next_state = cpuidle_select(drv, dev, &stop_tick); - if (stop_tick || tick_nohz_tick_stopped()) - tick_nohz_idle_stop_tick(); - else - tick_nohz_idle_retain_tick(); + idle_call_stop_or_retain_tick(stop_tick); entered_state = call_cpuidle(drv, dev, next_state); /* * Give the governor an opportunity to reflect on the outcome */ cpuidle_reflect(dev, entered_state); + } else { + idle_call_stop_or_retain_tick(stop_tick); + + /* + * If there is only a single idle state (or none), there is + * nothing meaningful for the governor to choose. Skip the + * governor and always use state 0. + */ + call_cpuidle(drv, dev, 0); } exit_idle: @@ -252,6 +278,7 @@ exit_idle: static void do_idle(void) { int cpu = smp_processor_id(); + bool got_tick = false; /* * Check if we need to update blocked load @@ -322,8 +349,9 @@ static void do_idle(void) tick_nohz_idle_restart_tick(); cpu_idle_poll(); } else { - cpuidle_idle_call(); + cpuidle_idle_call(got_tick); } + got_tick = tick_nohz_idle_got_tick(); arch_cpu_idle_exit(); } @@ -427,7 +455,6 @@ void cpu_startup_entry(enum cpuhp_state state) * idle-task scheduling class. */ -#ifdef CONFIG_SMP static int select_task_rq_idle(struct task_struct *p, int cpu, int flags) { @@ -439,7 +466,6 @@ balance_idle(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { return WARN_ON_ONCE(1); } -#endif /* * Idle tasks are unconditionally rescheduled: @@ -449,10 +475,13 @@ static void wakeup_preempt_idle(struct rq *rq, struct task_struct *p, int flags) resched_curr(rq); } +static void update_curr_idle(struct rq *rq); + static void put_prev_task_idle(struct rq *rq, struct task_struct *prev, struct task_struct *next) { - dl_server_update_idle_time(rq, prev); + update_curr_idle(rq); scx_update_idle(rq, false, true); + update_rq_avg_idle(rq); } static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool first) @@ -461,9 +490,15 @@ static void set_next_task_idle(struct rq *rq, struct task_struct *next, bool fir scx_update_idle(rq, true, true); schedstat_inc(rq->sched_goidle); next->se.exec_start = rq_clock_task(rq); + + /* + * rq is about to be idle, check if we need to update the + * lost_idle_time of clock_pelt + */ + update_idle_rq_clock_pelt(rq); } -struct task_struct *pick_task_idle(struct rq *rq) +struct task_struct *pick_task_idle(struct rq *rq, struct rq_flags *rf) { scx_update_idle(rq, true, false); return rq->idle; @@ -493,28 +528,45 @@ dequeue_task_idle(struct rq *rq, struct task_struct *p, int flags) */ static void task_tick_idle(struct rq *rq, struct task_struct *curr, int queued) { + update_curr_idle(rq); } -static void switched_to_idle(struct rq *rq, struct task_struct *p) +static void switching_to_idle(struct rq *rq, struct task_struct *p) { BUG(); } static void -prio_changed_idle(struct rq *rq, struct task_struct *p, int oldprio) +prio_changed_idle(struct rq *rq, struct task_struct *p, u64 oldprio) { + if (p->prio == oldprio) + return; + BUG(); } static void update_curr_idle(struct rq *rq) { + struct sched_entity *se = &rq->idle->se; + u64 now = rq_clock_task(rq); + s64 delta_exec; + + delta_exec = now - se->exec_start; + if (unlikely(delta_exec <= 0)) + return; + + se->exec_start = now; + + dl_server_update_idle(&rq->fair_server, delta_exec); +#ifdef CONFIG_SCHED_CLASS_EXT + dl_server_update_idle(&rq->ext_server, delta_exec); +#endif } /* * Simple, special scheduling class for the per-CPU idle tasks: */ DEFINE_SCHED_CLASS(idle) = { - /* no enqueue/yield_task for idle tasks */ /* dequeue is not valid, we print a debug message there: */ @@ -526,15 +578,13 @@ DEFINE_SCHED_CLASS(idle) = { .put_prev_task = put_prev_task_idle, .set_next_task = set_next_task_idle, -#ifdef CONFIG_SMP .balance = balance_idle, .select_task_rq = select_task_rq_idle, .set_cpus_allowed = set_cpus_allowed_common, -#endif .task_tick = task_tick_idle, .prio_changed = prio_changed_idle, - .switched_to = switched_to_idle, + .switching_to = switching_to_idle, .update_curr = update_curr_idle, }; diff --git a/kernel/sched/isolation.c b/kernel/sched/isolation.c index 81bc8b329ef1..ef152d401fe2 100644 --- a/kernel/sched/isolation.c +++ b/kernel/sched/isolation.c @@ -7,8 +7,12 @@ * Copyright (C) 2017-2018 SUSE, Frederic Weisbecker * */ +#include <linux/sched/isolation.h> +#include <linux/pci.h> +#include "sched.h" enum hk_flags { + HK_FLAG_DOMAIN_BOOT = BIT(HK_TYPE_DOMAIN_BOOT), HK_FLAG_DOMAIN = BIT(HK_TYPE_DOMAIN), HK_FLAG_MANAGED_IRQ = BIT(HK_TYPE_MANAGED_IRQ), HK_FLAG_KERNEL_NOISE = BIT(HK_TYPE_KERNEL_NOISE), @@ -18,7 +22,7 @@ DEFINE_STATIC_KEY_FALSE(housekeeping_overridden); EXPORT_SYMBOL_GPL(housekeeping_overridden); struct housekeeping { - cpumask_var_t cpumasks[HK_TYPE_MAX]; + struct cpumask __rcu *cpumasks[HK_TYPE_MAX]; unsigned long flags; }; @@ -26,21 +30,62 @@ static struct housekeeping housekeeping; bool housekeeping_enabled(enum hk_type type) { - return !!(housekeeping.flags & BIT(type)); + return !!(READ_ONCE(housekeeping.flags) & BIT(type)); } EXPORT_SYMBOL_GPL(housekeeping_enabled); +static bool housekeeping_dereference_check(enum hk_type type) +{ + if (IS_ENABLED(CONFIG_LOCKDEP) && type == HK_TYPE_DOMAIN) { + /* Cpuset isn't even writable yet? */ + if (system_state <= SYSTEM_SCHEDULING) + return true; + + /* CPU hotplug write locked, so cpuset partition can't be overwritten */ + if (IS_ENABLED(CONFIG_HOTPLUG_CPU) && lockdep_is_cpus_write_held()) + return true; + + /* Cpuset lock held, partitions not writable */ + if (IS_ENABLED(CONFIG_CPUSETS) && lockdep_is_cpuset_held()) + return true; + + return false; + } + + return true; +} + +static inline struct cpumask *housekeeping_cpumask_dereference(enum hk_type type) +{ + return rcu_dereference_all_check(housekeeping.cpumasks[type], + housekeeping_dereference_check(type)); +} + +const struct cpumask *housekeeping_cpumask(enum hk_type type) +{ + const struct cpumask *mask = NULL; + + if (static_branch_unlikely(&housekeeping_overridden)) { + if (READ_ONCE(housekeeping.flags) & BIT(type)) + mask = housekeeping_cpumask_dereference(type); + } + if (!mask) + mask = cpu_possible_mask; + return mask; +} +EXPORT_SYMBOL_GPL(housekeeping_cpumask); + int housekeeping_any_cpu(enum hk_type type) { int cpu; if (static_branch_unlikely(&housekeeping_overridden)) { if (housekeeping.flags & BIT(type)) { - cpu = sched_numa_find_closest(housekeeping.cpumasks[type], smp_processor_id()); + cpu = sched_numa_find_closest(housekeeping_cpumask(type), smp_processor_id()); if (cpu < nr_cpu_ids) return cpu; - cpu = cpumask_any_and(housekeeping.cpumasks[type], cpu_online_mask); + cpu = cpumask_any_and_distribute(housekeeping_cpumask(type), cpu_online_mask); if (likely(cpu < nr_cpu_ids)) return cpu; /* @@ -56,32 +101,67 @@ int housekeeping_any_cpu(enum hk_type type) } EXPORT_SYMBOL_GPL(housekeeping_any_cpu); -const struct cpumask *housekeeping_cpumask(enum hk_type type) -{ - if (static_branch_unlikely(&housekeeping_overridden)) - if (housekeeping.flags & BIT(type)) - return housekeeping.cpumasks[type]; - return cpu_possible_mask; -} -EXPORT_SYMBOL_GPL(housekeeping_cpumask); - void housekeeping_affine(struct task_struct *t, enum hk_type type) { if (static_branch_unlikely(&housekeeping_overridden)) if (housekeeping.flags & BIT(type)) - set_cpus_allowed_ptr(t, housekeeping.cpumasks[type]); + set_cpus_allowed_ptr(t, housekeeping_cpumask(type)); } EXPORT_SYMBOL_GPL(housekeeping_affine); bool housekeeping_test_cpu(int cpu, enum hk_type type) { - if (static_branch_unlikely(&housekeeping_overridden)) - if (housekeeping.flags & BIT(type)) - return cpumask_test_cpu(cpu, housekeeping.cpumasks[type]); + if (static_branch_unlikely(&housekeeping_overridden) && + READ_ONCE(housekeeping.flags) & BIT(type)) + return cpumask_test_cpu(cpu, housekeeping_cpumask(type)); return true; } EXPORT_SYMBOL_GPL(housekeeping_test_cpu); +int housekeeping_update(struct cpumask *isol_mask) +{ + struct cpumask *trial, *old = NULL; + int err; + + trial = kmalloc(cpumask_size(), GFP_KERNEL); + if (!trial) + return -ENOMEM; + + cpumask_andnot(trial, housekeeping_cpumask(HK_TYPE_DOMAIN_BOOT), isol_mask); + if (!cpumask_intersects(trial, cpu_online_mask)) { + kfree(trial); + return -EINVAL; + } + + if (!housekeeping.flags) + static_branch_enable(&housekeeping_overridden); + + if (housekeeping.flags & HK_FLAG_DOMAIN) + old = housekeeping_cpumask_dereference(HK_TYPE_DOMAIN); + else + WRITE_ONCE(housekeeping.flags, housekeeping.flags | HK_FLAG_DOMAIN); + rcu_assign_pointer(housekeeping.cpumasks[HK_TYPE_DOMAIN], trial); + + synchronize_rcu(); + + pci_probe_flush_workqueue(); + mem_cgroup_flush_workqueue(); + vmstat_flush_workqueue(); + + err = workqueue_unbound_housekeeping_update(housekeeping_cpumask(HK_TYPE_DOMAIN)); + WARN_ON_ONCE(err < 0); + + err = tmigr_isolated_exclude_cpumask(isol_mask); + WARN_ON_ONCE(err < 0); + + err = kthreads_update_housekeeping(); + WARN_ON_ONCE(err < 0); + + kfree(old); + + return 0; +} + void __init housekeeping_init(void) { enum hk_type type; @@ -93,20 +173,33 @@ void __init housekeeping_init(void) if (housekeeping.flags & HK_FLAG_KERNEL_NOISE) sched_tick_offload_init(); - + /* + * Realloc with a proper allocator so that any cpumask update + * can indifferently free the old version with kfree(). + */ for_each_set_bit(type, &housekeeping.flags, HK_TYPE_MAX) { + struct cpumask *omask, *nmask = kmalloc(cpumask_size(), GFP_KERNEL); + + if (WARN_ON_ONCE(!nmask)) + return; + + omask = rcu_dereference(housekeeping.cpumasks[type]); + /* We need at least one CPU to handle housekeeping work */ - WARN_ON_ONCE(cpumask_empty(housekeeping.cpumasks[type])); + WARN_ON_ONCE(cpumask_empty(omask)); + cpumask_copy(nmask, omask); + RCU_INIT_POINTER(housekeeping.cpumasks[type], nmask); + memblock_free(omask, cpumask_size()); } } static void __init housekeeping_setup_type(enum hk_type type, cpumask_var_t housekeeping_staging) { + struct cpumask *mask = memblock_alloc_or_panic(cpumask_size(), SMP_CACHE_BYTES); - alloc_bootmem_cpumask_var(&housekeeping.cpumasks[type]); - cpumask_copy(housekeeping.cpumasks[type], - housekeeping_staging); + cpumask_copy(mask, housekeeping_staging); + RCU_INIT_POINTER(housekeeping.cpumasks[type], mask); } static int __init housekeeping_setup(char *str, unsigned long flags) @@ -159,12 +252,35 @@ static int __init housekeeping_setup(char *str, unsigned long flags) for_each_set_bit(type, &iter_flags, HK_TYPE_MAX) { if (!cpumask_equal(housekeeping_staging, - housekeeping.cpumasks[type])) { + housekeeping_cpumask(type))) { pr_warn("Housekeeping: nohz_full= must match isolcpus=\n"); goto free_housekeeping_staging; } } + /* + * Check the combination of nohz_full and isolcpus=domain, + * necessary to avoid problems with the timer migration + * hierarchy. managed_irq is ignored by this check since it + * isn't considered in the timer migration logic. + */ + iter_flags = housekeeping.flags & (HK_FLAG_KERNEL_NOISE | HK_FLAG_DOMAIN); + type = find_first_bit(&iter_flags, HK_TYPE_MAX); + /* + * Pass the check if none of these flags were previously set or + * are not in the current selection. + */ + iter_flags = flags & (HK_FLAG_KERNEL_NOISE | HK_FLAG_DOMAIN); + first_cpu = (type == HK_TYPE_MAX || !iter_flags) ? 0 : + cpumask_first_and_and(cpu_present_mask, + housekeeping_staging, housekeeping_cpumask(type)); + if (first_cpu >= min(nr_cpu_ids, setup_max_cpus)) { + pr_warn("Housekeeping: must include one present CPU " + "neither in nohz_full= nor in isolcpus=domain, " + "ignoring setting %s\n", str); + goto free_housekeeping_staging; + } + iter_flags = flags & ~housekeeping.flags; for_each_set_bit(type, &iter_flags, HK_TYPE_MAX) @@ -214,7 +330,7 @@ static int __init housekeeping_isolcpus_setup(char *str) if (!strncmp(str, "domain,", 7)) { str += 7; - flags |= HK_FLAG_DOMAIN; + flags |= HK_FLAG_DOMAIN | HK_FLAG_DOMAIN_BOOT; continue; } @@ -244,7 +360,7 @@ static int __init housekeeping_isolcpus_setup(char *str) /* Default behaviour for isolcpus without flags */ if (!flags) - flags |= HK_FLAG_DOMAIN; + flags |= HK_FLAG_DOMAIN | HK_FLAG_DOMAIN_BOOT; return housekeeping_setup(str, flags); } diff --git a/kernel/sched/loadavg.c b/kernel/sched/loadavg.c index c48900b856a2..b601e0243d0e 100644 --- a/kernel/sched/loadavg.c +++ b/kernel/sched/loadavg.c @@ -6,6 +6,8 @@ * figure. Its a silly number but people think its important. We go through * great pains to make it work on big machines and tickless kernels. */ +#include <linux/sched/nohz.h> +#include "sched.h" /* * Global load-average calculations @@ -80,7 +82,7 @@ long calc_load_fold_active(struct rq *this_rq, long adjust) long nr_active, delta = 0; nr_active = this_rq->nr_running - adjust; - nr_active += (int)this_rq->nr_uninterruptible; + nr_active += (long)this_rq->nr_uninterruptible; if (nr_active != this_rq->calc_load_active) { delta = nr_active - this_rq->calc_load_active; @@ -333,12 +335,12 @@ static void calc_global_nohz(void) smp_wmb(); calc_load_idx++; } -#else /* !CONFIG_NO_HZ_COMMON */ +#else /* !CONFIG_NO_HZ_COMMON: */ static inline long calc_load_nohz_read(void) { return 0; } static inline void calc_global_nohz(void) { } -#endif /* CONFIG_NO_HZ_COMMON */ +#endif /* !CONFIG_NO_HZ_COMMON */ /* * calc_load - update the avenrun load estimates 10 ticks after the diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c index 809194cd779f..226a6329f3e9 100644 --- a/kernel/sched/membarrier.c +++ b/kernel/sched/membarrier.c @@ -4,6 +4,8 @@ * * membarrier system call */ +#include <uapi/linux/membarrier.h> +#include "sched.h" /* * For documentation purposes, here are some membarrier ordering @@ -197,7 +199,16 @@ static void ipi_rseq(void *info) * is negligible. */ smp_mb(); - rseq_preempt(current); + /* + * Legacy mode requires that IDs are written and the critical section is + * evaluated. V2 optimized mode handles the critical section and IDs are + * only updated if they change as a consequence of preemption after + * return from this IPI. + */ + if (rseq_v2(current)) + rseq_sched_switch_event(current); + else + rseq_force_update(); } static void ipi_sync_rq_state(void *info) @@ -405,9 +416,9 @@ static int membarrier_private_expedited(int flags, int cpu_id) * membarrier, we will end up with some thread in the mm * running without a core sync. * - * For RSEQ, don't rseq_preempt() the caller. User code - * is not supposed to issue syscalls at all from inside an - * rseq critical section. + * For RSEQ, don't invoke rseq_sched_switch_event() on the + * caller. User code is not supposed to issue syscalls at + * all from inside an rseq critical section. */ if (flags != MEMBARRIER_FLAG_SYNC_CORE) { preempt_disable(); diff --git a/kernel/sched/pelt.c b/kernel/sched/pelt.c index 7a8534a2deff..897790889ba3 100644 --- a/kernel/sched/pelt.c +++ b/kernel/sched/pelt.c @@ -15,7 +15,7 @@ * Author: Srivatsa Vaddagiri <vatsa@linux.vnet.ibm.com> * * Scaled math optimizations by Thomas Gleixner - * Copyright (C) 2007, Thomas Gleixner <tglx@linutronix.de> + * Copyright (C) 2007, Linutronix GmbH, Thomas Gleixner <tglx@kernel.org> * * Adaptive scheduling granularity, math enhancements by Peter Zijlstra * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra @@ -23,6 +23,7 @@ * Move PELT related code from fair.c into this pelt.c file * Author: Vincent Guittot <vincent.guittot@linaro.org> */ +#include "pelt.h" /* * Approximate: @@ -413,7 +414,7 @@ int update_hw_load_avg(u64 now, struct rq *rq, u64 capacity) return 0; } -#endif +#endif /* CONFIG_SCHED_HW_PRESSURE */ #ifdef CONFIG_HAVE_SCHED_AVG_IRQ /* @@ -466,7 +467,7 @@ int update_irq_load_avg(struct rq *rq, u64 running) return ret; } -#endif +#endif /* CONFIG_HAVE_SCHED_AVG_IRQ */ /* * Load avg and utiliztion metrics need to be updated periodically and before diff --git a/kernel/sched/pelt.h b/kernel/sched/pelt.h index f4f6a0875c66..f921302dc40f 100644 --- a/kernel/sched/pelt.h +++ b/kernel/sched/pelt.h @@ -1,4 +1,8 @@ -#ifdef CONFIG_SMP +// SPDX-License-Identifier: GPL-2.0 +#ifndef _KERNEL_SCHED_PELT_H +#define _KERNEL_SCHED_PELT_H +#include "sched.h" + #include "sched-pelt.h" int __update_load_avg_blocked_se(u64 now, struct sched_entity *se); @@ -15,7 +19,7 @@ static inline u64 hw_load_avg(struct rq *rq) { return READ_ONCE(rq->avg_hw.load_avg); } -#else +#else /* !CONFIG_SCHED_HW_PRESSURE: */ static inline int update_hw_load_avg(u64 now, struct rq *rq, u64 capacity) { @@ -26,7 +30,7 @@ static inline u64 hw_load_avg(struct rq *rq) { return 0; } -#endif +#endif /* !CONFIG_SCHED_HW_PRESSURE */ #ifdef CONFIG_HAVE_SCHED_AVG_IRQ int update_irq_load_avg(struct rq *rq, u64 running); @@ -158,7 +162,7 @@ static inline void update_idle_cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) { u64 throttled; - if (unlikely(cfs_rq->throttle_count)) + if (unlikely(cfs_rq->pelt_clock_throttled)) throttled = U64_MAX; else throttled = cfs_rq->throttled_clock_pelt_time; @@ -169,68 +173,17 @@ static inline void update_idle_cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) /* rq->task_clock normalized against any time this cfs_rq has spent throttled */ static inline u64 cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) { - if (unlikely(cfs_rq->throttle_count)) + if (unlikely(cfs_rq->pelt_clock_throttled)) return cfs_rq->throttled_clock_pelt - cfs_rq->throttled_clock_pelt_time; return rq_clock_pelt(rq_of(cfs_rq)) - cfs_rq->throttled_clock_pelt_time; } -#else +#else /* !CONFIG_CFS_BANDWIDTH: */ static inline void update_idle_cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) { } static inline u64 cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) { return rq_clock_pelt(rq_of(cfs_rq)); } -#endif - -#else - -static inline int -update_cfs_rq_load_avg(u64 now, struct cfs_rq *cfs_rq) -{ - return 0; -} - -static inline int -update_rt_rq_load_avg(u64 now, struct rq *rq, int running) -{ - return 0; -} - -static inline int -update_dl_rq_load_avg(u64 now, struct rq *rq, int running) -{ - return 0; -} - -static inline int -update_hw_load_avg(u64 now, struct rq *rq, u64 capacity) -{ - return 0; -} - -static inline u64 hw_load_avg(struct rq *rq) -{ - return 0; -} - -static inline int -update_irq_load_avg(struct rq *rq, u64 running) -{ - return 0; -} - -static inline u64 rq_clock_pelt(struct rq *rq) -{ - return rq_clock_task(rq); -} - -static inline void -update_rq_clock_pelt(struct rq *rq, s64 delta) { } - -static inline void -update_idle_rq_clock_pelt(struct rq *rq) { } - -static inline void update_idle_cfs_rq_clock_pelt(struct cfs_rq *cfs_rq) { } -#endif - +#endif /* !CONFIG_CFS_BANDWIDTH */ +#endif /* _KERNEL_SCHED_PELT_H */ diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index bb56805e3d47..d9c9d9480a45 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -136,6 +136,10 @@ * cost-wise, yet way more sensitive and accurate than periodic * sampling of the aggregate task states would be. */ +#include <linux/sched/clock.h> +#include <linux/workqueue.h> +#include <linux/psi.h> +#include "sched.h" static int psi_bug __read_mostly; @@ -172,17 +176,35 @@ struct psi_group psi_system = { .pcpu = &system_group_pcpu, }; +static DEFINE_PER_CPU(seqcount_t, psi_seq) = SEQCNT_ZERO(psi_seq); + +static inline void psi_write_begin(int cpu) +{ + write_seqcount_begin(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline void psi_write_end(int cpu) +{ + write_seqcount_end(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline u32 psi_read_begin(int cpu) +{ + return read_seqcount_begin(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline bool psi_read_retry(int cpu, u32 seq) +{ + return read_seqcount_retry(per_cpu_ptr(&psi_seq, cpu), seq); +} + static void psi_avgs_work(struct work_struct *work); static void poll_timer_fn(struct timer_list *t); static void group_init(struct psi_group *group) { - int cpu; - group->enabled = true; - for_each_possible_cpu(cpu) - seqcount_init(&per_cpu_ptr(group->pcpu, cpu)->seq); group->avg_last_update = sched_clock(); group->avg_next_update = group->avg_last_update + psi_period; mutex_init(&group->avgs_lock); @@ -262,14 +284,14 @@ static void get_recent_times(struct psi_group *group, int cpu, /* Snapshot a coherent view of the CPU state */ do { - seq = read_seqcount_begin(&groupc->seq); + seq = psi_read_begin(cpu); now = cpu_clock(cpu); memcpy(times, groupc->times, sizeof(groupc->times)); state_mask = groupc->state_mask; state_start = groupc->state_start; if (cpu == current_cpu) memcpy(tasks, groupc->tasks, sizeof(groupc->tasks)); - } while (read_seqcount_retry(&groupc->seq, seq)); + } while (psi_read_retry(cpu, seq)); /* Calculate state time deltas against the previous snapshot */ for (s = 0; s < NR_PSI_STATES; s++) { @@ -733,7 +755,7 @@ static int psi_rtpoll_worker(void *data) static void poll_timer_fn(struct timer_list *t) { - struct psi_group *group = from_timer(group, t, rtpoll_timer); + struct psi_group *group = timer_container_of(group, t, rtpoll_timer); atomic_set(&group->rtpoll_wakeup, 1); wake_up_interruptible(&group->rtpoll_wait); @@ -768,31 +790,21 @@ static void record_times(struct psi_group_cpu *groupc, u64 now) groupc->times[PSI_NONIDLE] += delta; } +#define for_each_group(iter, group) \ + for (typeof(group) iter = group; iter; iter = iter->parent) + static void psi_group_change(struct psi_group *group, int cpu, unsigned int clear, unsigned int set, - bool wake_clock) + u64 now, bool wake_clock) { struct psi_group_cpu *groupc; unsigned int t, m; u32 state_mask; - u64 now; lockdep_assert_rq_held(cpu_rq(cpu)); groupc = per_cpu_ptr(group->pcpu, cpu); /* - * First we update the task counts according to the state - * change requested through the @clear and @set bits. - * - * Then if the cgroup PSI stats accounting enabled, we - * assess the aggregate resource states this CPU's tasks - * have been in since the last change, and account any - * SOME and FULL time these may have resulted in. - */ - write_seqcount_begin(&groupc->seq); - now = cpu_clock(cpu); - - /* * Start with TSK_ONCPU, which doesn't have a corresponding * task count - it's just a boolean flag directly encoded in * the state mask. Clear, set, or carry the current state if @@ -843,7 +855,6 @@ static void psi_group_change(struct psi_group *group, int cpu, groupc->state_mask = state_mask; - write_seqcount_end(&groupc->seq); return; } @@ -864,8 +875,6 @@ static void psi_group_change(struct psi_group *group, int cpu, groupc->state_mask = state_mask; - write_seqcount_end(&groupc->seq); - if (state_mask & group->rtpoll_states) psi_schedule_rtpoll_work(group, 1, false); @@ -900,24 +909,29 @@ static void psi_flags_change(struct task_struct *task, int clear, int set) void psi_task_change(struct task_struct *task, int clear, int set) { int cpu = task_cpu(task); - struct psi_group *group; + u64 now; if (!task->pid) return; psi_flags_change(task, clear, set); - group = task_psi_group(task); - do { - psi_group_change(group, cpu, clear, set, true); - } while ((group = group->parent)); + psi_write_begin(cpu); + now = cpu_clock(cpu); + for_each_group(group, task_psi_group(task)) + psi_group_change(group, cpu, clear, set, now, true); + psi_write_end(cpu); } void psi_task_switch(struct task_struct *prev, struct task_struct *next, bool sleep) { - struct psi_group *group, *common = NULL; + struct psi_group *common = NULL; int cpu = task_cpu(prev); + u64 now; + + psi_write_begin(cpu); + now = cpu_clock(cpu); if (next->pid) { psi_flags_change(next, 0, TSK_ONCPU); @@ -926,16 +940,15 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, * ancestors with @prev, those will already have @prev's * TSK_ONCPU bit set, and we can stop the iteration there. */ - group = task_psi_group(next); - do { - if (per_cpu_ptr(group->pcpu, cpu)->state_mask & - PSI_ONCPU) { + for_each_group(group, task_psi_group(next)) { + struct psi_group_cpu *groupc = per_cpu_ptr(group->pcpu, cpu); + + if (groupc->state_mask & PSI_ONCPU) { common = group; break; } - - psi_group_change(group, cpu, 0, TSK_ONCPU, true); - } while ((group = group->parent)); + psi_group_change(group, cpu, 0, TSK_ONCPU, now, true); + } } if (prev->pid) { @@ -968,12 +981,11 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, psi_flags_change(prev, clear, set); - group = task_psi_group(prev); - do { + for_each_group(group, task_psi_group(prev)) { if (group == common) break; - psi_group_change(group, cpu, clear, set, wake_clock); - } while ((group = group->parent)); + psi_group_change(group, cpu, clear, set, now, wake_clock); + } /* * TSK_ONCPU is handled up to the common ancestor. If there are @@ -983,20 +995,21 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, */ if ((prev->psi_flags ^ next->psi_flags) & ~TSK_ONCPU) { clear &= ~TSK_ONCPU; - for (; group; group = group->parent) - psi_group_change(group, cpu, clear, set, wake_clock); + for_each_group(group, common) + psi_group_change(group, cpu, clear, set, now, wake_clock); } } + psi_write_end(cpu); } #ifdef CONFIG_IRQ_TIME_ACCOUNTING void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_struct *prev) { int cpu = task_cpu(curr); - struct psi_group *group; struct psi_group_cpu *groupc; s64 delta; u64 irq; + u64 now; if (static_branch_likely(&psi_disabled) || !irqtime_enabled()) return; @@ -1005,8 +1018,7 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st return; lockdep_assert_rq_held(rq); - group = task_psi_group(curr); - if (prev && task_psi_group(prev) == group) + if (prev && task_psi_group(prev) == task_psi_group(curr)) return; irq = irq_time_read(cpu); @@ -1015,27 +1027,24 @@ void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_st return; rq->psi_irq_time = irq; - do { - u64 now; + psi_write_begin(cpu); + now = cpu_clock(cpu); + for_each_group(group, task_psi_group(curr)) { if (!group->enabled) continue; groupc = per_cpu_ptr(group->pcpu, cpu); - write_seqcount_begin(&groupc->seq); - now = cpu_clock(cpu); - record_times(groupc, now); groupc->times[PSI_IRQ_FULL] += delta; - write_seqcount_end(&groupc->seq); - if (group->rtpoll_states & (1 << PSI_IRQ_FULL)) psi_schedule_rtpoll_work(group, 1, false); - } while ((group = group->parent)); + } + psi_write_end(cpu); } -#endif +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ /** * psi_memstall_enter - mark the beginning of a memory stall section @@ -1105,7 +1114,7 @@ int psi_cgroup_alloc(struct cgroup *cgroup) if (!static_branch_likely(&psi_cgroups_enabled)) return 0; - cgroup->psi = kzalloc(sizeof(struct psi_group), GFP_KERNEL); + cgroup->psi = kzalloc_obj(struct psi_group); if (!cgroup->psi) return -ENOMEM; @@ -1221,12 +1230,14 @@ void psi_cgroup_restart(struct psi_group *group) return; for_each_possible_cpu(cpu) { - struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; + u64 now; - rq_lock_irq(rq, &rf); - psi_group_change(group, cpu, 0, 0, true); - rq_unlock_irq(rq, &rf); + guard(rq_lock_irq)(cpu_rq(cpu)); + + psi_write_begin(cpu); + now = cpu_clock(cpu); + psi_group_change(group, cpu, 0, 0, now, true); + psi_write_end(cpu); } } #endif /* CONFIG_CGROUPS */ @@ -1329,7 +1340,7 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, char *buf, if (threshold_us == 0 || threshold_us > window_us) return ERR_PTR(-EINVAL); - t = kmalloc(sizeof(*t), GFP_KERNEL); + t = kmalloc_obj(*t); if (!t) return ERR_PTR(-ENOMEM); @@ -1440,7 +1451,7 @@ void psi_trigger_destroy(struct psi_trigger *t) group->rtpoll_task, lockdep_is_held(&group->rtpoll_trigger_lock)); rcu_assign_pointer(group->rtpoll_task, NULL); - del_timer(&group->rtpoll_timer); + timer_delete(&group->rtpoll_timer); } } mutex_unlock(&group->rtpoll_trigger_lock); @@ -1651,7 +1662,7 @@ static const struct proc_ops psi_irq_proc_ops = { .proc_poll = psi_fop_poll, .proc_release = psi_fop_release, }; -#endif +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ static int __init psi_proc_init(void) { diff --git a/kernel/sched/rq-offsets.c b/kernel/sched/rq-offsets.c new file mode 100644 index 000000000000..a23747bbe25b --- /dev/null +++ b/kernel/sched/rq-offsets.c @@ -0,0 +1,12 @@ +// SPDX-License-Identifier: GPL-2.0 +#define COMPILE_OFFSETS +#include <linux/kbuild.h> +#include <linux/types.h> +#include "sched.h" + +int main(void) +{ + DEFINE(RQ_nr_pinned, offsetof(struct rq, nr_pinned)); + + return 0; +} diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index 4b8e33c615b1..4ee8faf01441 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -4,6 +4,9 @@ * policies) */ +#include "sched.h" +#include "pelt.h" + int sched_rr_timeslice = RR_TIMESLICE; /* More than 4 hours if BW_SHIFT equals 20. */ static const u64 max_rt_runtime = MAX_BW; @@ -60,7 +63,7 @@ static int __init sched_rt_sysctl_init(void) return 0; } late_initcall(sched_rt_sysctl_init); -#endif +#endif /* CONFIG_SYSCTL */ void init_rt_rq(struct rt_rq *rt_rq) { @@ -75,12 +78,10 @@ void init_rt_rq(struct rt_rq *rt_rq) /* delimiter for bitsearch: */ __set_bit(MAX_RT_PRIO, array->bitmap); -#if defined CONFIG_SMP rt_rq->highest_prio.curr = MAX_RT_PRIO-1; rt_rq->highest_prio.next = MAX_RT_PRIO-1; rt_rq->overloaded = 0; plist_head_init(&rt_rq->pushable_tasks); -#endif /* CONFIG_SMP */ /* We start is dequeued state, because no RT tasks are queued */ rt_rq->rt_queued = 0; @@ -89,6 +90,7 @@ void init_rt_rq(struct rt_rq *rt_rq) rt_rq->rt_throttled = 0; rt_rq->rt_runtime = 0; raw_spin_lock_init(&rt_rq->rt_runtime_lock); + rt_rq->tg = &root_task_group; #endif } @@ -127,9 +129,8 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime) raw_spin_lock_init(&rt_b->rt_runtime_lock); - hrtimer_init(&rt_b->rt_period_timer, CLOCK_MONOTONIC, - HRTIMER_MODE_REL_HARD); - rt_b->rt_period_timer.function = sched_rt_period_timer; + hrtimer_setup(&rt_b->rt_period_timer, sched_rt_period_timer, CLOCK_MONOTONIC, + HRTIMER_MODE_REL_HARD); } static inline void do_start_rt_bandwidth(struct rt_bandwidth *rt_b) @@ -169,19 +170,21 @@ static void destroy_rt_bandwidth(struct rt_bandwidth *rt_b) static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) { -#ifdef CONFIG_SCHED_DEBUG WARN_ON_ONCE(!rt_entity_is_task(rt_se)); -#endif + return container_of(rt_se, struct task_struct, rt); } static inline struct rq *rq_of_rt_rq(struct rt_rq *rt_rq) { + /* Cannot fold with non-CONFIG_RT_GROUP_SCHED version, layout */ + WARN_ON(!rt_group_sched_enabled() && rt_rq->tg != &root_task_group); return rt_rq->rq; } static inline struct rt_rq *rt_rq_of_se(struct sched_rt_entity *rt_se) { + WARN_ON(!rt_group_sched_enabled() && rt_se->rt_rq->tg != &root_task_group); return rt_se->rt_rq; } @@ -189,11 +192,15 @@ static inline struct rq *rq_of_rt_se(struct sched_rt_entity *rt_se) { struct rt_rq *rt_rq = rt_se->rt_rq; + WARN_ON(!rt_group_sched_enabled() && rt_rq->tg != &root_task_group); return rt_rq->rq; } void unregister_rt_sched_group(struct task_group *tg) { + if (!rt_group_sched_enabled()) + return; + if (tg->rt_se) destroy_rt_bandwidth(&tg->rt_bandwidth); } @@ -202,6 +209,9 @@ void free_rt_sched_group(struct task_group *tg) { int i; + if (!rt_group_sched_enabled()) + return; + for_each_possible_cpu(i) { if (tg->rt_rq) kfree(tg->rt_rq[i]); @@ -246,10 +256,13 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) struct sched_rt_entity *rt_se; int i; - tg->rt_rq = kcalloc(nr_cpu_ids, sizeof(rt_rq), GFP_KERNEL); + if (!rt_group_sched_enabled()) + return 1; + + tg->rt_rq = kzalloc_objs(rt_rq, nr_cpu_ids); if (!tg->rt_rq) goto err; - tg->rt_se = kcalloc(nr_cpu_ids, sizeof(rt_se), GFP_KERNEL); + tg->rt_se = kzalloc_objs(rt_se, nr_cpu_ids); if (!tg->rt_se) goto err; @@ -279,7 +292,7 @@ err: return 0; } -#else /* CONFIG_RT_GROUP_SCHED */ +#else /* !CONFIG_RT_GROUP_SCHED: */ #define rt_entity_is_task(rt_se) (1) @@ -315,9 +328,7 @@ int alloc_rt_sched_group(struct task_group *tg, struct task_group *parent) { return 1; } -#endif /* CONFIG_RT_GROUP_SCHED */ - -#ifdef CONFIG_SMP +#endif /* !CONFIG_RT_GROUP_SCHED */ static inline bool need_pull_rt_task(struct rq *rq, struct task_struct *prev) { @@ -418,21 +429,6 @@ static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) } } -#else - -static inline void enqueue_pushable_task(struct rq *rq, struct task_struct *p) -{ -} - -static inline void dequeue_pushable_task(struct rq *rq, struct task_struct *p) -{ -} - -static inline void rt_queue_push_tasks(struct rq *rq) -{ -} -#endif /* CONFIG_SMP */ - static void enqueue_top_rt_rq(struct rt_rq *rt_rq); static void dequeue_top_rt_rq(struct rt_rq *rt_rq, unsigned int count); @@ -473,20 +469,17 @@ static inline bool rt_task_fits_capacity(struct task_struct *p, int cpu) return cpu_cap >= min(min_cap, max_cap); } -#else +#else /* !CONFIG_UCLAMP_TASK: */ static inline bool rt_task_fits_capacity(struct task_struct *p, int cpu) { return true; } -#endif +#endif /* !CONFIG_UCLAMP_TASK */ #ifdef CONFIG_RT_GROUP_SCHED static inline u64 sched_rt_runtime(struct rt_rq *rt_rq) { - if (!rt_rq->tg) - return RUNTIME_INF; - return rt_rq->rt_runtime; } @@ -499,6 +492,11 @@ typedef struct task_group *rt_rq_iter_t; static inline struct task_group *next_task_group(struct task_group *tg) { + if (!rt_group_sched_enabled()) { + WARN_ON(tg != &root_task_group); + return NULL; + } + do { tg = list_entry_rcu(tg->list.next, typeof(struct task_group), list); @@ -511,9 +509,9 @@ static inline struct task_group *next_task_group(struct task_group *tg) } #define for_each_rt_rq(rt_rq, iter, rq) \ - for (iter = container_of(&task_groups, typeof(*iter), list); \ - (iter = next_task_group(iter)) && \ - (rt_rq = iter->rt_rq[cpu_of(rq)]);) + for (iter = &root_task_group; \ + iter && (rt_rq = iter->rt_rq[cpu_of(rq)]); \ + iter = next_task_group(iter)) #define for_each_sched_rt_entity(rt_se) \ for (; rt_se; rt_se = rt_se->parent) @@ -580,17 +578,10 @@ static int rt_se_boosted(struct sched_rt_entity *rt_se) return p->prio != p->normal_prio; } -#ifdef CONFIG_SMP static inline const struct cpumask *sched_rt_period_mask(void) { return this_rq()->rd->span; } -#else -static inline const struct cpumask *sched_rt_period_mask(void) -{ - return cpu_online_mask; -} -#endif static inline struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu) @@ -611,7 +602,6 @@ bool sched_rt_bandwidth_account(struct rt_rq *rt_rq) rt_rq->rt_time < rt_b->rt_runtime); } -#ifdef CONFIG_SMP /* * We ran out of runtime, see if we can borrow some from our neighbours. */ @@ -784,9 +774,6 @@ static void balance_runtime(struct rt_rq *rt_rq) raw_spin_lock(&rt_rq->rt_runtime_lock); } } -#else /* !CONFIG_SMP */ -static inline void balance_runtime(struct rt_rq *rt_rq) {} -#endif /* CONFIG_SMP */ static int do_sched_rt_period_timer(struct rt_bandwidth *rt_b, int overrun) { @@ -916,7 +903,7 @@ static int sched_rt_runtime_exceeded(struct rt_rq *rt_rq) return 0; } -#else /* !CONFIG_RT_GROUP_SCHED */ +#else /* !CONFIG_RT_GROUP_SCHED: */ typedef struct rt_rq *rt_rq_iter_t; @@ -963,12 +950,10 @@ struct rt_rq *sched_rt_period_rt_rq(struct rt_bandwidth *rt_b, int cpu) return &cpu_rq(cpu)->rt; } -#ifdef CONFIG_SMP static void __enable_runtime(struct rq *rq) { } static void __disable_runtime(struct rq *rq) { } -#endif -#endif /* CONFIG_RT_GROUP_SCHED */ +#endif /* !CONFIG_RT_GROUP_SCHED */ static inline int rt_se_prio(struct sched_rt_entity *rt_se) { @@ -1019,7 +1004,7 @@ static void update_curr_rt(struct rq *rq) do_start_rt_bandwidth(sched_rt_bandwidth(rt_rq)); } } -#endif +#endif /* CONFIG_RT_GROUP_SCHED */ } static void @@ -1061,20 +1046,17 @@ enqueue_top_rt_rq(struct rt_rq *rt_rq) cpufreq_update_util(rq, 0); } -#if defined CONFIG_SMP - static void inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); -#ifdef CONFIG_RT_GROUP_SCHED /* * Change rq's cpupri only if rt_rq is the top queue. */ - if (&rq->rt != rt_rq) + if (IS_ENABLED(CONFIG_RT_GROUP_SCHED) && &rq->rt != rt_rq) return; -#endif + if (rq->online && prio < prev_prio) cpupri_set(&rq->rd->cpupri, rq->cpu, prio); } @@ -1084,27 +1066,16 @@ dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) { struct rq *rq = rq_of_rt_rq(rt_rq); -#ifdef CONFIG_RT_GROUP_SCHED /* * Change rq's cpupri only if rt_rq is the top queue. */ - if (&rq->rt != rt_rq) + if (IS_ENABLED(CONFIG_RT_GROUP_SCHED) && &rq->rt != rt_rq) return; -#endif + if (rq->online && rt_rq->highest_prio.curr != prev_prio) cpupri_set(&rq->rd->cpupri, rq->cpu, rt_rq->highest_prio.curr); } -#else /* CONFIG_SMP */ - -static inline -void inc_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) {} -static inline -void dec_rt_prio_smp(struct rt_rq *rt_rq, int prio, int prev_prio) {} - -#endif /* CONFIG_SMP */ - -#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED static void inc_rt_prio(struct rt_rq *rt_rq, int prio) { @@ -1143,13 +1114,6 @@ dec_rt_prio(struct rt_rq *rt_rq, int prio) dec_rt_prio_smp(rt_rq, prio, prev_prio); } -#else - -static inline void inc_rt_prio(struct rt_rq *rt_rq, int prio) {} -static inline void dec_rt_prio(struct rt_rq *rt_rq, int prio) {} - -#endif /* CONFIG_SMP || CONFIG_RT_GROUP_SCHED */ - #ifdef CONFIG_RT_GROUP_SCHED static void @@ -1158,8 +1122,7 @@ inc_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) if (rt_se_boosted(rt_se)) rt_rq->rt_nr_boosted++; - if (rt_rq->tg) - start_rt_bandwidth(&rt_rq->tg->rt_bandwidth); + start_rt_bandwidth(&rt_rq->tg->rt_bandwidth); } static void @@ -1171,7 +1134,7 @@ dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) WARN_ON(!rt_rq->rt_nr_running && rt_rq->rt_nr_boosted); } -#else /* CONFIG_RT_GROUP_SCHED */ +#else /* !CONFIG_RT_GROUP_SCHED: */ static void inc_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) @@ -1181,7 +1144,7 @@ inc_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) static inline void dec_rt_group(struct sched_rt_entity *rt_se, struct rt_rq *rt_rq) {} -#endif /* CONFIG_RT_GROUP_SCHED */ +#endif /* !CONFIG_RT_GROUP_SCHED */ static inline unsigned int rt_se_nr_running(struct sched_rt_entity *rt_se) @@ -1259,11 +1222,9 @@ static void __delist_rt_entity(struct sched_rt_entity *rt_se, struct rt_prio_arr static inline struct sched_statistics * __schedstats_from_rt_se(struct sched_rt_entity *rt_se) { -#ifdef CONFIG_RT_GROUP_SCHED /* schedstats is not supported for rt group. */ if (!rt_entity_is_task(rt_se)) return NULL; -#endif return &rt_task_of(rt_se)->stats; } @@ -1341,13 +1302,18 @@ update_stats_dequeue_rt(struct rt_rq *rt_rq, struct sched_rt_entity *rt_se, int flags) { struct task_struct *p = NULL; + struct rq *rq = rq_of_rt_rq(rt_rq); if (!schedstat_enabled()) return; - if (rt_entity_is_task(rt_se)) + if (rt_entity_is_task(rt_se)) { p = rt_task_of(rt_se); + if (p != rq->curr) + update_stats_wait_end_rt(rt_rq, rt_se); + } + if ((flags & DEQUEUE_SLEEP) && p) { unsigned int state; @@ -1479,6 +1445,9 @@ enqueue_task_rt(struct rq *rq, struct task_struct *p, int flags) enqueue_rt_entity(rt_se, flags); + if (task_is_blocked(p)) + return; + if (!task_current(rq, p) && p->nr_cpus_allowed > 1) enqueue_pushable_task(rq, p); } @@ -1526,10 +1495,9 @@ static void requeue_task_rt(struct rq *rq, struct task_struct *p, int head) static void yield_task_rt(struct rq *rq) { - requeue_task_rt(rq, rq->curr, 0); + requeue_task_rt(rq, rq->donor, 0); } -#ifdef CONFIG_SMP static int find_lowest_rq(struct task_struct *task); static int @@ -1644,7 +1612,6 @@ static int balance_rt(struct rq *rq, struct task_struct *p, struct rq_flags *rf) return sched_stop_runnable(rq) || sched_dl_runnable(rq) || sched_rt_runnable(rq); } -#endif /* CONFIG_SMP */ /* * Preempt the current task with a newly woken task if needed: @@ -1653,12 +1620,17 @@ static void wakeup_preempt_rt(struct rq *rq, struct task_struct *p, int flags) { struct task_struct *donor = rq->donor; + /* + * XXX If we're preempted by DL, queue a push? + */ + if (p->sched_class != &rt_sched_class) + return; + if (p->prio < donor->prio) { resched_curr(rq); return; } -#ifdef CONFIG_SMP /* * If: * @@ -1673,7 +1645,6 @@ static void wakeup_preempt_rt(struct rq *rq, struct task_struct *p, int flags) */ if (p->prio == donor->prio && !test_tsk_need_resched(rq->curr)) check_preempt_equal_prio(rq, p); -#endif } static inline void set_next_task_rt(struct rq *rq, struct task_struct *p, bool first) @@ -1713,7 +1684,7 @@ static struct sched_rt_entity *pick_next_rt_entity(struct rt_rq *rt_rq) BUG_ON(idx >= MAX_RT_PRIO); queue = array->queue + idx; - if (SCHED_WARN_ON(list_empty(queue))) + if (WARN_ON_ONCE(list_empty(queue))) return NULL; next = list_entry(queue->next, struct sched_rt_entity, run_list); @@ -1735,7 +1706,7 @@ static struct task_struct *_pick_next_task_rt(struct rq *rq) return rt_task_of(rt_se); } -static struct task_struct *pick_task_rt(struct rq *rq) +static struct task_struct *pick_task_rt(struct rq *rq, struct rq_flags *rf) { struct task_struct *p; @@ -1759,6 +1730,8 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p, struct task_s update_rt_rq_load_avg(rq_clock_pelt(rq), rq, 1); + if (task_is_blocked(p)) + return; /* * The previous task needs to be made eligible for pushing * if it is still active @@ -1767,8 +1740,6 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p, struct task_s enqueue_pushable_task(rq, p); } -#ifdef CONFIG_SMP - /* Only try algorithms three times */ #define RT_MAX_TRIES 3 @@ -1885,6 +1856,36 @@ static int find_lowest_rq(struct task_struct *task) return -1; } +static struct task_struct *pick_next_pushable_task(struct rq *rq) +{ + struct plist_head *head = &rq->rt.pushable_tasks; + struct task_struct *i, *p = NULL; + + if (!has_pushable_tasks(rq)) + return NULL; + + plist_for_each_entry(i, head, pushable_tasks) { + /* make sure task isn't on_cpu (possible with proxy-exec) */ + if (!task_on_cpu(rq, i)) { + p = i; + break; + } + } + + if (!p) + return NULL; + + BUG_ON(rq->cpu != task_cpu(p)); + BUG_ON(task_current(rq, p)); + BUG_ON(task_current_donor(rq, p)); + BUG_ON(p->nr_cpus_allowed <= 1); + + BUG_ON(!task_on_rq_queued(p)); + BUG_ON(!rt_task(p)); + + return p; +} + /* Will lock the rq it finds */ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) { @@ -1915,18 +1916,16 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) /* * We had to unlock the run queue. In * the mean time, task could have - * migrated already or had its affinity changed. - * Also make sure that it wasn't scheduled on its rq. + * migrated already or had its affinity changed, + * therefore check if the task is still at the + * head of the pushable tasks list. * It is possible the task was scheduled, set * "migrate_disabled" and then got preempted, so we must * check the task migration disable flag here too. */ - if (unlikely(task_rq(task) != rq || + if (unlikely(is_migration_disabled(task) || !cpumask_test_cpu(lowest_rq->cpu, &task->cpus_mask) || - task_on_cpu(rq, task) || - !rt_task(task) || - is_migration_disabled(task) || - !task_on_rq_queued(task))) { + task != pick_next_pushable_task(rq))) { double_unlock_balance(rq, lowest_rq); lowest_rq = NULL; @@ -1946,27 +1945,6 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) return lowest_rq; } -static struct task_struct *pick_next_pushable_task(struct rq *rq) -{ - struct task_struct *p; - - if (!has_pushable_tasks(rq)) - return NULL; - - p = plist_first_entry(&rq->rt.pushable_tasks, - struct task_struct, pushable_tasks); - - BUG_ON(rq->cpu != task_cpu(p)); - BUG_ON(task_current(rq, p)); - BUG_ON(task_current_donor(rq, p)); - BUG_ON(p->nr_cpus_allowed <= 1); - - BUG_ON(!task_on_rq_queued(p)); - BUG_ON(!rt_task(p)); - - return p; -} - /* * If the current CPU has more than one RT task, see if the non * running task can migrate over to a CPU that is running a task @@ -2142,6 +2120,7 @@ static void push_rt_tasks(struct rq *rq) */ static int rto_next_cpu(struct root_domain *rd) { + int this_cpu = smp_processor_id(); int next; int cpu; @@ -2165,6 +2144,10 @@ static int rto_next_cpu(struct root_domain *rd) rd->rto_cpu = cpu; + /* Do not send IPI to self */ + if (cpu == this_cpu) + continue; + if (cpu < nr_cpu_ids) return cpu; @@ -2444,7 +2427,6 @@ void __init init_sched_rt_class(void) GFP_KERNEL, cpu_to_node(i)); } } -#endif /* CONFIG_SMP */ /* * When switching a task to RT, we may overload the runqueue @@ -2468,10 +2450,8 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p) * then see if we can move to another run queue. */ if (task_on_rq_queued(p)) { -#ifdef CONFIG_SMP if (p->nr_cpus_allowed > 1 && rq->rt.overloaded) rt_queue_push_tasks(rq); -#endif /* CONFIG_SMP */ if (p->prio < rq->donor->prio && cpu_online(cpu_of(rq))) resched_curr(rq); } @@ -2482,13 +2462,15 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p) * us to initiate a push or pull. */ static void -prio_changed_rt(struct rq *rq, struct task_struct *p, int oldprio) +prio_changed_rt(struct rq *rq, struct task_struct *p, u64 oldprio) { if (!task_on_rq_queued(p)) return; + if (p->prio == oldprio) + return; + if (task_current_donor(rq, p)) { -#ifdef CONFIG_SMP /* * If our priority decreases while running, we * may need to pull tasks to this runqueue. @@ -2502,11 +2484,6 @@ prio_changed_rt(struct rq *rq, struct task_struct *p, int oldprio) */ if (p->prio > rq->rt.highest_prio.curr) resched_curr(rq); -#else - /* For UP simply resched on drop of prio */ - if (oldprio < p->prio) - resched_curr(rq); -#endif /* CONFIG_SMP */ } else { /* * This task is not running, but if it is @@ -2542,9 +2519,9 @@ static void watchdog(struct rq *rq, struct task_struct *p) } } } -#else +#else /* !CONFIG_POSIX_TIMERS: */ static inline void watchdog(struct rq *rq, struct task_struct *p) { } -#endif +#endif /* !CONFIG_POSIX_TIMERS */ /* * scheduler tick hitting a task of our scheduling class. @@ -2604,18 +2581,18 @@ static int task_is_throttled_rt(struct task_struct *p, int cpu) { struct rt_rq *rt_rq; -#ifdef CONFIG_RT_GROUP_SCHED +#ifdef CONFIG_RT_GROUP_SCHED // XXX maybe add task_rt_rq(), see also sched_rt_period_rt_rq rt_rq = task_group(p)->rt_rq[cpu]; + WARN_ON(!rt_group_sched_enabled() && rt_rq->tg != &root_task_group); #else rt_rq = &cpu_rq(cpu)->rt; #endif return rt_rq_throttled(rt_rq); } -#endif +#endif /* CONFIG_SCHED_CORE */ DEFINE_SCHED_CLASS(rt) = { - .enqueue_task = enqueue_task_rt, .dequeue_task = dequeue_task_rt, .yield_task = yield_task_rt, @@ -2626,7 +2603,6 @@ DEFINE_SCHED_CLASS(rt) = { .put_prev_task = put_prev_task_rt, .set_next_task = set_next_task_rt, -#ifdef CONFIG_SMP .balance = balance_rt, .select_task_rq = select_task_rq_rt, .set_cpus_allowed = set_cpus_allowed_common, @@ -2635,14 +2611,13 @@ DEFINE_SCHED_CLASS(rt) = { .task_woken = task_woken_rt, .switched_from = switched_from_rt, .find_lock_rq = find_lock_lowest_rq, -#endif .task_tick = task_tick_rt, .get_rr_interval = get_rr_interval_rt, - .prio_changed = prio_changed_rt, .switched_to = switched_to_rt, + .prio_changed = prio_changed_rt, .update_curr = update_curr_rt, @@ -2691,7 +2666,7 @@ static int tg_rt_schedulable(struct task_group *tg, void *data) { struct rt_schedulable_data *d = data; struct task_group *child; - unsigned long total, sum = 0; + u64 total, sum = 0; u64 period, runtime; period = ktime_to_ns(tg->rt_bandwidth.rt_period); @@ -2854,37 +2829,16 @@ long sched_group_rt_period(struct task_group *tg) return rt_period_us; } -#ifdef CONFIG_SYSCTL -static int sched_rt_global_constraints(void) -{ - int ret = 0; - - mutex_lock(&rt_constraints_mutex); - ret = __rt_schedulable(NULL, 0, 0); - mutex_unlock(&rt_constraints_mutex); - - return ret; -} -#endif /* CONFIG_SYSCTL */ - int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) { /* Don't accept real-time tasks when there is no way for them to run */ - if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) + if (rt_group_sched_enabled() && rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) return 0; return 1; } -#else /* !CONFIG_RT_GROUP_SCHED */ - -#ifdef CONFIG_SYSCTL -static int sched_rt_global_constraints(void) -{ - return 0; -} -#endif /* CONFIG_SYSCTL */ -#endif /* CONFIG_RT_GROUP_SCHED */ +#endif /* !CONFIG_RT_GROUP_SCHED */ #ifdef CONFIG_SYSCTL static int sched_rt_global_validate(void) @@ -2895,11 +2849,14 @@ static int sched_rt_global_validate(void) NSEC_PER_USEC > max_rt_runtime))) return -EINVAL; - return 0; -} +#ifdef CONFIG_RT_GROUP_SCHED + if (!rt_group_sched_enabled()) + return 0; -static void sched_rt_do_global(void) -{ + scoped_guard(mutex, &rt_constraints_mutex) + return __rt_schedulable(NULL, 0, 0); +#endif + return 0; } static int sched_rt_handler(const struct ctl_table *table, int write, void *buffer, @@ -2910,6 +2867,7 @@ static int sched_rt_handler(const struct ctl_table *table, int write, void *buff int ret; mutex_lock(&mutex); + sched_domains_mutex_lock(); old_period = sysctl_sched_rt_period; old_runtime = sysctl_sched_rt_runtime; @@ -2924,11 +2882,6 @@ static int sched_rt_handler(const struct ctl_table *table, int write, void *buff if (ret) goto undo; - ret = sched_rt_global_constraints(); - if (ret) - goto undo; - - sched_rt_do_global(); sched_dl_do_global(); } if (0) { @@ -2936,8 +2889,15 @@ undo: sysctl_sched_rt_period = old_period; sysctl_sched_rt_runtime = old_runtime; } + sched_domains_mutex_unlock(); mutex_unlock(&mutex); + /* + * After changing maximum available bandwidth for DEADLINE, we need to + * recompute per root domain and per cpus variables accordingly. + */ + rebuild_sched_domains(); + return ret; } @@ -2967,7 +2927,6 @@ static int sched_rr_handler(const struct ctl_table *table, int write, void *buff } #endif /* CONFIG_SYSCTL */ -#ifdef CONFIG_SCHED_DEBUG void print_rt_stats(struct seq_file *m, int cpu) { rt_rq_iter_t iter; @@ -2978,4 +2937,3 @@ void print_rt_stats(struct seq_file *m, int cpu) print_rt_rq(m, cpu, rt_rq); rcu_read_unlock(); } -#endif /* CONFIG_SCHED_DEBUG */ diff --git a/kernel/sched/sched-pelt.h b/kernel/sched/sched-pelt.h index c529706bed11..6803cfec7a1e 100644 --- a/kernel/sched/sched-pelt.h +++ b/kernel/sched/sched-pelt.h @@ -1,5 +1,6 @@ /* SPDX-License-Identifier: GPL-2.0 */ /* Generated by Documentation/scheduler/sched-pelt; do not modify. */ +#include <linux/types.h> static const u32 runnable_avg_yN_inv[] __maybe_unused = { 0xffffffff, 0xfa83b2da, 0xf5257d14, 0xefe4b99a, 0xeac0c6e6, 0xe5b906e6, diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 023b844159c9..9f63b15d309d 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -5,6 +5,7 @@ #ifndef _KERNEL_SCHED_SCHED_H #define _KERNEL_SCHED_SCHED_H +#include <linux/prandom.h> #include <linux/sched/affinity.h> #include <linux/sched/autogroup.h> #include <linux/sched/cpufreq.h> @@ -20,7 +21,6 @@ #include <linux/sched/task_flags.h> #include <linux/sched/task.h> #include <linux/sched/topology.h> - #include <linux/atomic.h> #include <linux/bitmap.h> #include <linux/bug.h> @@ -30,6 +30,7 @@ #include <linux/context_tracking.h> #include <linux/cpufreq.h> #include <linux/cpumask_api.h> +#include <linux/cpuset.h> #include <linux/ctype.h> #include <linux/file.h> #include <linux/fs_api.h> @@ -42,6 +43,8 @@ #include <linux/ktime_api.h> #include <linux/lockdep_api.h> #include <linux/lockdep.h> +#include <linux/memblock.h> +#include <linux/memcontrol.h> #include <linux/minmax.h> #include <linux/mm.h> #include <linux/module.h> @@ -65,10 +68,12 @@ #include <linux/types.h> #include <linux/u64_stats_sync_api.h> #include <linux/uaccess.h> +#include <linux/vmstat.h> #include <linux/wait_api.h> #include <linux/wait_bit.h> #include <linux/workqueue_api.h> #include <linux/delayacct.h> +#include <linux/mmu_context.h> #include <trace/events/power.h> #include <trace/events/sched.h> @@ -81,9 +86,8 @@ struct rt_rq; struct sched_group; struct cpuidle_state; -#ifdef CONFIG_PARAVIRT +#if defined(CONFIG_PARAVIRT) && !defined(CONFIG_HAVE_PV_STEAL_CLOCK_GEN) # include <asm/paravirt.h> -# include <asm/paravirt_api_clock.h> #endif #include <asm/barrier.h> @@ -91,12 +95,6 @@ struct cpuidle_state; #include "cpupri.h" #include "cpudeadline.h" -#ifdef CONFIG_SCHED_DEBUG -# define SCHED_WARN_ON(x) WARN_ONCE(x, #x) -#else -# define SCHED_WARN_ON(x) ({ (void)(x), 0; }) -#endif - /* task_struct::on_rq states: */ #define TASK_ON_RQ_QUEUED 1 #define TASK_ON_RQ_MIGRATING 2 @@ -358,7 +356,7 @@ extern int sched_dl_global_validate(void); extern void sched_dl_do_global(void); extern int sched_dl_overflow(struct task_struct *p, int policy, const struct sched_attr *attr); extern void __setparam_dl(struct task_struct *p, const struct sched_attr *attr); -extern void __getparam_dl(struct task_struct *p, struct sched_attr *attr); +extern void __getparam_dl(struct task_struct *p, struct sched_attr *attr, unsigned int flags); extern bool __checkparam_dl(const struct sched_attr *attr); extern bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr); extern int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial); @@ -370,30 +368,56 @@ extern s64 dl_scaled_delta_exec(struct rq *rq, struct sched_dl_entity *dl_se, s6 * * dl_se::rq -- runqueue we belong to. * - * dl_se::server_has_tasks() -- used on bandwidth enforcement; we 'stop' the - * server when it runs out of tasks to run. - * * dl_se::server_pick() -- nested pick_next_task(); we yield the period if this * returns NULL. * * dl_server_update() -- called from update_curr_common(), propagates runtime * to the server. * - * dl_server_start() - * dl_server_stop() -- start/stop the server when it has (no) tasks. + * dl_server_start() -- start the server when it has tasks; it will stop + * automatically when there are no more tasks, per + * dl_se::server_pick() returning NULL. + * + * dl_server_stop() -- (force) stop the server; use when updating + * parameters. * * dl_server_init() -- initializes the server. + * + * When started the dl_server will (per dl_defer) schedule a timer for its + * zero-laxity point -- that is, unlike regular EDF tasks which run ASAP, a + * server will run at the very end of its period. + * + * This is done such that any runtime from the target class can be accounted + * against the server -- through dl_server_update() above -- such that when it + * becomes time to run, it might already be out of runtime and get deferred + * until the next period. In this case dl_server_timer() will alternate + * between defer and replenish but never actually enqueue the server. + * + * Only when the target class does not manage to exhaust the server's runtime + * (there's actualy starvation in the given period), will the dl_server get on + * the runqueue. Once queued it will pick tasks from the target class and run + * them until either its runtime is exhaused, at which point its back to + * dl_server_timer, or until there are no more tasks to run, at which point + * the dl_server stops itself. + * + * By stopping at this point the dl_server retains bandwidth, which, if a new + * task wakes up imminently (starting the server again), can be used -- + * subject to CBS wakeup rules -- without having to wait for the next period. + * + * Additionally, because of the dl_defer behaviour the start/stop behaviour is + * naturally thottled to once per period, avoiding high context switch + * workloads from spamming the hrtimer program/cancel paths. */ +extern void dl_server_update_idle(struct sched_dl_entity *dl_se, s64 delta_exec); extern void dl_server_update(struct sched_dl_entity *dl_se, s64 delta_exec); extern void dl_server_start(struct sched_dl_entity *dl_se); extern void dl_server_stop(struct sched_dl_entity *dl_se); extern void dl_server_init(struct sched_dl_entity *dl_se, struct rq *rq, - dl_server_has_tasks_f has_tasks, dl_server_pick_f pick_task); +extern void sched_init_dl_servers(void); -extern void dl_server_update_idle_time(struct rq *rq, - struct task_struct *p); extern void fair_server_init(struct rq *rq); +extern void ext_server_init(struct rq *rq); extern void __dl_server_attach_root(struct sched_dl_entity *dl_se, struct rq *rq); extern int dl_server_apply_params(struct sched_dl_entity *dl_se, u64 runtime, u64 period, bool init); @@ -407,6 +431,19 @@ static inline bool dl_server_active(struct sched_dl_entity *dl_se) extern struct list_head task_groups; +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH +extern const u64 max_bw_quota_period_us; + +/* + * default period for group bandwidth. + * default: 0.1s, units: microseconds + */ +static inline u64 default_bw_period_us(void) +{ + return 100000ULL; +} +#endif /* CONFIG_GROUP_SCHED_BANDWIDTH */ + struct cfs_bandwidth { #ifdef CONFIG_CFS_BANDWIDTH raw_spinlock_t lock; @@ -430,7 +467,7 @@ struct cfs_bandwidth { int nr_burst; u64 throttled_time; u64 burst_time; -#endif +#endif /* CONFIG_CFS_BANDWIDTH */ }; /* Task group related information */ @@ -448,15 +485,13 @@ struct task_group { /* runqueue "owned" by this group on each CPU */ struct cfs_rq **cfs_rq; unsigned long shares; -#ifdef CONFIG_SMP /* * load_avg can be heavily contended at clock tick time, so put * it in its own cache-line separated from the fields above which * will also be accessed at each tick. */ atomic_long_t load_avg ____cacheline_aligned; -#endif -#endif +#endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED struct sched_rt_entity **rt_se; @@ -465,10 +500,7 @@ struct task_group { struct rt_bandwidth rt_bandwidth; #endif -#ifdef CONFIG_EXT_GROUP_SCHED - u32 scx_flags; /* SCX_TG_* */ - u32 scx_weight; -#endif + struct scx_task_group scx; struct rcu_head rcu; struct list_head list; @@ -537,7 +569,7 @@ extern void free_fair_sched_group(struct task_group *tg); extern int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent); extern void online_fair_sched_group(struct task_group *tg); extern void unregister_fair_sched_group(struct task_group *tg); -#else +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static inline void free_fair_sched_group(struct task_group *tg) { } static inline int alloc_fair_sched_group(struct task_group *tg, struct task_group *parent) { @@ -545,7 +577,7 @@ static inline int alloc_fair_sched_group(struct task_group *tg, struct task_grou } static inline void online_fair_sched_group(struct task_group *tg) { } static inline void unregister_fair_sched_group(struct task_group *tg) { } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ extern void init_tg_cfs_entry(struct task_group *tg, struct cfs_rq *cfs_rq, struct sched_entity *se, int cpu, @@ -579,25 +611,20 @@ extern int sched_group_set_shares(struct task_group *tg, unsigned long shares); extern int sched_group_set_idle(struct task_group *tg, long idle); -#ifdef CONFIG_SMP extern void set_task_rq_fair(struct sched_entity *se, struct cfs_rq *prev, struct cfs_rq *next); -#else /* !CONFIG_SMP */ -static inline void set_task_rq_fair(struct sched_entity *se, - struct cfs_rq *prev, struct cfs_rq *next) { } -#endif /* CONFIG_SMP */ -#else /* !CONFIG_FAIR_GROUP_SCHED */ +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static inline int sched_group_set_shares(struct task_group *tg, unsigned long shares) { return 0; } static inline int sched_group_set_idle(struct task_group *tg, long idle) { return 0; } -#endif /* CONFIG_FAIR_GROUP_SCHED */ +#endif /* !CONFIG_FAIR_GROUP_SCHED */ -#else /* CONFIG_CGROUP_SCHED */ +#else /* !CONFIG_CGROUP_SCHED: */ struct cfs_bandwidth { }; static inline bool cfs_task_bw_constrained(struct task_struct *p) { return false; } -#endif /* CONFIG_CGROUP_SCHED */ +#endif /* !CONFIG_CGROUP_SCHED */ extern void unregister_rt_sched_group(struct task_group *tg); extern void free_rt_sched_group(struct task_group *tg); @@ -647,33 +674,33 @@ struct balance_callback { void (*func)(struct rq *rq); }; -/* CFS-related fields in a runqueue */ +/* Fair scheduling SCHED_{NORMAL,BATCH,IDLE} related fields in a runqueue: */ struct cfs_rq { struct load_weight load; unsigned int nr_queued; - unsigned int h_nr_queued; /* SCHED_{NORMAL,BATCH,IDLE} */ - unsigned int h_nr_runnable; /* SCHED_{NORMAL,BATCH,IDLE} */ - unsigned int h_nr_idle; /* SCHED_IDLE */ + unsigned int h_nr_queued; /* SCHED_{NORMAL,BATCH,IDLE} */ + unsigned int h_nr_runnable; /* SCHED_{NORMAL,BATCH,IDLE} */ + unsigned int h_nr_idle; /* SCHED_IDLE */ - s64 avg_vruntime; - u64 avg_load; + s64 sum_w_vruntime; + u64 sum_weight; + u64 zero_vruntime; + unsigned int sum_shift; - u64 min_vruntime; #ifdef CONFIG_SCHED_CORE unsigned int forceidle_seq; - u64 min_vruntime_fi; + u64 zero_vruntime_fi; #endif struct rb_root_cached tasks_timeline; /* - * 'curr' points to currently running entity on this cfs_rq. + * 'curr' points to the currently running entity on this cfs_rq. * It is set to NULL otherwise (i.e when none are currently running). */ struct sched_entity *curr; struct sched_entity *next; -#ifdef CONFIG_SMP /* * CFS load tracking */ @@ -704,10 +731,7 @@ struct cfs_rq { unsigned long h_load; u64 last_h_load_update; struct sched_entity *h_load_next; -#endif /* CONFIG_FAIR_GROUP_SCHED */ -#endif /* CONFIG_SMP */ -#ifdef CONFIG_FAIR_GROUP_SCHED struct rq *rq; /* CPU runqueue to which this cfs_rq is attached */ /* @@ -720,29 +744,31 @@ struct cfs_rq { */ int on_list; struct list_head leaf_cfs_rq_list; - struct task_group *tg; /* group that "owns" this runqueue */ + struct task_group *tg; /* Group that "owns" this runqueue */ /* Locally cached copy of our task_group's idle value */ int idle; -#ifdef CONFIG_CFS_BANDWIDTH +# ifdef CONFIG_CFS_BANDWIDTH int runtime_enabled; s64 runtime_remaining; u64 throttled_pelt_idle; -#ifndef CONFIG_64BIT +# ifndef CONFIG_64BIT u64 throttled_pelt_idle_copy; -#endif +# endif u64 throttled_clock; u64 throttled_clock_pelt; u64 throttled_clock_pelt_time; u64 throttled_clock_self; u64 throttled_clock_self_time; - int throttled; + bool throttled:1; + bool pelt_clock_throttled:1; int throttle_count; struct list_head throttled_list; struct list_head throttled_csd_list; -#endif /* CONFIG_CFS_BANDWIDTH */ + struct list_head throttled_limbo_list; +# endif /* CONFIG_CFS_BANDWIDTH */ #endif /* CONFIG_FAIR_GROUP_SCHED */ }; @@ -756,10 +782,9 @@ enum scx_rq_flags { */ SCX_RQ_ONLINE = 1 << 0, SCX_RQ_CAN_STOP_TICK = 1 << 1, - SCX_RQ_BAL_PENDING = 1 << 2, /* balance hasn't run yet */ SCX_RQ_BAL_KEEP = 1 << 3, /* balance decided to keep current */ - SCX_RQ_BYPASSING = 1 << 4, SCX_RQ_CLK_VALID = 1 << 5, /* RQ clock is fresh and valid */ + SCX_RQ_BAL_CB_PENDING = 1 << 6, /* must queue a cb after dispatching */ SCX_RQ_IN_WAKEUP = 1 << 16, SCX_RQ_IN_BALANCE = 1 << 17, @@ -773,15 +798,27 @@ struct scx_rq { u64 extra_enq_flags; /* see move_task_to_local_dsq() */ u32 nr_running; u32 cpuperf_target; /* [0, SCHED_CAPACITY_SCALE] */ + bool in_select_cpu; bool cpu_released; u32 flags; + u32 nr_immed; /* ENQ_IMMED tasks on local_dsq */ u64 clock; /* current per-rq clock -- see scx_bpf_now() */ cpumask_var_t cpus_to_kick; cpumask_var_t cpus_to_kick_if_idle; cpumask_var_t cpus_to_preempt; cpumask_var_t cpus_to_wait; - unsigned long pnt_seq; + cpumask_var_t cpus_to_sync; + bool kick_sync_pending; + unsigned long kick_sync; + + struct task_struct *sub_dispatch_prev; + + raw_spinlock_t deferred_reenq_lock; + u64 deferred_reenq_locals_seq; + struct list_head deferred_reenq_locals; /* scheds requesting reenq of local DSQ */ + struct list_head deferred_reenq_users; /* user DSQs requesting reenq */ struct balance_callback deferred_bal_cb; + struct balance_callback kick_sync_bal_cb; struct irq_work deferred_irq_work; struct irq_work kick_cpus_irq_work; }; @@ -802,32 +839,28 @@ struct rt_rq { struct rt_prio_array active; unsigned int rt_nr_running; unsigned int rr_nr_running; -#if defined CONFIG_SMP || defined CONFIG_RT_GROUP_SCHED struct { int curr; /* highest queued rt task prio */ -#ifdef CONFIG_SMP int next; /* next highest */ -#endif } highest_prio; -#endif -#ifdef CONFIG_SMP bool overloaded; struct plist_head pushable_tasks; -#endif /* CONFIG_SMP */ int rt_queued; #ifdef CONFIG_RT_GROUP_SCHED int rt_throttled; - u64 rt_time; - u64 rt_runtime; + u64 rt_time; /* consumed RT time, goes up in update_curr_rt */ + u64 rt_runtime; /* allotted RT time, "slice" from rt_bandwidth, RT sharing/balancing */ /* Nests inside the rq lock: */ raw_spinlock_t rt_runtime_lock; unsigned int rt_nr_boosted; - struct rq *rq; - struct task_group *tg; + struct rq *rq; /* this is always top-level rq, cache? */ +#endif +#ifdef CONFIG_CGROUP_SCHED + struct task_group *tg; /* this tg has "this" rt_rq on given CPU for runnable entities */ #endif }; @@ -843,7 +876,6 @@ struct dl_rq { unsigned int dl_nr_running; -#ifdef CONFIG_SMP /* * Deadline values of the currently executing and the * earliest ready task on this rq. Caching these facilitates @@ -863,9 +895,7 @@ struct dl_rq { * of the leftmost (earliest deadline) element. */ struct rb_root_cached pushable_dl_tasks_root; -#else - struct dl_bw dl_bw; -#endif + /* * "Active utilization" for this runqueue: increased when a * task wakes up (becomes TASK_RUNNING) and decreased when a @@ -936,7 +966,6 @@ static inline long se_runnable(struct sched_entity *se) #endif /* !CONFIG_FAIR_GROUP_SCHED */ -#ifdef CONFIG_SMP /* * XXX we want to get rid of these helpers and use the full load resolution. */ @@ -998,7 +1027,7 @@ struct root_domain { * Also, some corner cases, like 'wrap around' is dangerous, but given * that u64 is 'big enough'. So that shouldn't be a concern. */ - u64 visit_gen; + u64 visit_cookie; #ifdef HAVE_RT_PUSH_IPI /* @@ -1012,7 +1041,7 @@ struct root_domain { /* These atomics are updated outside of a lock */ atomic_t rto_loop_next; atomic_t rto_loop_start; -#endif +#endif /* HAVE_RT_PUSH_IPI */ /* * The "RT overload" flag: it gets set if a CPU has more than * one runnable RT task. @@ -1047,7 +1076,6 @@ static inline void set_rd_overloaded(struct root_domain *rd, int status) #ifdef HAVE_RT_PUSH_IPI extern void rto_push_irq_work_func(struct irq_work *work); #endif -#endif /* CONFIG_SMP */ #ifdef CONFIG_UCLAMP_TASK /* @@ -1101,30 +1129,50 @@ DECLARE_STATIC_KEY_FALSE(sched_uclamp_used); * acquire operations must be ordered by ascending &runqueue. */ struct rq { - /* runqueue lock: */ - raw_spinlock_t __lock; - + /* + * The following members are loaded together, without holding the + * rq->lock, in an extremely hot loop in update_sg_lb_stats() + * (called from pick_next_task()). To reduce cache pollution from + * this operation, they are placed together on this dedicated cache + * line. Even though some of them are frequently modified, they are + * loaded much more frequently than they are stored. + */ unsigned int nr_running; #ifdef CONFIG_NUMA_BALANCING unsigned int nr_numa_running; unsigned int nr_preferred_running; - unsigned int numa_migrate_on; #endif + unsigned int ttwu_pending; + unsigned long cpu_capacity; +#ifdef CONFIG_SCHED_PROXY_EXEC + struct task_struct __rcu *donor; /* Scheduling context */ + struct task_struct __rcu *curr; /* Execution context */ +#else + union { + struct task_struct __rcu *donor; /* Scheduler context */ + struct task_struct __rcu *curr; /* Execution context */ + }; +#endif + struct task_struct *idle; + /* padding left here deliberately */ + + /* + * The next cacheline holds the (hot) runqueue lock, as well as + * some other less performance-critical fields. + */ + u64 nr_switches ____cacheline_aligned; + + /* runqueue lock: */ + raw_spinlock_t __lock; + #ifdef CONFIG_NO_HZ_COMMON -#ifdef CONFIG_SMP - unsigned long last_blocked_load_update_tick; - unsigned int has_blocked_load; - call_single_data_t nohz_csd; -#endif /* CONFIG_SMP */ unsigned int nohz_tick_stopped; atomic_t nohz_flags; + unsigned int has_blocked_load; + unsigned long last_blocked_load_update_tick; + call_single_data_t nohz_csd; #endif /* CONFIG_NO_HZ_COMMON */ -#ifdef CONFIG_SMP - unsigned int ttwu_pending; -#endif - u64 nr_switches; - #ifdef CONFIG_UCLAMP_TASK /* Utilization clamp values based on CPU's RUNNABLE tasks */ struct uclamp_rq uclamp[UCLAMP_CNT] ____cacheline_aligned; @@ -1137,6 +1185,7 @@ struct rq { struct dl_rq dl; #ifdef CONFIG_SCHED_CLASS_EXT struct scx_rq scx; + struct sched_dl_entity ext_server; #endif struct sched_dl_entity fair_server; @@ -1147,54 +1196,50 @@ struct rq { struct list_head *tmp_alone_branch; #endif /* CONFIG_FAIR_GROUP_SCHED */ +#ifdef CONFIG_NUMA_BALANCING + unsigned int numa_migrate_on; +#endif /* * This is part of a global counter where only the total sum * over all CPUs matters. A task can increase this counter on * one CPU and if it got migrated afterwards it may decrease * it on another CPU. Always updated under the runqueue lock: */ - unsigned int nr_uninterruptible; + unsigned long nr_uninterruptible; - union { - struct task_struct __rcu *donor; /* Scheduler context */ - struct task_struct __rcu *curr; /* Execution context */ - }; struct sched_dl_entity *dl_server; - struct task_struct *idle; struct task_struct *stop; + const struct sched_class *next_class; unsigned long next_balance; struct mm_struct *prev_mm; - unsigned int clock_update_flags; - u64 clock; - /* Ensure that all clocks are in the same cache line */ + /* + * The following fields of clock data are frequently referenced + * and updated together, and should go on their own cache line. + */ u64 clock_task ____cacheline_aligned; u64 clock_pelt; + u64 clock; unsigned long lost_idle_time; + unsigned int clock_update_flags; u64 clock_pelt_idle; u64 clock_idle; + #ifndef CONFIG_64BIT u64 clock_pelt_idle_copy; u64 clock_idle_copy; #endif - atomic_t nr_iowait; - -#ifdef CONFIG_SCHED_DEBUG u64 last_seen_need_resched_ns; int ticks_without_resched; -#endif #ifdef CONFIG_MEMBARRIER int membarrier_state; #endif -#ifdef CONFIG_SMP struct root_domain *rd; struct sched_domain __rcu *sd; - unsigned long cpu_capacity; - struct balance_callback *balance_callback; unsigned char nohz_idle_balance; @@ -1230,7 +1275,6 @@ struct rq { #ifdef CONFIG_HOTPLUG_CPU struct rcuwait hotplug_wait; #endif -#endif /* CONFIG_SMP */ #ifdef CONFIG_IRQ_TIME_ACCOUNTING u64 prev_irq_time; @@ -1248,11 +1292,11 @@ struct rq { long calc_load_active; #ifdef CONFIG_SCHED_HRTICK -#ifdef CONFIG_SMP call_single_data_t hrtick_csd; -#endif struct hrtimer hrtick_timer; ktime_t hrtick_time; + ktime_t hrtick_delay; + unsigned int hrtick_sched; #endif #ifdef CONFIG_SCHEDSTATS @@ -1277,9 +1321,7 @@ struct rq { struct cpuidle_state *idle_state; #endif -#ifdef CONFIG_SMP unsigned int nr_pinned; -#endif unsigned int push_busy; struct cpu_stop_work push_work; @@ -1300,16 +1342,18 @@ struct rq { unsigned int core_forceidle_seq; unsigned int core_forceidle_occupation; u64 core_forceidle_start; -#endif +#endif /* CONFIG_SCHED_CORE */ /* Scratch cpumask to be temporarily used under rq_lock */ cpumask_var_t scratch_mask; -#if defined(CONFIG_CFS_BANDWIDTH) && defined(CONFIG_SMP) +#ifdef CONFIG_CFS_BANDWIDTH call_single_data_t cfsb_csd; struct list_head cfsb_csd_list; #endif -}; + + atomic_t nr_iowait; +} __no_randomize_layout; #ifdef CONFIG_FAIR_GROUP_SCHED @@ -1319,46 +1363,78 @@ static inline struct rq *rq_of(struct cfs_rq *cfs_rq) return cfs_rq->rq; } -#else +#else /* !CONFIG_FAIR_GROUP_SCHED: */ static inline struct rq *rq_of(struct cfs_rq *cfs_rq) { return container_of(cfs_rq, struct rq, cfs); } -#endif +#endif /* !CONFIG_FAIR_GROUP_SCHED */ static inline int cpu_of(struct rq *rq) { -#ifdef CONFIG_SMP return rq->cpu; -#else - return 0; -#endif } #define MDF_PUSH 0x01 static inline bool is_migration_disabled(struct task_struct *p) { -#ifdef CONFIG_SMP return p->migration_disabled; -#else - return false; -#endif } DECLARE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); +DECLARE_PER_CPU(struct rnd_state, sched_rnd_state); + +static inline u32 sched_rng(void) +{ + return prandom_u32_state(this_cpu_ptr(&sched_rnd_state)); +} + +static __always_inline struct rq *__this_rq(void) +{ + return this_cpu_ptr(&runqueues); +} #define cpu_rq(cpu) (&per_cpu(runqueues, (cpu))) -#define this_rq() this_cpu_ptr(&runqueues) +#define this_rq() __this_rq() #define task_rq(p) cpu_rq(task_cpu(p)) #define cpu_curr(cpu) (cpu_rq(cpu)->curr) #define raw_rq() raw_cpu_ptr(&runqueues) +static inline bool idle_rq(struct rq *rq) +{ + return rq->curr == rq->idle && !rq->nr_running && !rq->ttwu_pending; +} + +/** + * available_idle_cpu - is a given CPU idle for enqueuing work. + * @cpu: the CPU in question. + * + * Return: 1 if the CPU is currently idle. 0 otherwise. + */ +static inline bool available_idle_cpu(int cpu) +{ + if (!idle_rq(cpu_rq(cpu))) + return 0; + + if (vcpu_is_preempted(cpu)) + return 0; + + return 1; +} + +#ifdef CONFIG_SCHED_PROXY_EXEC +static inline void rq_set_donor(struct rq *rq, struct task_struct *t) +{ + rcu_assign_pointer(rq->donor, t); +} +#else static inline void rq_set_donor(struct rq *rq, struct task_struct *t) { /* Do nothing */ } +#endif #ifdef CONFIG_SCHED_CORE static inline struct cpumask *sched_group_span(struct sched_group *sg); @@ -1388,6 +1464,7 @@ static inline raw_spinlock_t *rq_lockp(struct rq *rq) } static inline raw_spinlock_t *__rq_lockp(struct rq *rq) + __returns_ctx_lock(rq_lockp(rq)) /* alias them */ { if (rq->core_enabled) return &rq->core->__lock; @@ -1424,6 +1501,9 @@ static inline bool sched_core_cookie_match(struct rq *rq, struct task_struct *p) if (!sched_core_enabled(rq)) return true; + if (rq->core->core_cookie == p->core_cookie) + return true; + for_each_cpu(cpu, cpu_smt_mask(cpu_of(rq))) { if (!available_idle_cpu(cpu)) { idle_core = false; @@ -1435,7 +1515,7 @@ static inline bool sched_core_cookie_match(struct rq *rq, struct task_struct *p) * A CPU in an idle core is always the best choice for tasks with * cookies. */ - return idle_core || rq->core->core_cookie == p->core_cookie; + return idle_core; } static inline bool sched_group_cookie_match(struct rq *rq, @@ -1484,6 +1564,7 @@ static inline raw_spinlock_t *rq_lockp(struct rq *rq) } static inline raw_spinlock_t *__rq_lockp(struct rq *rq) + __returns_ctx_lock(rq_lockp(rq)) /* alias them */ { return &rq->__lock; } @@ -1507,33 +1588,64 @@ static inline bool sched_group_cookie_match(struct rq *rq, #endif /* !CONFIG_SCHED_CORE */ +#ifdef CONFIG_RT_GROUP_SCHED +# ifdef CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED +DECLARE_STATIC_KEY_FALSE(rt_group_sched); +static inline bool rt_group_sched_enabled(void) +{ + return static_branch_unlikely(&rt_group_sched); +} +# else /* !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED: */ +DECLARE_STATIC_KEY_TRUE(rt_group_sched); +static inline bool rt_group_sched_enabled(void) +{ + return static_branch_likely(&rt_group_sched); +} +# endif /* !CONFIG_RT_GROUP_SCHED_DEFAULT_DISABLED */ +#else /* !CONFIG_RT_GROUP_SCHED: */ +# define rt_group_sched_enabled() false +#endif /* !CONFIG_RT_GROUP_SCHED */ + static inline void lockdep_assert_rq_held(struct rq *rq) + __assumes_ctx_lock(__rq_lockp(rq)) { lockdep_assert_held(__rq_lockp(rq)); } -extern void raw_spin_rq_lock_nested(struct rq *rq, int subclass); -extern bool raw_spin_rq_trylock(struct rq *rq); -extern void raw_spin_rq_unlock(struct rq *rq); +extern void raw_spin_rq_lock_nested(struct rq *rq, int subclass) + __acquires(__rq_lockp(rq)); + +extern bool raw_spin_rq_trylock(struct rq *rq) + __cond_acquires(true, __rq_lockp(rq)); static inline void raw_spin_rq_lock(struct rq *rq) + __acquires(__rq_lockp(rq)) { raw_spin_rq_lock_nested(rq, 0); } +static inline void raw_spin_rq_unlock(struct rq *rq) + __releases(__rq_lockp(rq)) +{ + raw_spin_unlock(rq_lockp(rq)); +} + static inline void raw_spin_rq_lock_irq(struct rq *rq) + __acquires(__rq_lockp(rq)) { local_irq_disable(); raw_spin_rq_lock(rq); } static inline void raw_spin_rq_unlock_irq(struct rq *rq) + __releases(__rq_lockp(rq)) { raw_spin_rq_unlock(rq); local_irq_enable(); } static inline unsigned long _raw_spin_rq_lock_irqsave(struct rq *rq) + __acquires(__rq_lockp(rq)) { unsigned long flags; @@ -1544,6 +1656,7 @@ static inline unsigned long _raw_spin_rq_lock_irqsave(struct rq *rq) } static inline void raw_spin_rq_unlock_irqrestore(struct rq *rq, unsigned long flags) + __releases(__rq_lockp(rq)) { raw_spin_rq_unlock(rq); local_irq_restore(flags); @@ -1563,15 +1676,15 @@ static inline void update_idle_core(struct rq *rq) __update_idle_core(rq); } -#else +#else /* !CONFIG_SCHED_SMT: */ static inline void update_idle_core(struct rq *rq) { } -#endif +#endif /* !CONFIG_SCHED_SMT */ #ifdef CONFIG_FAIR_GROUP_SCHED static inline struct task_struct *task_of(struct sched_entity *se) { - SCHED_WARN_ON(!entity_is_task(se)); + WARN_ON_ONCE(!entity_is_task(se)); return container_of(se, struct task_struct, se); } @@ -1617,6 +1730,7 @@ static inline struct cfs_rq *group_cfs_rq(struct sched_entity *grp) #endif /* !CONFIG_FAIR_GROUP_SCHED */ +extern void update_rq_avg_idle(struct rq *rq); extern void update_rq_clock(struct rq *rq); /* @@ -1652,7 +1766,7 @@ static inline void assert_clock_updated(struct rq *rq) * The only reason for not seeing a clock update since the * last rq_pin_lock() is if we're currently skipping updates. */ - SCHED_WARN_ON(rq->clock_update_flags < RQCF_ACT_SKIP); + WARN_ON_ONCE(rq->clock_update_flags < RQCF_ACT_SKIP); } static inline u64 rq_clock(struct rq *rq) @@ -1699,7 +1813,7 @@ static inline void rq_clock_cancel_skipupdate(struct rq *rq) static inline void rq_clock_start_loop_update(struct rq *rq) { lockdep_assert_rq_held(rq); - SCHED_WARN_ON(rq->clock_update_flags & RQCF_ACT_SKIP); + WARN_ON_ONCE(rq->clock_update_flags & RQCF_ACT_SKIP); rq->clock_update_flags |= RQCF_ACT_SKIP; } @@ -1712,14 +1826,12 @@ static inline void rq_clock_stop_loop_update(struct rq *rq) struct rq_flags { unsigned long flags; struct pin_cookie cookie; -#ifdef CONFIG_SCHED_DEBUG /* * A copy of (rq::clock_update_flags & RQCF_UPDATED) for the * current pin context is stashed here in case it needs to be * restored in rq_repin_lock(). */ unsigned int clock_update_flags; -#endif }; extern struct balance_callback balance_push_callback; @@ -1727,10 +1839,10 @@ extern struct balance_callback balance_push_callback; #ifdef CONFIG_SCHED_CLASS_EXT extern const struct sched_class ext_sched_class; -DECLARE_STATIC_KEY_FALSE(__scx_ops_enabled); /* SCX BPF scheduler loaded */ +DECLARE_STATIC_KEY_FALSE(__scx_enabled); /* SCX BPF scheduler loaded */ DECLARE_STATIC_KEY_FALSE(__scx_switched_all); /* all fair class tasks on SCX */ -#define scx_enabled() static_branch_unlikely(&__scx_ops_enabled) +#define scx_enabled() static_branch_unlikely(&__scx_enabled) #define scx_switched_all() static_branch_unlikely(&__scx_switched_all) static inline void scx_rq_clock_update(struct rq *rq, u64 clock) @@ -1748,7 +1860,7 @@ static inline void scx_rq_clock_invalidate(struct rq *rq) WRITE_ONCE(rq->scx.flags, rq->scx.flags & ~SCX_RQ_CLK_VALID); } -#else /* !CONFIG_SCHED_CLASS_EXT */ +#else /* !CONFIG_SCHED_CLASS_EXT: */ #define scx_enabled() false #define scx_switched_all() false @@ -1756,6 +1868,13 @@ static inline void scx_rq_clock_update(struct rq *rq, u64 clock) {} static inline void scx_rq_clock_invalidate(struct rq *rq) {} #endif /* !CONFIG_SCHED_CLASS_EXT */ +static inline void assert_balance_callbacks_empty(struct rq *rq) +{ + WARN_ON_ONCE(IS_ENABLED(CONFIG_PROVE_LOCKING) && + rq->balance_callback && + rq->balance_callback != &balance_push_callback); +} + /* * Lockdep annotation that avoids accidental unlocks; it's like a * sticky/continuous lockdep_assert_held(). @@ -1770,21 +1889,16 @@ static inline void rq_pin_lock(struct rq *rq, struct rq_flags *rf) { rf->cookie = lockdep_pin_lock(__rq_lockp(rq)); -#ifdef CONFIG_SCHED_DEBUG rq->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP); rf->clock_update_flags = 0; -# ifdef CONFIG_SMP - SCHED_WARN_ON(rq->balance_callback && rq->balance_callback != &balance_push_callback); -# endif -#endif + assert_balance_callbacks_empty(rq); } static inline void rq_unpin_lock(struct rq *rq, struct rq_flags *rf) { -#ifdef CONFIG_SCHED_DEBUG if (rq->clock_update_flags > RQCF_ACT_SKIP) rf->clock_update_flags = RQCF_UPDATED; -#endif + scx_rq_clock_invalidate(rq); lockdep_unpin_lock(__rq_lockp(rq), rf->cookie); } @@ -1793,25 +1907,22 @@ static inline void rq_repin_lock(struct rq *rq, struct rq_flags *rf) { lockdep_repin_lock(__rq_lockp(rq), rf->cookie); -#ifdef CONFIG_SCHED_DEBUG /* * Restore the value we stashed in @rf for this pin context. */ rq->clock_update_flags |= rf->clock_update_flags; -#endif } -extern -struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) - __acquires(rq->lock); +#define __task_rq_lock(...) __acquire_ret(___task_rq_lock(__VA_ARGS__), __rq_lockp(__ret)) +extern struct rq *___task_rq_lock(struct task_struct *p, struct rq_flags *rf) __acquires_ret; -extern -struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) - __acquires(p->pi_lock) - __acquires(rq->lock); +#define task_rq_lock(...) __acquire_ret(_task_rq_lock(__VA_ARGS__), __rq_lockp(__ret)) +extern struct rq *_task_rq_lock(struct task_struct *p, struct rq_flags *rf) + __acquires(&p->pi_lock) __acquires_ret; -static inline void __task_rq_unlock(struct rq *rq, struct rq_flags *rf) - __releases(rq->lock) +static inline void +__task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf) + __releases(__rq_lockp(rq)) { rq_unpin_lock(rq, rf); raw_spin_rq_unlock(rq); @@ -1819,11 +1930,9 @@ static inline void __task_rq_unlock(struct rq *rq, struct rq_flags *rf) static inline void task_rq_unlock(struct rq *rq, struct task_struct *p, struct rq_flags *rf) - __releases(rq->lock) - __releases(p->pi_lock) + __releases(__rq_lockp(rq), &p->pi_lock) { - rq_unpin_lock(rq, rf); - raw_spin_rq_unlock(rq); + __task_rq_unlock(rq, p, rf); raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags); } @@ -1831,44 +1940,51 @@ DEFINE_LOCK_GUARD_1(task_rq_lock, struct task_struct, _T->rq = task_rq_lock(_T->lock, &_T->rf), task_rq_unlock(_T->rq, _T->lock, &_T->rf), struct rq *rq; struct rq_flags rf) +DECLARE_LOCK_GUARD_1_ATTRS(task_rq_lock, __acquires(_T->pi_lock), __releases((*(struct task_struct **)_T)->pi_lock)) +#define class_task_rq_lock_constructor(_T) WITH_LOCK_GUARD_1_ATTRS(task_rq_lock, _T) + +DEFINE_LOCK_GUARD_1(__task_rq_lock, struct task_struct, + _T->rq = __task_rq_lock(_T->lock, &_T->rf), + __task_rq_unlock(_T->rq, _T->lock, &_T->rf), + struct rq *rq; struct rq_flags rf) static inline void rq_lock_irqsave(struct rq *rq, struct rq_flags *rf) - __acquires(rq->lock) + __acquires(__rq_lockp(rq)) { raw_spin_rq_lock_irqsave(rq, rf->flags); rq_pin_lock(rq, rf); } static inline void rq_lock_irq(struct rq *rq, struct rq_flags *rf) - __acquires(rq->lock) + __acquires(__rq_lockp(rq)) { raw_spin_rq_lock_irq(rq); rq_pin_lock(rq, rf); } static inline void rq_lock(struct rq *rq, struct rq_flags *rf) - __acquires(rq->lock) + __acquires(__rq_lockp(rq)) { raw_spin_rq_lock(rq); rq_pin_lock(rq, rf); } static inline void rq_unlock_irqrestore(struct rq *rq, struct rq_flags *rf) - __releases(rq->lock) + __releases(__rq_lockp(rq)) { rq_unpin_lock(rq, rf); raw_spin_rq_unlock_irqrestore(rq, rf->flags); } static inline void rq_unlock_irq(struct rq *rq, struct rq_flags *rf) - __releases(rq->lock) + __releases(__rq_lockp(rq)) { rq_unpin_lock(rq, rf); raw_spin_rq_unlock_irq(rq); } static inline void rq_unlock(struct rq *rq, struct rq_flags *rf) - __releases(rq->lock) + __releases(__rq_lockp(rq)) { rq_unpin_lock(rq, rf); raw_spin_rq_unlock(rq); @@ -1879,18 +1995,27 @@ DEFINE_LOCK_GUARD_1(rq_lock, struct rq, rq_unlock(_T->lock, &_T->rf), struct rq_flags rf) +DECLARE_LOCK_GUARD_1_ATTRS(rq_lock, __acquires(__rq_lockp(_T)), __releases(__rq_lockp(*(struct rq **)_T))); +#define class_rq_lock_constructor(_T) WITH_LOCK_GUARD_1_ATTRS(rq_lock, _T) + DEFINE_LOCK_GUARD_1(rq_lock_irq, struct rq, rq_lock_irq(_T->lock, &_T->rf), rq_unlock_irq(_T->lock, &_T->rf), struct rq_flags rf) +DECLARE_LOCK_GUARD_1_ATTRS(rq_lock_irq, __acquires(__rq_lockp(_T)), __releases(__rq_lockp(*(struct rq **)_T))); +#define class_rq_lock_irq_constructor(_T) WITH_LOCK_GUARD_1_ATTRS(rq_lock_irq, _T) + DEFINE_LOCK_GUARD_1(rq_lock_irqsave, struct rq, rq_lock_irqsave(_T->lock, &_T->rf), rq_unlock_irqrestore(_T->lock, &_T->rf), struct rq_flags rf) -static inline struct rq *this_rq_lock_irq(struct rq_flags *rf) - __acquires(rq->lock) +DECLARE_LOCK_GUARD_1_ATTRS(rq_lock_irqsave, __acquires(__rq_lockp(_T)), __releases(__rq_lockp(*(struct rq **)_T))); +#define class_rq_lock_irqsave_constructor(_T) WITH_LOCK_GUARD_1_ATTRS(rq_lock_irqsave, _T) + +#define this_rq_lock_irq(...) __acquire_ret(_this_rq_lock_irq(__VA_ARGS__), __rq_lockp(__ret)) +static inline struct rq *_this_rq_lock_irq(struct rq_flags *rf) __acquires_ret { struct rq *rq; @@ -1946,19 +2071,17 @@ extern void sched_setnuma(struct task_struct *p, int node); extern int migrate_task_to(struct task_struct *p, int cpu); extern int migrate_swap(struct task_struct *p, struct task_struct *t, int cpu, int scpu); -extern void init_numa_balancing(unsigned long clone_flags, struct task_struct *p); +extern void init_numa_balancing(u64 clone_flags, struct task_struct *p); #else /* !CONFIG_NUMA_BALANCING: */ static inline void -init_numa_balancing(unsigned long clone_flags, struct task_struct *p) +init_numa_balancing(u64 clone_flags, struct task_struct *p) { } #endif /* !CONFIG_NUMA_BALANCING */ -#ifdef CONFIG_SMP - static inline void queue_balance_callback(struct rq *rq, struct balance_callback *head, @@ -1979,8 +2102,8 @@ queue_balance_callback(struct rq *rq, rq->balance_callback = head; } -#define rcu_dereference_check_sched_domain(p) \ - rcu_dereference_check((p), lockdep_is_held(&sched_domains_mutex)) +#define rcu_dereference_sched_domain(p) \ + rcu_dereference_all_check((p), lockdep_is_held(&sched_domains_mutex)) /* * The domain tree (rq->sd) is protected by RCU's quiescent state transition. @@ -1990,7 +2113,7 @@ queue_balance_callback(struct rq *rq, * preempt-disabled sections. */ #define for_each_domain(cpu, __sd) \ - for (__sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd); \ + for (__sd = rcu_dereference_sched_domain(cpu_rq(cpu)->sd); \ __sd; __sd = __sd->parent) /* A mask of all the SD flags that have the SDF_SHARED_CHILD metaflag */ @@ -2072,9 +2195,7 @@ struct sched_group_capacity { unsigned long next_update; int imbalance; /* XXX unrelated to capacity but shared group state */ -#ifdef CONFIG_SCHED_DEBUG int id; -#endif unsigned long cpumask[]; /* Balance mask */ }; @@ -2114,13 +2235,8 @@ static inline struct cpumask *group_balance_mask(struct sched_group *sg) extern int group_balance_cpu(struct sched_group *sg); -#ifdef CONFIG_SCHED_DEBUG extern void update_sched_domain_debugfs(void); extern void dirty_sched_domain_sysctl(int cpu); -#else -static inline void update_sched_domain_debugfs(void) { } -static inline void dirty_sched_domain_sysctl(int cpu) { } -#endif extern int sched_update_scaling(void); @@ -2131,8 +2247,6 @@ static inline const struct cpumask *task_user_cpus(struct task_struct *p) return p->user_cpus_ptr; } -#endif /* CONFIG_SMP */ - #ifdef CONFIG_CGROUP_SCHED /* @@ -2168,9 +2282,16 @@ static inline void set_task_rq(struct task_struct *p, unsigned int cpu) #endif #ifdef CONFIG_RT_GROUP_SCHED + /* + * p->rt.rt_rq is NULL initially and it is easier to assign + * root_task_group's rt_rq than switching in rt_rq_of_se() + * Clobbers tg(!) + */ + if (!rt_group_sched_enabled()) + tg = &root_task_group; p->rt.rt_rq = tg->rt_rq[cpu]; p->rt.parent = tg->rt_se[cpu]; -#endif +#endif /* CONFIG_RT_GROUP_SCHED */ } #else /* !CONFIG_CGROUP_SCHED: */ @@ -2196,17 +2317,13 @@ static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu) smp_wmb(); WRITE_ONCE(task_thread_info(p)->cpu, cpu); p->wake_cpu = cpu; -#endif + rseq_sched_set_ids_changed(p); +#endif /* CONFIG_SMP */ } /* - * Tunables that become constants when CONFIG_SCHED_DEBUG is off: + * Tunables: */ -#ifdef CONFIG_SCHED_DEBUG -# define const_debug __read_mostly -#else -# define const_debug const -#endif #define SCHED_FEAT(name, enabled) \ __SCHED_FEAT_##name , @@ -2218,13 +2335,11 @@ enum { #undef SCHED_FEAT -#ifdef CONFIG_SCHED_DEBUG - /* * To support run-time toggling of sched features, all the translation units * (but core.c) reference the sysctl_sched_features defined in core.c. */ -extern const_debug unsigned int sysctl_sched_features; +extern __read_mostly unsigned int sysctl_sched_features; #ifdef CONFIG_JUMP_LABEL @@ -2246,24 +2361,6 @@ extern struct static_key sched_feat_keys[__SCHED_FEAT_NR]; #endif /* !CONFIG_JUMP_LABEL */ -#else /* !SCHED_DEBUG: */ - -/* - * Each translation unit has its own copy of sysctl_sched_features to allow - * constants propagation at compile time and compiler optimization based on - * features default. - */ -#define SCHED_FEAT(name, enabled) \ - (1UL << __SCHED_FEAT_##name) * enabled | -static const_debug __maybe_unused unsigned int sysctl_sched_features = -#include "features.h" - 0; -#undef SCHED_FEAT - -#define sched_feat(x) !!(sysctl_sched_features & (1UL << __SCHED_FEAT_##x)) - -#endif /* !SCHED_DEBUG */ - extern struct static_key_false sched_numa_balancing; extern struct static_key_false sched_schedstats; @@ -2299,13 +2396,17 @@ static inline int task_current_donor(struct rq *rq, struct task_struct *p) return rq->donor == p; } +static inline bool task_is_blocked(struct task_struct *p) +{ + if (!sched_proxy_exec()) + return false; + + return !!p->blocked_on; +} + static inline int task_on_cpu(struct rq *rq, struct task_struct *p) { -#ifdef CONFIG_SMP return p->on_cpu; -#else - return task_current(rq, p); -#endif } static inline int task_on_rq_queued(struct task_struct *p) @@ -2328,11 +2429,9 @@ static inline int task_on_rq_migrating(struct task_struct *p) #define WF_CURRENT_CPU 0x40 /* Prefer to move the wakee to the current CPU. */ #define WF_RQ_SELECTED 0x80 /* ->select_task_rq() was called */ -#ifdef CONFIG_SMP static_assert(WF_EXEC == SD_BALANCE_EXEC); static_assert(WF_FORK == SD_BALANCE_FORK); static_assert(WF_TTWU == SD_BALANCE_WAKE); -#endif /* * To aid in avoiding the subversion of "niceness" due to uneven distribution @@ -2352,51 +2451,60 @@ extern const u32 sched_prio_to_wmult[40]; /* * {de,en}queue flags: * - * DEQUEUE_SLEEP - task is no longer runnable - * ENQUEUE_WAKEUP - task just became runnable + * SLEEP/WAKEUP - task is no-longer/just-became runnable * * SAVE/RESTORE - an otherwise spurious dequeue/enqueue, done to ensure tasks * are in a known state which allows modification. Such pairs * should preserve as much state as possible. * * MOVE - paired with SAVE/RESTORE, explicitly does not preserve the location - * in the runqueue. + * in the runqueue. IOW the priority is allowed to change. Callers + * must expect to deal with balance callbacks. * * NOCLOCK - skip the update_rq_clock() (avoids double updates) * * MIGRATION - p->on_rq == TASK_ON_RQ_MIGRATING (used for DEADLINE) * + * DELAYED - de/re-queue a sched_delayed task + * + * CLASS - going to update p->sched_class; makes sched_change call the + * various switch methods. + * * ENQUEUE_HEAD - place at front of runqueue (tail if not specified) * ENQUEUE_REPLENISH - CBS (replenish runtime and postpone deadline) * ENQUEUE_MIGRATED - the task was migrated during wakeup * ENQUEUE_RQ_SELECTED - ->select_task_rq() was called * + * XXX SAVE/RESTORE in combination with CLASS doesn't really make sense, but + * SCHED_DEADLINE seems to rely on this for now. */ -#define DEQUEUE_SLEEP 0x01 /* Matches ENQUEUE_WAKEUP */ -#define DEQUEUE_SAVE 0x02 /* Matches ENQUEUE_RESTORE */ -#define DEQUEUE_MOVE 0x04 /* Matches ENQUEUE_MOVE */ -#define DEQUEUE_NOCLOCK 0x08 /* Matches ENQUEUE_NOCLOCK */ -#define DEQUEUE_SPECIAL 0x10 -#define DEQUEUE_MIGRATING 0x100 /* Matches ENQUEUE_MIGRATING */ -#define DEQUEUE_DELAYED 0x200 /* Matches ENQUEUE_DELAYED */ - -#define ENQUEUE_WAKEUP 0x01 -#define ENQUEUE_RESTORE 0x02 -#define ENQUEUE_MOVE 0x04 -#define ENQUEUE_NOCLOCK 0x08 - -#define ENQUEUE_HEAD 0x10 -#define ENQUEUE_REPLENISH 0x20 -#ifdef CONFIG_SMP -#define ENQUEUE_MIGRATED 0x40 -#else -#define ENQUEUE_MIGRATED 0x00 -#endif -#define ENQUEUE_INITIAL 0x80 -#define ENQUEUE_MIGRATING 0x100 -#define ENQUEUE_DELAYED 0x200 -#define ENQUEUE_RQ_SELECTED 0x400 +#define DEQUEUE_SLEEP 0x0001 /* Matches ENQUEUE_WAKEUP */ +#define DEQUEUE_SAVE 0x0002 /* Matches ENQUEUE_RESTORE */ +#define DEQUEUE_MOVE 0x0004 /* Matches ENQUEUE_MOVE */ +#define DEQUEUE_NOCLOCK 0x0008 /* Matches ENQUEUE_NOCLOCK */ + +#define DEQUEUE_MIGRATING 0x0010 /* Matches ENQUEUE_MIGRATING */ +#define DEQUEUE_DELAYED 0x0020 /* Matches ENQUEUE_DELAYED */ +#define DEQUEUE_CLASS 0x0040 /* Matches ENQUEUE_CLASS */ + +#define DEQUEUE_SPECIAL 0x00010000 +#define DEQUEUE_THROTTLE 0x00020000 + +#define ENQUEUE_WAKEUP 0x0001 +#define ENQUEUE_RESTORE 0x0002 +#define ENQUEUE_MOVE 0x0004 +#define ENQUEUE_NOCLOCK 0x0008 + +#define ENQUEUE_MIGRATING 0x0010 +#define ENQUEUE_DELAYED 0x0020 +#define ENQUEUE_CLASS 0x0040 + +#define ENQUEUE_HEAD 0x00010000 +#define ENQUEUE_REPLENISH 0x00020000 +#define ENQUEUE_MIGRATED 0x00040000 +#define ENQUEUE_INITIAL 0x00080000 +#define ENQUEUE_RQ_SELECTED 0x00100000 #define RETRY_TASK ((void *)-1UL) @@ -2414,15 +2522,51 @@ struct sched_class { int uclamp_enabled; #endif + /* + * move_queued_task/activate_task/enqueue_task: rq->lock + * ttwu_do_activate/activate_task/enqueue_task: rq->lock + * wake_up_new_task/activate_task/enqueue_task: task_rq_lock + * ttwu_runnable/enqueue_task: task_rq_lock + * proxy_task_current: rq->lock + * sched_change_end + */ void (*enqueue_task) (struct rq *rq, struct task_struct *p, int flags); + /* + * move_queued_task/deactivate_task/dequeue_task: rq->lock + * __schedule/block_task/dequeue_task: rq->lock + * proxy_task_current: rq->lock + * wait_task_inactive: task_rq_lock + * sched_change_begin + */ bool (*dequeue_task) (struct rq *rq, struct task_struct *p, int flags); + + /* + * do_sched_yield: rq->lock + */ void (*yield_task) (struct rq *rq); + /* + * yield_to: rq->lock (double) + */ bool (*yield_to_task)(struct rq *rq, struct task_struct *p); + /* + * move_queued_task: rq->lock + * __migrate_swap_task: rq->lock + * ttwu_do_activate: rq->lock + * ttwu_runnable: task_rq_lock + * wake_up_new_task: task_rq_lock + */ void (*wakeup_preempt)(struct rq *rq, struct task_struct *p, int flags); + /* + * schedule/pick_next_task/prev_balance: rq->lock + */ int (*balance)(struct rq *rq, struct task_struct *prev, struct rq_flags *rf); - struct task_struct *(*pick_task)(struct rq *rq); + + /* + * schedule/pick_next_task: rq->lock + */ + struct task_struct *(*pick_task)(struct rq *rq, struct rq_flags *rf); /* * Optional! When implemented pick_next_task() should be equivalent to: * @@ -2432,53 +2576,105 @@ struct sched_class { * set_next_task_first(next); * } */ - struct task_struct *(*pick_next_task)(struct rq *rq, struct task_struct *prev); + struct task_struct *(*pick_next_task)(struct rq *rq, struct task_struct *prev, + struct rq_flags *rf); + /* + * sched_change: + * __schedule: rq->lock + */ void (*put_prev_task)(struct rq *rq, struct task_struct *p, struct task_struct *next); void (*set_next_task)(struct rq *rq, struct task_struct *p, bool first); -#ifdef CONFIG_SMP + /* + * select_task_rq: p->pi_lock + * sched_exec: p->pi_lock + */ int (*select_task_rq)(struct task_struct *p, int task_cpu, int flags); + /* + * set_task_cpu: p->pi_lock || rq->lock (ttwu like) + */ void (*migrate_task_rq)(struct task_struct *p, int new_cpu); + /* + * ttwu_do_activate: rq->lock + * wake_up_new_task: task_rq_lock + */ void (*task_woken)(struct rq *this_rq, struct task_struct *task); + /* + * do_set_cpus_allowed: task_rq_lock + sched_change + */ void (*set_cpus_allowed)(struct task_struct *p, struct affinity_context *ctx); + /* + * sched_set_rq_{on,off}line: rq->lock + */ void (*rq_online)(struct rq *rq); void (*rq_offline)(struct rq *rq); + /* + * push_cpu_stop: p->pi_lock && rq->lock + */ struct rq *(*find_lock_rq)(struct task_struct *p, struct rq *rq); -#endif + /* + * hrtick: rq->lock + * sched_tick: rq->lock + * sched_tick_remote: rq->lock + */ void (*task_tick)(struct rq *rq, struct task_struct *p, int queued); + /* + * sched_cgroup_fork: p->pi_lock + */ void (*task_fork)(struct task_struct *p); + /* + * finish_task_switch: no locks + */ void (*task_dead)(struct task_struct *p); /* - * The switched_from() call is allowed to drop rq->lock, therefore we - * cannot assume the switched_from/switched_to pair is serialized by - * rq->lock. They are however serialized by p->pi_lock. + * sched_change + */ + void (*switching_from)(struct rq *this_rq, struct task_struct *task); + void (*switched_from) (struct rq *this_rq, struct task_struct *task); + void (*switching_to) (struct rq *this_rq, struct task_struct *task); + void (*switched_to) (struct rq *this_rq, struct task_struct *task); + u64 (*get_prio) (struct rq *this_rq, struct task_struct *task); + void (*prio_changed) (struct rq *this_rq, struct task_struct *task, + u64 oldprio); + + /* + * set_load_weight: task_rq_lock + sched_change + * __setscheduler_parms: task_rq_lock + sched_change */ - void (*switching_to) (struct rq *this_rq, struct task_struct *task); - void (*switched_from)(struct rq *this_rq, struct task_struct *task); - void (*switched_to) (struct rq *this_rq, struct task_struct *task); void (*reweight_task)(struct rq *this_rq, struct task_struct *task, const struct load_weight *lw); - void (*prio_changed) (struct rq *this_rq, struct task_struct *task, - int oldprio); + /* + * sched_rr_get_interval: task_rq_lock + */ unsigned int (*get_rr_interval)(struct rq *rq, struct task_struct *task); + /* + * task_sched_runtime: task_rq_lock + */ void (*update_curr)(struct rq *rq); #ifdef CONFIG_FAIR_GROUP_SCHED + /* + * sched_change_group: task_rq_lock + sched_change + */ void (*task_change_group)(struct task_struct *p); #endif #ifdef CONFIG_SCHED_CORE + /* + * pick_next_task: rq->lock + * try_steal_cookie: rq->lock (double) + */ int (*task_is_throttled)(struct task_struct *p, int cpu); #endif }; @@ -2508,7 +2704,7 @@ static inline void put_prev_set_next_task(struct rq *rq, struct task_struct *prev, struct task_struct *next) { - WARN_ON_ONCE(rq->curr != prev); + WARN_ON_ONCE(rq->donor != prev); __put_prev_set_next_dl_server(rq, prev, next); @@ -2574,6 +2770,17 @@ static inline const struct sched_class *next_active_class(const struct sched_cla #define sched_class_above(_a, _b) ((_a) < (_b)) +static inline void rq_modified_begin(struct rq *rq, const struct sched_class *class) +{ + if (sched_class_above(rq->next_class, class)) + rq->next_class = class; +} + +static inline bool rq_modified_above(struct rq *rq, const struct sched_class *class) +{ + return sched_class_above(rq->next_class, class); +} + static inline bool sched_stop_runnable(struct rq *rq) { return rq->stop && task_on_rq_queued(rq->stop); @@ -2594,16 +2801,15 @@ static inline bool sched_fair_runnable(struct rq *rq) return rq->cfs.nr_queued > 0; } -extern struct task_struct *pick_next_task_fair(struct rq *rq, struct task_struct *prev, struct rq_flags *rf); -extern struct task_struct *pick_task_idle(struct rq *rq); +extern struct task_struct *pick_next_task_fair(struct rq *rq, struct task_struct *prev, + struct rq_flags *rf); +extern struct task_struct *pick_task_idle(struct rq *rq, struct rq_flags *rf); #define SCA_CHECK 0x01 #define SCA_MIGRATE_DISABLE 0x02 #define SCA_MIGRATE_ENABLE 0x04 #define SCA_USER 0x08 -#ifdef CONFIG_SMP - extern void update_group_capacity(struct sched_domain *sd, int cpu); extern void sched_balance_trigger(struct rq *rq); @@ -2627,7 +2833,7 @@ static inline bool task_allowed_on_cpu(struct task_struct *p, int cpu) static inline cpumask_t *alloc_user_cpus_ptr(int node) { /* - * See do_set_cpus_allowed() above for the rcu_head usage. + * See set_cpus_allowed_force() above for the rcu_head usage. */ int size = max_t(int, cpumask_size(), sizeof(struct rcu_head)); @@ -2655,26 +2861,6 @@ static inline struct task_struct *get_push_task(struct rq *rq) extern int push_cpu_stop(void *arg); -#else /* !CONFIG_SMP: */ - -static inline bool task_allowed_on_cpu(struct task_struct *p, int cpu) -{ - return true; -} - -static inline int __set_cpus_allowed_ptr(struct task_struct *p, - struct affinity_context *ctx) -{ - return set_cpus_allowed_ptr(p, ctx->new_mask); -} - -static inline cpumask_t *alloc_user_cpus_ptr(int node) -{ - return NULL; -} - -#endif /* !CONFIG_SMP */ - #ifdef CONFIG_CPU_IDLE static inline void idle_set_state(struct rq *rq, @@ -2685,7 +2871,7 @@ static inline void idle_set_state(struct rq *rq, static inline struct cpuidle_state *idle_get_state(struct rq *rq) { - SCHED_WARN_ON(!rcu_read_lock_held()); + lockdep_assert(rcu_read_lock_any_held()); return rq->idle_state; } @@ -2724,13 +2910,15 @@ extern bool sched_rt_bandwidth_account(struct rt_rq *rt_rq); extern void init_dl_entity(struct sched_dl_entity *dl_se); +extern void init_cfs_throttle_work(struct task_struct *p); + #define BW_SHIFT 20 #define BW_UNIT (1 << BW_SHIFT) #define RATIO_SHIFT 8 #define MAX_BW_BITS (64 - BW_SHIFT) #define MAX_BW ((1ULL << MAX_BW_BITS) - 1) -extern unsigned long to_ratio(u64 period, u64 runtime); +extern u64 to_ratio(u64 period, u64 runtime); extern void init_entity_runnable_average(struct sched_entity *se); extern void post_init_entity_util_avg(struct task_struct *p); @@ -2770,10 +2958,8 @@ static inline void add_nr_running(struct rq *rq, unsigned count) call_trace_sched_update_nr_running(rq, count); } -#ifdef CONFIG_SMP if (prev_nr < 2 && rq->nr_running >= 2) set_rd_overloaded(rq->rd, 1); -#endif sched_update_tick_dependency(rq); } @@ -2837,18 +3023,40 @@ extern void deactivate_task(struct rq *rq, struct task_struct *p, int flags); extern void wakeup_preempt(struct rq *rq, struct task_struct *p, int flags); +/* + * attach_task() -- attach the task detached by detach_task() to its new rq. + */ +static inline void attach_task(struct rq *rq, struct task_struct *p) +{ + lockdep_assert_rq_held(rq); + + WARN_ON_ONCE(task_rq(p) != rq); + activate_task(rq, p, ENQUEUE_NOCLOCK); + wakeup_preempt(rq, p, 0); +} + +/* + * attach_one_task() -- attaches the task returned from detach_one_task() to + * its new rq. + */ +static inline void attach_one_task(struct rq *rq, struct task_struct *p) +{ + guard(rq_lock)(rq); + update_rq_clock(rq); + attach_task(rq, p); +} + #ifdef CONFIG_PREEMPT_RT # define SCHED_NR_MIGRATE_BREAK 8 #else # define SCHED_NR_MIGRATE_BREAK 32 #endif -extern const_debug unsigned int sysctl_sched_nr_migrate; -extern const_debug unsigned int sysctl_sched_migration_cost; +extern __read_mostly unsigned int sysctl_sched_nr_migrate; +extern __read_mostly unsigned int sysctl_sched_migration_cost; extern unsigned int sysctl_sched_base_slice; -#ifdef CONFIG_SCHED_DEBUG extern int sysctl_resched_latency_warn_ms; extern int sysctl_resched_latency_warn_once; @@ -2859,7 +3067,6 @@ extern unsigned int sysctl_numa_balancing_scan_period_min; extern unsigned int sysctl_numa_balancing_scan_period_max; extern unsigned int sysctl_numa_balancing_scan_size; extern unsigned int sysctl_numa_balancing_hot_threshold; -#endif #ifdef CONFIG_SCHED_HRTICK @@ -2868,46 +3075,31 @@ extern unsigned int sysctl_numa_balancing_hot_threshold; * - enabled by features * - hrtimer is actually high res */ -static inline int hrtick_enabled(struct rq *rq) +static inline bool hrtick_enabled(struct rq *rq) { - if (!cpu_active(cpu_of(rq))) - return 0; - return hrtimer_is_hres_active(&rq->hrtick_timer); + return cpu_active(cpu_of(rq)) && hrtimer_highres_enabled(); } -static inline int hrtick_enabled_fair(struct rq *rq) +static inline bool hrtick_enabled_fair(struct rq *rq) { - if (!sched_feat(HRTICK)) - return 0; - return hrtick_enabled(rq); + return sched_feat(HRTICK) && hrtick_enabled(rq); } -static inline int hrtick_enabled_dl(struct rq *rq) +static inline bool hrtick_enabled_dl(struct rq *rq) { - if (!sched_feat(HRTICK_DL)) - return 0; - return hrtick_enabled(rq); + return sched_feat(HRTICK_DL) && hrtick_enabled(rq); } extern void hrtick_start(struct rq *rq, u64 delay); - -#else /* !CONFIG_SCHED_HRTICK: */ - -static inline int hrtick_enabled_fair(struct rq *rq) -{ - return 0; -} - -static inline int hrtick_enabled_dl(struct rq *rq) +static inline bool hrtick_active(struct rq *rq) { - return 0; -} - -static inline int hrtick_enabled(struct rq *rq) -{ - return 0; + return hrtimer_active(&rq->hrtick_timer); } +#else /* !CONFIG_SCHED_HRTICK: */ +static inline bool hrtick_enabled_fair(struct rq *rq) { return false; } +static inline bool hrtick_enabled_dl(struct rq *rq) { return false; } +static inline bool hrtick_enabled(struct rq *rq) { return false; } #endif /* !CONFIG_SCHED_HRTICK */ #ifndef arch_scale_freq_tick @@ -2932,7 +3124,6 @@ unsigned long arch_scale_freq_capacity(int cpu) } #endif -#ifdef CONFIG_SCHED_DEBUG /* * In double_lock_balance()/double_rq_lock(), we use raw_spin_rq_lock() to * acquire rq lock instead of rq_lock(). So at the end of these two functions @@ -2942,22 +3133,26 @@ unsigned long arch_scale_freq_capacity(int cpu) static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2) { rq1->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP); - /* rq1 == rq2 for !CONFIG_SMP, so just clear RQCF_UPDATED once. */ -#ifdef CONFIG_SMP rq2->clock_update_flags &= (RQCF_REQ_SKIP|RQCF_ACT_SKIP); -#endif } -#else -static inline void double_rq_clock_clear_update(struct rq *rq1, struct rq *rq2) { } -#endif #define DEFINE_LOCK_GUARD_2(name, type, _lock, _unlock, ...) \ __DEFINE_UNLOCK_GUARD(name, type, _unlock, type *lock2; __VA_ARGS__) \ static inline class_##name##_t class_##name##_constructor(type *lock, type *lock2) \ + __no_context_analysis \ { class_##name##_t _t = { .lock = lock, .lock2 = lock2 }, *_T = &_t; \ _lock; return _t; } - -#ifdef CONFIG_SMP +#define DECLARE_LOCK_GUARD_2_ATTRS(_name, _lock, _unlock1, _unlock2) \ +static inline class_##_name##_t class_##_name##_constructor(lock_##_name##_t *_T1, \ + lock_##_name##_t *_T2) _lock; \ +static __always_inline void __class_##_name##_cleanup_ctx1(class_##_name##_t **_T1) \ + __no_context_analysis _unlock1 { } \ +static __always_inline void __class_##_name##_cleanup_ctx2(class_##_name##_t **_T2) \ + __no_context_analysis _unlock2 { } +#define WITH_LOCK_GUARD_2_ATTRS(_name, _T1, _T2) \ + class_##_name##_constructor(_T1, _T2), \ + *__UNIQUE_ID(unlock1) __cleanup(__class_##_name##_cleanup_ctx1) = (void *)(_T1),\ + *__UNIQUE_ID(unlock2) __cleanup(__class_##_name##_cleanup_ctx2) = (void *)(_T2) static inline bool rq_order_less(struct rq *rq1, struct rq *rq2) { @@ -2981,11 +3176,12 @@ static inline bool rq_order_less(struct rq *rq1, struct rq *rq2) /* * __sched_core_flip() relies on SMT having cpu-id lock order. */ -#endif +#endif /* CONFIG_SCHED_CORE */ return rq1->cpu < rq2->cpu; } -extern void double_rq_lock(struct rq *rq1, struct rq *rq2); +extern void double_rq_lock(struct rq *rq1, struct rq *rq2) + __acquires(__rq_lockp(rq1), __rq_lockp(rq2)); #ifdef CONFIG_PREEMPTION @@ -2998,9 +3194,8 @@ extern void double_rq_lock(struct rq *rq1, struct rq *rq2); * also adds more overhead and therefore may reduce throughput. */ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) - __releases(this_rq->lock) - __acquires(busiest->lock) - __acquires(this_rq->lock) + __must_hold(__rq_lockp(this_rq)) + __acquires(__rq_lockp(busiest)) { raw_spin_rq_unlock(this_rq); double_rq_lock(this_rq, busiest); @@ -3017,12 +3212,16 @@ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) * regardless of entry order into the function. */ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) - __releases(this_rq->lock) - __acquires(busiest->lock) - __acquires(this_rq->lock) + __must_hold(__rq_lockp(this_rq)) + __acquires(__rq_lockp(busiest)) { - if (__rq_lockp(this_rq) == __rq_lockp(busiest) || - likely(raw_spin_rq_trylock(busiest))) { + if (__rq_lockp(this_rq) == __rq_lockp(busiest)) { + __acquire(__rq_lockp(busiest)); /* already held */ + double_rq_clock_clear_update(this_rq, busiest); + return 0; + } + + if (likely(raw_spin_rq_trylock(busiest))) { double_rq_clock_clear_update(this_rq, busiest); return 0; } @@ -3045,6 +3244,8 @@ static inline int _double_lock_balance(struct rq *this_rq, struct rq *busiest) * double_lock_balance - lock the busiest runqueue, this_rq is locked already. */ static inline int double_lock_balance(struct rq *this_rq, struct rq *busiest) + __must_hold(__rq_lockp(this_rq)) + __acquires(__rq_lockp(busiest)) { lockdep_assert_irqs_disabled(); @@ -3052,14 +3253,17 @@ static inline int double_lock_balance(struct rq *this_rq, struct rq *busiest) } static inline void double_unlock_balance(struct rq *this_rq, struct rq *busiest) - __releases(busiest->lock) + __releases(__rq_lockp(busiest)) { if (__rq_lockp(this_rq) != __rq_lockp(busiest)) raw_spin_rq_unlock(busiest); + else + __release(__rq_lockp(busiest)); /* fake release */ lock_set_subclass(&__rq_lockp(this_rq)->dep_map, 0, _RET_IP_); } static inline void double_lock(spinlock_t *l1, spinlock_t *l2) + __acquires(l1, l2) { if (l1 > l2) swap(l1, l2); @@ -3069,6 +3273,7 @@ static inline void double_lock(spinlock_t *l1, spinlock_t *l2) } static inline void double_lock_irq(spinlock_t *l1, spinlock_t *l2) + __acquires(l1, l2) { if (l1 > l2) swap(l1, l2); @@ -3078,6 +3283,7 @@ static inline void double_lock_irq(spinlock_t *l1, spinlock_t *l2) } static inline void double_raw_lock(raw_spinlock_t *l1, raw_spinlock_t *l2) + __acquires(l1, l2) { if (l1 > l2) swap(l1, l2); @@ -3087,6 +3293,7 @@ static inline void double_raw_lock(raw_spinlock_t *l1, raw_spinlock_t *l2) } static inline void double_raw_unlock(raw_spinlock_t *l1, raw_spinlock_t *l2) + __releases(l1, l2) { raw_spin_unlock(l1); raw_spin_unlock(l2); @@ -3096,6 +3303,13 @@ DEFINE_LOCK_GUARD_2(double_raw_spinlock, raw_spinlock_t, double_raw_lock(_T->lock, _T->lock2), double_raw_unlock(_T->lock, _T->lock2)) +DECLARE_LOCK_GUARD_2_ATTRS(double_raw_spinlock, + __acquires(_T1, _T2), + __releases(*(raw_spinlock_t **)_T1), + __releases(*(raw_spinlock_t **)_T2)); +#define class_double_raw_spinlock_constructor(_T1, _T2) \ + WITH_LOCK_GUARD_2_ATTRS(double_raw_spinlock, _T1, _T2) + /* * double_rq_unlock - safely unlock two runqueues * @@ -3103,13 +3317,12 @@ DEFINE_LOCK_GUARD_2(double_raw_spinlock, raw_spinlock_t, * you need to do so manually after calling. */ static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) - __releases(rq1->lock) - __releases(rq2->lock) + __releases(__rq_lockp(rq1), __rq_lockp(rq2)) { if (__rq_lockp(rq1) != __rq_lockp(rq2)) raw_spin_rq_unlock(rq2); else - __release(rq2->lock); + __release(__rq_lockp(rq2)); /* fake release */ raw_spin_rq_unlock(rq1); } @@ -3118,42 +3331,6 @@ extern void set_rq_offline(struct rq *rq); extern bool sched_smp_initialized; -#else /* !CONFIG_SMP: */ - -/* - * double_rq_lock - safely lock two runqueues - * - * Note this does not disable interrupts like task_rq_lock, - * you need to do so manually before calling. - */ -static inline void double_rq_lock(struct rq *rq1, struct rq *rq2) - __acquires(rq1->lock) - __acquires(rq2->lock) -{ - WARN_ON_ONCE(!irqs_disabled()); - WARN_ON_ONCE(rq1 != rq2); - raw_spin_rq_lock(rq1); - __acquire(rq2->lock); /* Fake it out ;) */ - double_rq_clock_clear_update(rq1, rq2); -} - -/* - * double_rq_unlock - safely unlock two runqueues - * - * Note this does not restore interrupts like task_rq_unlock, - * you need to do so manually after calling. - */ -static inline void double_rq_unlock(struct rq *rq1, struct rq *rq2) - __releases(rq1->lock) - __releases(rq2->lock) -{ - WARN_ON_ONCE(rq1 != rq2); - raw_spin_rq_unlock(rq1); - __release(rq2->lock); -} - -#endif /* !CONFIG_SMP */ - DEFINE_LOCK_GUARD_2(double_rq_lock, struct rq, double_rq_lock(_T->lock, _T->lock2), double_rq_unlock(_T->lock, _T->lock2)) @@ -3162,7 +3339,6 @@ extern struct sched_entity *__pick_root_entity(struct cfs_rq *cfs_rq); extern struct sched_entity *__pick_first_entity(struct cfs_rq *cfs_rq); extern struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq); -#ifdef CONFIG_SCHED_DEBUG extern bool sched_debug_verbose; extern void print_cfs_stats(struct seq_file *m, int cpu); @@ -3173,15 +3349,13 @@ extern void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq); extern void print_dl_rq(struct seq_file *m, int cpu, struct dl_rq *dl_rq); extern void resched_latency_warn(int cpu, u64 latency); -# ifdef CONFIG_NUMA_BALANCING + +#ifdef CONFIG_NUMA_BALANCING extern void show_numa_stats(struct task_struct *p, struct seq_file *m); extern void print_numa_stats(struct seq_file *m, int node, unsigned long tsf, unsigned long tpf, unsigned long gsf, unsigned long gpf); -# endif /* CONFIG_NUMA_BALANCING */ -#else /* !CONFIG_SCHED_DEBUG: */ -static inline void resched_latency_warn(int cpu, u64 latency) { } -#endif /* !CONFIG_SCHED_DEBUG */ +#endif /* CONFIG_NUMA_BALANCING */ extern void init_cfs_rq(struct cfs_rq *cfs_rq); extern void init_rt_rq(struct rt_rq *rt_rq); @@ -3215,7 +3389,7 @@ extern void nohz_balance_exit_idle(struct rq *rq); static inline void nohz_balance_exit_idle(struct rq *rq) { } #endif /* !CONFIG_NO_HZ_COMMON */ -#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) +#ifdef CONFIG_NO_HZ_COMMON extern void nohz_run_idle_balance(int cpu); #else static inline void nohz_run_idle_balance(int cpu) { } @@ -3259,11 +3433,11 @@ struct irqtime { }; DECLARE_PER_CPU(struct irqtime, cpu_irqtime); -extern int sched_clock_irqtime; +DECLARE_STATIC_KEY_FALSE(sched_clock_irqtime); static inline int irqtime_enabled(void) { - return sched_clock_irqtime; + return static_branch_likely(&sched_clock_irqtime); } /* @@ -3285,14 +3459,14 @@ static inline u64 irq_time_read(int cpu) return total; } -#else +#else /* !CONFIG_IRQ_TIME_ACCOUNTING: */ static inline int irqtime_enabled(void) { return 0; } -#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ +#endif /* !CONFIG_IRQ_TIME_ACCOUNTING */ #ifdef CONFIG_CPU_FREQ @@ -3341,8 +3515,6 @@ static inline void cpufreq_update_util(struct rq *rq, unsigned int flags) { } # define arch_scale_freq_invariant() false #endif -#ifdef CONFIG_SMP - unsigned long effective_cpu_util(int cpu, unsigned long util_cfs, unsigned long *min, unsigned long *max); @@ -3386,14 +3558,35 @@ static inline unsigned long cpu_util_rt(struct rq *rq) return READ_ONCE(rq->avg_rt.util_avg); } -#else /* !CONFIG_SMP */ -static inline bool update_other_load_avgs(struct rq *rq) { return false; } -#endif /* CONFIG_SMP */ - #ifdef CONFIG_UCLAMP_TASK unsigned long uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id); +/* + * When uclamp is compiled in, the aggregation at rq level is 'turned off' + * by default in the fast path and only gets turned on once userspace performs + * an operation that requires it. + * + * Returns true if userspace opted-in to use uclamp and aggregation at rq level + * hence is active. + */ +static inline bool uclamp_is_used(void) +{ + return static_branch_likely(&sched_uclamp_used); +} + +/* + * Enabling static branches would get the cpus_read_lock(), + * check whether uclamp_is_used before enable it to avoid always + * calling cpus_read_lock(). Because we never disable this + * static key once enable it. + */ +static inline void sched_uclamp_enable(void) +{ + if (!uclamp_is_used()) + static_branch_enable(&sched_uclamp_used); +} + static inline unsigned long uclamp_rq_get(struct rq *rq, enum uclamp_id clamp_id) { @@ -3417,7 +3610,7 @@ static inline bool uclamp_rq_is_capped(struct rq *rq) unsigned long rq_util; unsigned long max_util; - if (!static_branch_likely(&sched_uclamp_used)) + if (!uclamp_is_used()) return false; rq_util = cpu_util_cfs(cpu_of(rq)) + cpu_util_rt(rq); @@ -3426,19 +3619,6 @@ static inline bool uclamp_rq_is_capped(struct rq *rq) return max_util != SCHED_CAPACITY_SCALE && rq_util >= max_util; } -/* - * When uclamp is compiled in, the aggregation at rq level is 'turned off' - * by default in the fast path and only gets turned on once userspace performs - * an operation that requires it. - * - * Returns true if userspace opted-in to use uclamp and aggregation at rq level - * hence is active. - */ -static inline bool uclamp_is_used(void) -{ - return static_branch_likely(&sched_uclamp_used); -} - #define for_each_clamp_id(clamp_id) \ for ((clamp_id) = 0; (clamp_id) < UCLAMP_CNT; (clamp_id)++) @@ -3486,6 +3666,8 @@ static inline bool uclamp_is_used(void) return false; } +static inline void sched_uclamp_enable(void) {} + static inline unsigned long uclamp_rq_get(struct rq *rq, enum uclamp_id clamp_id) { @@ -3552,15 +3734,13 @@ static inline bool sched_energy_enabled(void) return static_branch_unlikely(&sched_energy_present); } -extern struct cpufreq_governor schedutil_gov; - -#else /* ! (CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */ +#else /* !(CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL): */ #define perf_domain_span(pd) NULL static inline bool sched_energy_enabled(void) { return false; } -#endif /* CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL */ +#endif /* !(CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */ #ifdef CONFIG_MEMBARRIER @@ -3586,7 +3766,7 @@ static inline void membarrier_switch_mm(struct rq *rq, WRITE_ONCE(rq->membarrier_state, membarrier_state); } -#else /* !CONFIG_MEMBARRIER :*/ +#else /* !CONFIG_MEMBARRIER: */ static inline void membarrier_switch_mm(struct rq *rq, struct mm_struct *prev_mm, @@ -3596,7 +3776,6 @@ static inline void membarrier_switch_mm(struct rq *rq, #endif /* !CONFIG_MEMBARRIER */ -#ifdef CONFIG_SMP static inline bool is_per_cpu_kthread(struct task_struct *p) { if (!(p->flags & PF_KTHREAD)) @@ -3607,7 +3786,6 @@ static inline bool is_per_cpu_kthread(struct task_struct *p) return true; } -#endif extern void swake_up_all_locked(struct swait_queue_head *q); extern void __prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait); @@ -3619,293 +3797,248 @@ extern int preempt_dynamic_mode; extern int sched_dynamic_mode(const char *str); extern void sched_dynamic_update(int mode); #endif +extern const char *preempt_modes[]; #ifdef CONFIG_SCHED_MM_CID -#define SCHED_MM_CID_PERIOD_NS (100ULL * 1000000) /* 100ms */ -#define MM_CID_SCAN_DELAY 100 /* 100ms */ +static __always_inline bool cid_on_cpu(unsigned int cid) +{ + return cid & MM_CID_ONCPU; +} -extern raw_spinlock_t cid_lock; -extern int use_cid_lock; +static __always_inline bool cid_in_transit(unsigned int cid) +{ + return cid & MM_CID_TRANSIT; +} -extern void sched_mm_cid_migrate_from(struct task_struct *t); -extern void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t); -extern void task_tick_mm_cid(struct rq *rq, struct task_struct *curr); -extern void init_sched_mm_cid(struct task_struct *t); +static __always_inline unsigned int cpu_cid_to_cid(unsigned int cid) +{ + return cid & ~MM_CID_ONCPU; +} -static inline void __mm_cid_put(struct mm_struct *mm, int cid) +static __always_inline unsigned int cid_to_cpu_cid(unsigned int cid) { - if (cid < 0) - return; - cpumask_clear_cpu(cid, mm_cidmask(mm)); + return cid | MM_CID_ONCPU; } -/* - * The per-mm/cpu cid can have the MM_CID_LAZY_PUT flag set or transition to - * the MM_CID_UNSET state without holding the rq lock, but the rq lock needs to - * be held to transition to other states. - * - * State transitions synchronized with cmpxchg or try_cmpxchg need to be - * consistent across CPUs, which prevents use of this_cpu_cmpxchg. - */ -static inline void mm_cid_put_lazy(struct task_struct *t) +static __always_inline unsigned int cid_to_transit_cid(unsigned int cid) { - struct mm_struct *mm = t->mm; - struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid; - int cid; + return cid | MM_CID_TRANSIT; +} - lockdep_assert_irqs_disabled(); - cid = __this_cpu_read(pcpu_cid->cid); - if (!mm_cid_is_lazy_put(cid) || - !try_cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, &cid, MM_CID_UNSET)) - return; - __mm_cid_put(mm, mm_cid_clear_lazy_put(cid)); +static __always_inline unsigned int cid_from_transit_cid(unsigned int cid) +{ + return cid & ~MM_CID_TRANSIT; } -static inline int mm_cid_pcpu_unset(struct mm_struct *mm) +static __always_inline bool cid_on_task(unsigned int cid) { - struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid; - int cid, res; + /* True if none of the MM_CID_ONCPU, MM_CID_TRANSIT, MM_CID_UNSET bits is set */ + return cid < MM_CID_TRANSIT; +} - lockdep_assert_irqs_disabled(); - cid = __this_cpu_read(pcpu_cid->cid); - for (;;) { - if (mm_cid_is_unset(cid)) - return MM_CID_UNSET; - /* - * Attempt transition from valid or lazy-put to unset. - */ - res = cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, cid, MM_CID_UNSET); - if (res == cid) - break; - cid = res; +static __always_inline void mm_drop_cid(struct mm_struct *mm, unsigned int cid) +{ + clear_bit(cid, mm_cidmask(mm)); +} + +static __always_inline void mm_unset_cid_on_task(struct task_struct *t) +{ + unsigned int cid = t->mm_cid.cid; + + t->mm_cid.cid = MM_CID_UNSET; + if (cid_on_task(cid)) + mm_drop_cid(t->mm, cid); +} + +static __always_inline void mm_drop_cid_on_cpu(struct mm_struct *mm, struct mm_cid_pcpu *pcp) +{ + /* Clear the ONCPU bit, but do not set UNSET in the per CPU storage */ + if (cid_on_cpu(pcp->cid)) { + pcp->cid = cpu_cid_to_cid(pcp->cid); + mm_drop_cid(mm, pcp->cid); } - return cid; } -static inline void mm_cid_put(struct mm_struct *mm) +static inline unsigned int __mm_get_cid(struct mm_struct *mm, unsigned int max_cids) { - int cid; + unsigned int cid = find_first_zero_bit(mm_cidmask(mm), max_cids); - lockdep_assert_irqs_disabled(); - cid = mm_cid_pcpu_unset(mm); - if (cid == MM_CID_UNSET) - return; - __mm_cid_put(mm, mm_cid_clear_lazy_put(cid)); + if (cid >= max_cids) + return MM_CID_UNSET; + if (test_and_set_bit(cid, mm_cidmask(mm))) + return MM_CID_UNSET; + return cid; } -static inline int __mm_cid_try_get(struct task_struct *t, struct mm_struct *mm) +static inline unsigned int mm_get_cid(struct mm_struct *mm) { - struct cpumask *cidmask = mm_cidmask(mm); - struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid; - int cid, max_nr_cid, allowed_max_nr_cid; + unsigned int cid = __mm_get_cid(mm, READ_ONCE(mm->mm_cid.max_cids)); - /* - * After shrinking the number of threads or reducing the number - * of allowed cpus, reduce the value of max_nr_cid so expansion - * of cid allocation will preserve cache locality if the number - * of threads or allowed cpus increase again. - */ - max_nr_cid = atomic_read(&mm->max_nr_cid); - while ((allowed_max_nr_cid = min_t(int, READ_ONCE(mm->nr_cpus_allowed), - atomic_read(&mm->mm_users))), - max_nr_cid > allowed_max_nr_cid) { - /* atomic_try_cmpxchg loads previous mm->max_nr_cid into max_nr_cid. */ - if (atomic_try_cmpxchg(&mm->max_nr_cid, &max_nr_cid, allowed_max_nr_cid)) { - max_nr_cid = allowed_max_nr_cid; - break; - } - } - /* Try to re-use recent cid. This improves cache locality. */ - cid = __this_cpu_read(pcpu_cid->recent_cid); - if (!mm_cid_is_unset(cid) && cid < max_nr_cid && - !cpumask_test_and_set_cpu(cid, cidmask)) - return cid; - /* - * Expand cid allocation if the maximum number of concurrency - * IDs allocated (max_nr_cid) is below the number cpus allowed - * and number of threads. Expanding cid allocation as much as - * possible improves cache locality. - */ - cid = max_nr_cid; - while (cid < READ_ONCE(mm->nr_cpus_allowed) && cid < atomic_read(&mm->mm_users)) { - /* atomic_try_cmpxchg loads previous mm->max_nr_cid into cid. */ - if (!atomic_try_cmpxchg(&mm->max_nr_cid, &cid, cid + 1)) - continue; - if (!cpumask_test_and_set_cpu(cid, cidmask)) - return cid; - } - /* - * Find the first available concurrency id. - * Retry finding first zero bit if the mask is temporarily - * filled. This only happens during concurrent remote-clear - * which owns a cid without holding a rq lock. - */ - for (;;) { - cid = cpumask_first_zero(cidmask); - if (cid < READ_ONCE(mm->nr_cpus_allowed)) - break; + while (cid == MM_CID_UNSET) { cpu_relax(); + cid = __mm_get_cid(mm, num_possible_cpus()); } - if (cpumask_test_and_set_cpu(cid, cidmask)) - return -1; - return cid; } -/* - * Save a snapshot of the current runqueue time of this cpu - * with the per-cpu cid value, allowing to estimate how recently it was used. - */ -static inline void mm_cid_snapshot_time(struct rq *rq, struct mm_struct *mm) +static inline unsigned int mm_cid_converge(struct mm_struct *mm, unsigned int orig_cid, + unsigned int max_cids) { - struct mm_cid *pcpu_cid = per_cpu_ptr(mm->pcpu_cid, cpu_of(rq)); + unsigned int new_cid, cid = cpu_cid_to_cid(orig_cid); - lockdep_assert_rq_held(rq); - WRITE_ONCE(pcpu_cid->time, rq->clock); + /* Is it in the optimal CID space? */ + if (likely(cid < max_cids)) + return orig_cid; + + /* Try to find one in the optimal space. Otherwise keep the provided. */ + new_cid = __mm_get_cid(mm, max_cids); + if (new_cid != MM_CID_UNSET) { + mm_drop_cid(mm, cid); + /* Preserve the ONCPU mode of the original CID */ + return new_cid | (orig_cid & MM_CID_ONCPU); + } + return orig_cid; } -static inline int __mm_cid_get(struct rq *rq, struct task_struct *t, - struct mm_struct *mm) +static __always_inline void mm_cid_update_task_cid(struct task_struct *t, unsigned int cid) { - int cid; + if (t->mm_cid.cid != cid) { + t->mm_cid.cid = cid; + rseq_sched_set_ids_changed(t); + } +} - /* - * All allocations (even those using the cid_lock) are lock-free. If - * use_cid_lock is set, hold the cid_lock to perform cid allocation to - * guarantee forward progress. - */ - if (!READ_ONCE(use_cid_lock)) { - cid = __mm_cid_try_get(t, mm); - if (cid >= 0) - goto end; - raw_spin_lock(&cid_lock); +static __always_inline void mm_cid_update_pcpu_cid(struct mm_struct *mm, unsigned int cid) +{ + __this_cpu_write(mm->mm_cid.pcpu->cid, cid); +} + +static __always_inline void mm_cid_from_cpu(struct task_struct *t, unsigned int cpu_cid, + unsigned int mode) +{ + unsigned int max_cids, tcid = t->mm_cid.cid; + struct mm_struct *mm = t->mm; + + max_cids = READ_ONCE(mm->mm_cid.max_cids); + /* Optimize for the common case where both have the ONCPU bit set */ + if (likely(cid_on_cpu(cpu_cid & tcid))) { + if (likely(cpu_cid_to_cid(cpu_cid) < max_cids)) { + mm_cid_update_task_cid(t, cpu_cid); + return; + } + /* Try to converge into the optimal CID space */ + cpu_cid = mm_cid_converge(mm, cpu_cid, max_cids); } else { - raw_spin_lock(&cid_lock); - cid = __mm_cid_try_get(t, mm); - if (cid >= 0) - goto unlock; + /* Hand over or drop the task owned CID */ + if (cid_on_task(tcid)) { + if (cid_on_cpu(cpu_cid)) + mm_unset_cid_on_task(t); + else + cpu_cid = cid_to_cpu_cid(tcid); + } + /* Still nothing, allocate a new one */ + if (!cid_on_cpu(cpu_cid)) + cpu_cid = cid_to_cpu_cid(mm_get_cid(mm)); + + /* Handle the transition mode flag if required */ + if (mode & MM_CID_TRANSIT) + cpu_cid = cpu_cid_to_cid(cpu_cid) | MM_CID_TRANSIT; } + mm_cid_update_pcpu_cid(mm, cpu_cid); + mm_cid_update_task_cid(t, cpu_cid); +} - /* - * cid concurrently allocated. Retry while forcing following - * allocations to use the cid_lock to ensure forward progress. - */ - WRITE_ONCE(use_cid_lock, 1); - /* - * Set use_cid_lock before allocation. Only care about program order - * because this is only required for forward progress. - */ - barrier(); - /* - * Retry until it succeeds. It is guaranteed to eventually succeed once - * all newcoming allocations observe the use_cid_lock flag set. - */ - do { - cid = __mm_cid_try_get(t, mm); - cpu_relax(); - } while (cid < 0); - /* - * Allocate before clearing use_cid_lock. Only care about - * program order because this is for forward progress. - */ - barrier(); - WRITE_ONCE(use_cid_lock, 0); -unlock: - raw_spin_unlock(&cid_lock); -end: - mm_cid_snapshot_time(rq, mm); +static __always_inline void mm_cid_from_task(struct task_struct *t, unsigned int cpu_cid, + unsigned int mode) +{ + unsigned int max_cids, tcid = t->mm_cid.cid; + struct mm_struct *mm = t->mm; - return cid; + max_cids = READ_ONCE(mm->mm_cid.max_cids); + /* Optimize for the common case, where both have the ONCPU bit clear */ + if (likely(cid_on_task(tcid | cpu_cid))) { + if (likely(tcid < max_cids)) { + mm_cid_update_pcpu_cid(mm, tcid); + return; + } + /* Try to converge into the optimal CID space */ + tcid = mm_cid_converge(mm, tcid, max_cids); + } else { + /* Hand over or drop the CPU owned CID */ + if (cid_on_cpu(cpu_cid)) { + if (cid_on_task(tcid)) + mm_drop_cid_on_cpu(mm, this_cpu_ptr(mm->mm_cid.pcpu)); + else + tcid = cpu_cid_to_cid(cpu_cid); + } + /* Still nothing, allocate a new one */ + if (!cid_on_task(tcid)) + tcid = mm_get_cid(mm); + /* Set the transition mode flag if required */ + tcid |= mode & MM_CID_TRANSIT; + } + mm_cid_update_pcpu_cid(mm, tcid); + mm_cid_update_task_cid(t, tcid); } -static inline int mm_cid_get(struct rq *rq, struct task_struct *t, - struct mm_struct *mm) +static __always_inline void mm_cid_schedin(struct task_struct *next) { - struct mm_cid __percpu *pcpu_cid = mm->pcpu_cid; - struct cpumask *cpumask; - int cid; + struct mm_struct *mm = next->mm; + unsigned int cpu_cid, mode; - lockdep_assert_rq_held(rq); - cpumask = mm_cidmask(mm); - cid = __this_cpu_read(pcpu_cid->cid); - if (mm_cid_is_valid(cid)) { - mm_cid_snapshot_time(rq, mm); - return cid; - } - if (mm_cid_is_lazy_put(cid)) { - if (try_cmpxchg(&this_cpu_ptr(pcpu_cid)->cid, &cid, MM_CID_UNSET)) - __mm_cid_put(mm, mm_cid_clear_lazy_put(cid)); - } - cid = __mm_cid_get(rq, t, mm); - __this_cpu_write(pcpu_cid->cid, cid); - __this_cpu_write(pcpu_cid->recent_cid, cid); + if (!next->mm_cid.active) + return; - return cid; + cpu_cid = __this_cpu_read(mm->mm_cid.pcpu->cid); + mode = READ_ONCE(mm->mm_cid.mode); + if (likely(!cid_on_cpu(mode))) + mm_cid_from_task(next, cpu_cid, mode); + else + mm_cid_from_cpu(next, cpu_cid, mode); } -static inline void switch_mm_cid(struct rq *rq, - struct task_struct *prev, - struct task_struct *next) +static __always_inline void mm_cid_schedout(struct task_struct *prev) { + struct mm_struct *mm = prev->mm; + unsigned int mode, cid; + + /* During mode transitions CIDs are temporary and need to be dropped */ + if (likely(!cid_in_transit(prev->mm_cid.cid))) + return; + + mode = READ_ONCE(mm->mm_cid.mode); + cid = cid_from_transit_cid(prev->mm_cid.cid); + /* - * Provide a memory barrier between rq->curr store and load of - * {prev,next}->mm->pcpu_cid[cpu] on rq->curr->mm transition. - * - * Should be adapted if context_switch() is modified. + * If transition mode is done, transfer ownership when the CID is + * within the convergence range to optimize the next schedule in. */ - if (!next->mm) { // to kernel - /* - * user -> kernel transition does not guarantee a barrier, but - * we can use the fact that it performs an atomic operation in - * mmgrab(). - */ - if (prev->mm) // from user - smp_mb__after_mmgrab(); - /* - * kernel -> kernel transition does not change rq->curr->mm - * state. It stays NULL. - */ - } else { // to user - /* - * kernel -> user transition does not provide a barrier - * between rq->curr store and load of {prev,next}->mm->pcpu_cid[cpu]. - * Provide it here. - */ - if (!prev->mm) { // from kernel - smp_mb(); - } else { // from user - /* - * user->user transition relies on an implicit - * memory barrier in switch_mm() when - * current->mm changes. If the architecture - * switch_mm() does not have an implicit memory - * barrier, it is emitted here. If current->mm - * is unchanged, no barrier is needed. - */ - smp_mb__after_switch_mm(); - } - } - if (prev->mm_cid_active) { - mm_cid_snapshot_time(rq, prev->mm); - mm_cid_put_lazy(prev); - prev->mm_cid = -1; + if (!cid_in_transit(mode) && cid < READ_ONCE(mm->mm_cid.max_cids)) { + if (cid_on_cpu(mode)) + cid = cid_to_cpu_cid(cid); + + /* Update both so that the next schedule in goes into the fast path */ + mm_cid_update_pcpu_cid(mm, cid); + prev->mm_cid.cid = cid; + } else { + mm_drop_cid(mm, cid); + prev->mm_cid.cid = MM_CID_UNSET; } - if (next->mm_cid_active) - next->last_mm_cid = next->mm_cid = mm_cid_get(rq, next, next->mm); +} + +static inline void mm_cid_switch_to(struct task_struct *prev, struct task_struct *next) +{ + mm_cid_schedout(prev); + mm_cid_schedin(next); } #else /* !CONFIG_SCHED_MM_CID: */ -static inline void switch_mm_cid(struct rq *rq, struct task_struct *prev, struct task_struct *next) { } -static inline void sched_mm_cid_migrate_from(struct task_struct *t) { } -static inline void sched_mm_cid_migrate_to(struct rq *dst_rq, struct task_struct *t) { } -static inline void task_tick_mm_cid(struct rq *rq, struct task_struct *curr) { } -static inline void init_sched_mm_cid(struct task_struct *t) { } +static inline void mm_cid_switch_to(struct task_struct *prev, struct task_struct *next) { } #endif /* !CONFIG_SCHED_MM_CID */ extern u64 avg_vruntime(struct cfs_rq *cfs_rq); extern int entity_eligible(struct cfs_rq *cfs_rq, struct sched_entity *se); -#ifdef CONFIG_SMP static inline void move_queued_task_locked(struct rq *src_rq, struct rq *dst_rq, struct task_struct *task) { @@ -3915,6 +4048,7 @@ void move_queued_task_locked(struct rq *src_rq, struct rq *dst_rq, struct task_s deactivate_task(src_rq, task, 0); set_task_cpu(task, dst_rq->cpu); activate_task(dst_rq, task, 0); + wakeup_preempt(dst_rq, task, 0); } static inline @@ -3926,7 +4060,6 @@ bool task_is_pushable(struct rq *rq, struct task_struct *p, int cpu) return false; } -#endif #ifdef CONFIG_RT_MUTEXES @@ -3961,45 +4094,45 @@ extern void set_load_weight(struct task_struct *p, bool update_load); extern void enqueue_task(struct rq *rq, struct task_struct *p, int flags); extern bool dequeue_task(struct rq *rq, struct task_struct *p, int flags); -extern void check_class_changing(struct rq *rq, struct task_struct *p, - const struct sched_class *prev_class); -extern void check_class_changed(struct rq *rq, struct task_struct *p, - const struct sched_class *prev_class, - int oldprio); - -#ifdef CONFIG_SMP extern struct balance_callback *splice_balance_callbacks(struct rq *rq); -extern void balance_callbacks(struct rq *rq, struct balance_callback *head); -#else -static inline struct balance_callback *splice_balance_callbacks(struct rq *rq) -{ - return NULL; -} - -static inline void balance_callbacks(struct rq *rq, struct balance_callback *head) -{ -} +extern void __balance_callbacks(struct rq *rq, struct rq_flags *rf); +extern void balance_callbacks(struct rq *rq, struct balance_callback *head); -#endif +/* + * The 'sched_change' pattern is the safe, easy and slow way of changing a + * task's scheduling properties. It dequeues a task, such that the scheduler + * is fully unaware of it; at which point its properties can be modified; + * after which it is enqueued again. + * + * Typically this must be called while holding task_rq_lock, since most/all + * properties are serialized under those locks. There is currently one + * exception to this rule in sched/ext which only holds rq->lock. + */ -#ifdef CONFIG_SCHED_CLASS_EXT /* - * Used by SCX in the enable/disable paths to move tasks between sched_classes - * and establish invariants. + * This structure is a temporary, used to preserve/convey the queueing state + * of the task between sched_change_begin() and sched_change_end(). Ensuring + * the task's queueing state is idempotent across the operation. */ -struct sched_enq_and_set_ctx { +struct sched_change_ctx { + u64 prio; struct task_struct *p; - int queue_flags; + const struct sched_class *class; + int flags; bool queued; bool running; }; -void sched_deq_and_put_task(struct task_struct *p, int queue_flags, - struct sched_enq_and_set_ctx *ctx); -void sched_enq_and_set_task(struct sched_enq_and_set_ctx *ctx); +struct sched_change_ctx *sched_change_begin(struct task_struct *p, unsigned int flags); +void sched_change_end(struct sched_change_ctx *ctx); -#endif /* CONFIG_SCHED_CLASS_EXT */ +DEFINE_CLASS(sched_change, struct sched_change_ctx *, + sched_change_end(_T), + sched_change_begin(p, flags), + struct task_struct *p, unsigned int flags) + +DEFINE_CLASS_IS_UNCONDITIONAL(sched_change) #include "ext.h" diff --git a/kernel/sched/smp.h b/kernel/sched/smp.h index 21ac44428bb0..7f151d96dba9 100644 --- a/kernel/sched/smp.h +++ b/kernel/sched/smp.h @@ -1,8 +1,13 @@ /* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef _KERNEL_SCHED_SMP_H +#define _KERNEL_SCHED_SMP_H + /* * Scheduler internal SMP callback types and methods between the scheduler * and other internal parts of the core kernel: */ +#include <linux/types.h> extern void sched_ttwu_pending(void *arg); @@ -13,3 +18,5 @@ extern void flush_smp_call_function_queue(void); #else static inline void flush_smp_call_function_queue(void) { } #endif + +#endif /* _KERNEL_SCHED_SMP_H */ diff --git a/kernel/sched/stats.c b/kernel/sched/stats.c index 4346fd81c31f..d1c9429a4ac5 100644 --- a/kernel/sched/stats.c +++ b/kernel/sched/stats.c @@ -2,6 +2,7 @@ /* * /proc/schedstat implementation */ +#include "sched.h" void __update_stats_wait_start(struct rq *rq, struct task_struct *p, struct sched_statistics *stats) @@ -114,10 +115,8 @@ static int show_schedstat(struct seq_file *seq, void *v) seq_printf(seq, "timestamp %lu\n", jiffies); } else { struct rq *rq; -#ifdef CONFIG_SMP struct sched_domain *sd; int dcount = 0; -#endif cpu = (unsigned long)(v - 2); rq = cpu_rq(cpu); @@ -132,7 +131,6 @@ static int show_schedstat(struct seq_file *seq, void *v) seq_printf(seq, "\n"); -#ifdef CONFIG_SMP /* domain-specific stats */ rcu_read_lock(); for_each_domain(cpu, sd) { @@ -163,7 +161,6 @@ static int show_schedstat(struct seq_file *seq, void *v) sd->ttwu_move_balance); } rcu_read_unlock(); -#endif } return 0; } diff --git a/kernel/sched/stats.h b/kernel/sched/stats.h index 19cdbe96f93d..a612cf253c87 100644 --- a/kernel/sched/stats.h +++ b/kernel/sched/stats.h @@ -112,10 +112,10 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, bool sleep); #ifdef CONFIG_IRQ_TIME_ACCOUNTING void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_struct *prev); -#else +#else /* !CONFIG_IRQ_TIME_ACCOUNTING: */ static inline void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_struct *prev) {} -#endif /*CONFIG_IRQ_TIME_ACCOUNTING */ +#endif /* !CONFIG_IRQ_TIME_ACCOUNTING */ /* * PSI tracks state that persists across sleeps, such as iowaits and * memory stalls. As a result, it has to distinguish between sleeps, @@ -144,7 +144,7 @@ static inline void psi_enqueue(struct task_struct *p, int flags) if (p->se.sched_delayed) { /* CPU migration of "sleeping" task */ - SCHED_WARN_ON(!(flags & ENQUEUE_MIGRATED)); + WARN_ON_ONCE(!(flags & ENQUEUE_MIGRATED)); if (p->in_memstall) set |= TSK_MEMSTALL; if (p->in_iowait) @@ -180,8 +180,13 @@ static inline void psi_dequeue(struct task_struct *p, int flags) * avoid walking all ancestors twice, psi_task_switch() handles * TSK_RUNNING and TSK_IOWAIT for us when it moves TSK_ONCPU. * Do nothing here. + * + * In the SCHED_PROXY_EXECUTION case we may do sleeping + * dequeues that are not followed by a task switch, so check + * TSK_ONCPU is set to ensure the task switch is imminent. + * Otherwise clear the flags as usual. */ - if (flags & DEQUEUE_SLEEP) + if ((flags & DEQUEUE_SLEEP) && (p->psi_flags & TSK_ONCPU)) return; /* @@ -206,7 +211,7 @@ static inline void psi_ttwu_dequeue(struct task_struct *p) rq = __task_rq_lock(p, &rf); psi_task_change(p, p->psi_flags, 0); - __task_rq_unlock(rq, &rf); + __task_rq_unlock(rq, p, &rf); } } @@ -220,7 +225,7 @@ static inline void psi_sched_switch(struct task_struct *prev, psi_task_switch(prev, next, sleep); } -#else /* CONFIG_PSI */ +#else /* !CONFIG_PSI: */ static inline void psi_enqueue(struct task_struct *p, bool migrate) {} static inline void psi_dequeue(struct task_struct *p, bool migrate) {} static inline void psi_ttwu_dequeue(struct task_struct *p) {} @@ -229,7 +234,7 @@ static inline void psi_sched_switch(struct task_struct *prev, bool sleep) {} static inline void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_struct *prev) {} -#endif /* CONFIG_PSI */ +#endif /* !CONFIG_PSI */ #ifdef CONFIG_SCHED_INFO /* @@ -248,8 +253,10 @@ static inline void sched_info_dequeue(struct rq *rq, struct task_struct *t) delta = rq_clock(rq) - t->sched_info.last_queued; t->sched_info.last_queued = 0; t->sched_info.run_delay += delta; - if (delta > t->sched_info.max_run_delay) + if (delta > t->sched_info.max_run_delay) { t->sched_info.max_run_delay = delta; + ktime_get_real_ts64(&t->sched_info.max_run_delay_ts); + } if (delta && (!t->sched_info.min_run_delay || delta < t->sched_info.min_run_delay)) t->sched_info.min_run_delay = delta; rq_sched_info_dequeue(rq, delta); @@ -273,8 +280,10 @@ static void sched_info_arrive(struct rq *rq, struct task_struct *t) t->sched_info.run_delay += delta; t->sched_info.last_arrival = now; t->sched_info.pcount++; - if (delta > t->sched_info.max_run_delay) + if (delta > t->sched_info.max_run_delay) { t->sched_info.max_run_delay = delta; + ktime_get_real_ts64(&t->sched_info.max_run_delay_ts); + } if (delta && (!t->sched_info.min_run_delay || delta < t->sched_info.min_run_delay)) t->sched_info.min_run_delay = delta; @@ -334,6 +343,6 @@ sched_info_switch(struct rq *rq, struct task_struct *prev, struct task_struct *n # define sched_info_enqueue(rq, t) do { } while (0) # define sched_info_dequeue(rq, t) do { } while (0) # define sched_info_switch(rq, t, next) do { } while (0) -#endif /* CONFIG_SCHED_INFO */ +#endif /* !CONFIG_SCHED_INFO */ #endif /* _KERNEL_STATS_H */ diff --git a/kernel/sched/stop_task.c b/kernel/sched/stop_task.c index 058dd42e3d9b..f95798baddeb 100644 --- a/kernel/sched/stop_task.c +++ b/kernel/sched/stop_task.c @@ -7,8 +7,8 @@ * * See kernel/stop_machine.c */ +#include "sched.h" -#ifdef CONFIG_SMP static int select_task_rq_stop(struct task_struct *p, int cpu, int flags) { @@ -20,7 +20,6 @@ balance_stop(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { return sched_stop_runnable(rq); } -#endif /* CONFIG_SMP */ static void wakeup_preempt_stop(struct rq *rq, struct task_struct *p, int flags) @@ -33,7 +32,7 @@ static void set_next_task_stop(struct rq *rq, struct task_struct *stop, bool fir stop->se.exec_start = rq_clock_task(rq); } -static struct task_struct *pick_task_stop(struct rq *rq) +static struct task_struct *pick_task_stop(struct rq *rq, struct rq_flags *rf) { if (!sched_stop_runnable(rq)) return NULL; @@ -76,14 +75,17 @@ static void task_tick_stop(struct rq *rq, struct task_struct *curr, int queued) { } -static void switched_to_stop(struct rq *rq, struct task_struct *p) +static void switching_to_stop(struct rq *rq, struct task_struct *p) { BUG(); /* its impossible to change to this class */ } static void -prio_changed_stop(struct rq *rq, struct task_struct *p, int oldprio) +prio_changed_stop(struct rq *rq, struct task_struct *p, u64 oldprio) { + if (p->prio == oldprio) + return; + BUG(); /* how!?, what priority? */ } @@ -95,7 +97,6 @@ static void update_curr_stop(struct rq *rq) * Simple, special scheduling class for the per-CPU stop tasks: */ DEFINE_SCHED_CLASS(stop) = { - .enqueue_task = enqueue_task_stop, .dequeue_task = dequeue_task_stop, .yield_task = yield_task_stop, @@ -106,15 +107,13 @@ DEFINE_SCHED_CLASS(stop) = { .put_prev_task = put_prev_task_stop, .set_next_task = set_next_task_stop, -#ifdef CONFIG_SMP .balance = balance_stop, .select_task_rq = select_task_rq_stop, .set_cpus_allowed = set_cpus_allowed_common, -#endif .task_tick = task_tick_stop, .prio_changed = prio_changed_stop, - .switched_to = switched_to_stop, + .switching_to = switching_to_stop, .update_curr = update_curr_stop, }; diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c index 72505cd3b60a..0fef6496c4c8 100644 --- a/kernel/sched/swait.c +++ b/kernel/sched/swait.c @@ -2,6 +2,7 @@ /* * <linux/swait.h> (simple wait queues ) implementation: */ +#include "sched.h" void __init_swait_queue_head(struct swait_queue_head *q, const char *name, struct lock_class_key *key) diff --git a/kernel/sched/syscalls.c b/kernel/sched/syscalls.c index 456d339be98f..b215b0ead9a6 100644 --- a/kernel/sched/syscalls.c +++ b/kernel/sched/syscalls.c @@ -64,8 +64,6 @@ static int effective_prio(struct task_struct *p) void set_user_nice(struct task_struct *p, long nice) { - bool queued, running; - struct rq *rq; int old_prio; if (task_nice(p) == nice || nice < MIN_NICE || nice > MAX_NICE) @@ -74,10 +72,7 @@ void set_user_nice(struct task_struct *p, long nice) * We have to be careful, if called from sys_setpriority(), * the task might be in the middle of scheduling on another CPU. */ - CLASS(task_rq_lock, rq_guard)(p); - rq = rq_guard.rq; - - update_rq_clock(rq); + guard(task_rq_lock)(p); /* * The RT priorities are set via sched_setscheduler(), but we still @@ -90,28 +85,12 @@ void set_user_nice(struct task_struct *p, long nice) return; } - queued = task_on_rq_queued(p); - running = task_current_donor(rq, p); - if (queued) - dequeue_task(rq, p, DEQUEUE_SAVE | DEQUEUE_NOCLOCK); - if (running) - put_prev_task(rq, p); - - p->static_prio = NICE_TO_PRIO(nice); - set_load_weight(p, true); - old_prio = p->prio; - p->prio = effective_prio(p); - - if (queued) - enqueue_task(rq, p, ENQUEUE_RESTORE | ENQUEUE_NOCLOCK); - if (running) - set_next_task(rq, p); - - /* - * If the task increased its priority or is running and - * lowered its priority, then reschedule its CPU: - */ - p->sched_class->prio_changed(rq, p, old_prio); + scoped_guard (sched_change, p, DEQUEUE_SAVE) { + p->static_prio = NICE_TO_PRIO(nice); + set_load_weight(p, true); + old_prio = p->prio; + p->prio = effective_prio(p); + } } EXPORT_SYMBOL(set_user_nice); @@ -174,7 +153,7 @@ SYSCALL_DEFINE1(nice, int, increment) return 0; } -#endif +#endif /* __ARCH_WANT_SYS_NICE */ /** * task_prio - return the priority value of a given task. @@ -201,37 +180,7 @@ int task_prio(const struct task_struct *p) */ int idle_cpu(int cpu) { - struct rq *rq = cpu_rq(cpu); - - if (rq->curr != rq->idle) - return 0; - - if (rq->nr_running) - return 0; - -#ifdef CONFIG_SMP - if (rq->ttwu_pending) - return 0; -#endif - - return 1; -} - -/** - * available_idle_cpu - is a given CPU idle for enqueuing work. - * @cpu: the CPU in question. - * - * Return: 1 if the CPU is currently idle. 0 otherwise. - */ -int available_idle_cpu(int cpu) -{ - if (!idle_cpu(cpu)) - return 0; - - if (vcpu_is_preempted(cpu)) - return 0; - - return 1; + return idle_rq(cpu_rq(cpu)); } /** @@ -255,8 +204,7 @@ int sched_core_idle_cpu(int cpu) return idle_cpu(cpu); } - -#endif +#endif /* CONFIG_SCHED_CORE */ /** * find_process_by_pid - find a process with a matching PID value. @@ -336,6 +284,35 @@ static bool check_same_owner(struct task_struct *p) uid_eq(cred->euid, pcred->uid)); } +#ifdef CONFIG_RT_MUTEXES +static inline void __setscheduler_dl_pi(int newprio, int policy, + struct task_struct *p, + struct sched_change_ctx *scope) +{ + /* + * In case a DEADLINE task (either proper or boosted) gets + * setscheduled to a lower priority class, check if it neeeds to + * inherit parameters from a potential pi_task. In that case make + * sure replenishment happens with the next enqueue. + */ + + if (dl_prio(newprio) && !dl_policy(policy)) { + struct task_struct *pi_task = rt_mutex_get_top_task(p); + + if (pi_task) { + p->dl.pi_se = pi_task->dl.pi_se; + scope->flags |= ENQUEUE_REPLENISH; + } + } +} +#else /* !CONFIG_RT_MUTEXES */ +static inline void __setscheduler_dl_pi(int newprio, int policy, + struct task_struct *p, + struct sched_change_ctx *scope) +{ +} +#endif /* !CONFIG_RT_MUTEXES */ + #ifdef CONFIG_UCLAMP_TASK static int uclamp_validate(struct task_struct *p, @@ -368,7 +345,7 @@ static int uclamp_validate(struct task_struct *p, * blocking operation which obviously cannot be done while holding * scheduler locks. */ - static_branch_enable(&sched_uclamp_used); + sched_uclamp_enable(); return 0; } @@ -448,7 +425,7 @@ static inline int uclamp_validate(struct task_struct *p, } static void __setscheduler_uclamp(struct task_struct *p, const struct sched_attr *attr) { } -#endif +#endif /* !CONFIG_UCLAMP_TASK */ /* * Allow unprivileged RT tasks to decrease priority. @@ -518,7 +495,7 @@ int __sched_setscheduler(struct task_struct *p, bool user, bool pi) { int oldpolicy = -1, policy = attr->sched_policy; - int retval, oldprio, newprio, queued, running; + int retval, oldprio, newprio; const struct sched_class *prev_class, *next_class; struct balance_callback *head; struct rq_flags rf; @@ -634,14 +611,14 @@ change: * Do not allow real-time tasks into groups that have no runtime * assigned. */ - if (rt_bandwidth_enabled() && rt_policy(policy) && + if (rt_group_sched_enabled() && + rt_bandwidth_enabled() && rt_policy(policy) && task_group(p)->rt_bandwidth.rt_runtime == 0 && !task_group_is_autogroup(task_group(p))) { retval = -EPERM; goto unlock; } -#endif -#ifdef CONFIG_SMP +#endif /* CONFIG_RT_GROUP_SCHED */ if (dl_bandwidth_enabled() && dl_policy(policy) && !(attr->sched_flags & SCHED_FLAG_SUGOV)) { cpumask_t *span = rq->rd->span; @@ -657,7 +634,6 @@ change: goto unlock; } } -#endif } /* Re-check policy now with rq lock held: */ @@ -692,45 +668,35 @@ change: * itself. */ newprio = rt_effective_prio(p, newprio); - if (newprio == oldprio) + if (newprio == oldprio && !dl_prio(newprio)) queue_flags &= ~DEQUEUE_MOVE; } prev_class = p->sched_class; next_class = __setscheduler_class(policy, newprio); - if (prev_class != next_class && p->se.sched_delayed) - dequeue_task(rq, p, DEQUEUE_SLEEP | DEQUEUE_DELAYED | DEQUEUE_NOCLOCK); + if (prev_class != next_class) + queue_flags |= DEQUEUE_CLASS; - queued = task_on_rq_queued(p); - running = task_current_donor(rq, p); - if (queued) - dequeue_task(rq, p, queue_flags); - if (running) - put_prev_task(rq, p); - - if (!(attr->sched_flags & SCHED_FLAG_KEEP_PARAMS)) { - __setscheduler_params(p, attr); - p->sched_class = next_class; - p->prio = newprio; - } - __setscheduler_uclamp(p, attr); - check_class_changing(rq, p, prev_class); + scoped_guard (sched_change, p, queue_flags) { - if (queued) { - /* - * We enqueue to tail when the priority of a task is - * increased (user space view). - */ - if (oldprio < p->prio) - queue_flags |= ENQUEUE_HEAD; + if (!(attr->sched_flags & SCHED_FLAG_KEEP_PARAMS)) { + __setscheduler_params(p, attr); + p->sched_class = next_class; + p->prio = newprio; + __setscheduler_dl_pi(newprio, policy, p, scope); + } + __setscheduler_uclamp(p, attr); - enqueue_task(rq, p, queue_flags); + if (scope->queued) { + /* + * We enqueue to tail when the priority of a task is + * increased (user space view). + */ + if (oldprio < p->prio) + scope->flags |= ENQUEUE_HEAD; + } } - if (running) - set_next_task(rq, p); - - check_class_changed(rq, p, prev_class, oldprio); /* Avoid rq from going away on us: */ preempt_disable(); @@ -860,6 +826,19 @@ void sched_set_fifo_low(struct task_struct *p) } EXPORT_SYMBOL_GPL(sched_set_fifo_low); +/* + * Used when the primary interrupt handler is forced into a thread, in addition + * to the (always threaded) secondary handler. The secondary handler gets a + * slightly lower priority so that the primary handler can preempt it, thereby + * emulating the behavior of a non-PREEMPT_RT system where the primary handler + * runs in hard interrupt context. + */ +void sched_set_fifo_secondary(struct task_struct *p) +{ + struct sched_param sp = { .sched_priority = MAX_RT_PRIO / 2 - 1 }; + WARN_ON_ONCE(sched_setscheduler_nocheck(p, SCHED_FIFO, &sp) != 0); +} + void sched_set_normal(struct task_struct *p, int nice) { struct sched_attr attr = { @@ -875,7 +854,7 @@ do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) { struct sched_param lparam; - if (!param || pid < 0) + if (unlikely(!param || pid < 0)) return -EINVAL; if (copy_from_user(&lparam, param, sizeof(struct sched_param))) return -EFAULT; @@ -932,10 +911,10 @@ err_size: return -E2BIG; } -static void get_params(struct task_struct *p, struct sched_attr *attr) +static void get_params(struct task_struct *p, struct sched_attr *attr, unsigned int flags) { if (task_has_dl_policy(p)) { - __getparam_dl(p, attr); + __getparam_dl(p, attr, flags); } else if (task_has_rt_policy(p)) { attr->sched_priority = p->rt_priority; } else { @@ -984,7 +963,7 @@ SYSCALL_DEFINE3(sched_setattr, pid_t, pid, struct sched_attr __user *, uattr, struct sched_attr attr; int retval; - if (!uattr || pid < 0 || flags) + if (unlikely(!uattr || pid < 0 || flags)) return -EINVAL; retval = sched_copy_attr(uattr, &attr); @@ -1001,7 +980,7 @@ SYSCALL_DEFINE3(sched_setattr, pid_t, pid, struct sched_attr __user *, uattr, return -ESRCH; if (attr.sched_flags & SCHED_FLAG_KEEP_PARAMS) - get_params(p, &attr); + get_params(p, &attr, 0); return sched_setattr(p, &attr); } @@ -1049,7 +1028,7 @@ SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) struct task_struct *p; int retval; - if (!param || pid < 0) + if (unlikely(!param || pid < 0)) return -EINVAL; scoped_guard (rcu) { @@ -1085,8 +1064,8 @@ SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr, struct task_struct *p; int retval; - if (!uattr || pid < 0 || usize > PAGE_SIZE || - usize < SCHED_ATTR_SIZE_VER0 || flags) + if (unlikely(!uattr || pid < 0 || usize > PAGE_SIZE || + usize < SCHED_ATTR_SIZE_VER0)) return -EINVAL; scoped_guard (rcu) { @@ -1094,6 +1073,12 @@ SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr, if (!p) return -ESRCH; + if (flags) { + if (!task_has_dl_policy(p) || + flags != SCHED_GETATTR_FLAG_DL_DYNAMIC) + return -EINVAL; + } + retval = security_task_getscheduler(p); if (retval) return retval; @@ -1101,7 +1086,7 @@ SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr, kattr.sched_policy = p->policy; if (p->sched_reset_on_fork) kattr.sched_flags |= SCHED_FLAG_RESET_ON_FORK; - get_params(p, &kattr); + get_params(p, &kattr, flags); kattr.sched_flags &= SCHED_FLAG_ALL; #ifdef CONFIG_UCLAMP_TASK @@ -1119,7 +1104,6 @@ SYSCALL_DEFINE4(sched_getattr, pid_t, pid, struct sched_attr __user *, uattr, return copy_struct_to_user(uattr, usize, &kattr, sizeof(kattr), NULL); } -#ifdef CONFIG_SMP int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask) { /* @@ -1148,7 +1132,6 @@ int dl_task_check_affinity(struct task_struct *p, const struct cpumask *mask) return 0; } -#endif /* CONFIG_SMP */ int __sched_setaffinity(struct task_struct *p, struct affinity_context *ctx) { @@ -1241,7 +1224,7 @@ long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) user_mask = alloc_user_cpus_ptr(NUMA_NO_NODE); if (user_mask) { cpumask_copy(user_mask, in_mask); - } else if (IS_ENABLED(CONFIG_SMP)) { + } else { return -ENOMEM; } @@ -1357,7 +1340,7 @@ static void do_sched_yield(void) rq = this_rq_lock_irq(&rf); schedstat_inc(rq->yld_count); - current->sched_class->yield_task(rq); + rq->donor->sched_class->yield_task(rq); preempt_disable(); rq_unlock_irq(rq, &rf); @@ -1426,12 +1409,13 @@ EXPORT_SYMBOL(yield); */ int __sched yield_to(struct task_struct *p, bool preempt) { - struct task_struct *curr = current; + struct task_struct *curr; struct rq *rq, *p_rq; int yielded = 0; scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { rq = this_rq(); + curr = rq->donor; again: p_rq = task_rq(p); diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index c49aea8c1025..5847b83d9d55 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -3,16 +3,25 @@ * Scheduler topology setup/handling methods */ +#include <linux/sched/isolation.h> +#include <linux/sched/clock.h> #include <linux/bsearch.h> +#include "sched.h" DEFINE_MUTEX(sched_domains_mutex); +void sched_domains_mutex_lock(void) +{ + mutex_lock(&sched_domains_mutex); +} +void sched_domains_mutex_unlock(void) +{ + mutex_unlock(&sched_domains_mutex); +} /* Protected by sched_domains_mutex: */ static cpumask_var_t sched_domains_tmpmask; static cpumask_var_t sched_domains_tmpmask2; -#ifdef CONFIG_SCHED_DEBUG - static int __init sched_debug_setup(char *str) { sched_debug_verbose = true; @@ -81,7 +90,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, break; } - if (!(sd->flags & SD_OVERLAP) && + if (!(sd->flags & SD_NUMA) && cpumask_intersects(groupmask, sched_group_span(group))) { printk(KERN_CONT "\n"); printk(KERN_ERR "ERROR: repeated CPUs\n"); @@ -94,7 +103,7 @@ static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, group->sgc->id, cpumask_pr_args(sched_group_span(group))); - if ((sd->flags & SD_OVERLAP) && + if ((sd->flags & SD_NUMA) && !cpumask_equal(group_balance_mask(group), sched_group_span(group))) { printk(KERN_CONT " mask=%*pbl", cpumask_pr_args(group_balance_mask(group))); @@ -151,15 +160,6 @@ static void sched_domain_debug(struct sched_domain *sd, int cpu) break; } } -#else /* !CONFIG_SCHED_DEBUG */ - -# define sched_debug_verbose 0 -# define sched_domain_debug(sd, cpu) do { } while (0) -static inline bool sched_debug(void) -{ - return false; -} -#endif /* CONFIG_SCHED_DEBUG */ /* Generate a mask of SD flags with the SDF_NEEDS_GROUPS metaflag */ #define SD_FLAG(name, mflags) (name * !!((mflags) & SDF_NEEDS_GROUPS)) | @@ -215,8 +215,6 @@ static bool sched_energy_update; static bool sched_is_eas_possible(const struct cpumask *cpu_mask) { bool any_asym_capacity = false; - struct cpufreq_policy *policy; - struct cpufreq_governor *gov; int i; /* EAS is enabled for asymmetric CPU capacity topologies. */ @@ -251,25 +249,12 @@ static bool sched_is_eas_possible(const struct cpumask *cpu_mask) return false; } - /* Do not attempt EAS if schedutil is not being used. */ - for_each_cpu(i, cpu_mask) { - policy = cpufreq_cpu_get(i); - if (!policy) { - if (sched_debug()) { - pr_info("rd %*pbl: Checking EAS, cpufreq policy not set for CPU: %d", - cpumask_pr_args(cpu_mask), i); - } - return false; - } - gov = policy->governor; - cpufreq_cpu_put(policy); - if (gov != &schedutil_gov) { - if (sched_debug()) { - pr_info("rd %*pbl: Checking EAS, schedutil is mandatory\n", - cpumask_pr_args(cpu_mask)); - } - return false; + if (!cpufreq_ready_for_eas(cpu_mask)) { + if (sched_debug()) { + pr_info("rd %*pbl: Checking EAS: cpufreq is not ready\n", + cpumask_pr_args(cpu_mask)); } + return false; } return true; @@ -288,7 +273,7 @@ void rebuild_sched_domains_energy(void) static int sched_energy_aware_handler(const struct ctl_table *table, int write, void *buffer, size_t *lenp, loff_t *ppos) { - int ret, state; + int ret; if (write && !capable(CAP_SYS_ADMIN)) return -EPERM; @@ -304,8 +289,7 @@ static int sched_energy_aware_handler(const struct ctl_table *table, int write, ret = proc_dointvec_minmax(table, write, buffer, lenp, ppos); if (!ret && write) { - state = static_branch_unlikely(&sched_energy_present); - if (state != sysctl_sched_energy_aware) + if (sysctl_sched_energy_aware != sched_energy_enabled()) rebuild_sched_domains_energy(); } @@ -331,7 +315,7 @@ static int __init sched_energy_aware_sysctl_init(void) } late_initcall(sched_energy_aware_sysctl_init); -#endif +#endif /* CONFIG_PROC_SYSCTL */ static void free_pd(struct perf_domain *pd) { @@ -366,7 +350,7 @@ static struct perf_domain *pd_init(int cpu) return NULL; } - pd = kzalloc(sizeof(*pd), GFP_KERNEL); + pd = kzalloc_obj(*pd); if (!pd) return NULL; pd->em_pd = obj; @@ -403,11 +387,11 @@ static void destroy_perf_domain_rcu(struct rcu_head *rp) static void sched_energy_set(bool has_eas) { - if (!has_eas && static_branch_unlikely(&sched_energy_present)) { + if (!has_eas && sched_energy_enabled()) { if (sched_debug()) pr_info("%s: stopping EAS\n", __func__); static_branch_disable_cpuslocked(&sched_energy_present); - } else if (has_eas && !static_branch_unlikely(&sched_energy_present)) { + } else if (has_eas && !sched_energy_enabled()) { if (sched_debug()) pr_info("%s: starting EAS\n", __func__); static_branch_enable_cpuslocked(&sched_energy_present); @@ -467,9 +451,9 @@ free: return false; } -#else +#else /* !(CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL): */ static void free_pd(struct perf_domain *pd) { } -#endif /* CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL*/ +#endif /* !(CONFIG_ENERGY_MODEL && CONFIG_CPU_FREQ_GOV_SCHEDUTIL) */ static void free_rootdomain(struct rcu_head *rcu) { @@ -524,6 +508,11 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd) if (rq->fair_server.dl_server) __dl_server_attach_root(&rq->fair_server, rq); +#ifdef CONFIG_SCHED_CLASS_EXT + if (rq->ext_server.dl_server) + __dl_server_attach_root(&rq->ext_server, rq); +#endif + rq_unlock_irqrestore(rq, &rf); if (old_rd) @@ -560,7 +549,7 @@ static int init_rootdomain(struct root_domain *rd) rd->rto_push_work = IRQ_WORK_INIT_HARD(rto_push_irq_work_func); #endif - rd->visit_gen = 0; + rd->visit_cookie = 0; init_dl_bw(&rd->dl_bw); if (cpudl_init(&rd->cpudl) != 0) goto free_rto_mask; @@ -600,7 +589,7 @@ static struct root_domain *alloc_rootdomain(void) { struct root_domain *rd; - rd = kzalloc(sizeof(*rd), GFP_KERNEL); + rd = kzalloc_obj(*rd); if (!rd) return NULL; @@ -695,6 +684,9 @@ static void update_top_cache_domain(int cpu) if (sd) { id = cpumask_first(sched_domain_span(sd)); size = cpumask_weight(sched_domain_span(sd)); + + /* If sd_llc exists, sd_llc_shared should exist too. */ + WARN_ON_ONCE(!sd->shared); sds = sd->shared; } @@ -743,6 +735,13 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) if (sd_parent_degenerate(tmp, parent)) { tmp->parent = parent->parent; + /* Pick reference to parent->shared. */ + if (parent->shared) { + WARN_ON_ONCE(tmp->shared); + tmp->shared = parent->shared; + parent->shared = NULL; + } + if (parent->parent) { parent->parent->child = tmp; parent->parent->groups->flags = tmp->flags; @@ -792,6 +791,7 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) } struct s_data { + struct sched_domain_shared * __percpu *sds; struct sched_domain * __percpu *sd; struct root_domain *rd; }; @@ -799,6 +799,7 @@ struct s_data { enum s_alloc { sa_rootdomain, sa_sd, + sa_sd_shared, sa_sd_storage, sa_none, }; @@ -1336,6 +1337,60 @@ next: update_group_capacity(sd, cpu); } +/* Update the "asym_prefer_cpu" when arch_asym_cpu_priority() changes. */ +void sched_update_asym_prefer_cpu(int cpu, int old_prio, int new_prio) +{ + int asym_prefer_cpu = cpu; + struct sched_domain *sd; + + guard(rcu)(); + + for_each_domain(cpu, sd) { + struct sched_group *sg; + int group_cpu; + + if (!(sd->flags & SD_ASYM_PACKING)) + continue; + + /* + * Groups of overlapping domain are replicated per NUMA + * node and will require updating "asym_prefer_cpu" on + * each local copy. + * + * If you are hitting this warning, consider moving + * "sg->asym_prefer_cpu" to "sg->sgc->asym_prefer_cpu" + * which is shared by all the overlapping groups. + */ + WARN_ON_ONCE(sd->flags & SD_NUMA); + + sg = sd->groups; + if (cpu != sg->asym_prefer_cpu) { + /* + * Since the parent is a superset of the current group, + * if the cpu is not the "asym_prefer_cpu" at the + * current level, it cannot be the preferred CPU at a + * higher levels either. + */ + if (!sched_asym_prefer(cpu, sg->asym_prefer_cpu)) + return; + + WRITE_ONCE(sg->asym_prefer_cpu, cpu); + continue; + } + + /* Ranking has improved; CPU is still the preferred one. */ + if (new_prio >= old_prio) + continue; + + for_each_cpu(group_cpu, sched_group_span(sg)) { + if (sched_asym_prefer(group_cpu, asym_prefer_cpu)) + asym_prefer_cpu = group_cpu; + } + + WRITE_ONCE(sg->asym_prefer_cpu, asym_prefer_cpu); + } +} + /* * Set of available CPUs grouped by their corresponding capacities * Each list entry contains a CPU mask reflecting CPUs that share the same @@ -1491,6 +1546,9 @@ static void set_domain_attribute(struct sched_domain *sd, static void __sdt_free(const struct cpumask *cpu_map); static int __sdt_alloc(const struct cpumask *cpu_map); +static void __sds_free(struct s_data *d, const struct cpumask *cpu_map); +static int __sds_alloc(struct s_data *d, const struct cpumask *cpu_map); + static void __free_domain_allocs(struct s_data *d, enum s_alloc what, const struct cpumask *cpu_map) { @@ -1502,6 +1560,9 @@ static void __free_domain_allocs(struct s_data *d, enum s_alloc what, case sa_sd: free_percpu(d->sd); fallthrough; + case sa_sd_shared: + __sds_free(d, cpu_map); + fallthrough; case sa_sd_storage: __sdt_free(cpu_map); fallthrough; @@ -1517,9 +1578,11 @@ __visit_domain_allocation_hell(struct s_data *d, const struct cpumask *cpu_map) if (__sdt_alloc(cpu_map)) return sa_sd_storage; + if (__sds_alloc(d, cpu_map)) + return sa_sd_shared; d->sd = alloc_percpu(struct sched_domain *); if (!d->sd) - return sa_sd_storage; + return sa_sd_shared; d->rd = alloc_rootdomain(); if (!d->rd) return sa_sd; @@ -1532,33 +1595,43 @@ __visit_domain_allocation_hell(struct s_data *d, const struct cpumask *cpu_map) * sched_group structure so that the subsequent __free_domain_allocs() * will not free the data we're using. */ -static void claim_allocations(int cpu, struct sched_domain *sd) +static void claim_allocations(int cpu, struct s_data *d) { - struct sd_data *sdd = sd->private; + struct sched_domain *sd; - WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); - *per_cpu_ptr(sdd->sd, cpu) = NULL; + if (atomic_read(&(*per_cpu_ptr(d->sds, cpu))->ref)) + *per_cpu_ptr(d->sds, cpu) = NULL; - if (atomic_read(&(*per_cpu_ptr(sdd->sds, cpu))->ref)) - *per_cpu_ptr(sdd->sds, cpu) = NULL; + for (sd = *per_cpu_ptr(d->sd, cpu); sd; sd = sd->parent) { + struct sd_data *sdd = sd->private; - if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref)) - *per_cpu_ptr(sdd->sg, cpu) = NULL; + WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); + *per_cpu_ptr(sdd->sd, cpu) = NULL; - if (atomic_read(&(*per_cpu_ptr(sdd->sgc, cpu))->ref)) - *per_cpu_ptr(sdd->sgc, cpu) = NULL; + if (atomic_read(&(*per_cpu_ptr(sdd->sg, cpu))->ref)) + *per_cpu_ptr(sdd->sg, cpu) = NULL; + + if (atomic_read(&(*per_cpu_ptr(sdd->sgc, cpu))->ref)) + *per_cpu_ptr(sdd->sgc, cpu) = NULL; + } } #ifdef CONFIG_NUMA enum numa_topology_type sched_numa_topology_type; +/* + * sched_domains_numa_distance is derived from sched_numa_node_distance + * and provides a simplified view of NUMA distances used specifically + * for building NUMA scheduling domains. + */ static int sched_domains_numa_levels; -static int sched_domains_curr_level; +static int sched_numa_node_levels; int sched_max_numa_distance; static int *sched_domains_numa_distance; +static int *sched_numa_node_distance; static struct cpumask ***sched_domains_numa_masks; -#endif +#endif /* CONFIG_NUMA */ /* * SD_flags allowed in topology descriptions. @@ -1593,21 +1666,19 @@ sd_init(struct sched_domain_topology_level *tl, struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); int sd_id, sd_weight, sd_flags = 0; struct cpumask *sd_span; + u64 now = sched_clock(); -#ifdef CONFIG_NUMA - /* - * Ugly hack to pass state to sd_numa_mask()... - */ - sched_domains_curr_level = tl->numa_level; -#endif - - sd_weight = cpumask_weight(tl->mask(cpu)); + sd_span = sched_domain_span(sd); + cpumask_and(sd_span, cpu_map, tl->mask(tl, cpu)); + sd_weight = cpumask_weight(sd_span); + sd_id = cpumask_first(sd_span); if (tl->sd_flags) sd_flags = (*tl->sd_flags)(); if (WARN_ONCE(sd_flags & ~TOPOLOGY_SD_FLAGS, - "wrong sd_flags in topology description\n")) + "wrong sd_flags in topology description\n")) sd_flags &= TOPOLOGY_SD_FLAGS; + sd_flags |= asym_cpu_capacity_classify(sd_span, cpu_map); *sd = (struct sched_domain){ .min_interval = sd_weight, @@ -1632,18 +1703,19 @@ sd_init(struct sched_domain_topology_level *tl, .last_balance = jiffies, .balance_interval = sd_weight, + + /* 50% success rate */ + .newidle_call = 512, + .newidle_success = 256, + .newidle_ratio = 512, + .newidle_stamp = now, + .max_newidle_lb_cost = 0, .last_decay_max_lb_cost = jiffies, .child = child, .name = tl->name, }; - sd_span = sched_domain_span(sd); - cpumask_and(sd_span, cpu_map, tl->mask(cpu)); - sd_id = cpumask_first(sd_span); - - sd->flags |= asym_cpu_capacity_classify(sd_span, cpu_map); - WARN_ONCE((sd->flags & (SD_SHARE_CPUCAPACITY | SD_ASYM_CPUCAPACITY)) == (SD_SHARE_CPUCAPACITY | SD_ASYM_CPUCAPACITY), "CPU capacity asymmetry not supported on SMT\n"); @@ -1674,42 +1746,73 @@ sd_init(struct sched_domain_topology_level *tl, SD_WAKE_AFFINE); } -#endif +#endif /* CONFIG_NUMA */ } else { sd->cache_nice_tries = 1; } - /* - * For all levels sharing cache; connect a sched_domain_shared - * instance. - */ - if (sd->flags & SD_SHARE_LLC) { - sd->shared = *per_cpu_ptr(sdd->sds, sd_id); - atomic_inc(&sd->shared->ref); - atomic_set(&sd->shared->nr_busy_cpus, sd_weight); - } - sd->private = sdd; return sd; } +#ifdef CONFIG_SCHED_SMT +int cpu_smt_flags(void) +{ + return SD_SHARE_CPUCAPACITY | SD_SHARE_LLC; +} + +const struct cpumask *tl_smt_mask(struct sched_domain_topology_level *tl, int cpu) +{ + return cpu_smt_mask(cpu); +} +#endif + +#ifdef CONFIG_SCHED_CLUSTER +int cpu_cluster_flags(void) +{ + return SD_CLUSTER | SD_SHARE_LLC; +} + +const struct cpumask *tl_cls_mask(struct sched_domain_topology_level *tl, int cpu) +{ + return cpu_clustergroup_mask(cpu); +} +#endif + +#ifdef CONFIG_SCHED_MC +int cpu_core_flags(void) +{ + return SD_SHARE_LLC; +} + +const struct cpumask *tl_mc_mask(struct sched_domain_topology_level *tl, int cpu) +{ + return cpu_coregroup_mask(cpu); +} +#endif + +const struct cpumask *tl_pkg_mask(struct sched_domain_topology_level *tl, int cpu) +{ + return cpu_node_mask(cpu); +} + /* * Topology list, bottom-up. */ static struct sched_domain_topology_level default_topology[] = { #ifdef CONFIG_SCHED_SMT - { cpu_smt_mask, cpu_smt_flags, SD_INIT_NAME(SMT) }, + SDTL_INIT(tl_smt_mask, cpu_smt_flags, SMT), #endif #ifdef CONFIG_SCHED_CLUSTER - { cpu_clustergroup_mask, cpu_cluster_flags, SD_INIT_NAME(CLS) }, + SDTL_INIT(tl_cls_mask, cpu_cluster_flags, CLS), #endif #ifdef CONFIG_SCHED_MC - { cpu_coregroup_mask, cpu_core_flags, SD_INIT_NAME(MC) }, + SDTL_INIT(tl_mc_mask, cpu_core_flags, MC), #endif - { cpu_cpu_mask, SD_INIT_NAME(PKG) }, + SDTL_INIT(tl_pkg_mask, NULL, PKG), { NULL, }, }; @@ -1730,10 +1833,14 @@ void __init set_sched_topology(struct sched_domain_topology_level *tl) } #ifdef CONFIG_NUMA +static int cpu_numa_flags(void) +{ + return SD_NUMA; +} -static const struct cpumask *sd_numa_mask(int cpu) +static const struct cpumask *sd_numa_mask(struct sched_domain_topology_level *tl, int cpu) { - return sched_domains_numa_masks[sched_domains_curr_level][cpu_to_node(cpu)]; + return sched_domains_numa_masks[tl->numa_level][cpu_to_node(cpu)]; } static void sched_numa_warn(const char *str) @@ -1770,10 +1877,10 @@ bool find_numa_distance(int distance) return true; rcu_read_lock(); - distances = rcu_dereference(sched_domains_numa_distance); + distances = rcu_dereference(sched_numa_node_distance); if (!distances) goto unlock; - for (i = 0; i < sched_domains_numa_levels; i++) { + for (i = 0; i < sched_numa_node_levels; i++) { if (distances[i] == distance) { found = true; break; @@ -1849,14 +1956,34 @@ static void init_numa_topology_type(int offline_node) #define NR_DISTANCE_VALUES (1 << DISTANCE_BITS) -void sched_init_numa(int offline_node) +/* + * An architecture could modify its NUMA distance, to change + * grouping of NUMA nodes and number of NUMA levels when creating + * NUMA level sched domains. + * + * A NUMA level is created for each unique + * arch_sched_node_distance. + */ +static int numa_node_dist(int i, int j) { - struct sched_domain_topology_level *tl; - unsigned long *distance_map; + return node_distance(i, j); +} + +int arch_sched_node_distance(int from, int to) + __weak __alias(numa_node_dist); + +static bool modified_sched_node_distance(void) +{ + return numa_node_dist != arch_sched_node_distance; +} + +static int sched_record_numa_dist(int offline_node, int (*n_dist)(int, int), + int **dist, int *levels) +{ + unsigned long *distance_map __free(bitmap) = NULL; int nr_levels = 0; int i, j; int *distances; - struct cpumask ***masks; /* * O(nr_nodes^2) de-duplicating selection sort -- in order to find the @@ -1864,17 +1991,16 @@ void sched_init_numa(int offline_node) */ distance_map = bitmap_alloc(NR_DISTANCE_VALUES, GFP_KERNEL); if (!distance_map) - return; + return -ENOMEM; bitmap_zero(distance_map, NR_DISTANCE_VALUES); for_each_cpu_node_but(i, offline_node) { for_each_cpu_node_but(j, offline_node) { - int distance = node_distance(i, j); + int distance = n_dist(i, j); if (distance < LOCAL_DISTANCE || distance >= NR_DISTANCE_VALUES) { sched_numa_warn("Invalid distance value range"); - bitmap_free(distance_map); - return; + return -EINVAL; } bitmap_set(distance_map, distance, 1); @@ -1886,19 +2012,47 @@ void sched_init_numa(int offline_node) */ nr_levels = bitmap_weight(distance_map, NR_DISTANCE_VALUES); - distances = kcalloc(nr_levels, sizeof(int), GFP_KERNEL); - if (!distances) { - bitmap_free(distance_map); - return; - } + distances = kzalloc_objs(int, nr_levels); + if (!distances) + return -ENOMEM; for (i = 0, j = 0; i < nr_levels; i++, j++) { j = find_next_bit(distance_map, NR_DISTANCE_VALUES, j); distances[i] = j; } - rcu_assign_pointer(sched_domains_numa_distance, distances); + *dist = distances; + *levels = nr_levels; + + return 0; +} + +void sched_init_numa(int offline_node) +{ + struct sched_domain_topology_level *tl; + int nr_levels, nr_node_levels; + int i, j; + int *distances, *domain_distances; + struct cpumask ***masks; - bitmap_free(distance_map); + /* Record the NUMA distances from SLIT table */ + if (sched_record_numa_dist(offline_node, numa_node_dist, &distances, + &nr_node_levels)) + return; + + /* Record modified NUMA distances for building sched domains */ + if (modified_sched_node_distance()) { + if (sched_record_numa_dist(offline_node, arch_sched_node_distance, + &domain_distances, &nr_levels)) { + kfree(distances); + return; + } + } else { + domain_distances = distances; + nr_levels = nr_node_levels; + } + rcu_assign_pointer(sched_numa_node_distance, distances); + WRITE_ONCE(sched_max_numa_distance, distances[nr_node_levels - 1]); + WRITE_ONCE(sched_numa_node_levels, nr_node_levels); /* * 'nr_levels' contains the number of unique distances @@ -1916,6 +2070,8 @@ void sched_init_numa(int offline_node) * * We reset it to 'nr_levels' at the end of this function. */ + rcu_assign_pointer(sched_domains_numa_distance, domain_distances); + sched_domains_numa_levels = 0; masks = kzalloc(sizeof(void *) * nr_levels, GFP_KERNEL); @@ -1941,10 +2097,13 @@ void sched_init_numa(int offline_node) masks[i][j] = mask; for_each_cpu_node_but(k, offline_node) { - if (sched_debug() && (node_distance(j, k) != node_distance(k, j))) + if (sched_debug() && + (arch_sched_node_distance(j, k) != + arch_sched_node_distance(k, j))) sched_numa_warn("Node-distance not symmetric"); - if (node_distance(j, k) > sched_domains_numa_distance[i]) + if (arch_sched_node_distance(j, k) > + sched_domains_numa_distance[i]) continue; cpumask_or(mask, mask, cpumask_of_node(k)); @@ -1970,30 +2129,20 @@ void sched_init_numa(int offline_node) /* * Add the NUMA identity distance, aka single NODE. */ - tl[i++] = (struct sched_domain_topology_level){ - .mask = sd_numa_mask, - .numa_level = 0, - SD_INIT_NAME(NODE) - }; + tl[i++] = SDTL_INIT(sd_numa_mask, NULL, NODE); /* * .. and append 'j' levels of NUMA goodness. */ for (j = 1; j < nr_levels; i++, j++) { - tl[i] = (struct sched_domain_topology_level){ - .mask = sd_numa_mask, - .sd_flags = cpu_numa_flags, - .flags = SDTL_OVERLAP, - .numa_level = j, - SD_INIT_NAME(NUMA) - }; + tl[i] = SDTL_INIT(sd_numa_mask, cpu_numa_flags, NUMA); + tl[i].numa_level = j; } sched_domain_topology_saved = sched_domain_topology; sched_domain_topology = tl; sched_domains_numa_levels = nr_levels; - WRITE_ONCE(sched_max_numa_distance, sched_domains_numa_distance[nr_levels - 1]); init_numa_topology_type(offline_node); } @@ -2001,14 +2150,18 @@ void sched_init_numa(int offline_node) static void sched_reset_numa(void) { - int nr_levels, *distances; + int nr_levels, *distances, *dom_distances = NULL; struct cpumask ***masks; nr_levels = sched_domains_numa_levels; + sched_numa_node_levels = 0; sched_domains_numa_levels = 0; sched_max_numa_distance = 0; sched_numa_topology_type = NUMA_DIRECT; - distances = sched_domains_numa_distance; + distances = sched_numa_node_distance; + if (sched_numa_node_distance != sched_domains_numa_distance) + dom_distances = sched_domains_numa_distance; + rcu_assign_pointer(sched_numa_node_distance, NULL); rcu_assign_pointer(sched_domains_numa_distance, NULL); masks = sched_domains_numa_masks; rcu_assign_pointer(sched_domains_numa_masks, NULL); @@ -2017,6 +2170,7 @@ static void sched_reset_numa(void) synchronize_rcu(); kfree(distances); + kfree(dom_distances); for (i = 0; i < nr_levels && masks; i++) { if (!masks[i]) continue; @@ -2063,7 +2217,8 @@ void sched_domains_numa_masks_set(unsigned int cpu) continue; /* Set ourselves in the remote node's masks */ - if (node_distance(j, node) <= sched_domains_numa_distance[i]) + if (arch_sched_node_distance(j, node) <= + sched_domains_numa_distance[i]) cpumask_set_cpu(cpu, sched_domains_numa_masks[i][j]); } } @@ -2101,7 +2256,7 @@ int sched_numa_find_closest(const struct cpumask *cpus, int cpu) for (i = 0; i < sched_domains_numa_levels; i++) { if (!masks[i][j]) break; - cpu = cpumask_any_and(cpus, masks[i][j]); + cpu = cpumask_any_and_distribute(cpus, masks[i][j]); if (cpu < nr_cpu_ids) { found = cpu; break; @@ -2172,6 +2327,8 @@ int sched_numa_find_nth_cpu(const struct cpumask *cpus, int cpu, int node) goto unlock; hop_masks = bsearch(&k, k.masks, sched_domains_numa_levels, sizeof(k.masks[0]), hop_cmp); + if (!hop_masks) + goto unlock; hop = hop_masks - k.masks; ret = hop ? @@ -2229,10 +2386,6 @@ static int __sdt_alloc(const struct cpumask *cpu_map) if (!sdd->sd) return -ENOMEM; - sdd->sds = alloc_percpu(struct sched_domain_shared *); - if (!sdd->sds) - return -ENOMEM; - sdd->sg = alloc_percpu(struct sched_group *); if (!sdd->sg) return -ENOMEM; @@ -2243,7 +2396,6 @@ static int __sdt_alloc(const struct cpumask *cpu_map) for_each_cpu(j, cpu_map) { struct sched_domain *sd; - struct sched_domain_shared *sds; struct sched_group *sg; struct sched_group_capacity *sgc; @@ -2254,13 +2406,6 @@ static int __sdt_alloc(const struct cpumask *cpu_map) *per_cpu_ptr(sdd->sd, j) = sd; - sds = kzalloc_node(sizeof(struct sched_domain_shared), - GFP_KERNEL, cpu_to_node(j)); - if (!sds) - return -ENOMEM; - - *per_cpu_ptr(sdd->sds, j) = sds; - sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), GFP_KERNEL, cpu_to_node(j)); if (!sg) @@ -2275,9 +2420,7 @@ static int __sdt_alloc(const struct cpumask *cpu_map) if (!sgc) return -ENOMEM; -#ifdef CONFIG_SCHED_DEBUG sgc->id = j; -#endif *per_cpu_ptr(sdd->sgc, j) = sgc; } @@ -2299,13 +2442,11 @@ static void __sdt_free(const struct cpumask *cpu_map) if (sdd->sd) { sd = *per_cpu_ptr(sdd->sd, j); - if (sd && (sd->flags & SD_OVERLAP)) + if (sd && (sd->flags & SD_NUMA)) free_sched_groups(sd->groups, 0); kfree(*per_cpu_ptr(sdd->sd, j)); } - if (sdd->sds) - kfree(*per_cpu_ptr(sdd->sds, j)); if (sdd->sg) kfree(*per_cpu_ptr(sdd->sg, j)); if (sdd->sgc) @@ -2313,8 +2454,6 @@ static void __sdt_free(const struct cpumask *cpu_map) } free_percpu(sdd->sd); sdd->sd = NULL; - free_percpu(sdd->sds); - sdd->sds = NULL; free_percpu(sdd->sg); sdd->sg = NULL; free_percpu(sdd->sgc); @@ -2322,6 +2461,42 @@ static void __sdt_free(const struct cpumask *cpu_map) } } +static int __sds_alloc(struct s_data *d, const struct cpumask *cpu_map) +{ + int j; + + d->sds = alloc_percpu(struct sched_domain_shared *); + if (!d->sds) + return -ENOMEM; + + for_each_cpu(j, cpu_map) { + struct sched_domain_shared *sds; + + sds = kzalloc_node(sizeof(struct sched_domain_shared), + GFP_KERNEL, cpu_to_node(j)); + if (!sds) + return -ENOMEM; + + *per_cpu_ptr(d->sds, j) = sds; + } + + return 0; +} + +static void __sds_free(struct s_data *d, const struct cpumask *cpu_map) +{ + int j; + + if (!d->sds) + return; + + for_each_cpu(j, cpu_map) + kfree(*per_cpu_ptr(d->sds, j)); + + free_percpu(d->sds); + d->sds = NULL; +} + static struct sched_domain *build_sched_domain(struct sched_domain_topology_level *tl, const struct cpumask *cpu_map, struct sched_domain_attr *attr, struct sched_domain *child, int cpu) @@ -2352,39 +2527,130 @@ static struct sched_domain *build_sched_domain(struct sched_domain_topology_leve /* * Ensure topology masks are sane, i.e. there are no conflicts (overlaps) for - * any two given CPUs at this (non-NUMA) topology level. + * any two given CPUs on non-NUMA topology levels. */ -static bool topology_span_sane(struct sched_domain_topology_level *tl, - const struct cpumask *cpu_map, int cpu) +static bool topology_span_sane(const struct cpumask *cpu_map) { - int i = cpu + 1; + struct sched_domain_topology_level *tl; + struct cpumask *covered, *id_seen; + int cpu; - /* NUMA levels are allowed to overlap */ - if (tl->flags & SDTL_OVERLAP) - return true; + lockdep_assert_held(&sched_domains_mutex); + covered = sched_domains_tmpmask; + id_seen = sched_domains_tmpmask2; + + for_each_sd_topology(tl) { + int tl_common_flags = 0; + + if (tl->sd_flags) + tl_common_flags = (*tl->sd_flags)(); + + /* NUMA levels are allowed to overlap */ + if (tl_common_flags & SD_NUMA) + continue; + + cpumask_clear(covered); + cpumask_clear(id_seen); - /* - * Non-NUMA levels cannot partially overlap - they must be either - * completely equal or completely disjoint. Otherwise we can end up - * breaking the sched_group lists - i.e. a later get_group() pass - * breaks the linking done for an earlier span. - */ - for_each_cpu_from(i, cpu_map) { /* - * We should 'and' all those masks with 'cpu_map' to exactly - * match the topology we're about to build, but that can only - * remove CPUs, which only lessens our ability to detect - * overlaps + * Non-NUMA levels cannot partially overlap - they must be either + * completely equal or completely disjoint. Otherwise we can end up + * breaking the sched_group lists - i.e. a later get_group() pass + * breaks the linking done for an earlier span. */ - if (!cpumask_equal(tl->mask(cpu), tl->mask(i)) && - cpumask_intersects(tl->mask(cpu), tl->mask(i))) - return false; - } + for_each_cpu(cpu, cpu_map) { + const struct cpumask *tl_cpu_mask = tl->mask(tl, cpu); + int id; + + /* lowest bit set in this mask is used as a unique id */ + id = cpumask_first(tl_cpu_mask); + if (cpumask_test_cpu(id, id_seen)) { + /* First CPU has already been seen, ensure identical spans */ + if (!cpumask_equal(tl->mask(tl, id), tl_cpu_mask)) + return false; + } else { + /* First CPU hasn't been seen before, ensure it's a completely new span */ + if (cpumask_intersects(tl_cpu_mask, covered)) + return false; + + cpumask_or(covered, covered, tl_cpu_mask); + cpumask_set_cpu(id, id_seen); + } + } + } return true; } /* + * Calculate an allowed NUMA imbalance such that LLCs do not get + * imbalanced. + */ +static void adjust_numa_imbalance(struct sched_domain *sd_llc) +{ + struct sched_domain *parent; + unsigned int imb_span = 1; + unsigned int imb = 0; + unsigned int nr_llcs; + + WARN_ON(!(sd_llc->flags & SD_SHARE_LLC)); + WARN_ON(!sd_llc->parent); + + /* + * For a single LLC per node, allow an + * imbalance up to 12.5% of the node. This is + * arbitrary cutoff based two factors -- SMT and + * memory channels. For SMT-2, the intent is to + * avoid premature sharing of HT resources but + * SMT-4 or SMT-8 *may* benefit from a different + * cutoff. For memory channels, this is a very + * rough estimate of how many channels may be + * active and is based on recent CPUs with + * many cores. + * + * For multiple LLCs, allow an imbalance + * until multiple tasks would share an LLC + * on one node while LLCs on another node + * remain idle. This assumes that there are + * enough logical CPUs per LLC to avoid SMT + * factors and that there is a correlation + * between LLCs and memory channels. + */ + nr_llcs = sd_llc->parent->span_weight / sd_llc->span_weight; + if (nr_llcs == 1) + imb = sd_llc->parent->span_weight >> 3; + else + imb = nr_llcs; + + imb = max(1U, imb); + sd_llc->parent->imb_numa_nr = imb; + + /* + * Set span based on the first NUMA domain. + * + * NUMA systems always add a NODE domain before + * iterating the NUMA domains. Since this is before + * degeneration, start from sd_llc's parent's + * parent which is the lowest an SD_NUMA domain can + * be relative to sd_llc. + */ + parent = sd_llc->parent->parent; + while (parent && !(parent->flags & SD_NUMA)) + parent = parent->parent; + + imb_span = parent ? parent->span_weight : sd_llc->parent->span_weight; + + /* Update the upper remainder of the topology */ + parent = sd_llc->parent; + while (parent) { + int factor = max(1U, (parent->span_weight / imb_span)); + + parent->imb_numa_nr = imb * factor; + parent = parent->parent; + } +} + +/* * Build sched domains for a given set of CPUs and attach the sched domains * to the individual CPUs */ @@ -2413,27 +2679,25 @@ build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att sd = NULL; for_each_sd_topology(tl) { - if (WARN_ON(!topology_span_sane(tl, cpu_map, i))) - goto error; - sd = build_sched_domain(tl, cpu_map, attr, sd, i); has_asym |= sd->flags & SD_ASYM_CPUCAPACITY; if (tl == sched_domain_topology) *per_cpu_ptr(d.sd, i) = sd; - if (tl->flags & SDTL_OVERLAP) - sd->flags |= SD_OVERLAP; if (cpumask_equal(cpu_map, sched_domain_span(sd))) break; } } + if (WARN_ON(!topology_span_sane(cpu_map))) + goto error; + /* Build the groups for the domains */ for_each_cpu(i, cpu_map) { for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { sd->span_weight = cpumask_weight(sched_domain_span(sd)); - if (sd->flags & SD_OVERLAP) { + if (sd->flags & SD_NUMA) { if (build_overlap_sched_groups(sd, i)) goto error; } else { @@ -2443,61 +2707,28 @@ build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att } } - /* - * Calculate an allowed NUMA imbalance such that LLCs do not get - * imbalanced. - */ for_each_cpu(i, cpu_map) { - unsigned int imb = 0; - unsigned int imb_span = 1; + sd = *per_cpu_ptr(d.sd, i); + if (!sd) + continue; - for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { - struct sched_domain *child = sd->child; - - if (!(sd->flags & SD_SHARE_LLC) && child && - (child->flags & SD_SHARE_LLC)) { - struct sched_domain __rcu *top_p; - unsigned int nr_llcs; - - /* - * For a single LLC per node, allow an - * imbalance up to 12.5% of the node. This is - * arbitrary cutoff based two factors -- SMT and - * memory channels. For SMT-2, the intent is to - * avoid premature sharing of HT resources but - * SMT-4 or SMT-8 *may* benefit from a different - * cutoff. For memory channels, this is a very - * rough estimate of how many channels may be - * active and is based on recent CPUs with - * many cores. - * - * For multiple LLCs, allow an imbalance - * until multiple tasks would share an LLC - * on one node while LLCs on another node - * remain idle. This assumes that there are - * enough logical CPUs per LLC to avoid SMT - * factors and that there is a correlation - * between LLCs and memory channels. - */ - nr_llcs = sd->span_weight / child->span_weight; - if (nr_llcs == 1) - imb = sd->span_weight >> 3; - else - imb = nr_llcs; - imb = max(1U, imb); - sd->imb_numa_nr = imb; - - /* Set span based on the first NUMA domain. */ - top_p = sd->parent; - while (top_p && !(top_p->flags & SD_NUMA)) { - top_p = top_p->parent; - } - imb_span = top_p ? top_p->span_weight : sd->span_weight; - } else { - int factor = max(1U, (sd->span_weight / imb_span)); + /* First, find the topmost SD_SHARE_LLC domain */ + while (sd->parent && (sd->parent->flags & SD_SHARE_LLC)) + sd = sd->parent; - sd->imb_numa_nr = imb * factor; - } + if (sd->flags & SD_SHARE_LLC) { + int sd_id = cpumask_first(sched_domain_span(sd)); + + sd->shared = *per_cpu_ptr(d.sds, sd_id); + atomic_set(&sd->shared->nr_busy_cpus, sd->span_weight); + atomic_inc(&sd->shared->ref); + + /* + * In presence of higher domains, adjust the + * NUMA imbalance stats for the hierarchy. + */ + if (IS_ENABLED(CONFIG_NUMA) && sd->parent) + adjust_numa_imbalance(sd); } } @@ -2506,10 +2737,10 @@ build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att if (!cpumask_test_cpu(i, cpu_map)) continue; - for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { - claim_allocations(i, sd); + claim_allocations(i, &d); + + for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) init_sched_groups_capacity(i, sd); - } } /* Attach the domains */ @@ -2572,7 +2803,7 @@ cpumask_var_t *alloc_sched_domains(unsigned int ndoms) int i; cpumask_var_t *doms; - doms = kmalloc_array(ndoms, sizeof(*doms), GFP_KERNEL); + doms = kmalloc_objs(*doms, ndoms); if (!doms) return NULL; for (i = 0; i < ndoms; i++) { @@ -2680,7 +2911,7 @@ static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, * * Call with hotplug lock and sched_domains_mutex held */ -void partition_sched_domains_locked(int ndoms_new, cpumask_var_t doms_new[], +static void partition_sched_domains_locked(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new) { bool __maybe_unused has_eas = false; @@ -2712,21 +2943,8 @@ void partition_sched_domains_locked(int ndoms_new, cpumask_var_t doms_new[], for (i = 0; i < ndoms_cur; i++) { for (j = 0; j < n && !new_topology; j++) { if (cpumask_equal(doms_cur[i], doms_new[j]) && - dattrs_equal(dattr_cur, i, dattr_new, j)) { - struct root_domain *rd; - - /* - * This domain won't be destroyed and as such - * its dl_bw->total_bw needs to be cleared. - * Tasks contribution will be then recomputed - * in function dl_update_tasks_root_domain(), - * dl_servers contribution in function - * dl_restore_server_root_domain(). - */ - rd = cpu_rq(cpumask_any(doms_cur[i]))->rd; - dl_clear_root_domain(rd); + dattrs_equal(dattr_cur, i, dattr_new, j)) goto match1; - } } /* No match - a current sched domain not in new doms_new[] */ detach_destroy_domains(doms_cur[i]); @@ -2783,6 +3001,7 @@ match3: ndoms_cur = ndoms_new; update_sched_domain_debugfs(); + dl_rebuild_rd_accounting(); } /* @@ -2791,7 +3010,7 @@ match3: void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], struct sched_domain_attr *dattr_new) { - mutex_lock(&sched_domains_mutex); + sched_domains_mutex_lock(); partition_sched_domains_locked(ndoms_new, doms_new, dattr_new); - mutex_unlock(&sched_domains_mutex); + sched_domains_mutex_unlock(); } diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 51e38f5f4701..20f27e2cf7ae 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -4,6 +4,7 @@ * * (C) 2004 Nadia Yvette Chambers, Oracle */ +#include "sched.h" void __init_waitqueue_head(struct wait_queue_head *wq_head, const char *name, struct lock_class_key *key) { @@ -40,13 +41,31 @@ void add_wait_queue_priority(struct wait_queue_head *wq_head, struct wait_queue_ { unsigned long flags; - wq_entry->flags |= WQ_FLAG_EXCLUSIVE | WQ_FLAG_PRIORITY; + wq_entry->flags |= WQ_FLAG_PRIORITY; spin_lock_irqsave(&wq_head->lock, flags); __add_wait_queue(wq_head, wq_entry); spin_unlock_irqrestore(&wq_head->lock, flags); } EXPORT_SYMBOL_GPL(add_wait_queue_priority); +int add_wait_queue_priority_exclusive(struct wait_queue_head *wq_head, + struct wait_queue_entry *wq_entry) +{ + struct list_head *head = &wq_head->head; + + wq_entry->flags |= WQ_FLAG_EXCLUSIVE | WQ_FLAG_PRIORITY; + + guard(spinlock_irqsave)(&wq_head->lock); + + if (!list_empty(head) && + (list_first_entry(head, typeof(*wq_entry), entry)->flags & WQ_FLAG_PRIORITY)) + return -EBUSY; + + list_add(&wq_entry->entry, head); + return 0; +} +EXPORT_SYMBOL_GPL(add_wait_queue_priority_exclusive); + void remove_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq_entry) { unsigned long flags; @@ -64,7 +83,7 @@ EXPORT_SYMBOL(remove_wait_queue); * the non-exclusive tasks. Normally, exclusive tasks will be at the end of * the list and any non-exclusive tasks will be woken first. A priority task * may be at the head of the list, and can consume the event without any other - * tasks being woken. + * tasks being woken if it's also an exclusive task. * * There are circumstances in which we can try to wake a task which has already * started to run but is not in state TASK_RUNNING. try_to_wake_up() returns diff --git a/kernel/sched/wait_bit.c b/kernel/sched/wait_bit.c index b410b61cec95..1088d3b7012c 100644 --- a/kernel/sched/wait_bit.c +++ b/kernel/sched/wait_bit.c @@ -1,5 +1,8 @@ // SPDX-License-Identifier: GPL-2.0-only +#include <linux/sched/debug.h> +#include "sched.h" + /* * The implementation of the wait_bit*() and related waiting APIs: */ |