diff options
Diffstat (limited to 'kernel/sched/core.c')
| -rw-r--r-- | kernel/sched/core.c | 13557 |
1 files changed, 8438 insertions, 5119 deletions
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 0d8eb4525e76..41ba0be16911 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -1,307 +1,709 @@ +// SPDX-License-Identifier: GPL-2.0-only /* * kernel/sched/core.c * - * Kernel scheduler and related syscalls + * Core kernel CPU scheduler code * * Copyright (C) 1991-2002 Linus Torvalds - * - * 1996-12-23 Modified by Dave Grothe to fix bugs in semaphores and - * make semaphores SMP safe - * 1998-11-19 Implemented schedule_timeout() and related stuff - * by Andrea Arcangeli - * 2002-01-04 New ultra-scalable O(1) scheduler by Ingo Molnar: - * hybrid priority-list and round-robin design with - * an array-switch method of distributing timeslices - * and per-CPU runqueues. Cleanups and useful suggestions - * by Davide Libenzi, preemptible kernel bits by Robert Love. - * 2003-09-03 Interactivity tuning by Con Kolivas. - * 2004-04-02 Scheduler domains code by Nick Piggin - * 2007-04-15 Work begun on replacing all interactivity tuning with a - * fair scheduling design by Con Kolivas. - * 2007-05-05 Load balancing (smp-nice) and other improvements - * by Peter Williams - * 2007-05-06 Interactivity improvements to CFS by Mike Galbraith - * 2007-07-01 Group scheduling enhancements by Srivatsa Vaddagiri - * 2007-11-29 RT balancing improvements by Steven Rostedt, Gregory Haskins, - * Thomas Gleixner, Mike Kravetz + * Copyright (C) 1998-2024 Ingo Molnar, Red Hat */ - -#include <linux/mm.h> -#include <linux/module.h> -#include <linux/nmi.h> -#include <linux/init.h> -#include <linux/uaccess.h> +#define INSTANTIATE_EXPORTED_MIGRATE_DISABLE +#include <linux/sched.h> #include <linux/highmem.h> -#include <asm/mmu_context.h> -#include <linux/interrupt.h> -#include <linux/capability.h> -#include <linux/completion.h> -#include <linux/kernel_stat.h> +#include <linux/hrtimer_api.h> +#include <linux/ktime_api.h> +#include <linux/sched/signal.h> +#include <linux/syscalls_api.h> #include <linux/debug_locks.h> -#include <linux/perf_event.h> -#include <linux/security.h> -#include <linux/notifier.h> -#include <linux/profile.h> -#include <linux/freezer.h> -#include <linux/vmalloc.h> +#include <linux/prefetch.h> +#include <linux/capability.h> +#include <linux/pgtable_api.h> +#include <linux/wait_bit.h> +#include <linux/jiffies.h> +#include <linux/spinlock_api.h> +#include <linux/cpumask_api.h> +#include <linux/lockdep_api.h> +#include <linux/hardirq.h> +#include <linux/softirq.h> +#include <linux/refcount_api.h> +#include <linux/topology.h> +#include <linux/sched/clock.h> +#include <linux/sched/cond_resched.h> +#include <linux/sched/cputime.h> +#include <linux/sched/debug.h> +#include <linux/sched/hotplug.h> +#include <linux/sched/init.h> +#include <linux/sched/isolation.h> +#include <linux/sched/loadavg.h> +#include <linux/sched/mm.h> +#include <linux/sched/nohz.h> +#include <linux/sched/rseq_api.h> +#include <linux/sched/rt.h> + #include <linux/blkdev.h> -#include <linux/delay.h> -#include <linux/pid_namespace.h> -#include <linux/smp.h> -#include <linux/threads.h> -#include <linux/timer.h> -#include <linux/rcupdate.h> -#include <linux/cpu.h> +#include <linux/context_tracking.h> #include <linux/cpuset.h> -#include <linux/percpu.h> -#include <linux/proc_fs.h> -#include <linux/seq_file.h> -#include <linux/sysctl.h> -#include <linux/syscalls.h> -#include <linux/times.h> -#include <linux/tsacct_kern.h> -#include <linux/kprobes.h> #include <linux/delayacct.h> -#include <linux/unistd.h> -#include <linux/pagemap.h> -#include <linux/hrtimer.h> -#include <linux/tick.h> -#include <linux/debugfs.h> -#include <linux/ctype.h> -#include <linux/ftrace.h> -#include <linux/slab.h> #include <linux/init_task.h> -#include <linux/binfmts.h> -#include <linux/context_tracking.h> +#include <linux/interrupt.h> +#include <linux/ioprio.h> +#include <linux/kallsyms.h> +#include <linux/kcov.h> +#include <linux/kprobes.h> +#include <linux/llist_api.h> +#include <linux/mmu_context.h> +#include <linux/mmzone.h> +#include <linux/mutex_api.h> +#include <linux/nmi.h> +#include <linux/nospec.h> +#include <linux/perf_event_api.h> +#include <linux/profile.h> +#include <linux/psi.h> +#include <linux/rcuwait_api.h> +#include <linux/rseq.h> +#include <linux/sched/wake_q.h> +#include <linux/scs.h> +#include <linux/slab.h> +#include <linux/syscalls.h> +#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_IRQ_ENTRY +# include <linux/irq-entry-common.h> +# endif +#endif +#include <uapi/linux/sched/types.h> + +#include <asm/irq_regs.h> #include <asm/switch_to.h> #include <asm/tlb.h> -#include <asm/irq_regs.h> -#include <asm/mutex.h> -#ifdef CONFIG_PARAVIRT -#include <asm/paravirt.h> -#endif - -#include "sched.h" -#include "../workqueue_internal.h" -#include "../smpboot.h" #define CREATE_TRACE_POINTS +#include <linux/sched/rseq_api.h> #include <trace/events/sched.h> +#include <trace/events/ipi.h> +#undef CREATE_TRACE_POINTS -void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period) -{ - unsigned long delta; - ktime_t soft, hard, now; +#include "sched.h" +#include "stats.h" - for (;;) { - if (hrtimer_active(period_timer)) - break; +#include "autogroup.h" +#include "pelt.h" +#include "smp.h" - now = hrtimer_cb_get_time(period_timer); - hrtimer_forward(period_timer, now, period); +#include "../workqueue_internal.h" +#include "../../io_uring/io-wq.h" +#include "../smpboot.h" +#include "../locking/mutex.h" - soft = hrtimer_get_softexpires(period_timer); - hard = hrtimer_get_expires(period_timer); - delta = ktime_to_ns(ktime_sub(hard, soft)); - __hrtimer_start_range_ns(period_timer, soft, delta, - HRTIMER_MODE_ABS_PINNED, 0); - } -} +EXPORT_TRACEPOINT_SYMBOL_GPL(ipi_send_cpu); +EXPORT_TRACEPOINT_SYMBOL_GPL(ipi_send_cpumask); -DEFINE_MUTEX(sched_domains_mutex); -DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); +/* + * Export tracepoints that act as a bare tracehook (ie: have no trace event + * associated with them) to allow external modules to probe them. + */ +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_cfs_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_rt_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_dl_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_irq_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_se_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(pelt_hw_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_cpu_capacity_tp); +EXPORT_TRACEPOINT_SYMBOL_GPL(sched_overutilized_tp); +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); -static void update_rq_clock_task(struct rq *rq, s64 delta); +DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); +DEFINE_PER_CPU(struct rnd_state, sched_rnd_state); -void update_rq_clock(struct rq *rq) +#ifdef CONFIG_SCHED_PROXY_EXEC +DEFINE_STATIC_KEY_TRUE(__sched_proxy_exec); +static int __init setup_proxy_exec(char *str) { - s64 delta; + bool proxy_enable = true; - if (rq->skip_clock_update > 0) - return; + if (*str && kstrtobool(str + 1, &proxy_enable)) { + pr_warn("Unable to parse sched_proxy_exec=\n"); + return 0; + } - delta = sched_clock_cpu(cpu_of(rq)) - rq->clock; - rq->clock += delta; - update_rq_clock_task(rq, delta); + 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); /* * Debugging: various feature bits + * + * If SCHED_DEBUG is disabled, each compilation unit has its own copy of + * sysctl_sched_features, defined in sched.h, to allow constants propagation + * at compile time and compiler optimization based on features default. */ - #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 -#ifdef CONFIG_SCHED_DEBUG -#define SCHED_FEAT(name, enabled) \ - #name , +/* + * Print a warning if need_resched is set for the given duration (if + * LATENCY_WARN is enabled). + * + * If sysctl_resched_latency_warn_once is set, only one warning will be shown + * per boot. + */ +__read_mostly int sysctl_resched_latency_warn_ms = 100; +__read_mostly int sysctl_resched_latency_warn_once = 1; -static const char * const sched_feat_names[] = { -#include "features.h" -}; +/* + * Number of tasks to iterate in a single balance run. + * Limited because this is done with IRQs disabled. + */ +__read_mostly unsigned int sysctl_sched_nr_migrate = SCHED_NR_MIGRATE_BREAK; -#undef SCHED_FEAT +__read_mostly int scheduler_running; + +#ifdef CONFIG_SCHED_CORE -static int sched_feat_show(struct seq_file *m, void *v) +DEFINE_STATIC_KEY_FALSE(__sched_core_enabled); + +/* kernel prio, less is more */ +static inline int __task_prio(const struct task_struct *p) { - int i; + if (p->sched_class == &stop_sched_class) /* trumps deadline */ + return -2; - for (i = 0; i < __SCHED_FEAT_NR; i++) { - if (!(sysctl_sched_features & (1UL << i))) - seq_puts(m, "NO_"); - seq_printf(m, "%s ", sched_feat_names[i]); - } - seq_puts(m, "\n"); + if (p->dl_server) + return -1; /* deadline */ - return 0; + if (rt_or_dl_prio(p->prio)) + return p->prio; /* [-1, 99] */ + + if (p->sched_class == &idle_sched_class) + return MAX_RT_PRIO + NICE_WIDTH; /* 140 */ + + if (task_on_scx(p)) + return MAX_RT_PRIO + MAX_NICE + 1; /* 120, squash ext */ + + return MAX_RT_PRIO + MAX_NICE; /* 119, squash fair */ } -#ifdef HAVE_JUMP_LABEL +/* + * l(a,b) + * le(a,b) := !l(b,a) + * g(a,b) := l(b,a) + * ge(a,b) := !l(a,b) + */ -#define jump_label_key__true STATIC_KEY_INIT_TRUE -#define jump_label_key__false STATIC_KEY_INIT_FALSE +/* real prio, less is less */ +static inline bool prio_less(const struct task_struct *a, + const struct task_struct *b, bool in_fi) +{ -#define SCHED_FEAT(name, enabled) \ - jump_label_key__##enabled , + int pa = __task_prio(a), pb = __task_prio(b); -struct static_key sched_feat_keys[__SCHED_FEAT_NR] = { -#include "features.h" -}; + if (-pa < -pb) + return true; -#undef SCHED_FEAT + if (-pb < -pa) + return false; + + if (pa == -1) { /* dl_prio() doesn't work because of stop_class above */ + const struct sched_dl_entity *a_dl, *b_dl; + + a_dl = &a->dl; + /* + * Since,'a' and 'b' can be CFS tasks served by DL server, + * __task_prio() can return -1 (for DL) even for those. In that + * case, get to the dl_server's DL entity. + */ + if (a->dl_server) + a_dl = a->dl_server; + + b_dl = &b->dl; + if (b->dl_server) + b_dl = b->dl_server; + + return !dl_time_before(a_dl->deadline, b_dl->deadline); + } + + if (pa == MAX_RT_PRIO + MAX_NICE) /* fair */ + return cfs_prio_less(a, b, in_fi); + +#ifdef CONFIG_SCHED_CLASS_EXT + if (pa == MAX_RT_PRIO + MAX_NICE + 1) /* ext */ + return scx_prio_less(a, b, in_fi); +#endif -static void sched_feat_disable(int i) + return false; +} + +static inline bool __sched_core_less(const struct task_struct *a, + const struct task_struct *b) { - if (static_key_enabled(&sched_feat_keys[i])) - static_key_slow_dec(&sched_feat_keys[i]); + if (a->core_cookie < b->core_cookie) + return true; + + if (a->core_cookie > b->core_cookie) + return false; + + /* flip prio, so high prio is leftmost */ + if (prio_less(b, a, !!task_rq(a)->core->core_forceidle_count)) + return true; + + return false; } -static void sched_feat_enable(int i) +#define __node_2_sc(node) rb_entry((node), struct task_struct, core_node) + +static inline bool rb_sched_core_less(struct rb_node *a, const struct rb_node *b) { - if (!static_key_enabled(&sched_feat_keys[i])) - static_key_slow_inc(&sched_feat_keys[i]); + return __sched_core_less(__node_2_sc(a), __node_2_sc(b)); } -#else -static void sched_feat_disable(int i) { }; -static void sched_feat_enable(int i) { }; -#endif /* HAVE_JUMP_LABEL */ -static int sched_feat_set(char *cmp) +static inline int rb_sched_core_cmp(const void *key, const struct rb_node *node) { - int i; - int neg = 0; + const struct task_struct *p = __node_2_sc(node); + unsigned long cookie = (unsigned long)key; - if (strncmp(cmp, "NO_", 3) == 0) { - neg = 1; - cmp += 3; - } + if (cookie < p->core_cookie) + return -1; - for (i = 0; i < __SCHED_FEAT_NR; i++) { - if (strcmp(cmp, sched_feat_names[i]) == 0) { - if (neg) { - sysctl_sched_features &= ~(1UL << i); - sched_feat_disable(i); - } else { - sysctl_sched_features |= (1UL << i); - sched_feat_enable(i); - } - break; - } - } + if (cookie > p->core_cookie) + return 1; - return i; + return 0; } -static ssize_t -sched_feat_write(struct file *filp, const char __user *ubuf, - size_t cnt, loff_t *ppos) +void sched_core_enqueue(struct rq *rq, struct task_struct *p) { - char buf[64]; - char *cmp; - int i; + if (p->se.sched_delayed) + return; - if (cnt > 63) - cnt = 63; + rq->core->core_task_seq++; - if (copy_from_user(&buf, ubuf, cnt)) - return -EFAULT; + if (!p->core_cookie) + return; - buf[cnt] = 0; - cmp = strstrip(buf); + rb_add(&p->core_node, &rq->core_tree, rb_sched_core_less); +} - i = sched_feat_set(cmp); - if (i == __SCHED_FEAT_NR) - return -EINVAL; +void sched_core_dequeue(struct rq *rq, struct task_struct *p, int flags) +{ + if (p->se.sched_delayed) + return; - *ppos += cnt; + rq->core->core_task_seq++; - return cnt; + if (sched_core_enqueued(p)) { + rb_erase(&p->core_node, &rq->core_tree); + RB_CLEAR_NODE(&p->core_node); + } + + /* + * Migrating the last task off the cpu, with the cpu in forced idle + * state. Reschedule to create an accounting edge for forced idle, + * and re-examine whether the core is still in forced idle state. + */ + if (!(flags & DEQUEUE_SAVE) && rq->nr_running == 1 && + rq->core->core_forceidle_count && rq->curr == rq->idle) + resched_curr(rq); } -static int sched_feat_open(struct inode *inode, struct file *filp) +static int sched_task_is_throttled(struct task_struct *p, int cpu) { - return single_open(filp, sched_feat_show, NULL); -} + if (p->sched_class->task_is_throttled) + return p->sched_class->task_is_throttled(p, cpu); -static const struct file_operations sched_feat_fops = { - .open = sched_feat_open, - .write = sched_feat_write, - .read = seq_read, - .llseek = seq_lseek, - .release = single_release, -}; + return 0; +} -static __init int sched_init_debug(void) +static struct task_struct *sched_core_next(struct task_struct *p, unsigned long cookie) { - debugfs_create_file("sched_features", 0644, NULL, NULL, - &sched_feat_fops); + struct rb_node *node = &p->core_node; + int cpu = task_cpu(p); - return 0; + do { + node = rb_next(node); + if (!node) + return NULL; + + p = __node_2_sc(node); + if (p->core_cookie != cookie) + return NULL; + + } while (sched_task_is_throttled(p, cpu)); + + return p; } -late_initcall(sched_init_debug); -#endif /* CONFIG_SCHED_DEBUG */ /* - * Number of tasks to iterate in a single balance run. - * Limited because this is done with IRQs disabled. + * Find left-most (aka, highest priority) and unthrottled task matching @cookie. + * If no suitable task is found, NULL will be returned. */ -const_debug unsigned int sysctl_sched_nr_migrate = 32; +static struct task_struct *sched_core_find(struct rq *rq, unsigned long cookie) +{ + struct task_struct *p; + struct rb_node *node; + + node = rb_find_first((void *)cookie, &rq->core_tree, rb_sched_core_cmp); + if (!node) + return NULL; + + p = __node_2_sc(node); + if (!sched_task_is_throttled(p, rq->cpu)) + return p; + + return sched_core_next(p, cookie); +} /* - * period over which we average the RT time consumption, measured - * in ms. + * Magic required such that: * - * default: 1s + * raw_spin_rq_lock(rq); + * ... + * raw_spin_rq_unlock(rq); + * + * ends up locking and unlocking the _same_ lock, and all CPUs + * always agree on what rq has what lock. + * + * XXX entirely possible to selectively enable cores, don't bother for now. */ -const_debug unsigned int sysctl_sched_time_avg = MSEC_PER_SEC; + +static DEFINE_MUTEX(sched_core_mutex); +static atomic_t sched_core_count; +static struct cpumask sched_core_mask; + +static void sched_core_lock(int cpu, unsigned long *flags) +{ + const struct cpumask *smt_mask = cpu_smt_mask(cpu); + int t, i = 0; + + local_irq_save(*flags); + for_each_cpu(t, smt_mask) + raw_spin_lock_nested(&cpu_rq(t)->__lock, i++); +} + +static void sched_core_unlock(int cpu, unsigned long *flags) +{ + const struct cpumask *smt_mask = cpu_smt_mask(cpu); + int t; + + for_each_cpu(t, smt_mask) + raw_spin_unlock(&cpu_rq(t)->__lock); + local_irq_restore(*flags); +} + +static void __sched_core_flip(bool enabled) +{ + unsigned long flags; + int cpu, t; + + cpus_read_lock(); + + /* + * Toggle the online cores, one by one. + */ + cpumask_copy(&sched_core_mask, cpu_online_mask); + for_each_cpu(cpu, &sched_core_mask) { + const struct cpumask *smt_mask = cpu_smt_mask(cpu); + + sched_core_lock(cpu, &flags); + + for_each_cpu(t, smt_mask) + cpu_rq(t)->core_enabled = enabled; + + cpu_rq(cpu)->core->core_forceidle_start = 0; + + sched_core_unlock(cpu, &flags); + + cpumask_andnot(&sched_core_mask, &sched_core_mask, smt_mask); + } + + /* + * Toggle the offline CPUs. + */ + for_each_cpu_andnot(cpu, cpu_possible_mask, cpu_online_mask) + cpu_rq(cpu)->core_enabled = enabled; + + cpus_read_unlock(); +} + +static void sched_core_assert_empty(void) +{ + int cpu; + + for_each_possible_cpu(cpu) + WARN_ON_ONCE(!RB_EMPTY_ROOT(&cpu_rq(cpu)->core_tree)); +} + +static void __sched_core_enable(void) +{ + static_branch_enable(&__sched_core_enabled); + /* + * Ensure all previous instances of raw_spin_rq_*lock() have finished + * and future ones will observe !sched_core_disabled(). + */ + synchronize_rcu(); + __sched_core_flip(true); + sched_core_assert_empty(); +} + +static void __sched_core_disable(void) +{ + sched_core_assert_empty(); + __sched_core_flip(false); + static_branch_disable(&__sched_core_enabled); +} + +void sched_core_get(void) +{ + if (atomic_inc_not_zero(&sched_core_count)) + return; + + mutex_lock(&sched_core_mutex); + if (!atomic_read(&sched_core_count)) + __sched_core_enable(); + + smp_mb__before_atomic(); + atomic_inc(&sched_core_count); + mutex_unlock(&sched_core_mutex); +} + +static void __sched_core_put(struct work_struct *work) +{ + if (atomic_dec_and_mutex_lock(&sched_core_count, &sched_core_mutex)) { + __sched_core_disable(); + mutex_unlock(&sched_core_mutex); + } +} + +void sched_core_put(void) +{ + static DECLARE_WORK(_work, __sched_core_put); + + /* + * "There can be only one" + * + * Either this is the last one, or we don't actually need to do any + * 'work'. If it is the last *again*, we rely on + * WORK_STRUCT_PENDING_BIT. + */ + if (!atomic_add_unless(&sched_core_count, -1, 1)) + schedule_work(&_work); +} + +#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 */ + +/* 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); /* - * period over which we measure -rt task cpu usage in us. - * default: 1s + * Serialization rules: + * + * Lock order: + * + * p->pi_lock + * rq->lock + * hrtimer_cpu_base->lock (hrtimer_start() for bandwidth controls) + * + * rq1->lock + * rq2->lock where: rq1 < rq2 + * + * Regular state: + * + * Normal scheduling state is serialized by rq->lock. __schedule() takes the + * local CPU's rq->lock, it optionally removes the task from the runqueue and + * always looks at the local rq data structures to find the most eligible task + * to run next. + * + * Task enqueue is also under rq->lock, possibly taken from another CPU. + * Wakeups from another LLC domain might use an IPI to transfer the enqueue to + * the local CPU to avoid bouncing the runqueue state around [ see + * ttwu_queue_wakelist() ] + * + * Task wakeup, specifically wakeups that involve migration, are horribly + * complicated to avoid having to take two rq->locks. + * + * Special state: + * + * System-calls and anything external will use task_rq_lock() which acquires + * both p->pi_lock and rq->lock. As a consequence the state they change is + * stable while holding either lock: + * + * - sched_setaffinity()/ + * set_cpus_allowed_ptr(): p->cpus_ptr, p->nr_cpus_allowed + * - set_user_nice(): p->se.load, p->*prio + * - __sched_setscheduler(): p->sched_class, p->policy, p->*prio, + * p->se.load, p->rt_priority, + * p->dl.dl_{runtime, deadline, period, flags, bw, density} + * - sched_setnuma(): p->numa_preferred_nid + * - sched_move_task(): p->sched_task_group + * - uclamp_update_active() p->uclamp* + * + * p->state <- TASK_*: + * + * is changed locklessly using set_current_state(), __set_current_state() or + * set_special_state(), see their respective comments, or by + * try_to_wake_up(). This latter uses p->pi_lock to serialize against + * concurrent self. + * + * p->on_rq <- { 0, 1 = TASK_ON_RQ_QUEUED, 2 = TASK_ON_RQ_MIGRATING }: + * + * 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(). + * + * Additionally it is possible to be ->on_rq but still be considered not + * runnable when p->se.sched_delayed is true. These tasks are on the runqueue + * but will be dequeued as soon as they get picked again. See the + * task_is_runnable() helper. + * + * p->on_cpu <- { 0, 1 }: + * + * is set by prepare_task() and cleared by finish_task() such that it will be + * set before p is scheduled-in and cleared after p is scheduled-out, both + * under rq->lock. Non-zero indicates the task is running on its CPU. + * + * [ The astute reader will observe that it is possible for two tasks on one + * CPU to have ->on_cpu = 1 at the same time. ] + * + * task_cpu(p): is changed by set_task_cpu(), the rules are: + * + * - Don't call set_task_cpu() on a blocked task: + * + * We don't care what CPU we're not running on, this simplifies hotplug, + * the CPU assignment of blocked tasks isn't required to be valid. + * + * - for try_to_wake_up(), called under p->pi_lock: + * + * This allows try_to_wake_up() to only take one rq->lock, see its comment. + * + * - for migration called under rq->lock: + * [ see task_on_rq_migrating() in task_rq_lock() ] + * + * o move_queued_task() + * o detach_task() + * + * - for migration called under double_rq_lock(): + * + * o __migrate_swap_task() + * o push_rt_task() / pull_rt_task() + * o push_dl_task() / pull_dl_task() + * o dl_task_offline_migration() + * */ -unsigned int sysctl_sched_rt_period = 1000000; -__read_mostly int scheduler_running; +void raw_spin_rq_lock_nested(struct rq *rq, int subclass) +{ + raw_spinlock_t *lock; + + /* Matches synchronize_rcu() in __sched_core_enable() */ + preempt_disable(); + if (sched_core_disabled()) { + raw_spin_lock_nested(&rq->__lock, subclass); + /* preempt_count *MUST* be > 1 */ + preempt_enable_no_resched(); + return; + } + + for (;;) { + lock = __rq_lockp(rq); + raw_spin_lock_nested(lock, subclass); + if (likely(lock == __rq_lockp(rq))) { + /* preempt_count *MUST* be > 1 */ + preempt_enable_no_resched(); + return; + } + raw_spin_unlock(lock); + } +} + +bool raw_spin_rq_trylock(struct rq *rq) +{ + raw_spinlock_t *lock; + bool ret; + + /* Matches synchronize_rcu() in __sched_core_enable() */ + preempt_disable(); + if (sched_core_disabled()) { + ret = raw_spin_trylock(&rq->__lock); + preempt_enable(); + return ret; + } + + for (;;) { + lock = __rq_lockp(rq); + ret = raw_spin_trylock(lock); + if (!ret || (likely(lock == __rq_lockp(rq)))) { + preempt_enable(); + return ret; + } + raw_spin_unlock(lock); + } +} + +void raw_spin_rq_unlock(struct rq *rq) +{ + raw_spin_unlock(rq_lockp(rq)); +} /* - * part of the period that we allow rt tasks to run in us. - * default: 0.95s + * double_rq_lock - safely lock two runqueues */ -int sysctl_sched_rt_runtime = 950000; +void double_rq_lock(struct rq *rq1, struct rq *rq2) +{ + lockdep_assert_irqs_disabled(); + if (rq_order_less(rq2, rq1)) + swap(rq1, rq2); + raw_spin_rq_lock(rq1); + if (__rq_lockp(rq1) != __rq_lockp(rq2)) + raw_spin_rq_lock_nested(rq2, SINGLE_DEPTH_NESTING); + + double_rq_clock_clear_update(rq1, rq2); +} /* * __task_rq_lock - lock the rq @p resides on. */ -static inline struct rq *__task_rq_lock(struct task_struct *p) +struct rq *__task_rq_lock(struct task_struct *p, struct rq_flags *rf) __acquires(rq->lock) { struct rq *rq; @@ -310,61 +712,146 @@ static inline struct rq *__task_rq_lock(struct task_struct *p) for (;;) { rq = task_rq(p); - raw_spin_lock(&rq->lock); - if (likely(rq == task_rq(p))) + raw_spin_rq_lock(rq); + if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) { + rq_pin_lock(rq, rf); return rq; - raw_spin_unlock(&rq->lock); + } + raw_spin_rq_unlock(rq); + + while (unlikely(task_on_rq_migrating(p))) + cpu_relax(); } } /* * task_rq_lock - lock p->pi_lock and lock the rq @p resides on. */ -static struct rq *task_rq_lock(struct task_struct *p, unsigned long *flags) +struct rq *task_rq_lock(struct task_struct *p, struct rq_flags *rf) __acquires(p->pi_lock) __acquires(rq->lock) { struct rq *rq; for (;;) { - raw_spin_lock_irqsave(&p->pi_lock, *flags); + raw_spin_lock_irqsave(&p->pi_lock, rf->flags); rq = task_rq(p); - raw_spin_lock(&rq->lock); - if (likely(rq == task_rq(p))) + raw_spin_rq_lock(rq); + /* + * move_queued_task() task_rq_lock() + * + * ACQUIRE (rq->lock) + * [S] ->on_rq = MIGRATING [L] rq = task_rq() + * WMB (__set_task_cpu()) ACQUIRE (rq->lock); + * [S] ->cpu = new_cpu [L] task_rq() + * [L] ->on_rq + * RELEASE (rq->lock) + * + * If we observe the old CPU in task_rq_lock(), the acquire of + * the old rq->lock will fully serialize against the stores. + * + * If we observe the new CPU in task_rq_lock(), the address + * dependency headed by '[L] rq = task_rq()' and the acquire + * will pair with the WMB to ensure we then also see migrating. + */ + if (likely(rq == task_rq(p) && !task_on_rq_migrating(p))) { + rq_pin_lock(rq, rf); return rq; - raw_spin_unlock(&rq->lock); - raw_spin_unlock_irqrestore(&p->pi_lock, *flags); + } + raw_spin_rq_unlock(rq); + raw_spin_unlock_irqrestore(&p->pi_lock, rf->flags); + + while (unlikely(task_on_rq_migrating(p))) + cpu_relax(); } } -static void __task_rq_unlock(struct rq *rq) - __releases(rq->lock) -{ - raw_spin_unlock(&rq->lock); -} +/* + * RQ-clock updating methods: + */ -static inline void -task_rq_unlock(struct rq *rq, struct task_struct *p, unsigned long *flags) - __releases(rq->lock) - __releases(p->pi_lock) +static void update_rq_clock_task(struct rq *rq, s64 delta) { - raw_spin_unlock(&rq->lock); - raw_spin_unlock_irqrestore(&p->pi_lock, *flags); -} - /* - * this_rq_lock - lock this runqueue and disable interrupts. + * In theory, the compile should just see 0 here, and optimize out the call + * to sched_rt_avg_update. But I don't trust it... */ -static struct rq *this_rq_lock(void) - __acquires(rq->lock) + s64 __maybe_unused steal = 0, irq_delta = 0; + +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + if (irqtime_enabled()) { + irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; + + /* + * Since irq_time is only updated on {soft,}irq_exit, we might run into + * this case when a previous update_rq_clock() happened inside a + * {soft,}IRQ region. + * + * When this happens, we stop ->clock_task and only update the + * prev_irq_time stamp to account for the part that fit, so that a next + * update will consume the rest. This ensures ->clock_task is + * monotonic. + * + * It does however cause some slight miss-attribution of {soft,}IRQ + * time, a more accurate solution would be to update the irq_time using + * the current rq->clock timestamp, except that would require using + * atomic ops. + */ + if (irq_delta > delta) + irq_delta = delta; + + rq->prev_irq_time += irq_delta; + delta -= irq_delta; + delayacct_irq(rq->curr, irq_delta); + } +#endif +#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING + if (static_key_false((¶virt_steal_rq_enabled))) { + u64 prev_steal; + + steal = prev_steal = paravirt_steal_clock(cpu_of(rq)); + steal -= rq->prev_steal_time_rq; + + if (unlikely(steal > delta)) + steal = delta; + + rq->prev_steal_time_rq = prev_steal; + delta -= steal; + } +#endif + + rq->clock_task += delta; + +#ifdef CONFIG_HAVE_SCHED_AVG_IRQ + if ((irq_delta + steal) && sched_feat(NONTASK_CAPACITY)) + update_irq_load_avg(rq, irq_delta + steal); +#endif + update_rq_clock_pelt(rq, delta); +} + +void update_rq_clock(struct rq *rq) { - struct rq *rq; + s64 delta; + u64 clock; - local_irq_disable(); - rq = this_rq(); - raw_spin_lock(&rq->lock); + lockdep_assert_rq_held(rq); - return rq; + if (rq->clock_update_flags & RQCF_ACT_SKIP) + return; + + if (sched_feat(WARN_DOUBLE_CLOCK)) + WARN_ON_ONCE(rq->clock_update_flags & RQCF_UPDATED); + rq->clock_update_flags |= RQCF_UPDATED; + + clock = sched_clock_cpu(cpu_of(rq)); + scx_rq_clock_update(rq, clock); + + delta = clock - rq->clock; + if (delta < 0) + return; + rq->clock += delta; + + update_rq_clock_task(rq, delta); } #ifdef CONFIG_SCHED_HRTICK @@ -385,25 +872,24 @@ static void hrtick_clear(struct rq *rq) static enum hrtimer_restart hrtick(struct hrtimer *timer) { struct rq *rq = container_of(timer, struct rq, hrtick_timer); + struct rq_flags rf; WARN_ON_ONCE(cpu_of(rq) != smp_processor_id()); - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); update_rq_clock(rq); - rq->curr->sched_class->task_tick(rq, rq->curr, 1); - raw_spin_unlock(&rq->lock); + rq->donor->sched_class->task_tick(rq, rq->donor, 1); + rq_unlock(rq, &rf); return HRTIMER_NORESTART; } -#ifdef CONFIG_SMP - -static int __hrtick_restart(struct rq *rq) +static void __hrtick_restart(struct rq *rq) { struct hrtimer *timer = &rq->hrtick_timer; - ktime_t time = hrtimer_get_softexpires(timer); + ktime_t time = rq->hrtick_time; - return __hrtimer_start_range_ns(timer, time, 0, HRTIMER_MODE_ABS_PINNED, 0); + hrtimer_start(timer, time, HRTIMER_MODE_ABS_PINNED_HARD); } /* @@ -412,127 +898,274 @@ static int __hrtick_restart(struct rq *rq) static void __hrtick_start(void *arg) { struct rq *rq = arg; + struct rq_flags rf; - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); __hrtick_restart(rq); - rq->hrtick_csd_pending = 0; - raw_spin_unlock(&rq->lock); + rq_unlock(rq, &rf); } /* * Called to set the hrtick timer state. * - * called with rq->lock held and irqs disabled + * called with rq->lock held and IRQs disabled */ void hrtick_start(struct rq *rq, u64 delay) { struct hrtimer *timer = &rq->hrtick_timer; - ktime_t time = ktime_add_ns(timer->base->get_time(), delay); + s64 delta; - hrtimer_set_expires(timer, time); + /* + * Don't schedule slices shorter than 10000ns, that just + * doesn't make sense and can cause timer DoS. + */ + delta = max_t(s64, delay, 10000LL); + rq->hrtick_time = ktime_add_ns(hrtimer_cb_get_time(timer), delta); - if (rq == this_rq()) { + if (rq == this_rq()) __hrtick_restart(rq); - } else if (!rq->hrtick_csd_pending) { - __smp_call_function_single(cpu_of(rq), &rq->hrtick_csd, 0); - rq->hrtick_csd_pending = 1; - } + else + smp_call_function_single_async(cpu_of(rq), &rq->hrtick_csd); } -static int -hotplug_hrtick(struct notifier_block *nfb, unsigned long action, void *hcpu) +static void hrtick_rq_init(struct rq *rq) { - int cpu = (int)(long)hcpu; - - switch (action) { - case CPU_UP_CANCELED: - case CPU_UP_CANCELED_FROZEN: - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - case CPU_DEAD: - case CPU_DEAD_FROZEN: - hrtick_clear(cpu_rq(cpu)); - return NOTIFY_OK; - } + INIT_CSD(&rq->hrtick_csd, __hrtick_start, rq); + hrtimer_setup(&rq->hrtick_timer, hrtick, CLOCK_MONOTONIC, HRTIMER_MODE_REL_HARD); +} +#else /* !CONFIG_SCHED_HRTICK: */ +static inline void hrtick_clear(struct rq *rq) +{ +} - return NOTIFY_DONE; +static inline void hrtick_rq_init(struct rq *rq) +{ } +#endif /* !CONFIG_SCHED_HRTICK */ -static __init void init_hrtick(void) +/* + * try_cmpxchg based fetch_or() macro so it works for different integer types: + */ +#define fetch_or(ptr, mask) \ + ({ \ + typeof(ptr) _ptr = (ptr); \ + typeof(mask) _mask = (mask); \ + typeof(*_ptr) _val = *_ptr; \ + \ + do { \ + } while (!try_cmpxchg(_ptr, &_val, _val | _mask)); \ + _val; \ +}) + +#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 + * spurious IPIs. + */ +static inline bool set_nr_and_not_polling(struct thread_info *ti, int tif) { - hotcpu_notifier(hotplug_hrtick, 0); + return !(fetch_or(&ti->flags, 1 << tif) & _TIF_POLLING_NRFLAG); } -#else + /* - * Called to set the hrtick timer state. + * Atomically set TIF_NEED_RESCHED if TIF_POLLING_NRFLAG is set. * - * called with rq->lock held and irqs disabled + * If this returns true, then the idle task promises to call + * sched_ttwu_pending() and reschedule soon. */ -void hrtick_start(struct rq *rq, u64 delay) +static bool set_nr_if_polling(struct task_struct *p) +{ + struct thread_info *ti = task_thread_info(p); + typeof(ti->flags) val = READ_ONCE(ti->flags); + + do { + if (!(val & _TIF_POLLING_NRFLAG)) + return false; + if (val & _TIF_NEED_RESCHED) + return true; + } while (!try_cmpxchg(&ti->flags, &val, val | _TIF_NEED_RESCHED)); + + return true; +} + +#else +static inline bool set_nr_and_not_polling(struct thread_info *ti, int tif) { - __hrtimer_start_range_ns(&rq->hrtick_timer, ns_to_ktime(delay), 0, - HRTIMER_MODE_REL_PINNED, 0); + set_ti_thread_flag(ti, tif); + return true; } -static inline void init_hrtick(void) +static inline bool set_nr_if_polling(struct task_struct *p) { + return false; } -#endif /* CONFIG_SMP */ +#endif -static void init_rq_hrtick(struct rq *rq) +static bool __wake_q_add(struct wake_q_head *head, struct task_struct *task) { -#ifdef CONFIG_SMP - rq->hrtick_csd_pending = 0; + struct wake_q_node *node = &task->wake_q; - rq->hrtick_csd.flags = 0; - rq->hrtick_csd.func = __hrtick_start; - rq->hrtick_csd.info = rq; -#endif + /* + * Atomically grab the task, if ->wake_q is !nil already it means + * it's already queued (either by us or someone else) and will get the + * wakeup due to that. + * + * In order to ensure that a pending wakeup will observe our pending + * state, even in the failed case, an explicit smp_mb() must be used. + */ + smp_mb__before_atomic(); + if (unlikely(cmpxchg_relaxed(&node->next, NULL, WAKE_Q_TAIL))) + return false; - hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); - rq->hrtick_timer.function = hrtick; + /* + * The head is context local, there can be no concurrency. + */ + *head->lastp = node; + head->lastp = &node->next; + return true; } -#else /* CONFIG_SCHED_HRTICK */ -static inline void hrtick_clear(struct rq *rq) + +/** + * wake_q_add() - queue a wakeup for 'later' waking. + * @head: the wake_q_head to add @task to + * @task: the task to queue for 'later' wakeup + * + * Queue a task for later wakeup, most likely by the wake_up_q() call in the + * same context, _HOWEVER_ this is not guaranteed, the wakeup can come + * instantly. + * + * This function must be used as-if it were wake_up_process(); IOW the task + * must be ready to be woken at this location. + */ +void wake_q_add(struct wake_q_head *head, struct task_struct *task) { + if (__wake_q_add(head, task)) + get_task_struct(task); } -static inline void init_rq_hrtick(struct rq *rq) +/** + * wake_q_add_safe() - safely queue a wakeup for 'later' waking. + * @head: the wake_q_head to add @task to + * @task: the task to queue for 'later' wakeup + * + * Queue a task for later wakeup, most likely by the wake_up_q() call in the + * same context, _HOWEVER_ this is not guaranteed, the wakeup can come + * instantly. + * + * This function must be used as-if it were wake_up_process(); IOW the task + * must be ready to be woken at this location. + * + * This function is essentially a task-safe equivalent to wake_q_add(). Callers + * that already hold reference to @task can call the 'safe' version and trust + * wake_q to do the right thing depending whether or not the @task is already + * queued for wakeup. + */ +void wake_q_add_safe(struct wake_q_head *head, struct task_struct *task) { + if (!__wake_q_add(head, task)) + put_task_struct(task); } -static inline void init_hrtick(void) +void wake_up_q(struct wake_q_head *head) { + struct wake_q_node *node = head->first; + + while (node != WAKE_Q_TAIL) { + struct task_struct *task; + + task = container_of(node, struct task_struct, wake_q); + node = node->next; + /* pairs with cmpxchg_relaxed() in __wake_q_add() */ + WRITE_ONCE(task->wake_q.next, NULL); + /* Task can safely be re-inserted now. */ + + /* + * wake_up_process() executes a full barrier, which pairs with + * the queueing in wake_q_add() so as not to miss wakeups. + */ + wake_up_process(task); + put_task_struct(task); + } } -#endif /* CONFIG_SCHED_HRTICK */ /* - * resched_task - mark a task 'to be rescheduled now'. + * resched_curr - mark rq's current task 'to be rescheduled now'. * * On UP this means the setting of the need_resched flag, on SMP it * might also involve a cross-CPU call to trigger the scheduler on * the target CPU. */ -#ifdef CONFIG_SMP -void resched_task(struct task_struct *p) +static void __resched_curr(struct rq *rq, int tif) { + struct task_struct *curr = rq->curr; + struct thread_info *cti = task_thread_info(curr); int cpu; - assert_raw_spin_locked(&task_rq(p)->lock); + lockdep_assert_rq_held(rq); - if (test_tsk_need_resched(p)) + /* + * Always immediately preempt the idle task; no point in delaying doing + * actual work. + */ + if (is_idle_task(curr) && tif == TIF_NEED_RESCHED_LAZY) + tif = TIF_NEED_RESCHED; + + if (cti->flags & ((1 << tif) | _TIF_NEED_RESCHED)) return; - set_tsk_need_resched(p); + cpu = cpu_of(rq); - cpu = task_cpu(p); - if (cpu == smp_processor_id()) + 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) + set_preempt_need_resched(); return; + } - /* NEED_RESCHED must be visible before we test polling */ - smp_mb(); - if (!tsk_is_polling(p)) - smp_send_reschedule(cpu); + if (set_nr_and_not_polling(cti, tif)) { + if (tif == TIF_NEED_RESCHED) + smp_send_reschedule(cpu); + } else { + trace_sched_wake_idle_without_ipi(cpu); + } +} + +void __trace_set_need_resched(struct task_struct *curr, int tif) +{ + trace_sched_set_need_resched_tp(curr, smp_processor_id(), tif); +} + +void resched_curr(struct rq *rq) +{ + __resched_curr(rq, TIF_NEED_RESCHED); +} + +#ifdef CONFIG_PREEMPT_DYNAMIC +static DEFINE_STATIC_KEY_FALSE(sk_dynamic_preempt_lazy); +static __always_inline bool dynamic_preempt_lazy(void) +{ + return static_branch_unlikely(&sk_dynamic_preempt_lazy); +} +#else +static __always_inline bool dynamic_preempt_lazy(void) +{ + return IS_ENABLED(CONFIG_PREEMPT_LAZY); +} +#endif + +static __always_inline int get_lazy_tif_bit(void) +{ + if (dynamic_preempt_lazy()) + return TIF_NEED_RESCHED_LAZY; + + return TIF_NEED_RESCHED; +} + +void resched_curr_lazy(struct rq *rq) +{ + __resched_curr(rq, get_lazy_tif_bit()); } void resched_cpu(int cpu) @@ -540,40 +1173,53 @@ void resched_cpu(int cpu) struct rq *rq = cpu_rq(cpu); unsigned long flags; - if (!raw_spin_trylock_irqsave(&rq->lock, flags)) - return; - resched_task(cpu_curr(cpu)); - raw_spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_rq_lock_irqsave(rq, flags); + if (cpu_online(cpu) || cpu == smp_processor_id()) + resched_curr(rq); + raw_spin_rq_unlock_irqrestore(rq, flags); } #ifdef CONFIG_NO_HZ_COMMON /* - * In the semi idle case, use the nearest busy cpu for migrating timers - * from an idle cpu. This is good for power-savings. + * In the semi idle case, use the nearest busy CPU for migrating timers + * from an idle CPU. This is good for power-savings. * * We don't do similar optimization for completely idle system, as - * selecting an idle cpu will add more delays to the timers than intended - * (as that cpu's timer base may not be uptodate wrt jiffies etc). + * selecting an idle CPU will add more delays to the timers than intended + * (as that CPU's timer base may not be up to date wrt jiffies etc). */ int get_nohz_timer_target(void) { - int cpu = smp_processor_id(); - int i; + int i, cpu = smp_processor_id(), default_cpu = -1; struct sched_domain *sd; + const struct cpumask *hk_mask; + + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) { + if (!idle_cpu(cpu)) + return cpu; + default_cpu = cpu; + } + + hk_mask = housekeeping_cpumask(HK_TYPE_KERNEL_NOISE); + + guard(rcu)(); - rcu_read_lock(); for_each_domain(cpu, sd) { - for_each_cpu(i, sched_domain_span(sd)) { - if (!idle_cpu(i)) { - cpu = i; - goto unlock; - } + for_each_cpu_and(i, sched_domain_span(sd), hk_mask) { + if (cpu == i) + continue; + + if (!idle_cpu(i)) + return i; } } -unlock: - rcu_read_unlock(); - return cpu; + + if (default_cpu == -1) + default_cpu = housekeeping_any_cpu(HK_TYPE_KERNEL_NOISE); + + return default_cpu; } + /* * When add_timer_on() enqueues a timer into the timer wheel of an * idle CPU then this timer might expire before the next timer event @@ -592,117 +1238,155 @@ static void wake_up_idle_cpu(int cpu) return; /* - * This is safe, as this function is called with the timer - * wheel base lock of (cpu) held. When the CPU is on the way - * to idle and has not yet set rq->curr to idle then it will - * be serialized on the timer wheel base lock and take the new - * timer into account automatically. - */ - if (rq->curr != rq->idle) - return; - - /* - * We can set TIF_RESCHED on the idle task of the other CPU - * lockless. The worst case is that the other CPU runs the - * idle task through an additional NOOP schedule() + * Set TIF_NEED_RESCHED and send an IPI if in the non-polling + * part of the idle loop. This forces an exit from the idle loop + * and a round trip to schedule(). Now this could be optimized + * because a simple new idle loop iteration is enough to + * re-evaluate the next tick. Provided some re-ordering of tick + * nohz functions that would need to follow TIF_NR_POLLING + * clearing: + * + * - On most architectures, a simple fetch_or on ti::flags with a + * "0" value would be enough to know if an IPI needs to be sent. + * + * - x86 needs to perform a last need_resched() check between + * monitor and mwait which doesn't take timers into account. + * There a dedicated TIF_TIMER flag would be required to + * fetch_or here and be checked along with TIF_NEED_RESCHED + * before mwait(). + * + * However, remote timer enqueue is not such a frequent event + * and testing of the above solutions didn't appear to report + * much benefits. */ - set_tsk_need_resched(rq->idle); - - /* NEED_RESCHED must be visible before we test polling */ - smp_mb(); - if (!tsk_is_polling(rq->idle)) + if (set_nr_and_not_polling(task_thread_info(rq->idle), TIF_NEED_RESCHED)) smp_send_reschedule(cpu); + else + trace_sched_wake_idle_without_ipi(cpu); } static bool wake_up_full_nohz_cpu(int cpu) { + /* + * We just need the target to call irq_exit() and re-evaluate + * the next tick. The nohz full kick at least implies that. + * If needed we can still optimize that later with an + * empty IRQ. + */ + if (cpu_is_offline(cpu)) + return true; /* Don't try to wake offline CPUs. */ if (tick_nohz_full_cpu(cpu)) { if (cpu != smp_processor_id() || tick_nohz_tick_stopped()) - smp_send_reschedule(cpu); + tick_nohz_full_kick_cpu(cpu); return true; } return false; } +/* + * Wake up the specified CPU. If the CPU is going offline, it is the + * caller's responsibility to deal with the lost wakeup, for example, + * by hooking into the CPU_DEAD notifier like timers and hrtimers do. + */ void wake_up_nohz_cpu(int cpu) { if (!wake_up_full_nohz_cpu(cpu)) wake_up_idle_cpu(cpu); } -static inline bool got_nohz_idle_kick(void) +static void nohz_csd_func(void *info) { - int cpu = smp_processor_id(); - - if (!test_bit(NOHZ_BALANCE_KICK, nohz_flags(cpu))) - return false; - - if (idle_cpu(cpu) && !need_resched()) - return true; + struct rq *rq = info; + int cpu = cpu_of(rq); + unsigned int flags; /* - * We can't run Idle Load Balance on this CPU for this time so we - * cancel it and clear NOHZ_BALANCE_KICK + * Release the rq::nohz_csd. */ - clear_bit(NOHZ_BALANCE_KICK, nohz_flags(cpu)); - return false; -} - -#else /* CONFIG_NO_HZ_COMMON */ + flags = atomic_fetch_andnot(NOHZ_KICK_MASK | NOHZ_NEWILB_KICK, nohz_flags(cpu)); + WARN_ON(!(flags & NOHZ_KICK_MASK)); -static inline bool got_nohz_idle_kick(void) -{ - return false; + rq->idle_balance = idle_cpu(cpu); + if (rq->idle_balance) { + rq->nohz_idle_balance = flags; + __raise_softirq_irqoff(SCHED_SOFTIRQ); + } } #endif /* CONFIG_NO_HZ_COMMON */ #ifdef CONFIG_NO_HZ_FULL -bool sched_can_stop_tick(void) +static inline bool __need_bw_check(struct rq *rq, struct task_struct *p) { - struct rq *rq; - - rq = this_rq(); + if (rq->nr_running != 1) + return false; - /* Make sure rq->nr_running update is visible after the IPI */ - smp_rmb(); + if (p->sched_class != &fair_sched_class) + return false; - /* More than one running task need preemption */ - if (rq->nr_running > 1) - return false; + if (!task_on_rq_queued(p)) + return false; - return true; + return true; } -#endif /* CONFIG_NO_HZ_FULL */ -void sched_avg_update(struct rq *rq) +bool sched_can_stop_tick(struct rq *rq) { - s64 period = sched_avg_period(); + int fifo_nr_running; - while ((s64)(rq_clock(rq) - rq->age_stamp) > period) { - /* - * Inline assembly required to prevent the compiler - * optimising this loop into a divmod call. - * See __iter_div_u64_rem() for another example of this. - */ - asm("" : "+rm" (rq->age_stamp)); - rq->age_stamp += period; - rq->rt_avg /= 2; + /* Deadline tasks, even if single, need the tick */ + if (rq->dl.dl_nr_running) + return false; + + /* + * If there are more than one RR tasks, we need the tick to affect the + * actual RR behaviour. + */ + if (rq->rt.rr_nr_running) { + if (rq->rt.rr_nr_running == 1) + return true; + else + return false; } -} -#else /* !CONFIG_SMP */ -void resched_task(struct task_struct *p) -{ - assert_raw_spin_locked(&task_rq(p)->lock); - set_tsk_need_resched(p); + /* + * If there's no RR tasks, but FIFO tasks, we can skip the tick, no + * forced preemption between FIFO tasks. + */ + fifo_nr_running = rq->rt.rt_nr_running - rq->rt.rr_nr_running; + if (fifo_nr_running) + return true; + + /* + * If there are no DL,RR/FIFO tasks, there must only be CFS or SCX tasks + * left. For CFS, if there's more than one we need the tick for + * involuntary preemption. For SCX, ask. + */ + if (scx_enabled() && !scx_can_stop_tick(rq)) + return false; + + if (rq->cfs.h_nr_queued > 1) + return false; + + /* + * If there is one task and it has CFS runtime bandwidth constraints + * and it's on the cpu now we don't want to stop the tick. + * This check prevents clearing the bit if a newly enqueued task here is + * dequeued by migrating while the constrained task continues to run. + * E.g. going from 2->1 without going through pick_next_task(). + */ + if (__need_bw_check(rq, rq->curr)) { + if (cfs_task_bw_constrained(rq->curr)) + return false; + } + + return true; } -#endif /* CONFIG_SMP */ +#endif /* CONFIG_NO_HZ_FULL */ -#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. @@ -746,306 +1430,795 @@ int tg_nop(struct task_group *tg, void *data) } #endif -static void set_load_weight(struct task_struct *p) +void set_load_weight(struct task_struct *p, bool update_load) { int prio = p->static_prio - MAX_RT_PRIO; - struct load_weight *load = &p->se.load; + struct load_weight lw; + + if (task_has_idle_policy(p)) { + lw.weight = scale_load(WEIGHT_IDLEPRIO); + lw.inv_weight = WMULT_IDLEPRIO; + } else { + lw.weight = scale_load(sched_prio_to_weight[prio]); + lw.inv_weight = sched_prio_to_wmult[prio]; + } + + /* + * SCHED_OTHER tasks have to update their load when changing their + * weight + */ + if (update_load && p->sched_class->reweight_task) + p->sched_class->reweight_task(task_rq(p), p, &lw); + else + p->se.load = lw; +} + +#ifdef CONFIG_UCLAMP_TASK +/* + * Serializes updates of utilization clamp values + * + * The (slow-path) user-space triggers utilization clamp value updates which + * can require updates on (fast-path) scheduler's data structures used to + * support enqueue/dequeue operations. + * While the per-CPU rq lock protects fast-path update operations, user-space + * requests are serialized using a mutex to reduce the risk of conflicting + * updates or API abuses. + */ +static __maybe_unused DEFINE_MUTEX(uclamp_mutex); + +/* Max allowed minimum utilization */ +static unsigned int __maybe_unused sysctl_sched_uclamp_util_min = SCHED_CAPACITY_SCALE; + +/* Max allowed maximum utilization */ +static unsigned int __maybe_unused sysctl_sched_uclamp_util_max = SCHED_CAPACITY_SCALE; + +/* + * By default RT tasks run at the maximum performance point/capacity of the + * system. Uclamp enforces this by always setting UCLAMP_MIN of RT tasks to + * SCHED_CAPACITY_SCALE. + * + * This knob allows admins to change the default behavior when uclamp is being + * used. In battery powered devices, particularly, running at the maximum + * capacity and frequency will increase energy consumption and shorten the + * battery life. + * + * This knob only affects RT tasks that their uclamp_se->user_defined == false. + * + * This knob will not override the system default sched_util_clamp_min defined + * above. + */ +unsigned int sysctl_sched_uclamp_util_min_rt_default = SCHED_CAPACITY_SCALE; + +/* All clamps are required to be less or equal than these values */ +static struct uclamp_se uclamp_default[UCLAMP_CNT]; +/* + * This static key is used to reduce the uclamp overhead in the fast path. It + * primarily disables the call to uclamp_rq_{inc, dec}() in + * enqueue/dequeue_task(). + * + * This allows users to continue to enable uclamp in their kernel config with + * minimum uclamp overhead in the fast path. + * + * As soon as userspace modifies any of the uclamp knobs, the static key is + * enabled, since we have an actual users that make use of uclamp + * functionality. + * + * The knobs that would enable this static key are: + * + * * A task modifying its uclamp value with sched_setattr(). + * * An admin modifying the sysctl_sched_uclamp_{min, max} via procfs. + * * An admin modifying the cgroup cpu.uclamp.{min, max} + */ +DEFINE_STATIC_KEY_FALSE(sched_uclamp_used); + +static inline unsigned int +uclamp_idle_value(struct rq *rq, enum uclamp_id clamp_id, + unsigned int clamp_value) +{ /* - * SCHED_IDLE tasks get minimal weight: + * Avoid blocked utilization pushing up the frequency when we go + * idle (which drops the max-clamp) by retaining the last known + * max-clamp. */ - if (p->policy == SCHED_IDLE) { - load->weight = scale_load(WEIGHT_IDLEPRIO); - load->inv_weight = WMULT_IDLEPRIO; + if (clamp_id == UCLAMP_MAX) { + rq->uclamp_flags |= UCLAMP_FLAG_IDLE; + return clamp_value; + } + + return uclamp_none(UCLAMP_MIN); +} + +static inline void uclamp_idle_reset(struct rq *rq, enum uclamp_id clamp_id, + unsigned int clamp_value) +{ + /* Reset max-clamp retention only on idle exit */ + if (!(rq->uclamp_flags & UCLAMP_FLAG_IDLE)) return; + + uclamp_rq_set(rq, clamp_id, clamp_value); +} + +static inline +unsigned int uclamp_rq_max_value(struct rq *rq, enum uclamp_id clamp_id, + unsigned int clamp_value) +{ + struct uclamp_bucket *bucket = rq->uclamp[clamp_id].bucket; + int bucket_id = UCLAMP_BUCKETS - 1; + + /* + * Since both min and max clamps are max aggregated, find the + * top most bucket with tasks in. + */ + for ( ; bucket_id >= 0; bucket_id--) { + if (!bucket[bucket_id].tasks) + continue; + return bucket[bucket_id].value; } - load->weight = scale_load(prio_to_weight[prio]); - load->inv_weight = prio_to_wmult[prio]; + /* No tasks -- default clamp values */ + return uclamp_idle_value(rq, clamp_id, clamp_value); } -static void enqueue_task(struct rq *rq, struct task_struct *p, int flags) +static void __uclamp_update_util_min_rt_default(struct task_struct *p) { - update_rq_clock(rq); - sched_info_queued(p); - p->sched_class->enqueue_task(rq, p, flags); + unsigned int default_util_min; + struct uclamp_se *uc_se; + + lockdep_assert_held(&p->pi_lock); + + uc_se = &p->uclamp_req[UCLAMP_MIN]; + + /* Only sync if user didn't override the default */ + if (uc_se->user_defined) + return; + + default_util_min = sysctl_sched_uclamp_util_min_rt_default; + uclamp_se_set(uc_se, default_util_min, false); } -static void dequeue_task(struct rq *rq, struct task_struct *p, int flags) +static void uclamp_update_util_min_rt_default(struct task_struct *p) { - update_rq_clock(rq); - sched_info_dequeued(p); - p->sched_class->dequeue_task(rq, p, flags); + if (!rt_task(p)) + return; + + /* Protect updates to p->uclamp_* */ + guard(task_rq_lock)(p); + __uclamp_update_util_min_rt_default(p); } -void activate_task(struct rq *rq, struct task_struct *p, int flags) +static inline struct uclamp_se +uclamp_tg_restrict(struct task_struct *p, enum uclamp_id clamp_id) { - if (task_contributes_to_load(p)) - rq->nr_uninterruptible--; + /* Copy by value as we could modify it */ + struct uclamp_se uc_req = p->uclamp_req[clamp_id]; +#ifdef CONFIG_UCLAMP_TASK_GROUP + unsigned int tg_min, tg_max, value; - enqueue_task(rq, p, flags); + /* + * Tasks in autogroups or root task group will be + * restricted by system defaults. + */ + if (task_group_is_autogroup(task_group(p))) + return uc_req; + if (task_group(p) == &root_task_group) + return uc_req; + + tg_min = task_group(p)->uclamp[UCLAMP_MIN].value; + tg_max = task_group(p)->uclamp[UCLAMP_MAX].value; + value = uc_req.value; + value = clamp(value, tg_min, tg_max); + uclamp_se_set(&uc_req, value, false); +#endif + + return uc_req; } -void deactivate_task(struct rq *rq, struct task_struct *p, int flags) +/* + * The effective clamp bucket index of a task depends on, by increasing + * priority: + * - the task specific clamp value, when explicitly requested from userspace + * - the task group effective clamp value, for tasks not either in the root + * group or in an autogroup + * - the system default clamp value, defined by the sysadmin + */ +static inline struct uclamp_se +uclamp_eff_get(struct task_struct *p, enum uclamp_id clamp_id) { - if (task_contributes_to_load(p)) - rq->nr_uninterruptible++; + struct uclamp_se uc_req = uclamp_tg_restrict(p, clamp_id); + struct uclamp_se uc_max = uclamp_default[clamp_id]; - dequeue_task(rq, p, flags); + /* System default restrictions always apply */ + if (unlikely(uc_req.value > uc_max.value)) + return uc_max; + + return uc_req; } -static void update_rq_clock_task(struct rq *rq, s64 delta) +unsigned long uclamp_eff_value(struct task_struct *p, enum uclamp_id clamp_id) { + struct uclamp_se uc_eff; + + /* Task currently refcounted: use back-annotated (effective) value */ + if (p->uclamp[clamp_id].active) + return (unsigned long)p->uclamp[clamp_id].value; + + uc_eff = uclamp_eff_get(p, clamp_id); + + return (unsigned long)uc_eff.value; +} + /* - * In theory, the compile should just see 0 here, and optimize out the call - * to sched_rt_avg_update. But I don't trust it... + * When a task is enqueued on a rq, the clamp bucket currently defined by the + * task's uclamp::bucket_id is refcounted on that rq. This also immediately + * updates the rq's clamp value if required. + * + * Tasks can have a task-specific value requested from user-space, track + * within each bucket the maximum value for tasks refcounted in it. + * This "local max aggregation" allows to track the exact "requested" value + * for each bucket when all its RUNNABLE tasks require the same clamp. */ -#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) - s64 steal = 0, irq_delta = 0; -#endif -#ifdef CONFIG_IRQ_TIME_ACCOUNTING - irq_delta = irq_time_read(cpu_of(rq)) - rq->prev_irq_time; +static inline void uclamp_rq_inc_id(struct rq *rq, struct task_struct *p, + enum uclamp_id clamp_id) +{ + struct uclamp_rq *uc_rq = &rq->uclamp[clamp_id]; + struct uclamp_se *uc_se = &p->uclamp[clamp_id]; + struct uclamp_bucket *bucket; + + lockdep_assert_rq_held(rq); + + /* Update task effective clamp */ + p->uclamp[clamp_id] = uclamp_eff_get(p, clamp_id); + + bucket = &uc_rq->bucket[uc_se->bucket_id]; + bucket->tasks++; + uc_se->active = true; + + uclamp_idle_reset(rq, clamp_id, uc_se->value); /* - * Since irq_time is only updated on {soft,}irq_exit, we might run into - * this case when a previous update_rq_clock() happened inside a - * {soft,}irq region. - * - * When this happens, we stop ->clock_task and only update the - * prev_irq_time stamp to account for the part that fit, so that a next - * update will consume the rest. This ensures ->clock_task is - * monotonic. - * - * It does however cause some slight miss-attribution of {soft,}irq - * time, a more accurate solution would be to update the irq_time using - * the current rq->clock timestamp, except that would require using - * atomic ops. + * Local max aggregation: rq buckets always track the max + * "requested" clamp value of its RUNNABLE tasks. */ - if (irq_delta > delta) - irq_delta = delta; + if (bucket->tasks == 1 || uc_se->value > bucket->value) + bucket->value = uc_se->value; - rq->prev_irq_time += irq_delta; - delta -= irq_delta; -#endif -#ifdef CONFIG_PARAVIRT_TIME_ACCOUNTING - if (static_key_false((¶virt_steal_rq_enabled))) { - u64 st; + if (uc_se->value > uclamp_rq_get(rq, clamp_id)) + uclamp_rq_set(rq, clamp_id, uc_se->value); +} - steal = paravirt_steal_clock(cpu_of(rq)); - steal -= rq->prev_steal_time_rq; +/* + * When a task is dequeued from a rq, the clamp bucket refcounted by the task + * is released. If this is the last task reference counting the rq's max + * active clamp value, then the rq's clamp value is updated. + * + * Both refcounted tasks and rq's cached clamp values are expected to be + * always valid. If it's detected they are not, as defensive programming, + * enforce the expected state and warn. + */ +static inline void uclamp_rq_dec_id(struct rq *rq, struct task_struct *p, + enum uclamp_id clamp_id) +{ + struct uclamp_rq *uc_rq = &rq->uclamp[clamp_id]; + struct uclamp_se *uc_se = &p->uclamp[clamp_id]; + struct uclamp_bucket *bucket; + unsigned int bkt_clamp; + unsigned int rq_clamp; - if (unlikely(steal > delta)) - steal = delta; + lockdep_assert_rq_held(rq); - st = steal_ticks(steal); - steal = st * TICK_NSEC; + /* + * If sched_uclamp_used was enabled after task @p was enqueued, + * we could end up with unbalanced call to uclamp_rq_dec_id(). + * + * In this case the uc_se->active flag should be false since no uclamp + * accounting was performed at enqueue time and we can just return + * here. + * + * Need to be careful of the following enqueue/dequeue ordering + * problem too + * + * enqueue(taskA) + * // sched_uclamp_used gets enabled + * enqueue(taskB) + * dequeue(taskA) + * // Must not decrement bucket->tasks here + * dequeue(taskB) + * + * where we could end up with stale data in uc_se and + * bucket[uc_se->bucket_id]. + * + * The following check here eliminates the possibility of such race. + */ + if (unlikely(!uc_se->active)) + return; - rq->prev_steal_time_rq += steal; + bucket = &uc_rq->bucket[uc_se->bucket_id]; - delta -= steal; - } -#endif + WARN_ON_ONCE(!bucket->tasks); + if (likely(bucket->tasks)) + bucket->tasks--; - rq->clock_task += delta; + uc_se->active = false; -#if defined(CONFIG_IRQ_TIME_ACCOUNTING) || defined(CONFIG_PARAVIRT_TIME_ACCOUNTING) - if ((irq_delta + steal) && sched_feat(NONTASK_POWER)) - sched_rt_avg_update(rq, irq_delta + steal); -#endif + /* + * Keep "local max aggregation" simple and accept to (possibly) + * overboost some RUNNABLE tasks in the same bucket. + * The rq clamp bucket value is reset to its base value whenever + * there are no more RUNNABLE tasks refcounting it. + */ + if (likely(bucket->tasks)) + return; + + rq_clamp = uclamp_rq_get(rq, clamp_id); + /* + * Defensive programming: this should never happen. If it happens, + * e.g. due to future modification, warn and fix up the expected value. + */ + 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); + } } -void sched_set_stop_task(int cpu, struct task_struct *stop) +static inline void uclamp_rq_inc(struct rq *rq, struct task_struct *p, int flags) { - struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; - struct task_struct *old_stop = cpu_rq(cpu)->stop; + enum uclamp_id clamp_id; - if (stop) { - /* - * Make it appear like a SCHED_FIFO task, its something - * userspace knows about and won't get confused about. - * - * Also, it will make PI more or less work without too - * much confusion -- but then, stop work should not - * rely on PI working anyway. - */ - sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); + /* + * Avoid any overhead until uclamp is actually used by the userspace. + * + * The condition is constructed such that a NOP is generated when + * sched_uclamp_used is disabled. + */ + if (!uclamp_is_used()) + return; - stop->sched_class = &stop_sched_class; - } + if (unlikely(!p->sched_class->uclamp_enabled)) + return; - cpu_rq(cpu)->stop = stop; + /* Only inc the delayed task which being woken up. */ + if (p->se.sched_delayed && !(flags & ENQUEUE_DELAYED)) + return; - if (old_stop) { - /* - * Reset it back to a normal scheduling class so that - * it can die in pieces. - */ - old_stop->sched_class = &rt_sched_class; - } + for_each_clamp_id(clamp_id) + uclamp_rq_inc_id(rq, p, clamp_id); + + /* Reset clamp idle holding when there is one RUNNABLE task */ + if (rq->uclamp_flags & UCLAMP_FLAG_IDLE) + rq->uclamp_flags &= ~UCLAMP_FLAG_IDLE; } -/* - * __normal_prio - return the priority that is based on the static prio - */ -static inline int __normal_prio(struct task_struct *p) +static inline void uclamp_rq_dec(struct rq *rq, struct task_struct *p) { - return p->static_prio; + enum uclamp_id clamp_id; + + /* + * Avoid any overhead until uclamp is actually used by the userspace. + * + * The condition is constructed such that a NOP is generated when + * sched_uclamp_used is disabled. + */ + if (!uclamp_is_used()) + return; + + if (unlikely(!p->sched_class->uclamp_enabled)) + return; + + if (p->se.sched_delayed) + return; + + for_each_clamp_id(clamp_id) + uclamp_rq_dec_id(rq, p, clamp_id); } -/* - * Calculate the expected normal priority: i.e. priority - * without taking RT-inheritance into account. Might be - * boosted by interactivity modifiers. Changes upon fork, - * setprio syscalls, and whenever the interactivity - * estimator recalculates. - */ -static inline int normal_prio(struct task_struct *p) +static inline void uclamp_rq_reinc_id(struct rq *rq, struct task_struct *p, + enum uclamp_id clamp_id) { - int prio; + if (!p->uclamp[clamp_id].active) + return; - if (task_has_rt_policy(p)) - prio = MAX_RT_PRIO-1 - p->rt_priority; - else - prio = __normal_prio(p); - return prio; + uclamp_rq_dec_id(rq, p, clamp_id); + uclamp_rq_inc_id(rq, p, clamp_id); + + /* + * Make sure to clear the idle flag if we've transiently reached 0 + * active tasks on rq. + */ + if (clamp_id == UCLAMP_MAX && (rq->uclamp_flags & UCLAMP_FLAG_IDLE)) + rq->uclamp_flags &= ~UCLAMP_FLAG_IDLE; } -/* - * Calculate the current priority, i.e. the priority - * taken into account by the scheduler. This value might - * be boosted by RT tasks, or might be boosted by - * interactivity modifiers. Will be RT if the task got - * RT-boosted. If not then it returns p->normal_prio. - */ -static int effective_prio(struct task_struct *p) +static inline void +uclamp_update_active(struct task_struct *p) { - p->normal_prio = normal_prio(p); + enum uclamp_id clamp_id; + struct rq_flags rf; + struct rq *rq; + /* - * If we are RT tasks or we were boosted to RT priority, - * keep the priority unchanged. Otherwise, update priority - * to the normal priority: + * Lock the task and the rq where the task is (or was) queued. + * + * We might lock the (previous) rq of a !RUNNABLE task, but that's the + * price to pay to safely serialize util_{min,max} updates with + * enqueues, dequeues and migration operations. + * This is the same locking schema used by __set_cpus_allowed_ptr(). */ - if (!rt_prio(p->prio)) - return p->normal_prio; - return p->prio; + rq = task_rq_lock(p, &rf); + + /* + * Setting the clamp bucket is serialized by task_rq_lock(). + * If the task is not yet RUNNABLE and its task_struct is not + * affecting a valid clamp bucket, the next time it's enqueued, + * it will already see the updated clamp bucket value. + */ + for_each_clamp_id(clamp_id) + uclamp_rq_reinc_id(rq, p, clamp_id); + + task_rq_unlock(rq, p, &rf); } -/** - * task_curr - is this task currently executing on a CPU? - * @p: the task in question. - */ -inline int task_curr(const struct task_struct *p) +#ifdef CONFIG_UCLAMP_TASK_GROUP +static inline void +uclamp_update_active_tasks(struct cgroup_subsys_state *css) { - return cpu_curr(task_cpu(p)) == p; + struct css_task_iter it; + struct task_struct *p; + + css_task_iter_start(css, 0, &it); + while ((p = css_task_iter_next(&it))) + uclamp_update_active(p); + css_task_iter_end(&it); } -static inline void check_class_changed(struct rq *rq, struct task_struct *p, - const struct sched_class *prev_class, - int oldprio) +static void cpu_util_update_eff(struct cgroup_subsys_state *css); +#endif + +#ifdef CONFIG_SYSCTL +#ifdef CONFIG_UCLAMP_TASK_GROUP +static void uclamp_update_root_tg(void) { - 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) - p->sched_class->prio_changed(rq, p, oldprio); + struct task_group *tg = &root_task_group; + + uclamp_se_set(&tg->uclamp_req[UCLAMP_MIN], + sysctl_sched_uclamp_util_min, false); + uclamp_se_set(&tg->uclamp_req[UCLAMP_MAX], + sysctl_sched_uclamp_util_max, false); + + guard(rcu)(); + cpu_util_update_eff(&root_task_group.css); } +#else +static void uclamp_update_root_tg(void) { } +#endif -void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) +static void uclamp_sync_util_min_rt_default(void) { - const struct sched_class *class; + struct task_struct *g, *p; - if (p->sched_class == rq->curr->sched_class) { - rq->curr->sched_class->check_preempt_curr(rq, p, flags); - } else { - for_each_class(class) { - if (class == rq->curr->sched_class) - break; - if (class == p->sched_class) { - resched_task(rq->curr); - break; - } - } + /* + * copy_process() sysctl_uclamp + * uclamp_min_rt = X; + * write_lock(&tasklist_lock) read_lock(&tasklist_lock) + * // link thread smp_mb__after_spinlock() + * write_unlock(&tasklist_lock) read_unlock(&tasklist_lock); + * sched_post_fork() for_each_process_thread() + * __uclamp_sync_rt() __uclamp_sync_rt() + * + * Ensures that either sched_post_fork() will observe the new + * uclamp_min_rt or for_each_process_thread() will observe the new + * task. + */ + read_lock(&tasklist_lock); + smp_mb__after_spinlock(); + read_unlock(&tasklist_lock); + + guard(rcu)(); + for_each_process_thread(g, p) + uclamp_update_util_min_rt_default(p); +} + +static int sysctl_sched_uclamp_handler(const struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + bool update_root_tg = false; + int old_min, old_max, old_min_rt; + int result; + + guard(mutex)(&uclamp_mutex); + + old_min = sysctl_sched_uclamp_util_min; + old_max = sysctl_sched_uclamp_util_max; + old_min_rt = sysctl_sched_uclamp_util_min_rt_default; + + result = proc_dointvec(table, write, buffer, lenp, ppos); + if (result) + goto undo; + if (!write) + return 0; + + if (sysctl_sched_uclamp_util_min > sysctl_sched_uclamp_util_max || + sysctl_sched_uclamp_util_max > SCHED_CAPACITY_SCALE || + sysctl_sched_uclamp_util_min_rt_default > SCHED_CAPACITY_SCALE) { + + result = -EINVAL; + goto undo; + } + + if (old_min != sysctl_sched_uclamp_util_min) { + uclamp_se_set(&uclamp_default[UCLAMP_MIN], + sysctl_sched_uclamp_util_min, false); + update_root_tg = true; + } + if (old_max != sysctl_sched_uclamp_util_max) { + uclamp_se_set(&uclamp_default[UCLAMP_MAX], + sysctl_sched_uclamp_util_max, false); + update_root_tg = true; + } + + if (update_root_tg) { + sched_uclamp_enable(); + uclamp_update_root_tg(); + } + + if (old_min_rt != sysctl_sched_uclamp_util_min_rt_default) { + sched_uclamp_enable(); + uclamp_sync_util_min_rt_default(); } /* - * A queue event has occurred, and we're going to schedule. In - * this case, we can save a useless back to back clock update. + * We update all RUNNABLE tasks only when task groups are in use. + * Otherwise, keep it simple and do just a lazy update at each next + * task enqueue time. + */ + return 0; + +undo: + sysctl_sched_uclamp_util_min = old_min; + sysctl_sched_uclamp_util_max = old_max; + sysctl_sched_uclamp_util_min_rt_default = old_min_rt; + return result; +} +#endif /* CONFIG_SYSCTL */ + +static void uclamp_fork(struct task_struct *p) +{ + enum uclamp_id clamp_id; + + /* + * We don't need to hold task_rq_lock() when updating p->uclamp_* here + * as the task is still at its early fork stages. */ - if (rq->curr->on_rq && test_tsk_need_resched(rq->curr)) - rq->skip_clock_update = 1; + for_each_clamp_id(clamp_id) + p->uclamp[clamp_id].active = false; + + if (likely(!p->sched_reset_on_fork)) + return; + + for_each_clamp_id(clamp_id) { + uclamp_se_set(&p->uclamp_req[clamp_id], + uclamp_none(clamp_id), false); + } } -static ATOMIC_NOTIFIER_HEAD(task_migration_notifier); +static void uclamp_post_fork(struct task_struct *p) +{ + uclamp_update_util_min_rt_default(p); +} -void register_task_migration_notifier(struct notifier_block *n) +static void __init init_uclamp_rq(struct rq *rq) { - atomic_notifier_chain_register(&task_migration_notifier, n); + enum uclamp_id clamp_id; + struct uclamp_rq *uc_rq = rq->uclamp; + + for_each_clamp_id(clamp_id) { + uc_rq[clamp_id] = (struct uclamp_rq) { + .value = uclamp_none(clamp_id) + }; + } + + rq->uclamp_flags = UCLAMP_FLAG_IDLE; } -#ifdef CONFIG_SMP -void set_task_cpu(struct task_struct *p, unsigned int new_cpu) +static void __init init_uclamp(void) +{ + struct uclamp_se uc_max = {}; + enum uclamp_id clamp_id; + int cpu; + + for_each_possible_cpu(cpu) + init_uclamp_rq(cpu_rq(cpu)); + + for_each_clamp_id(clamp_id) { + uclamp_se_set(&init_task.uclamp_req[clamp_id], + uclamp_none(clamp_id), false); + } + + /* System defaults allow max clamp values for both indexes */ + uclamp_se_set(&uc_max, uclamp_none(UCLAMP_MAX), false); + for_each_clamp_id(clamp_id) { + uclamp_default[clamp_id] = uc_max; +#ifdef CONFIG_UCLAMP_TASK_GROUP + root_task_group.uclamp_req[clamp_id] = uc_max; + root_task_group.uclamp[clamp_id] = uc_max; +#endif + } +} + +#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 */ + +bool sched_task_on_rq(struct task_struct *p) +{ + return task_on_rq_queued(p); +} + +unsigned long get_wchan(struct task_struct *p) +{ + unsigned long ip = 0; + unsigned int state; + + if (!p || p == current) + return 0; + + /* Only get wchan if task is blocked and we can keep it that way. */ + raw_spin_lock_irq(&p->pi_lock); + state = READ_ONCE(p->__state); + smp_rmb(); /* see try_to_wake_up() */ + if (state != TASK_RUNNING && state != TASK_WAKING && !p->on_rq) + ip = __get_wchan(p); + raw_spin_unlock_irq(&p->pi_lock); + + return ip; +} + +void enqueue_task(struct rq *rq, struct task_struct *p, int flags) +{ + if (!(flags & ENQUEUE_NOCLOCK)) + update_rq_clock(rq); + + /* + * 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, flags); + + rq->queue_mask |= p->sched_class->queue_mask; + p->sched_class->enqueue_task(rq, p, flags); + + psi_enqueue(p, flags); + + if (!(flags & ENQUEUE_RESTORE)) + sched_info_enqueue(rq, p); + + if (sched_core_enabled(rq)) + sched_core_enqueue(rq, p); +} + +/* + * Must only return false when DEQUEUE_SLEEP. + */ +inline bool dequeue_task(struct rq *rq, struct task_struct *p, int flags) { -#ifdef CONFIG_SCHED_DEBUG + if (sched_core_enabled(rq)) + sched_core_dequeue(rq, p, flags); + + if (!(flags & DEQUEUE_NOCLOCK)) + update_rq_clock(rq); + + if (!(flags & DEQUEUE_SAVE)) + sched_info_dequeue(rq, p); + + psi_dequeue(p, flags); + /* - * We should never call set_task_cpu() on a blocked task, - * ttwu() will sort out the placement. + * Must be before ->dequeue_task() because ->dequeue_task() can 'fail' + * and mark the task ->sched_delayed. */ - WARN_ON_ONCE(p->state != TASK_RUNNING && p->state != TASK_WAKING && - !(task_thread_info(p)->preempt_count & PREEMPT_ACTIVE)); + uclamp_rq_dec(rq, p); + rq->queue_mask |= p->sched_class->queue_mask; + return p->sched_class->dequeue_task(rq, p, flags); +} + +void activate_task(struct rq *rq, struct task_struct *p, int flags) +{ + if (task_on_rq_migrating(p)) + flags |= ENQUEUE_MIGRATED; + + enqueue_task(rq, p, flags); + + WRITE_ONCE(p->on_rq, TASK_ON_RQ_QUEUED); + ASSERT_EXCLUSIVE_WRITER(p->on_rq); +} + +void deactivate_task(struct rq *rq, struct task_struct *p, int flags) +{ + WARN_ON_ONCE(flags & DEQUEUE_SLEEP); + + WRITE_ONCE(p->on_rq, TASK_ON_RQ_MIGRATING); + ASSERT_EXCLUSIVE_WRITER(p->on_rq); -#ifdef CONFIG_LOCKDEP /* - * The caller should hold either p->pi_lock or rq->lock, when changing - * a task's CPU. ->pi_lock for waking tasks, rq->lock for runnable tasks. - * - * sched_move_task() holds both and thus holding either pins the cgroup, - * see task_group(). - * - * Furthermore, all task_rq users should acquire both locks, see - * task_rq_lock(). + * Code explicitly relies on TASK_ON_RQ_MIGRATING begin set *before* + * dequeue_task() and cleared *after* enqueue_task(). */ - WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) || - lockdep_is_held(&task_rq(p)->lock))); -#endif -#endif - trace_sched_migrate_task(p, new_cpu); + dequeue_task(rq, p, flags); +} - if (task_cpu(p) != new_cpu) { - struct task_migration_notifier tmn; +static void block_task(struct rq *rq, struct task_struct *p, int flags) +{ + if (dequeue_task(rq, p, DEQUEUE_SLEEP | flags)) + __block_task(rq, p); +} - if (p->sched_class->migrate_task_rq) - p->sched_class->migrate_task_rq(p, new_cpu); - p->se.nr_migrations++; - perf_sw_event(PERF_COUNT_SW_CPU_MIGRATIONS, 1, NULL, 0); +/** + * task_curr - is this task currently executing on a CPU? + * @p: the task in question. + * + * Return: 1 if the task is currently executing. 0 otherwise. + */ +inline int task_curr(const struct task_struct *p) +{ + return cpu_curr(task_cpu(p)) == p; +} - tmn.task = p; - tmn.from_cpu = task_cpu(p); - tmn.to_cpu = new_cpu; +void wakeup_preempt(struct rq *rq, struct task_struct *p, int flags) +{ + struct task_struct *donor = rq->donor; - atomic_notifier_call_chain(&task_migration_notifier, 0, &tmn); - } + 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)) + resched_curr(rq); - __set_task_cpu(p, new_cpu); + /* + * A queue event has occurred, and we're going to schedule. In + * this case, we can save a useless back to back clock update. + */ + if (task_on_rq_queued(donor) && test_tsk_need_resched(rq->curr)) + rq_clock_skip_update(rq); } -struct migration_arg { - struct task_struct *task; - int dest_cpu; -}; +static __always_inline +int __task_state_match(struct task_struct *p, unsigned int state) +{ + if (READ_ONCE(p->__state) & state) + return 1; -static int migration_cpu_stop(void *data); + if (READ_ONCE(p->saved_state) & state) + return -1; + + return 0; +} + +static __always_inline +int task_state_match(struct task_struct *p, unsigned int state) +{ + /* + * Serialize against current_save_and_set_rtlock_wait_state(), + * current_restore_rtlock_saved_state(), and __refrigerator(). + */ + guard(raw_spinlock_irq)(&p->pi_lock); + return __task_state_match(p, state); +} /* * wait_task_inactive - wait for a thread to unschedule. * - * If @match_state is nonzero, it's the @p->state value just checked and - * not expected to change. If it changes, i.e. @p might have woken up, - * then return zero. When we succeed in waiting for @p to be off its CPU, - * we return a positive number (its total switch count). If a second call - * a short while later returns the same number, the caller can be sure that - * @p has remained unscheduled the whole time. + * Wait for the thread to block in any of the states set in @match_state. + * If it changes, i.e. @p might have woken up, then return zero. When we + * succeed in waiting for @p to be off its CPU, we return a positive number + * (its total switch count). If a second call a short while later returns the + * same number, the caller can be sure that @p has remained unscheduled the + * whole time. * * The caller must ensure that the task *will* unschedule sometime soon, * else this function might spin for a *long* time. This function can't @@ -1053,10 +2226,10 @@ static int migration_cpu_stop(void *data); * smp_call_function() if an IPI is sent by the same process we are * waiting to become inactive. */ -unsigned long wait_task_inactive(struct task_struct *p, long match_state) +unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state) { - unsigned long flags; - int running, on_rq; + int running, queued, match; + struct rq_flags rf; unsigned long ncsw; struct rq *rq; @@ -1076,12 +2249,12 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) * * NOTE! Since we don't hold any locks, it's not * even sure that "rq" stays as the right runqueue! - * But we don't care, since "task_running()" will + * But we don't care, since "task_on_cpu()" will * return false if the runqueue has changed and p * is actually now running somewhere else! */ - while (task_running(rq, p)) { - if (match_state && unlikely(p->state != match_state)) + while (task_on_cpu(rq, p)) { + if (!task_state_match(p, match_state)) return 0; cpu_relax(); } @@ -1091,14 +2264,27 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) * lock now, to be *sure*. If we're wrong, we'll * just go back and repeat. */ - rq = task_rq_lock(p, &flags); + 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_running(rq, p); - on_rq = p->on_rq; + running = task_on_cpu(rq, p); + queued = task_on_rq_queued(p); ncsw = 0; - if (!match_state || p->state == match_state) + if ((match = __task_state_match(p, match_state))) { + /* + * When matching on p->saved_state, consider this task + * still queued so it will wait. + */ + if (match < 0) + queued = 1; ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ - task_rq_unlock(rq, p, &flags); + } + task_rq_unlock(rq, p, &rf); /* * If it changed from the expected state, bail out now. @@ -1126,11 +2312,11 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) * running right now), it's preempted, and we should * yield - it could be a while. */ - if (unlikely(on_rq)) { - ktime_t to = ktime_set(0, NSEC_PER_SEC/HZ); + if (unlikely(queued)) { + ktime_t to = NSEC_PER_SEC / HZ; set_current_state(TASK_UNINTERRUPTIBLE); - schedule_hrtimeout(&to, HRTIMER_MODE_REL); + schedule_hrtimeout(&to, HRTIMER_MODE_REL_HARD); continue; } @@ -1145,6 +2331,1050 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) return ncsw; } +static void +do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx); + +static void migrate_disable_switch(struct rq *rq, struct task_struct *p) +{ + struct affinity_context ac = { + .new_mask = cpumask_of(rq->cpu), + .flags = SCA_MIGRATE_DISABLE, + }; + + if (likely(!p->migration_disabled)) + return; + + if (p->cpus_ptr != &p->cpus_mask) + return; + + scoped_guard (task_rq_lock, p) + do_set_cpus_allowed(p, &ac); +} + +void ___migrate_enable(void) +{ + struct task_struct *p = current; + struct affinity_context ac = { + .new_mask = &p->cpus_mask, + .flags = SCA_MIGRATE_ENABLE, + }; + + __set_cpus_allowed_ptr(p, &ac); +} +EXPORT_SYMBOL_GPL(___migrate_enable); + +void migrate_disable(void) +{ + __migrate_disable(); +} +EXPORT_SYMBOL_GPL(migrate_disable); + +void migrate_enable(void) +{ + __migrate_enable(); +} +EXPORT_SYMBOL_GPL(migrate_enable); + +static inline bool rq_has_pinned_tasks(struct rq *rq) +{ + return rq->nr_pinned; +} + +/* + * Per-CPU kthreads are allowed to run on !active && online CPUs, see + * __set_cpus_allowed_ptr() and select_fallback_rq(). + */ +static inline bool is_cpu_allowed(struct task_struct *p, int cpu) +{ + /* When not in the task's cpumask, no point in looking further. */ + if (!task_allowed_on_cpu(p, cpu)) + return false; + + /* migrate_disabled() must be allowed to finish. */ + if (is_migration_disabled(p)) + return cpu_online(cpu); + + /* Non kernel threads are not allowed during either online or offline. */ + if (!(p->flags & PF_KTHREAD)) + return cpu_active(cpu); + + /* KTHREAD_IS_PER_CPU is always allowed. */ + if (kthread_is_per_cpu(p)) + return cpu_online(cpu); + + /* Regular kernel threads don't get to stay during offline. */ + if (cpu_dying(cpu)) + return false; + + /* But are allowed during online. */ + return cpu_online(cpu); +} + +/* + * This is how migration works: + * + * 1) we invoke migration_cpu_stop() on the target CPU using + * stop_one_cpu(). + * 2) stopper starts to run (implicitly forcing the migrated thread + * off the CPU) + * 3) it checks whether the migrated task is still in the wrong runqueue. + * 4) if it's in the wrong runqueue then the migration thread removes + * it and puts it into the right queue. + * 5) stopper completes and stop_one_cpu() returns and the migration + * is done. + */ + +/* + * move_queued_task - move a queued task to new rq. + * + * Returns (locked) new rq. Old rq's lock is released. + */ +static struct rq *move_queued_task(struct rq *rq, struct rq_flags *rf, + struct task_struct *p, int new_cpu) +{ + lockdep_assert_rq_held(rq); + + deactivate_task(rq, p, DEQUEUE_NOCLOCK); + set_task_cpu(p, new_cpu); + rq_unlock(rq, rf); + + rq = cpu_rq(new_cpu); + + rq_lock(rq, rf); + WARN_ON_ONCE(task_cpu(p) != new_cpu); + activate_task(rq, p, 0); + wakeup_preempt(rq, p, 0); + + return rq; +} + +struct migration_arg { + struct task_struct *task; + int dest_cpu; + struct set_affinity_pending *pending; +}; + +/* + * @refs: number of wait_for_completion() + * @stop_pending: is @stop_work in use + */ +struct set_affinity_pending { + refcount_t refs; + unsigned int stop_pending; + struct completion done; + struct cpu_stop_work stop_work; + struct migration_arg arg; +}; + +/* + * Move (not current) task off this CPU, onto the destination CPU. We're doing + * this because either it can't run here any more (set_cpus_allowed() + * away from this CPU, or CPU going down), or because we're + * attempting to rebalance this task on exec (sched_exec). + * + * So we race with normal scheduler movements, but that's OK, as long + * as the task is no longer on this CPU. + */ +static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf, + struct task_struct *p, int dest_cpu) +{ + /* Affinity changed (again). */ + if (!is_cpu_allowed(p, dest_cpu)) + return rq; + + rq = move_queued_task(rq, rf, p, dest_cpu); + + return rq; +} + +/* + * migration_cpu_stop - this will be executed by a high-prio stopper thread + * and performs thread migration by bumping thread off CPU then + * 'pushing' onto another runqueue. + */ +static int migration_cpu_stop(void *data) +{ + struct migration_arg *arg = data; + struct set_affinity_pending *pending = arg->pending; + struct task_struct *p = arg->task; + struct rq *rq = this_rq(); + bool complete = false; + struct rq_flags rf; + + /* + * The original target CPU might have gone down and we might + * be on another CPU but it doesn't matter. + */ + local_irq_save(rf.flags); + /* + * We need to explicitly wake pending tasks before running + * __migrate_task() such that we will not miss enforcing cpus_ptr + * during wakeups, see set_cpus_allowed_ptr()'s TASK_WAKING test. + */ + flush_smp_call_function_queue(); + + raw_spin_lock(&p->pi_lock); + rq_lock(rq, &rf); + + /* + * If we were passed a pending, then ->stop_pending was set, thus + * p->migration_pending must have remained stable. + */ + WARN_ON_ONCE(pending && pending != p->migration_pending); + + /* + * If task_rq(p) != rq, it cannot be migrated here, because we're + * holding rq->lock, if p->on_rq == 0 it cannot get enqueued because + * we're holding p->pi_lock. + */ + if (task_rq(p) == rq) { + if (is_migration_disabled(p)) + goto out; + + if (pending) { + p->migration_pending = NULL; + complete = true; + + if (cpumask_test_cpu(task_cpu(p), &p->cpus_mask)) + goto out; + } + + if (task_on_rq_queued(p)) { + update_rq_clock(rq); + rq = __migrate_task(rq, &rf, p, arg->dest_cpu); + } else { + p->wake_cpu = arg->dest_cpu; + } + + /* + * XXX __migrate_task() can fail, at which point we might end + * up running on a dodgy CPU, AFAICT this can only happen + * during CPU hotplug, at which point we'll get pushed out + * anyway, so it's probably not a big deal. + */ + + } else if (pending) { + /* + * This happens when we get migrated between migrate_enable()'s + * preempt_enable() and scheduling the stopper task. At that + * point we're a regular task again and not current anymore. + * + * A !PREEMPT kernel has a giant hole here, which makes it far + * more likely. + */ + + /* + * The task moved before the stopper got to run. We're holding + * ->pi_lock, so the allowed mask is stable - if it got + * somewhere allowed, we're done. + */ + if (cpumask_test_cpu(task_cpu(p), p->cpus_ptr)) { + p->migration_pending = NULL; + complete = true; + goto out; + } + + /* + * When migrate_enable() hits a rq mis-match we can't reliably + * determine is_migration_disabled() and so have to chase after + * it. + */ + WARN_ON_ONCE(!pending->stop_pending); + preempt_disable(); + 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(); + return 0; + } +out: + if (pending) + pending->stop_pending = false; + rq_unlock(rq, &rf); + raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); + + if (complete) + complete_all(&pending->done); + + return 0; +} + +int push_cpu_stop(void *arg) +{ + struct rq *lowest_rq = NULL, *rq = this_rq(); + struct task_struct *p = arg; + + raw_spin_lock_irq(&p->pi_lock); + raw_spin_rq_lock(rq); + + if (task_rq(p) != rq) + goto out_unlock; + + if (is_migration_disabled(p)) { + p->migration_flags |= MDF_PUSH; + goto out_unlock; + } + + p->migration_flags &= ~MDF_PUSH; + + if (p->sched_class->find_lock_rq) + lowest_rq = p->sched_class->find_lock_rq(p, rq); + + if (!lowest_rq) + goto out_unlock; + + // XXX validate p is still the highest prio task + if (task_rq(p) == rq) { + move_queued_task_locked(rq, lowest_rq, p); + resched_curr(lowest_rq); + } + + double_unlock_balance(rq, lowest_rq); + +out_unlock: + rq->push_busy = false; + raw_spin_rq_unlock(rq); + raw_spin_unlock_irq(&p->pi_lock); + + put_task_struct(p); + 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. + */ +void set_cpus_allowed_common(struct task_struct *p, struct affinity_context *ctx) +{ + if (ctx->flags & (SCA_MIGRATE_ENABLE | SCA_MIGRATE_DISABLE)) { + p->cpus_ptr = ctx->new_mask; + return; + } + + 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 + */ + if (ctx->flags & SCA_USER) + swap(p->user_cpus_ptr, ctx->user_mask); +} + +static void +do_set_cpus_allowed(struct task_struct *p, struct affinity_context *ctx) +{ + 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 set_cpus_allowed_force(struct task_struct *p, const struct cpumask *new_mask) +{ + struct affinity_context ac = { + .new_mask = new_mask, + .user_mask = NULL, + .flags = SCA_USER, /* clear the user requested mask */ + }; + union cpumask_rcuhead { + cpumask_t cpumask; + struct rcu_head rcu; + }; + + 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 + * to use kfree() here (when PREEMPT_RT=y), therefore punt to using + * kfree_rcu(). + */ + kfree_rcu((union cpumask_rcuhead *)ac.user_mask, rcu); +} + +int dup_user_cpus_ptr(struct task_struct *dst, struct task_struct *src, + int node) +{ + cpumask_t *user_mask; + unsigned long flags; + + /* + * Always clear dst->user_cpus_ptr first as their user_cpus_ptr's + * may differ by now due to racing. + */ + dst->user_cpus_ptr = NULL; + + /* + * This check is racy and losing the race is a valid situation. + * It is not worth the extra overhead of taking the pi_lock on + * every fork/clone. + */ + if (data_race(!src->user_cpus_ptr)) + return 0; + + user_mask = alloc_user_cpus_ptr(node); + if (!user_mask) + return -ENOMEM; + + /* + * Use pi_lock to protect content of user_cpus_ptr + * + * Though unlikely, user_cpus_ptr can be reset to NULL by a concurrent + * set_cpus_allowed_force(). + */ + raw_spin_lock_irqsave(&src->pi_lock, flags); + if (src->user_cpus_ptr) { + swap(dst->user_cpus_ptr, user_mask); + cpumask_copy(dst->user_cpus_ptr, src->user_cpus_ptr); + } + raw_spin_unlock_irqrestore(&src->pi_lock, flags); + + if (unlikely(user_mask)) + kfree(user_mask); + + return 0; +} + +static inline struct cpumask *clear_user_cpus_ptr(struct task_struct *p) +{ + struct cpumask *user_mask = NULL; + + swap(p->user_cpus_ptr, user_mask); + + return user_mask; +} + +void release_user_cpus_ptr(struct task_struct *p) +{ + kfree(clear_user_cpus_ptr(p)); +} + +/* + * This function is wildly self concurrent; here be dragons. + * + * + * When given a valid mask, __set_cpus_allowed_ptr() must block until the + * designated task is enqueued on an allowed CPU. If that task is currently + * running, we have to kick it out using the CPU stopper. + * + * Migrate-Disable comes along and tramples all over our nice sandcastle. + * Consider: + * + * Initial conditions: P0->cpus_mask = [0, 1] + * + * P0@CPU0 P1 + * + * migrate_disable(); + * <preempted> + * set_cpus_allowed_ptr(P0, [1]); + * + * P1 *cannot* return from this set_cpus_allowed_ptr() call until P0 executes + * its outermost migrate_enable() (i.e. it exits its Migrate-Disable region). + * This means we need the following scheme: + * + * P0@CPU0 P1 + * + * migrate_disable(); + * <preempted> + * set_cpus_allowed_ptr(P0, [1]); + * <blocks> + * <resumes> + * migrate_enable(); + * __set_cpus_allowed_ptr(); + * <wakes local stopper> + * `--> <woken on migration completion> + * + * Now the fun stuff: there may be several P1-like tasks, i.e. multiple + * concurrent set_cpus_allowed_ptr(P0, [*]) calls. CPU affinity changes of any + * task p are serialized by p->pi_lock, which we can leverage: the one that + * should come into effect at the end of the Migrate-Disable region is the last + * one. This means we only need to track a single cpumask (i.e. p->cpus_mask), + * but we still need to properly signal those waiting tasks at the appropriate + * moment. + * + * This is implemented using struct set_affinity_pending. The first + * __set_cpus_allowed_ptr() caller within a given Migrate-Disable region will + * setup an instance of that struct and install it on the targeted task_struct. + * Any and all further callers will reuse that instance. Those then wait for + * a completion signaled at the tail of the CPU stopper callback (1), triggered + * on the end of the Migrate-Disable region (i.e. outermost migrate_enable()). + * + * + * (1) In the cases covered above. There is one more where the completion is + * signaled within affine_move_task() itself: when a subsequent affinity request + * occurs after the stopper bailed out due to the targeted task still being + * Migrate-Disable. Consider: + * + * Initial conditions: P0->cpus_mask = [0, 1] + * + * CPU0 P1 P2 + * <P0> + * migrate_disable(); + * <preempted> + * set_cpus_allowed_ptr(P0, [1]); + * <blocks> + * <migration/0> + * migration_cpu_stop() + * is_migration_disabled() + * <bails> + * set_cpus_allowed_ptr(P0, [0, 1]); + * <signal completion> + * <awakes> + * + * Note that the above is safe vs a concurrent migrate_enable(), as any + * pending affinity completion is preceded by an uninstallation of + * p->migration_pending done with p->pi_lock held. + */ +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) +{ + 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 + * + * 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) && + (p->migration_flags & MDF_PUSH) && !rq->push_busy) { + rq->push_busy = true; + push_task = get_task_struct(p); + } + + /* + * If there are pending waiters, but no pending stop_work, + * then complete now. + */ + pending = p->migration_pending; + if (pending && !pending->stop_pending) { + p->migration_pending = NULL; + complete = true; + } + + preempt_disable(); + task_rq_unlock(rq, p, rf); + if (push_task) { + stop_one_cpu_nowait(rq->cpu, push_cpu_stop, + p, &rq->push_work); + } + preempt_enable(); + + if (complete) + complete_all(&pending->done); + + return 0; + } + + if (!(flags & SCA_MIGRATE_ENABLE)) { + /* serialized by p->pi_lock */ + if (!p->migration_pending) { + /* Install the request */ + refcount_set(&my_pending.refs, 1); + init_completion(&my_pending.done); + my_pending.arg = (struct migration_arg) { + .task = p, + .dest_cpu = dest_cpu, + .pending = &my_pending, + }; + + p->migration_pending = &my_pending; + } else { + pending = p->migration_pending; + refcount_inc(&pending->refs); + /* + * Affinity has changed, but we've already installed a + * pending. migration_cpu_stop() *must* see this, else + * we risk a completion of the pending despite having a + * task on a disallowed CPU. + * + * Serialized by p->pi_lock, so this is safe. + */ + pending->arg.dest_cpu = dest_cpu; + } + } + pending = p->migration_pending; + /* + * - !MIGRATE_ENABLE: + * we'll have installed a pending if there wasn't one already. + * + * - MIGRATE_ENABLE: + * we're here because the current CPU isn't matching anymore, + * the only way that can happen is because of a concurrent + * set_cpus_allowed_ptr() call, which should then still be + * pending completion. + * + * Either way, we really should have a @pending here. + */ + if (WARN_ON_ONCE(!pending)) { + task_rq_unlock(rq, p, rf); + return -EINVAL; + } + + if (task_on_cpu(rq, p) || READ_ONCE(p->__state) == TASK_WAKING) { + /* + * MIGRATE_ENABLE gets here because 'p == current', but for + * anything else we cannot do is_migration_disabled(), punt + * and have the stopper function handle it all race-free. + */ + stop_pending = pending->stop_pending; + if (!stop_pending) + pending->stop_pending = true; + + if (flags & SCA_MIGRATE_ENABLE) + p->migration_flags &= ~MDF_PUSH; + + preempt_disable(); + task_rq_unlock(rq, p, rf); + if (!stop_pending) { + stop_one_cpu_nowait(cpu_of(rq), migration_cpu_stop, + &pending->arg, &pending->stop_work); + } + preempt_enable(); + + if (flags & SCA_MIGRATE_ENABLE) + return 0; + } else { + + if (!is_migration_disabled(p)) { + if (task_on_rq_queued(p)) + rq = move_queued_task(rq, rf, p, dest_cpu); + + if (!pending->stop_pending) { + p->migration_pending = NULL; + complete = true; + } + } + task_rq_unlock(rq, p, rf); + + if (complete) + complete_all(&pending->done); + } + + wait_for_completion(&pending->done); + + if (refcount_dec_and_test(&pending->refs)) + wake_up_var(&pending->refs); /* No UaF, just an address */ + + /* + * Block the original owner of &pending until all subsequent callers + * have seen the completion and decremented the refcount + */ + wait_var_event(&my_pending.refs, !refcount_read(&my_pending.refs)); + + /* ARGH */ + WARN_ON_ONCE(my_pending.stop_pending); + + return 0; +} + +/* + * Called with both p->pi_lock and rq->lock held; drops both before returning. + */ +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) +{ + const struct cpumask *cpu_allowed_mask = task_cpu_possible_mask(p); + const struct cpumask *cpu_valid_mask = cpu_active_mask; + bool kthread = p->flags & PF_KTHREAD; + unsigned int dest_cpu; + int ret = 0; + + if (kthread || is_migration_disabled(p)) { + /* + * Kernel threads are allowed on online && !active CPUs, + * however, during cpu-hot-unplug, even these might get pushed + * away if not KTHREAD_IS_PER_CPU. + * + * Specifically, migration_disabled() tasks must not fail the + * cpumask_any_and_distribute() pick below, esp. so on + * SCA_MIGRATE_ENABLE, otherwise we'll not call + * set_cpus_allowed_common() and actually reset p->cpus_ptr. + */ + cpu_valid_mask = cpu_online_mask; + } + + if (!kthread && !cpumask_subset(ctx->new_mask, cpu_allowed_mask)) { + ret = -EINVAL; + goto out; + } + + /* + * Must re-check here, to close a race against __kthread_bind(), + * sched_setaffinity() is not guaranteed to observe the flag. + */ + if ((ctx->flags & SCA_CHECK) && (p->flags & PF_NO_SETAFFINITY)) { + ret = -EINVAL; + goto out; + } + + if (!(ctx->flags & SCA_MIGRATE_ENABLE)) { + if (cpumask_equal(&p->cpus_mask, ctx->new_mask)) { + if (ctx->flags & SCA_USER) + swap(p->user_cpus_ptr, ctx->user_mask); + goto out; + } + + if (WARN_ON_ONCE(p == current && + is_migration_disabled(p) && + !cpumask_test_cpu(task_cpu(p), ctx->new_mask))) { + ret = -EBUSY; + goto out; + } + } + + /* + * Picking a ~random cpu helps in cases where we are changing affinity + * for groups of tasks (ie. cpuset), so that load balancing is not + * immediately required to distribute the tasks within their new mask. + */ + dest_cpu = cpumask_any_and_distribute(cpu_valid_mask, ctx->new_mask); + if (dest_cpu >= nr_cpu_ids) { + ret = -EINVAL; + goto out; + } + + do_set_cpus_allowed(p, ctx); + + return affine_move_task(rq, p, rf, dest_cpu, ctx->flags); + +out: + task_rq_unlock(rq, p, rf); + + return ret; +} + +/* + * Change a given task's CPU affinity. Migrate the thread to a + * proper CPU and schedule it away if the CPU it's executing on + * is removed from the allowed bitmask. + * + * NOTE: the caller must have a valid reference to the task, the + * task must not exit() & deallocate itself prematurely. The + * call is not atomic; no spinlocks may be held. + */ +int __set_cpus_allowed_ptr(struct task_struct *p, struct affinity_context *ctx) +{ + struct rq_flags rf; + struct rq *rq; + + rq = task_rq_lock(p, &rf); + /* + * Masking should be skipped if SCA_USER or any of the SCA_MIGRATE_* + * flags are set. + */ + if (p->user_cpus_ptr && + !(ctx->flags & (SCA_USER | SCA_MIGRATE_ENABLE | SCA_MIGRATE_DISABLE)) && + cpumask_and(rq->scratch_mask, ctx->new_mask, p->user_cpus_ptr)) + ctx->new_mask = rq->scratch_mask; + + return __set_cpus_allowed_ptr_locked(p, ctx, rq, &rf); +} + +int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) +{ + struct affinity_context ac = { + .new_mask = new_mask, + .flags = 0, + }; + + return __set_cpus_allowed_ptr(p, &ac); +} +EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr); + +/* + * Change a given task's CPU affinity to the intersection of its current + * affinity mask and @subset_mask, writing the resulting mask to @new_mask. + * If user_cpus_ptr is defined, use it as the basis for restricting CPU + * affinity or use cpu_online_mask instead. + * + * If the resulting mask is empty, leave the affinity unchanged and return + * -EINVAL. + */ +static int restrict_cpus_allowed_ptr(struct task_struct *p, + struct cpumask *new_mask, + const struct cpumask *subset_mask) +{ + struct affinity_context ac = { + .new_mask = new_mask, + .flags = 0, + }; + struct rq_flags rf; + struct rq *rq; + int err; + + rq = task_rq_lock(p, &rf); + + /* + * Forcefully restricting the affinity of a deadline task is + * likely to cause problems, so fail and noisily override the + * mask entirely. + */ + if (task_has_dl_policy(p) && dl_bandwidth_enabled()) { + err = -EPERM; + goto err_unlock; + } + + if (!cpumask_and(new_mask, task_user_cpus(p), subset_mask)) { + err = -EINVAL; + goto err_unlock; + } + + return __set_cpus_allowed_ptr_locked(p, &ac, rq, &rf); + +err_unlock: + task_rq_unlock(rq, p, &rf); + return err; +} + +/* + * Restrict the CPU affinity of task @p so that it is a subset of + * task_cpu_possible_mask() and point @p->user_cpus_ptr to a copy of the + * old affinity mask. If the resulting mask is empty, we warn and walk + * up the cpuset hierarchy until we find a suitable mask. + */ +void force_compatible_cpus_allowed_ptr(struct task_struct *p) +{ + cpumask_var_t new_mask; + const struct cpumask *override_mask = task_cpu_possible_mask(p); + + alloc_cpumask_var(&new_mask, GFP_KERNEL); + + /* + * __migrate_task() can fail silently in the face of concurrent + * offlining of the chosen destination CPU, so take the hotplug + * lock to ensure that the migration succeeds. + */ + cpus_read_lock(); + if (!cpumask_available(new_mask)) + goto out_set_mask; + + if (!restrict_cpus_allowed_ptr(p, new_mask, override_mask)) + goto out_free_mask; + + /* + * We failed to find a valid subset of the affinity mask for the + * task, so override it based on its cpuset hierarchy. + */ + cpuset_cpus_allowed(p, new_mask); + override_mask = new_mask; + +out_set_mask: + if (printk_ratelimit()) { + printk_deferred("Overriding affinity for process %d (%s) to CPUs %*pbl\n", + task_pid_nr(p), p->comm, + cpumask_pr_args(override_mask)); + } + + WARN_ON(set_cpus_allowed_ptr(p, override_mask)); +out_free_mask: + cpus_read_unlock(); + free_cpumask_var(new_mask); +} + +/* + * Restore the affinity of a task @p which was previously restricted by a + * call to force_compatible_cpus_allowed_ptr(). + * + * It is the caller's responsibility to serialise this with any calls to + * force_compatible_cpus_allowed_ptr(@p). + */ +void relax_compatible_cpus_allowed_ptr(struct task_struct *p) +{ + struct affinity_context ac = { + .new_mask = task_user_cpus(p), + .flags = 0, + }; + int ret; + + /* + * Try to restore the old affinity mask with __sched_setaffinity(). + * Cpuset masking will be done there too. + */ + ret = __sched_setaffinity(p, &ac); + WARN_ON_ONCE(ret); +} + +#ifdef CONFIG_SMP + +void set_task_cpu(struct task_struct *p, unsigned int new_cpu) +{ + unsigned int state = READ_ONCE(p->__state); + + /* + * We should never call set_task_cpu() on a blocked task, + * ttwu() will sort out the placement. + */ + WARN_ON_ONCE(state != TASK_RUNNING && state != TASK_WAKING && !p->on_rq); + + /* + * Migrating fair class task must have p->on_rq = TASK_ON_RQ_MIGRATING, + * because schedstat_wait_{start,end} rebase migrating task's wait_start + * time relying on p->on_rq. + */ + WARN_ON_ONCE(state == TASK_RUNNING && + p->sched_class == &fair_sched_class && + (p->on_rq && !task_on_rq_migrating(p))); + +#ifdef CONFIG_LOCKDEP + /* + * The caller should hold either p->pi_lock or rq->lock, when changing + * a task's CPU. ->pi_lock for waking tasks, rq->lock for runnable tasks. + * + * sched_move_task() holds both and thus holding either pins the cgroup, + * see task_group(). + * + * Furthermore, all task_rq users should acquire both locks, see + * task_rq_lock(). + */ + WARN_ON_ONCE(debug_locks && !(lockdep_is_held(&p->pi_lock) || + lockdep_is_held(__rq_lockp(task_rq(p))))); +#endif + /* + * Clearly, migrating tasks to offline CPUs is a fairly daft thing. + */ + WARN_ON_ONCE(!cpu_online(new_cpu)); + + WARN_ON_ONCE(is_migration_disabled(p)); + + trace_sched_migrate_task(p, new_cpu); + + if (task_cpu(p) != new_cpu) { + if (p->sched_class->migrate_task_rq) + p->sched_class->migrate_task_rq(p, new_cpu); + p->se.nr_migrations++; + 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) +{ + if (task_on_rq_queued(p)) { + struct rq *src_rq, *dst_rq; + struct rq_flags srf, drf; + + src_rq = task_rq(p); + dst_rq = cpu_rq(cpu); + + rq_pin_lock(src_rq, &srf); + rq_pin_lock(dst_rq, &drf); + + move_queued_task_locked(src_rq, dst_rq, p); + wakeup_preempt(dst_rq, p, 0); + + rq_unpin_lock(dst_rq, &drf); + rq_unpin_lock(src_rq, &srf); + + } else { + /* + * Task isn't running anymore; make it appear like we migrated + * it before it went to sleep. This means on wakeup we make the + * previous CPU our target instead of where it really is. + */ + p->wake_cpu = cpu; + } +} + +struct migration_swap_arg { + struct task_struct *src_task, *dst_task; + int src_cpu, dst_cpu; +}; + +static int migrate_swap_stop(void *data) +{ + struct migration_swap_arg *arg = data; + struct rq *src_rq, *dst_rq; + + if (!cpu_active(arg->src_cpu) || !cpu_active(arg->dst_cpu)) + return -EAGAIN; + + src_rq = cpu_rq(arg->src_cpu); + dst_rq = cpu_rq(arg->dst_cpu); + + guard(double_raw_spinlock)(&arg->src_task->pi_lock, &arg->dst_task->pi_lock); + guard(double_rq_lock)(src_rq, dst_rq); + + if (task_cpu(arg->dst_task) != arg->dst_cpu) + return -EAGAIN; + + if (task_cpu(arg->src_task) != arg->src_cpu) + return -EAGAIN; + + if (!cpumask_test_cpu(arg->dst_cpu, arg->src_task->cpus_ptr)) + return -EAGAIN; + + if (!cpumask_test_cpu(arg->src_cpu, arg->dst_task->cpus_ptr)) + return -EAGAIN; + + __migrate_swap_task(arg->src_task, arg->dst_cpu); + __migrate_swap_task(arg->dst_task, arg->src_cpu); + + return 0; +} + +/* + * Cross migrate two tasks + */ +int migrate_swap(struct task_struct *cur, struct task_struct *p, + int target_cpu, int curr_cpu) +{ + struct migration_swap_arg arg; + int ret = -EINVAL; + + arg = (struct migration_swap_arg){ + .src_task = cur, + .src_cpu = curr_cpu, + .dst_task = p, + .dst_cpu = target_cpu, + }; + + if (arg.src_cpu == arg.dst_cpu) + goto out; + + /* + * These three tests are all lockless; this is OK since all of them + * will be re-checked with proper locks held further down the line. + */ + if (!cpu_active(arg.src_cpu) || !cpu_active(arg.dst_cpu)) + goto out; + + if (!cpumask_test_cpu(arg.dst_cpu, arg.src_task->cpus_ptr)) + goto out; + + if (!cpumask_test_cpu(arg.src_cpu, arg.dst_task->cpus_ptr)) + goto out; + + trace_sched_swap_numa(cur, arg.src_cpu, p, arg.dst_cpu); + ret = stop_two_cpus(arg.dst_cpu, arg.src_cpu, migrate_swap_stop, &arg); + +out: + return ret; +} +#endif /* CONFIG_NUMA_BALANCING */ + /*** * kick_process - kick a running thread to enter/exit the kernel * @p: the to-be-kicked thread @@ -1160,20 +3390,35 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state) */ void kick_process(struct task_struct *p) { - int cpu; + guard(preempt)(); + int cpu = task_cpu(p); - preempt_disable(); - cpu = task_cpu(p); if ((cpu != smp_processor_id()) && task_curr(p)) smp_send_reschedule(cpu); - preempt_enable(); } EXPORT_SYMBOL_GPL(kick_process); -#endif /* CONFIG_SMP */ -#ifdef CONFIG_SMP /* - * ->cpus_allowed is protected by both rq->lock and p->pi_lock + * ->cpus_ptr is protected by both rq->lock and p->pi_lock + * + * A few notes on cpu_active vs cpu_online: + * + * - cpu_active must be a subset of cpu_online + * + * - on CPU-up we allow per-CPU kthreads on the online && !active CPU, + * see __set_cpus_allowed_ptr(). At this point the newly online + * CPU isn't yet part of the sched domains, and balancing will not + * see it. + * + * - on CPU-down we clear cpu_active() to mask the sched domains and + * avoid the load balancer to place new tasks on the to be removed + * CPU. Existing tasks will remain running there and will be taken + * off. + * + * This means that fallback selection must not select !active CPUs. + * And can assume that any active CPU must be online. Conversely + * select_task_rq() below may allow selection of !active CPUs in order + * to satisfy the above rules. */ static int select_fallback_rq(int cpu, struct task_struct *p) { @@ -1183,46 +3428,41 @@ static int select_fallback_rq(int cpu, struct task_struct *p) int dest_cpu; /* - * If the node that the cpu is on has been offlined, cpu_to_node() - * will return -1. There is no cpu on the node, and we should - * select the cpu on the other node. + * If the node that the CPU is on has been offlined, cpu_to_node() + * will return -1. There is no CPU on the node, and we should + * select the CPU on the other node. */ if (nid != -1) { nodemask = cpumask_of_node(nid); /* Look for allowed, online CPU in same node. */ for_each_cpu(dest_cpu, nodemask) { - if (!cpu_online(dest_cpu)) - continue; - if (!cpu_active(dest_cpu)) - continue; - if (cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) + if (is_cpu_allowed(p, dest_cpu)) return dest_cpu; } } for (;;) { /* Any allowed, online CPU? */ - for_each_cpu(dest_cpu, tsk_cpus_allowed(p)) { - if (!cpu_online(dest_cpu)) - continue; - if (!cpu_active(dest_cpu)) + for_each_cpu(dest_cpu, p->cpus_ptr) { + if (!is_cpu_allowed(p, dest_cpu)) continue; + goto out; } + /* No more Mr. Nice Guy. */ switch (state) { case cpuset: - /* No more Mr. Nice Guy. */ - cpuset_cpus_allowed_fallback(p); - state = possible; - break; - + if (cpuset_cpus_allowed_fallback(p)) { + state = possible; + break; + } + fallthrough; case possible: - do_set_cpus_allowed(p, cpu_possible_mask); + set_cpus_allowed_force(p, task_cpu_fallback_mask(p)); state = fail; break; - case fail: BUG(); break; @@ -1237,7 +3477,7 @@ out: * leave kernel. */ if (p->mm && printk_ratelimit()) { - printk_sched("process %d (%s) no longer affine to cpu%d\n", + printk_deferred("process %d (%s) no longer affine to cpu%d\n", task_pid_nr(p), p->comm, cpu); } } @@ -1246,330 +3486,851 @@ out: } /* - * The caller (fork, wakeup) owns p->pi_lock, ->cpus_allowed is stable. + * The caller (fork, wakeup) owns p->pi_lock, ->cpus_ptr is stable. */ static inline -int select_task_rq(struct task_struct *p, int sd_flags, int wake_flags) +int select_task_rq(struct task_struct *p, int cpu, int *wake_flags) { - int cpu = p->sched_class->select_task_rq(p, sd_flags, wake_flags); + lockdep_assert_held(&p->pi_lock); + + if (p->nr_cpus_allowed > 1 && !is_migration_disabled(p)) { + cpu = p->sched_class->select_task_rq(p, cpu, *wake_flags); + *wake_flags |= WF_RQ_SELECTED; + } else { + cpu = cpumask_any(p->cpus_ptr); + } /* * In order not to call set_task_cpu() on a blocking task we need - * to rely on ttwu() to place the task on a valid ->cpus_allowed - * cpu. + * to rely on ttwu() to place the task on a valid ->cpus_ptr + * CPU. * * Since this is common to all placement strategies, this lives here. * * [ this allows ->select_task() to simply return task_cpu(p) and * not worry about this generic constraint ] */ - if (unlikely(!cpumask_test_cpu(cpu, tsk_cpus_allowed(p)) || - !cpu_online(cpu))) + if (unlikely(!is_cpu_allowed(p, cpu))) cpu = select_fallback_rq(task_cpu(p), p); return cpu; } -static void update_avg(u64 *avg, u64 sample) +void sched_set_stop_task(int cpu, struct task_struct *stop) { - s64 diff = sample - *avg; - *avg += diff >> 3; + static struct lock_class_key stop_pi_lock; + struct sched_param param = { .sched_priority = MAX_RT_PRIO - 1 }; + struct task_struct *old_stop = cpu_rq(cpu)->stop; + + if (stop) { + /* + * Make it appear like a SCHED_FIFO task, its something + * userspace knows about and won't get confused about. + * + * Also, it will make PI more or less work without too + * much confusion -- but then, stop work should not + * rely on PI working anyway. + */ + sched_setscheduler_nocheck(stop, SCHED_FIFO, ¶m); + + stop->sched_class = &stop_sched_class; + + /* + * The PI code calls rt_mutex_setprio() with ->pi_lock held to + * adjust the effective priority of a task. As a result, + * rt_mutex_setprio() can trigger (RT) balancing operations, + * which can then trigger wakeups of the stop thread to push + * around the current task. + * + * The stop task itself will never be part of the PI-chain, it + * never blocks, therefore that ->pi_lock recursion is safe. + * Tell lockdep about this by placing the stop->pi_lock in its + * own class. + */ + lockdep_set_class(&stop->pi_lock, &stop_pi_lock); + } + + cpu_rq(cpu)->stop = stop; + + if (old_stop) { + /* + * Reset it back to a normal scheduling class so that + * it can die in pieces. + */ + old_stop->sched_class = &rt_sched_class; + } } -#endif static void ttwu_stat(struct task_struct *p, int cpu, int wake_flags) { -#ifdef CONFIG_SCHEDSTATS - struct rq *rq = this_rq(); + struct rq *rq; -#ifdef CONFIG_SMP - int this_cpu = smp_processor_id(); + if (!schedstat_enabled()) + return; - if (cpu == this_cpu) { - schedstat_inc(rq, ttwu_local); - schedstat_inc(p, se.statistics.nr_wakeups_local); + rq = this_rq(); + + if (cpu == rq->cpu) { + __schedstat_inc(rq->ttwu_local); + __schedstat_inc(p->stats.nr_wakeups_local); } else { struct sched_domain *sd; - schedstat_inc(p, se.statistics.nr_wakeups_remote); - rcu_read_lock(); - for_each_domain(this_cpu, sd) { + __schedstat_inc(p->stats.nr_wakeups_remote); + + guard(rcu)(); + for_each_domain(rq->cpu, sd) { if (cpumask_test_cpu(cpu, sched_domain_span(sd))) { - schedstat_inc(sd, ttwu_wake_remote); + __schedstat_inc(sd->ttwu_wake_remote); break; } } - rcu_read_unlock(); } if (wake_flags & WF_MIGRATED) - schedstat_inc(p, se.statistics.nr_wakeups_migrate); + __schedstat_inc(p->stats.nr_wakeups_migrate); -#endif /* CONFIG_SMP */ - - schedstat_inc(rq, ttwu_count); - schedstat_inc(p, se.statistics.nr_wakeups); + __schedstat_inc(rq->ttwu_count); + __schedstat_inc(p->stats.nr_wakeups); if (wake_flags & WF_SYNC) - schedstat_inc(p, se.statistics.nr_wakeups_sync); - -#endif /* CONFIG_SCHEDSTATS */ + __schedstat_inc(p->stats.nr_wakeups_sync); } -static void ttwu_activate(struct rq *rq, struct task_struct *p, int en_flags) +/* + * Mark the task runnable. + */ +static inline void ttwu_do_wakeup(struct task_struct *p) { - activate_task(rq, p, en_flags); - p->on_rq = 1; - - /* if a worker is waking up, notify workqueue */ - if (p->flags & PF_WQ_WORKER) - wq_worker_waking_up(p, cpu_of(rq)); + WRITE_ONCE(p->__state, TASK_RUNNING); + trace_sched_wakeup(p); } -/* - * Mark the task runnable and perform wakeup-preemption. - */ static void -ttwu_do_wakeup(struct rq *rq, struct task_struct *p, int wake_flags) +ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags, + struct rq_flags *rf) { - check_preempt_curr(rq, p, wake_flags); - trace_sched_wakeup(p, true); + int en_flags = ENQUEUE_WAKEUP | ENQUEUE_NOCLOCK; - p->state = TASK_RUNNING; -#ifdef CONFIG_SMP - if (p->sched_class->task_woken) + lockdep_assert_rq_held(rq); + + if (p->sched_contributes_to_load) + rq->nr_uninterruptible--; + + if (wake_flags & WF_RQ_SELECTED) + en_flags |= ENQUEUE_RQ_SELECTED; + if (wake_flags & WF_MIGRATED) + en_flags |= ENQUEUE_MIGRATED; + else + if (p->in_iowait) { + delayacct_blkio_end(p); + atomic_dec(&task_rq(p)->nr_iowait); + } + + activate_task(rq, p, en_flags); + wakeup_preempt(rq, p, wake_flags); + + ttwu_do_wakeup(p); + + if (p->sched_class->task_woken) { + /* + * Our task @p is fully woken up and running; so it's safe to + * drop the rq->lock, hereafter rq is only used for statistics. + */ + rq_unpin_lock(rq, rf); 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*sysctl_sched_migration_cost; + u64 max = 2*rq->max_idle_balance_cost; + + update_avg(&rq->avg_idle, delta); - if (delta > max) + if (rq->avg_idle > max) rq->avg_idle = max; - else - update_avg(&rq->avg_idle, delta); + rq->idle_stamp = 0; } -#endif -} - -static void -ttwu_do_activate(struct rq *rq, struct task_struct *p, int wake_flags) -{ -#ifdef CONFIG_SMP - if (p->sched_contributes_to_load) - rq->nr_uninterruptible--; -#endif - - ttwu_activate(rq, p, ENQUEUE_WAKEUP | ENQUEUE_WAKING); - ttwu_do_wakeup(rq, p, wake_flags); } /* - * Called in case the task @p isn't fully descheduled from its runqueue, - * in this case we must do a remote wakeup. Its a 'light' wakeup though, - * since all we need to do is flip p->state to TASK_RUNNING, since - * the task is still ->on_rq. + * Consider @p being inside a wait loop: + * + * for (;;) { + * set_current_state(TASK_UNINTERRUPTIBLE); + * + * if (CONDITION) + * break; + * + * schedule(); + * } + * __set_current_state(TASK_RUNNING); + * + * between set_current_state() and schedule(). In this case @p is still + * runnable, so all that needs doing is change p->state back to TASK_RUNNING in + * an atomic manner. + * + * By taking task_rq(p)->lock we serialize against schedule(), if @p->on_rq + * then schedule() must still happen and p->state can be changed to + * TASK_RUNNING. Otherwise we lost the race, schedule() has happened, and we + * need to do a full wakeup with enqueue. + * + * Returns: %true when the wakeup is done, + * %false otherwise. */ -static int ttwu_remote(struct task_struct *p, int wake_flags) +static int ttwu_runnable(struct task_struct *p, int wake_flags) { + struct rq_flags rf; struct rq *rq; int ret = 0; - rq = __task_rq_lock(p); - if (p->on_rq) { - /* check_preempt_curr() may use rq clock */ + rq = __task_rq_lock(p, &rf); + if (task_on_rq_queued(p)) { update_rq_clock(rq); - ttwu_do_wakeup(rq, p, wake_flags); + if (p->se.sched_delayed) + enqueue_task(rq, p, ENQUEUE_NOCLOCK | ENQUEUE_DELAYED); + if (!task_on_cpu(rq, p)) { + /* + * When on_rq && !on_cpu the task is preempted, see if + * it should preempt the task that is current now. + */ + wakeup_preempt(rq, p, wake_flags); + } + ttwu_do_wakeup(p); ret = 1; } - __task_rq_unlock(rq); + __task_rq_unlock(rq, p, &rf); return ret; } -#ifdef CONFIG_SMP -static void sched_ttwu_pending(void) +void sched_ttwu_pending(void *arg) { + struct llist_node *llist = arg; struct rq *rq = this_rq(); - struct llist_node *llist = llist_del_all(&rq->wake_list); - struct task_struct *p; + struct task_struct *p, *t; + struct rq_flags rf; - raw_spin_lock(&rq->lock); + if (!llist) + return; - while (llist) { - p = llist_entry(llist, struct task_struct, wake_entry); - llist = llist_next(llist); - ttwu_do_activate(rq, p, 0); - } + rq_lock_irqsave(rq, &rf); + update_rq_clock(rq); - raw_spin_unlock(&rq->lock); -} + llist_for_each_entry_safe(p, t, llist, wake_entry.llist) { + if (WARN_ON_ONCE(p->on_cpu)) + smp_cond_load_acquire(&p->on_cpu, !VAL); -void scheduler_ipi(void) -{ - if (llist_empty(&this_rq()->wake_list) - && !tick_nohz_full_cpu(smp_processor_id()) - && !got_nohz_idle_kick()) - return; + if (WARN_ON_ONCE(task_cpu(p) != cpu_of(rq))) + set_task_cpu(p, cpu_of(rq)); + + ttwu_do_activate(rq, p, p->sched_remote_wakeup ? WF_MIGRATED : 0, &rf); + } /* - * Not all reschedule IPI handlers call irq_enter/irq_exit, since - * traditionally all their work was done from the interrupt return - * path. Now that we actually do some work, we need to make sure - * we do call them. - * - * Some archs already do call them, luckily irq_enter/exit nest - * properly. + * Must be after enqueueing at least once task such that + * idle_cpu() does not observe a false-negative -- if it does, + * it is possible for select_idle_siblings() to stack a number + * of tasks on this CPU during that window. * - * Arguably we should visit all archs and update all handlers, - * however a fair share of IPIs are still resched only so this would - * somewhat pessimize the simple resched case. + * It is OK to clear ttwu_pending when another task pending. + * We will receive IPI after local IRQ enabled and then enqueue it. + * Since now nr_running > 0, idle_cpu() will always get correct result. */ - irq_enter(); - tick_nohz_full_check(); - sched_ttwu_pending(); + WRITE_ONCE(rq->ttwu_pending, 0); + rq_unlock_irqrestore(rq, &rf); +} - /* - * Check if someone kicked us for doing the nohz idle load balance. - */ - if (unlikely(got_nohz_idle_kick())) { - this_rq()->idle_balance = 1; - raise_softirq_irqoff(SCHED_SOFTIRQ); +/* + * Prepare the scene for sending an IPI for a remote smp_call + * + * Returns true if the caller can proceed with sending the IPI. + * Returns false otherwise. + */ +bool call_function_single_prep_ipi(int cpu) +{ + if (set_nr_if_polling(cpu_rq(cpu)->idle)) { + trace_sched_wake_idle_without_ipi(cpu); + return false; } - irq_exit(); + + return true; } -static void ttwu_queue_remote(struct task_struct *p, int cpu) +/* + * Queue a task on the target CPUs wake_list and wake the CPU via IPI if + * necessary. The wakee CPU on receipt of the IPI will queue the task + * via sched_ttwu_wakeup() for activation so the wakee incurs the cost + * of the wakeup instead of the waker. + */ +static void __ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags) { - if (llist_add(&p->wake_entry, &cpu_rq(cpu)->wake_list)) - smp_send_reschedule(cpu); + struct rq *rq = cpu_rq(cpu); + + 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) +{ + struct rq *rq = cpu_rq(cpu); + + guard(rcu)(); + if (is_idle_task(rcu_dereference(rq->curr))) { + guard(rq_lock_irqsave)(rq); + if (is_idle_task(rq->curr)) + resched_curr(rq); + } +} + +bool cpus_equal_capacity(int this_cpu, int that_cpu) +{ + if (!sched_asym_cpucap_active()) + return true; + + if (this_cpu == that_cpu) + return true; + + return arch_scale_cpu_capacity(this_cpu) == arch_scale_cpu_capacity(that_cpu); } bool cpus_share_cache(int this_cpu, int that_cpu) { + if (this_cpu == that_cpu) + return true; + return per_cpu(sd_llc_id, this_cpu) == per_cpu(sd_llc_id, that_cpu); } -#endif /* CONFIG_SMP */ -static void ttwu_queue(struct task_struct *p, int cpu) +/* + * Whether CPUs are share cache resources, which means LLC on non-cluster + * machines and LLC tag or L2 on machines with clusters. + */ +bool cpus_share_resources(int this_cpu, int that_cpu) +{ + if (this_cpu == that_cpu) + return true; + + return per_cpu(sd_share_id, this_cpu) == per_cpu(sd_share_id, that_cpu); +} + +static inline bool ttwu_queue_cond(struct task_struct *p, int cpu) +{ + /* 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 + * in hotplug state. + */ + if (!cpu_active(cpu)) + return false; + + /* Ensure the task will still be allowed to run on the CPU. */ + if (!cpumask_test_cpu(cpu, p->cpus_ptr)) + return false; + + /* + * 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)) + return true; + + if (cpu == smp_processor_id()) + return false; + + /* + * If the wakee cpu is idle, or the task is descheduling and the + * only running task on the CPU, then use the wakelist to offload + * the task activation to the idle (or soon-to-be-idle) CPU as + * the current CPU is likely busy. nr_running is checked to + * avoid unnecessary task stacking. + * + * Note that we can only get here with (wakee) p->on_rq=0, + * p->on_cpu can be whatever, we've done the dequeue, so + * the wakee has been accounted out of ->nr_running. + */ + if (!cpu_rq(cpu)->nr_running) + return true; + + return false; +} + +static bool ttwu_queue_wakelist(struct task_struct *p, int cpu, int wake_flags) +{ + if (sched_feat(TTWU_QUEUE) && ttwu_queue_cond(p, cpu)) { + sched_clock_cpu(cpu); /* Sync clocks across CPUs */ + __ttwu_queue_wakelist(p, cpu, wake_flags); + return true; + } + + return false; +} + +static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) { struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; -#if defined(CONFIG_SMP) - if (sched_feat(TTWU_QUEUE) && !cpus_share_cache(smp_processor_id(), cpu)) { - sched_clock_cpu(cpu); /* sync clocks x-cpu */ - ttwu_queue_remote(p, cpu); + if (ttwu_queue_wakelist(p, cpu, wake_flags)) return; + + rq_lock(rq, &rf); + update_rq_clock(rq); + ttwu_do_activate(rq, p, wake_flags, &rf); + rq_unlock(rq, &rf); +} + +/* + * Invoked from try_to_wake_up() to check whether the task can be woken up. + * + * The caller holds p::pi_lock if p != current or has preemption + * disabled when p == current. + * + * The rules of saved_state: + * + * The related locking code always holds p::pi_lock when updating + * p::saved_state, which means the code is fully serialized in both cases. + * + * For PREEMPT_RT, the lock wait and lock wakeups happen via TASK_RTLOCK_WAIT. + * No other bits set. This allows to distinguish all wakeup scenarios. + * + * For FREEZER, the wakeup happens via TASK_FROZEN. No other bits set. This + * allows us to prevent early wakeup of tasks before they can be run on + * asymmetric ISA architectures (eg ARMv9). + */ +static __always_inline +bool ttwu_state_match(struct task_struct *p, unsigned int state, int *success) +{ + int match; + + if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)) { + WARN_ON_ONCE((state & TASK_RTLOCK_WAIT) && + state != TASK_RTLOCK_WAIT); } -#endif - raw_spin_lock(&rq->lock); - ttwu_do_activate(rq, p, 0); - raw_spin_unlock(&rq->lock); + *success = !!(match = __task_state_match(p, state)); + + /* + * Saved state preserves the task state across blocking on + * an RT lock or TASK_FREEZABLE tasks. If the state matches, + * set p::saved_state to TASK_RUNNING, but do not wake the task + * because it waits for a lock wakeup or __thaw_task(). Also + * indicate success because from the regular waker's point of + * view this has succeeded. + * + * After acquiring the lock the task will restore p::__state + * from p::saved_state which ensures that the regular + * wakeup is not lost. The restore will also set + * p::saved_state to TASK_RUNNING so any further tests will + * not result in false positives vs. @success + */ + if (match < 0) + p->saved_state = TASK_RUNNING; + + return match > 0; } +/* + * Notes on Program-Order guarantees on SMP systems. + * + * MIGRATION + * + * The basic program-order guarantee on SMP systems is that when a task [t] + * migrates, all its activity on its old CPU [c0] happens-before any subsequent + * execution on its new CPU [c1]. + * + * For migration (of runnable tasks) this is provided by the following means: + * + * A) UNLOCK of the rq(c0)->lock scheduling out task t + * B) migration for t is required to synchronize *both* rq(c0)->lock and + * rq(c1)->lock (if not at the same time, then in that order). + * C) LOCK of the rq(c1)->lock scheduling in task + * + * Release/acquire chaining guarantees that B happens after A and C after B. + * Note: the CPU doing B need not be c0 or c1 + * + * Example: + * + * CPU0 CPU1 CPU2 + * + * LOCK rq(0)->lock + * sched-out X + * sched-in Y + * UNLOCK rq(0)->lock + * + * LOCK rq(0)->lock // orders against CPU0 + * dequeue X + * UNLOCK rq(0)->lock + * + * LOCK rq(1)->lock + * enqueue X + * UNLOCK rq(1)->lock + * + * LOCK rq(1)->lock // orders against CPU2 + * sched-out Z + * sched-in X + * UNLOCK rq(1)->lock + * + * + * BLOCKING -- aka. SLEEP + WAKEUP + * + * For blocking we (obviously) need to provide the same guarantee as for + * migration. However the means are completely different as there is no lock + * chain to provide order. Instead we do: + * + * 1) smp_store_release(X->on_cpu, 0) -- finish_task() + * 2) smp_cond_load_acquire(!X->on_cpu) -- try_to_wake_up() + * + * Example: + * + * CPU0 (schedule) CPU1 (try_to_wake_up) CPU2 (schedule) + * + * LOCK rq(0)->lock LOCK X->pi_lock + * dequeue X + * sched-out X + * smp_store_release(X->on_cpu, 0); + * + * smp_cond_load_acquire(&X->on_cpu, !VAL); + * X->state = WAKING + * set_task_cpu(X,2) + * + * LOCK rq(2)->lock + * enqueue X + * X->state = RUNNING + * UNLOCK rq(2)->lock + * + * LOCK rq(2)->lock // orders against CPU1 + * sched-out Z + * sched-in X + * UNLOCK rq(2)->lock + * + * UNLOCK X->pi_lock + * UNLOCK rq(0)->lock + * + * + * However, for wakeups there is a second guarantee we must provide, namely we + * must ensure that CONDITION=1 done by the caller can not be reordered with + * accesses to the task state; see try_to_wake_up() and set_current_state(). + */ + /** * try_to_wake_up - wake up a thread * @p: the thread to be awakened * @state: the mask of task states that can be woken * @wake_flags: wake modifier flags (WF_*) * - * Put it on the run-queue if it's not already there. The "current" - * thread is always on the run-queue (except when the actual - * re-schedule is in progress), and as such you're allowed to do - * the simpler "current->state = TASK_RUNNING" to mark yourself - * runnable without the overhead of this. + * Conceptually does: * - * Returns %true if @p was woken up, %false if it was already running - * or @state didn't match @p's state. + * If (@state & @p->state) @p->state = TASK_RUNNING. + * + * If the task was not queued/runnable, also place it back on a runqueue. + * + * This function is atomic against schedule() which would dequeue the task. + * + * It issues a full memory barrier before accessing @p->state, see the comment + * with set_current_state(). + * + * Uses p->pi_lock to serialize against concurrent wake-ups. + * + * Relies on p->pi_lock stabilizing: + * - p->sched_class + * - p->cpus_ptr + * - p->sched_task_group + * in order to do migration, see its use of select_task_rq()/set_task_cpu(). + * + * Tries really hard to only take one task_rq(p)->lock for performance. + * Takes rq->lock in: + * - ttwu_runnable() -- old rq, unavoidable, see comment there; + * - ttwu_queue() -- new rq, for enqueue of the task; + * - psi_ttwu_dequeue() -- much sadness :-( accounting will kill us. + * + * As a consequence we race really badly with just about everything. See the + * many memory barriers and their comments for details. + * + * Return: %true if @p->state changes (an actual wakeup was done), + * %false otherwise. */ -static int -try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) +int try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags) { - unsigned long flags; + guard(preempt)(); int cpu, success = 0; - smp_wmb(); - raw_spin_lock_irqsave(&p->pi_lock, flags); - if (!(p->state & state)) - goto out; + wake_flags |= WF_TTWU; - success = 1; /* we're going to change ->state */ - cpu = task_cpu(p); + if (p == current) { + /* + * We're waking current, this means 'p->on_rq' and 'task_cpu(p) + * == smp_processor_id()'. Together this means we can special + * case the whole 'p->on_rq && ttwu_runnable()' case below + * without taking any locks. + * + * Specifically, given current runs ttwu() we must be before + * schedule()'s block_task(), as such this must not observe + * sched_delayed. + * + * In particular: + * - we rely on Program-Order guarantees for all the ordering, + * - we're serialized against set_special_state() by virtue of + * it disabling IRQs (this allows not taking ->pi_lock). + */ + WARN_ON_ONCE(p->se.sched_delayed); + if (!ttwu_state_match(p, state, &success)) + goto out; - if (p->on_rq && ttwu_remote(p, wake_flags)) - goto stat; + trace_sched_waking(p); + ttwu_do_wakeup(p); + goto out; + } -#ifdef CONFIG_SMP /* - * If the owning (remote) cpu is still in the middle of schedule() with - * this task as prev, wait until its done referencing the task. + * If we are going to wake up a thread waiting for CONDITION we + * need to ensure that CONDITION=1 done by the caller can not be + * reordered with p->state check below. This pairs with smp_store_mb() + * in set_current_state() that the waiting thread does. */ - while (p->on_cpu) - cpu_relax(); - /* - * Pairs with the smp_wmb() in finish_lock_switch(). - */ - smp_rmb(); + scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { + smp_mb__after_spinlock(); + if (!ttwu_state_match(p, state, &success)) + break; - p->sched_contributes_to_load = !!task_contributes_to_load(p); - p->state = TASK_WAKING; + trace_sched_waking(p); - if (p->sched_class->task_waking) - p->sched_class->task_waking(p); + /* + * Ensure we load p->on_rq _after_ p->state, otherwise it would + * be possible to, falsely, observe p->on_rq == 0 and get stuck + * in smp_cond_load_acquire() below. + * + * sched_ttwu_pending() try_to_wake_up() + * STORE p->on_rq = 1 LOAD p->state + * UNLOCK rq->lock + * + * __schedule() (switch to task 'p') + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * UNLOCK rq->lock + * + * [task p] + * STORE p->state = UNINTERRUPTIBLE LOAD p->on_rq + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + * + * A similar smp_rmb() lives in __task_needs_rq_lock(). + */ + smp_rmb(); + if (READ_ONCE(p->on_rq) && ttwu_runnable(p, wake_flags)) + break; - cpu = select_task_rq(p, SD_BALANCE_WAKE, wake_flags); - if (task_cpu(p) != cpu) { - wake_flags |= WF_MIGRATED; - set_task_cpu(p, cpu); - } -#endif /* CONFIG_SMP */ + /* + * Ensure we load p->on_cpu _after_ p->on_rq, otherwise it would be + * possible to, falsely, observe p->on_cpu == 0. + * + * One must be running (->on_cpu == 1) in order to remove oneself + * from the runqueue. + * + * __schedule() (switch to task 'p') try_to_wake_up() + * STORE p->on_cpu = 1 LOAD p->on_rq + * UNLOCK rq->lock + * + * __schedule() (put 'p' to sleep) + * LOCK rq->lock smp_rmb(); + * smp_mb__after_spinlock(); + * STORE p->on_rq = 0 LOAD p->on_cpu + * + * Pairs with the LOCK+smp_mb__after_spinlock() on rq->lock in + * __schedule(). See the comment for smp_mb__after_spinlock(). + * + * Form a control-dep-acquire with p->on_rq == 0 above, to ensure + * 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(); + + /* + * We're doing the wakeup (@success == 1), they did a dequeue (p->on_rq + * == 0), which means we need to do an enqueue, change p->state to + * TASK_WAKING such that we can unlock p->pi_lock before doing the + * enqueue, such as ttwu_queue_wakelist(). + */ + WRITE_ONCE(p->__state, TASK_WAKING); + + /* + * If the owning (remote) CPU is still in the middle of schedule() with + * this task as prev, considering queueing p on the remote CPUs wake_list + * which potentially sends an IPI instead of spinning on p->on_cpu to + * let the waker make forward progress. This is safe because IRQs are + * disabled and the IPI will deliver after on_cpu is cleared. + * + * Ensure we load task_cpu(p) after p->on_cpu: + * + * set_task_cpu(p, cpu); + * STORE p->cpu = @cpu + * __schedule() (switch to task 'p') + * LOCK rq->lock + * smp_mb__after_spin_lock() smp_cond_load_acquire(&p->on_cpu) + * STORE p->on_cpu = 1 LOAD p->cpu + * + * to ensure we observe the correct CPU on which the task is currently + * scheduling. + */ + if (smp_load_acquire(&p->on_cpu) && + ttwu_queue_wakelist(p, task_cpu(p), wake_flags)) + break; + + /* + * If the owning (remote) CPU is still in the middle of schedule() with + * this task as prev, wait until it's done referencing the task. + * + * Pairs with the smp_store_release() in finish_task(). + * + * This ensures that tasks getting woken will be fully ordered against + * their previous state and preserve Program Order. + */ + smp_cond_load_acquire(&p->on_cpu, !VAL); + + cpu = select_task_rq(p, p->wake_cpu, &wake_flags); + if (task_cpu(p) != cpu) { + if (p->in_iowait) { + delayacct_blkio_end(p); + atomic_dec(&task_rq(p)->nr_iowait); + } + + wake_flags |= WF_MIGRATED; + psi_ttwu_dequeue(p); + set_task_cpu(p, cpu); + } - ttwu_queue(p, cpu); -stat: - ttwu_stat(p, cpu, wake_flags); + ttwu_queue(p, cpu, wake_flags); + } out: - raw_spin_unlock_irqrestore(&p->pi_lock, flags); + if (success) + ttwu_stat(p, task_cpu(p), wake_flags); return success; } +static bool __task_needs_rq_lock(struct task_struct *p) +{ + unsigned int state = READ_ONCE(p->__state); + + /* + * Since pi->lock blocks try_to_wake_up(), we don't need rq->lock when + * the task is blocked. Make sure to check @state since ttwu() can drop + * locks at the end, see ttwu_queue_wakelist(). + */ + if (state == TASK_RUNNING || state == TASK_WAKING) + return true; + + /* + * Ensure we load p->on_rq after p->__state, otherwise it would be + * possible to, falsely, observe p->on_rq == 0. + * + * See try_to_wake_up() for a longer comment. + */ + smp_rmb(); + if (p->on_rq) + return true; + + /* + * 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); + + return false; +} + /** - * try_to_wake_up_local - try to wake up a local task with rq lock held - * @p: the thread to be awakened + * task_call_func - Invoke a function on task in fixed state + * @p: Process for which the function is to be invoked, can be @current. + * @func: Function to invoke. + * @arg: Argument to function. + * + * Fix the task in it's current state by avoiding wakeups and or rq operations + * and call @func(@arg) on it. This function can use task_is_runnable() and + * task_curr() to work out what the state is, if required. Given that @func + * can be invoked with a runqueue lock held, it had better be quite + * lightweight. * - * Put @p on the run-queue if it's not already there. The caller must - * ensure that this_rq() is locked, @p is bound to this_rq() and not - * the current task. + * Returns: + * Whatever @func returns */ -static void try_to_wake_up_local(struct task_struct *p) +int task_call_func(struct task_struct *p, task_call_f func, void *arg) { - struct rq *rq = task_rq(p); + struct rq *rq = NULL; + struct rq_flags rf; + int ret; - if (WARN_ON_ONCE(rq != this_rq()) || - WARN_ON_ONCE(p == current)) - return; + raw_spin_lock_irqsave(&p->pi_lock, rf.flags); - lockdep_assert_held(&rq->lock); + if (__task_needs_rq_lock(p)) + rq = __task_rq_lock(p, &rf); - if (!raw_spin_trylock(&p->pi_lock)) { - raw_spin_unlock(&rq->lock); - raw_spin_lock(&p->pi_lock); - raw_spin_lock(&rq->lock); - } + /* + * 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 (!(p->state & TASK_NORMAL)) - goto out; + if (rq) + __task_rq_unlock(rq, p, &rf); - if (!p->on_rq) - ttwu_activate(rq, p, ENQUEUE_WAKEUP); + raw_spin_unlock_irqrestore(&p->pi_lock, rf.flags); + return ret; +} - ttwu_do_wakeup(rq, p, 0); - ttwu_stat(p, smp_processor_id(), 0); -out: - raw_spin_unlock(&p->pi_lock); +/** + * cpu_curr_snapshot - Return a snapshot of the currently running task + * @cpu: The CPU on which to snapshot the task. + * + * Returns the task_struct pointer of the task "currently" running on + * the specified CPU. + * + * If the specified CPU was offline, the return value is whatever it + * is, perhaps a pointer to the task_struct structure of that CPU's idle + * task, but there is no guarantee. Callers wishing a useful return + * value must take some action to ensure that the specified CPU remains + * online throughout. + * + * This function executes full memory barriers before and after fetching + * the pointer, which permits the caller to confine this function's fetch + * with respect to the caller's accesses to other shared variables. + */ +struct task_struct *cpu_curr_snapshot(int cpu) +{ + struct rq *rq = cpu_rq(cpu); + struct task_struct *t; + struct rq_flags rf; + + rq_lock_irqsave(rq, &rf); + smp_mb__after_spinlock(); /* Pairing determined by caller's synchronization design. */ + t = rcu_dereference(cpu_curr(cpu)); + rq_unlock_irqrestore(rq, &rf); + smp_mb(); /* Pairing determined by caller's synchronization design. */ + + return t; } /** @@ -1577,15 +4338,14 @@ out: * @p: The process to be woken up. * * Attempt to wake up the nominated process and move it to the set of runnable - * processes. Returns 1 if the process was woken up, 0 if it was already - * running. + * processes. + * + * Return: 1 if the process was woken up, 0 if it was already running. * - * It may be assumed that this function implies a write memory barrier before - * changing the task state if and only if any tasks are woken up. + * This function executes a full memory barrier before accessing the task state. */ int wake_up_process(struct task_struct *p) { - WARN_ON(task_is_stopped_or_traced(p)); return try_to_wake_up(p, TASK_NORMAL, 0); } EXPORT_SYMBOL(wake_up_process); @@ -1599,9 +4359,10 @@ int wake_up_state(struct task_struct *p, unsigned int state) * Perform scheduler related setup for a newly forked process p. * p is forked by current. * - * __sched_fork() is basic setup used by init_idle() too: + * __sched_fork() is basic setup which is also used by sched_init() to + * initialize the boot CPU's idle task. */ -static void __sched_fork(struct task_struct *p) +static void __sched_fork(u64 clone_flags, struct task_struct *p) { p->on_rq = 0; @@ -1611,86 +4372,265 @@ static void __sched_fork(struct task_struct *p) p->se.prev_sum_exec_runtime = 0; p->se.nr_migrations = 0; p->se.vruntime = 0; + p->se.vlag = 0; INIT_LIST_HEAD(&p->se.group_node); + /* A delayed task cannot be in clone(). */ + 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 - memset(&p->se.statistics, 0, sizeof(p->se.statistics)); + /* Even if schedstat is disabled, there should not be garbage */ + memset(&p->stats, 0, sizeof(p->stats)); #endif + init_dl_entity(&p->dl); + INIT_LIST_HEAD(&p->rt.run_list); + p->rt.timeout = 0; + p->rt.time_slice = sched_rr_timeslice; + p->rt.on_rq = 0; + p->rt.on_list = 0; + +#ifdef CONFIG_SCHED_CLASS_EXT + init_scx_entity(&p->scx); +#endif #ifdef CONFIG_PREEMPT_NOTIFIERS INIT_HLIST_HEAD(&p->preempt_notifiers); #endif +#ifdef CONFIG_COMPACTION + p->capture_control = NULL; +#endif + init_numa_balancing(clone_flags, p); + p->wake_entry.u_flags = CSD_TYPE_TTWU; + p->migration_pending = NULL; +} + +DEFINE_STATIC_KEY_FALSE(sched_numa_balancing); + #ifdef CONFIG_NUMA_BALANCING - if (p->mm && atomic_read(&p->mm->mm_users) == 1) { - p->mm->numa_next_scan = jiffies; - p->mm->numa_next_reset = jiffies; - p->mm->numa_scan_seq = 0; - } - p->node_stamp = 0ULL; - p->numa_scan_seq = p->mm ? p->mm->numa_scan_seq : 0; - p->numa_migrate_seq = p->mm ? p->mm->numa_scan_seq - 1 : 0; - p->numa_scan_period = sysctl_numa_balancing_scan_delay; - p->numa_work.next = &p->numa_work; -#endif /* CONFIG_NUMA_BALANCING */ +int sysctl_numa_balancing_mode; + +static void __set_numabalancing_state(bool enabled) +{ + if (enabled) + static_branch_enable(&sched_numa_balancing); + else + static_branch_disable(&sched_numa_balancing); } -#ifdef CONFIG_NUMA_BALANCING -#ifdef CONFIG_SCHED_DEBUG void set_numabalancing_state(bool enabled) { if (enabled) - sched_feat_set("NUMA"); + sysctl_numa_balancing_mode = NUMA_BALANCING_NORMAL; else - sched_feat_set("NO_NUMA"); + sysctl_numa_balancing_mode = NUMA_BALANCING_DISABLED; + __set_numabalancing_state(enabled); } -#else -__read_mostly bool numabalancing_enabled; -void set_numabalancing_state(bool enabled) +#ifdef CONFIG_PROC_SYSCTL +static void reset_memory_tiering(void) { - numabalancing_enabled = enabled; + struct pglist_data *pgdat; + + for_each_online_pgdat(pgdat) { + pgdat->nbp_threshold = 0; + pgdat->nbp_th_nr_cand = node_page_state(pgdat, PGPROMOTE_CANDIDATE); + pgdat->nbp_th_start = jiffies_to_msecs(jiffies); + } } -#endif /* CONFIG_SCHED_DEBUG */ + +static int sysctl_numa_balancing(const struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + struct ctl_table t; + int err; + int state = sysctl_numa_balancing_mode; + + if (write && !capable(CAP_SYS_ADMIN)) + return -EPERM; + + t = *table; + t.data = &state; + err = proc_dointvec_minmax(&t, write, buffer, lenp, ppos); + if (err < 0) + return err; + if (write) { + if (!(sysctl_numa_balancing_mode & NUMA_BALANCING_MEMORY_TIERING) && + (state & NUMA_BALANCING_MEMORY_TIERING)) + reset_memory_tiering(); + sysctl_numa_balancing_mode = state; + __set_numabalancing_state(state); + } + return err; +} +#endif /* CONFIG_PROC_SYSCTL */ #endif /* CONFIG_NUMA_BALANCING */ +#ifdef CONFIG_SCHEDSTATS + +DEFINE_STATIC_KEY_FALSE(sched_schedstats); + +static void set_schedstats(bool enabled) +{ + if (enabled) + static_branch_enable(&sched_schedstats); + else + static_branch_disable(&sched_schedstats); +} + +void force_schedstat_enabled(void) +{ + if (!schedstat_enabled()) { + pr_info("kernel profiling enabled schedstats, disable via kernel.sched_schedstats.\n"); + static_branch_enable(&sched_schedstats); + } +} + +static int __init setup_schedstats(char *str) +{ + int ret = 0; + if (!str) + goto out; + + if (!strcmp(str, "enable")) { + set_schedstats(true); + ret = 1; + } else if (!strcmp(str, "disable")) { + set_schedstats(false); + ret = 1; + } +out: + if (!ret) + pr_warn("Unable to parse schedstats=\n"); + + return ret; +} +__setup("schedstats=", setup_schedstats); + +#ifdef CONFIG_PROC_SYSCTL +static int sysctl_schedstats(const struct ctl_table *table, int write, void *buffer, + size_t *lenp, loff_t *ppos) +{ + struct ctl_table t; + int err; + int state = static_branch_likely(&sched_schedstats); + + if (write && !capable(CAP_SYS_ADMIN)) + return -EPERM; + + t = *table; + t.data = &state; + err = proc_dointvec_minmax(&t, write, buffer, lenp, ppos); + if (err < 0) + return err; + if (write) + set_schedstats(state); + return err; +} +#endif /* CONFIG_PROC_SYSCTL */ +#endif /* CONFIG_SCHEDSTATS */ + +#ifdef CONFIG_SYSCTL +static const struct ctl_table sched_core_sysctls[] = { +#ifdef CONFIG_SCHEDSTATS + { + .procname = "sched_schedstats", + .data = NULL, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = sysctl_schedstats, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, +#endif /* CONFIG_SCHEDSTATS */ +#ifdef CONFIG_UCLAMP_TASK + { + .procname = "sched_util_clamp_min", + .data = &sysctl_sched_uclamp_util_min, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = sysctl_sched_uclamp_handler, + }, + { + .procname = "sched_util_clamp_max", + .data = &sysctl_sched_uclamp_util_max, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = sysctl_sched_uclamp_handler, + }, + { + .procname = "sched_util_clamp_min_rt_default", + .data = &sysctl_sched_uclamp_util_min_rt_default, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = sysctl_sched_uclamp_handler, + }, +#endif /* CONFIG_UCLAMP_TASK */ +#ifdef CONFIG_NUMA_BALANCING + { + .procname = "numa_balancing", + .data = NULL, /* filled in by handler */ + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = sysctl_numa_balancing, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_FOUR, + }, +#endif /* CONFIG_NUMA_BALANCING */ +}; +static int __init sched_core_sysctl_init(void) +{ + register_sysctl_init("kernel", sched_core_sysctls); + return 0; +} +late_initcall(sched_core_sysctl_init); +#endif /* CONFIG_SYSCTL */ + /* * fork()/clone()-time setup: */ -void sched_fork(struct task_struct *p) +int sched_fork(u64 clone_flags, struct task_struct *p) { - unsigned long flags; - int cpu = get_cpu(); - - __sched_fork(p); + __sched_fork(clone_flags, p); /* - * We mark the process as running here. This guarantees that + * We mark the process as NEW here. This guarantees that * nobody will actually run it, and a signal or other external * event cannot wake it up and insert it on the runqueue either. */ - p->state = TASK_RUNNING; + p->__state = TASK_NEW; /* * Make sure we do not leak PI boosting priority to the child. */ p->prio = current->normal_prio; + uclamp_fork(p); + /* * Revert to default priority/policy on fork if requested. */ if (unlikely(p->sched_reset_on_fork)) { - if (task_has_rt_policy(p)) { + if (task_has_dl_policy(p) || task_has_rt_policy(p)) { p->policy = SCHED_NORMAL; p->static_prio = NICE_TO_PRIO(0); p->rt_priority = 0; } else if (PRIO_TO_NICE(p->static_prio) < 0) p->static_prio = NICE_TO_PRIO(0); - p->prio = p->normal_prio = __normal_prio(p); - set_load_weight(p); + p->prio = p->normal_prio = p->static_prio; + set_load_weight(p, false); + p->se.custom_slice = 0; + p->se.slice = sysctl_sched_base_slice; /* * We don't need the reset flag anymore after the fork. It has @@ -1699,39 +4639,91 @@ void sched_fork(struct task_struct *p) p->sched_reset_on_fork = 0; } - if (!rt_prio(p->prio)) + if (dl_prio(p->prio)) + return -EAGAIN; + + scx_pre_fork(p); + + if (rt_prio(p->prio)) { + p->sched_class = &rt_sched_class; +#ifdef CONFIG_SCHED_CLASS_EXT + } else if (task_should_scx(p->policy)) { + p->sched_class = &ext_sched_class; +#endif + } else { p->sched_class = &fair_sched_class; + } - if (p->sched_class->task_fork) - p->sched_class->task_fork(p); + init_entity_runnable_average(&p->se); - /* - * The child is not yet in the pid-hash so no cgroup attach races, - * and the cgroup is pinned to this child due to cgroup_fork() - * is ran before sched_fork(). - * - * Silence PROVE_RCU. - */ - raw_spin_lock_irqsave(&p->pi_lock, flags); - set_task_cpu(p, cpu); - raw_spin_unlock_irqrestore(&p->pi_lock, flags); -#if defined(CONFIG_SCHEDSTATS) || defined(CONFIG_TASK_DELAY_ACCT) +#ifdef CONFIG_SCHED_INFO if (likely(sched_info_on())) memset(&p->sched_info, 0, sizeof(p->sched_info)); #endif -#if defined(CONFIG_SMP) p->on_cpu = 0; -#endif -#ifdef CONFIG_PREEMPT_COUNT - /* Want to start with kernel preemption disabled. */ - task_thread_info(p)->preempt_count = 1; -#endif -#ifdef CONFIG_SMP + init_task_preempt_count(p); plist_node_init(&p->pushable_tasks, MAX_PRIO); + RB_CLEAR_NODE(&p->pushable_dl_tasks); + + return 0; +} + +int sched_cgroup_fork(struct task_struct *p, struct kernel_clone_args *kargs) +{ + unsigned long flags; + + /* + * Because we're not yet on the pid-hash, p->pi_lock isn't strictly + * required yet, but lockdep gets upset if rules are violated. + */ + raw_spin_lock_irqsave(&p->pi_lock, flags); +#ifdef CONFIG_CGROUP_SCHED + if (1) { + struct task_group *tg; + tg = container_of(kargs->cset->subsys[cpu_cgrp_id], + struct task_group, css); + tg = autogroup_task_group(p, tg); + p->sched_task_group = tg; + } #endif + /* + * We're setting the CPU for the first time, we don't migrate, + * so use __set_task_cpu(). + */ + __set_task_cpu(p, smp_processor_id()); + if (p->sched_class->task_fork) + p->sched_class->task_fork(p); + raw_spin_unlock_irqrestore(&p->pi_lock, flags); + + return scx_fork(p); +} + +void sched_cancel_fork(struct task_struct *p) +{ + scx_cancel_fork(p); +} - put_cpu(); +void sched_post_fork(struct task_struct *p) +{ + uclamp_post_fork(p); + scx_post_fork(p); +} + +unsigned long to_ratio(u64 period, u64 runtime) +{ + if (runtime == RUNTIME_INF) + return BW_UNIT; + + /* + * Doing this here saves a lot of checks in all + * the calling paths, and returning zero seems + * safe for them anyway. + */ + if (period == 0) + return 0; + + return div64_u64(runtime << BW_SHIFT, period); } /* @@ -1743,41 +4735,66 @@ void sched_fork(struct task_struct *p) */ void wake_up_new_task(struct task_struct *p) { - unsigned long flags; + struct rq_flags rf; struct rq *rq; + int wake_flags = WF_FORK; - raw_spin_lock_irqsave(&p->pi_lock, flags); -#ifdef CONFIG_SMP + raw_spin_lock_irqsave(&p->pi_lock, rf.flags); + WRITE_ONCE(p->__state, TASK_RUNNING); /* * Fork balancing, do it here and not earlier because: - * - cpus_allowed can change in the fork path - * - any previously selected cpu might disappear through hotplug + * - cpus_ptr can change in the fork path + * - any previously selected CPU might disappear through hotplug + * + * Use __set_task_cpu() to avoid calling sched_class::migrate_task_rq, + * as we're not fully set-up yet. */ - set_task_cpu(p, select_task_rq(p, SD_BALANCE_FORK, 0)); -#endif + p->recent_used_cpu = task_cpu(p); + __set_task_cpu(p, select_task_rq(p, task_cpu(p), &wake_flags)); + rq = __task_rq_lock(p, &rf); + update_rq_clock(rq); + post_init_entity_util_avg(p); - /* Initialize new task's runnable average */ - init_task_runnable_average(p); - rq = __task_rq_lock(p); - activate_task(rq, p, 0); - p->on_rq = 1; - trace_sched_wakeup_new(p, true); - check_preempt_curr(rq, p, WF_FORK); -#ifdef CONFIG_SMP - if (p->sched_class->task_woken) + activate_task(rq, p, ENQUEUE_NOCLOCK | ENQUEUE_INITIAL); + trace_sched_wakeup_new(p); + wakeup_preempt(rq, p, wake_flags); + if (p->sched_class->task_woken) { + /* + * Nothing relies on rq->lock after this, so it's fine to + * drop it. + */ + rq_unpin_lock(rq, &rf); p->sched_class->task_woken(rq, p); -#endif - task_rq_unlock(rq, p, &flags); + rq_repin_lock(rq, &rf); + } + task_rq_unlock(rq, p, &rf); } #ifdef CONFIG_PREEMPT_NOTIFIERS +static DEFINE_STATIC_KEY_FALSE(preempt_notifier_key); + +void preempt_notifier_inc(void) +{ + static_branch_inc(&preempt_notifier_key); +} +EXPORT_SYMBOL_GPL(preempt_notifier_inc); + +void preempt_notifier_dec(void) +{ + static_branch_dec(&preempt_notifier_key); +} +EXPORT_SYMBOL_GPL(preempt_notifier_dec); + /** * preempt_notifier_register - tell me when current is being preempted & rescheduled * @notifier: notifier struct to register */ void preempt_notifier_register(struct preempt_notifier *notifier) { + if (!static_branch_unlikely(&preempt_notifier_key)) + WARN(1, "registering preempt_notifier while notifiers disabled\n"); + hlist_add_head(¬ifier->link, ¤t->preempt_notifiers); } EXPORT_SYMBOL_GPL(preempt_notifier_register); @@ -1786,7 +4803,7 @@ EXPORT_SYMBOL_GPL(preempt_notifier_register); * preempt_notifier_unregister - no longer interested in preemption notifications * @notifier: notifier struct to unregister * - * This is safe to call from within a preemption notifier. + * This is *not* safe to call from within a preemption notifier. */ void preempt_notifier_unregister(struct preempt_notifier *notifier) { @@ -1794,7 +4811,7 @@ void preempt_notifier_unregister(struct preempt_notifier *notifier) } EXPORT_SYMBOL_GPL(preempt_notifier_unregister); -static void fire_sched_in_preempt_notifiers(struct task_struct *curr) +static void __fire_sched_in_preempt_notifiers(struct task_struct *curr) { struct preempt_notifier *notifier; @@ -1802,9 +4819,15 @@ static void fire_sched_in_preempt_notifiers(struct task_struct *curr) notifier->ops->sched_in(notifier, raw_smp_processor_id()); } +static __always_inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) +{ + if (static_branch_unlikely(&preempt_notifier_key)) + __fire_sched_in_preempt_notifiers(curr); +} + static void -fire_sched_out_preempt_notifiers(struct task_struct *curr, - struct task_struct *next) +__fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) { struct preempt_notifier *notifier; @@ -1812,19 +4835,193 @@ fire_sched_out_preempt_notifiers(struct task_struct *curr, notifier->ops->sched_out(notifier, next); } -#else /* !CONFIG_PREEMPT_NOTIFIERS */ +static __always_inline void +fire_sched_out_preempt_notifiers(struct task_struct *curr, + struct task_struct *next) +{ + if (static_branch_unlikely(&preempt_notifier_key)) + __fire_sched_out_preempt_notifiers(curr, next); +} + +#else /* !CONFIG_PREEMPT_NOTIFIERS: */ -static void fire_sched_in_preempt_notifiers(struct task_struct *curr) +static inline void fire_sched_in_preempt_notifiers(struct task_struct *curr) { } -static void +static inline void fire_sched_out_preempt_notifiers(struct task_struct *curr, struct task_struct *next) { } -#endif /* CONFIG_PREEMPT_NOTIFIERS */ +#endif /* !CONFIG_PREEMPT_NOTIFIERS */ + +static inline void prepare_task(struct task_struct *next) +{ + /* + * Claim the task as running, we do this before switching to it + * such that any running task will have this set. + * + * See the smp_load_acquire(&p->on_cpu) case in ttwu() and + * its ordering comment. + */ + WRITE_ONCE(next->on_cpu, 1); +} + +static inline void finish_task(struct task_struct *prev) +{ + /* + * 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 + * must ensure this doesn't happen until the switch is completely + * finished. + * + * In particular, the load of prev->state in finish_task_switch() must + * happen before this. + * + * Pairs with the smp_cond_load_acquire() in try_to_wake_up(). + */ + smp_store_release(&prev->on_cpu, 0); +} + +static void do_balance_callbacks(struct rq *rq, struct balance_callback *head) +{ + void (*func)(struct rq *rq); + struct balance_callback *next; + + lockdep_assert_rq_held(rq); + + while (head) { + func = (void (*)(struct rq *))head->func; + next = head->next; + head->next = NULL; + head = next; + + func(rq); + } +} + +static void balance_push(struct rq *rq); + +/* + * balance_push_callback is a right abuse of the callback interface and plays + * by significantly different rules. + * + * Where the normal balance_callback's purpose is to be ran in the same context + * that queued it (only later, when it's safe to drop rq->lock again), + * balance_push_callback is specifically targeted at __schedule(). + * + * This abuse is tolerated because it places all the unlikely/odd cases behind + * a single test, namely: rq->balance_callback == NULL. + */ +struct balance_callback balance_push_callback = { + .next = NULL, + .func = balance_push, +}; + +static inline struct balance_callback * +__splice_balance_callbacks(struct rq *rq, bool split) +{ + struct balance_callback *head = rq->balance_callback; + + if (likely(!head)) + return NULL; + + lockdep_assert_rq_held(rq); + /* + * Must not take balance_push_callback off the list when + * splice_balance_callbacks() and balance_callbacks() are not + * in the same rq->lock section. + * + * In that case it would be possible for __schedule() to interleave + * and observe the list empty. + */ + if (split && head == &balance_push_callback) + head = NULL; + else + rq->balance_callback = NULL; + + return head; +} + +struct balance_callback *splice_balance_callbacks(struct rq *rq) +{ + return __splice_balance_callbacks(rq, true); +} + +static void __balance_callbacks(struct rq *rq) +{ + do_balance_callbacks(rq, __splice_balance_callbacks(rq, false)); +} + +void balance_callbacks(struct rq *rq, struct balance_callback *head) +{ + unsigned long flags; + + if (unlikely(head)) { + raw_spin_rq_lock_irqsave(rq, flags); + do_balance_callbacks(rq, head); + raw_spin_rq_unlock_irqrestore(rq, flags); + } +} + +static inline void +prepare_lock_switch(struct rq *rq, struct task_struct *next, struct rq_flags *rf) +{ + /* + * Since the runqueue lock will be released by the next + * task (which is an invalid locking op but in the case + * of the scheduler it's an obvious special-case), so we + * do an early lockdep release here: + */ + rq_unpin_lock(rq, rf); + spin_release(&__rq_lockp(rq)->dep_map, _THIS_IP_); +#ifdef CONFIG_DEBUG_SPINLOCK + /* this is a valid case when another task releases the spinlock */ + rq_lockp(rq)->owner = next; +#endif +} + +static inline void finish_lock_switch(struct rq *rq) +{ + /* + * If we are tracking spinlock dependencies then we have to + * fix up the runqueue lock - which gets 'carried over' from + * prev into current: + */ + spin_acquire(&__rq_lockp(rq)->dep_map, 0, 0, _THIS_IP_); + __balance_callbacks(rq); + raw_spin_rq_unlock_irq(rq); +} + +/* + * NOP if the arch has not defined these: + */ + +#ifndef prepare_arch_switch +# define prepare_arch_switch(next) do { } while (0) +#endif + +#ifndef finish_arch_post_lock_switch +# define finish_arch_post_lock_switch() do { } while (0) +#endif + +static inline void kmap_local_sched_out(void) +{ +#ifdef CONFIG_KMAP_LOCAL + if (unlikely(current->kmap_ctrl.idx)) + __kmap_local_sched_out(); +#endif +} + +static inline void kmap_local_sched_in(void) +{ +#ifdef CONFIG_KMAP_LOCAL + if (unlikely(current->kmap_ctrl.idx)) + __kmap_local_sched_in(); +#endif +} /** * prepare_task_switch - prepare to switch tasks @@ -1843,17 +5040,17 @@ static inline void prepare_task_switch(struct rq *rq, struct task_struct *prev, struct task_struct *next) { - trace_sched_switch(prev, next); - sched_info_switch(prev, next); + kcov_prepare_switch(prev); + sched_info_switch(rq, prev, next); perf_event_task_sched_out(prev, next); fire_sched_out_preempt_notifiers(prev, next); - prepare_lock_switch(rq, next); + kmap_local_sched_out(); + prepare_task(next); prepare_arch_switch(next); } /** * finish_task_switch - clean up after a task-switch - * @rq: runqueue associated with task-switch * @prev: the thread we just switched away from. * * finish_task_switch must be called after the context switch, paired @@ -1865,12 +5062,34 @@ prepare_task_switch(struct rq *rq, struct task_struct *prev, * so, we finish that here outside of the runqueue lock. (Doing it * with the lock held can cause deadlocks; see schedule() for * details.) + * + * The context switch have flipped the stack from under us and restored the + * local variables which were saved when this task called schedule() in the + * past. 'prev == current' is still correct but we need to recalculate this_rq + * because prev may have moved to another CPU. */ -static void finish_task_switch(struct rq *rq, struct task_struct *prev) +static struct rq *finish_task_switch(struct task_struct *prev) __releases(rq->lock) { + struct rq *rq = this_rq(); struct mm_struct *mm = rq->prev_mm; - long prev_state; + unsigned int prev_state; + + /* + * The previous task will have left us with a preempt_count of 2 + * because it left us after: + * + * schedule() + * preempt_disable(); // 1 + * __schedule() + * raw_spin_lock_irq(&rq->lock) // 2 + * + * Also, see FORK_PREEMPT_COUNT. + */ + if (WARN_ONCE(preempt_count() != 2*PREEMPT_DISABLE_OFFSET, + "corrupted preempt_count: %s/%d/0x%x\n", + current->comm, current->pid, preempt_count())) + preempt_count_set(FORK_PREEMPT_COUNT); rq->prev_mm = NULL; @@ -1879,109 +5098,108 @@ static void finish_task_switch(struct rq *rq, struct task_struct *prev) * If a task dies, then it sets TASK_DEAD in tsk->state and calls * schedule one last time. The schedule call will never return, and * the scheduled task must drop that reference. - * The test for TASK_DEAD must occur while the runqueue locks are - * still held, otherwise prev could be scheduled on another cpu, die - * there before we look at prev->state, and then the reference would - * be dropped twice. - * Manfred Spraul <manfred@colorfullife.com> + * + * We must observe prev->state before clearing prev->on_cpu (in + * finish_task), otherwise a concurrent wakeup can get prev + * running on another CPU and we could rave with its RUNNING -> DEAD + * transition, resulting in a double drop. */ - prev_state = prev->state; + prev_state = READ_ONCE(prev->__state); vtime_task_switch(prev); - finish_arch_switch(prev); perf_event_task_sched_in(prev, current); - finish_lock_switch(rq, prev); + finish_task(prev); + tick_nohz_task_switch(); + finish_lock_switch(rq); finish_arch_post_lock_switch(); + kcov_finish_switch(current); + /* + * kmap_local_sched_out() is invoked with rq::lock held and + * interrupts disabled. There is no requirement for that, but the + * sched out code does not have an interrupt enabled section. + * Restoring the maps on sched in does not require interrupts being + * disabled either. + */ + kmap_local_sched_in(); fire_sched_in_preempt_notifiers(current); - if (mm) - mmdrop(mm); - if (unlikely(prev_state == TASK_DEAD)) { - /* - * Remove function-return probe instances associated with this - * task and put them back on the free list. - */ - kprobe_flush_task(prev); - put_task_struct(prev); + /* + * When switching through a kernel thread, the loop in + * membarrier_{private,global}_expedited() may have observed that + * kernel thread and not issued an IPI. It is therefore possible to + * schedule between user->kernel->user threads without passing though + * switch_mm(). Membarrier requires a barrier after storing to + * rq->curr, before returning to userspace, so provide them here: + * + * - a full memory barrier for {PRIVATE,GLOBAL}_EXPEDITED, implicitly + * provided by mmdrop_lazy_tlb(), + * - a sync_core for SYNC_CORE. + */ + if (mm) { + membarrier_mm_sync_core_before_usermode(mm); + mmdrop_lazy_tlb_sched(mm); } - tick_nohz_task_switch(current); -} - -#ifdef CONFIG_SMP - -/* assumes rq->lock is held */ -static inline void pre_schedule(struct rq *rq, struct task_struct *prev) -{ - if (prev->sched_class->pre_schedule) - prev->sched_class->pre_schedule(rq, prev); -} + if (unlikely(prev_state == TASK_DEAD)) { + if (prev->sched_class->task_dead) + prev->sched_class->task_dead(prev); -/* rq->lock is NOT held, but preemption is disabled */ -static inline void post_schedule(struct rq *rq) -{ - if (rq->post_schedule) { - unsigned long flags; + /* + * 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); - raw_spin_lock_irqsave(&rq->lock, flags); - if (rq->curr->sched_class->post_schedule) - rq->curr->sched_class->post_schedule(rq); - raw_spin_unlock_irqrestore(&rq->lock, flags); + /* Task is done with its stack. */ + put_task_stack(prev); - rq->post_schedule = 0; + put_task_struct_rcu_user(prev); } -} - -#else -static inline void pre_schedule(struct rq *rq, struct task_struct *p) -{ -} - -static inline void post_schedule(struct rq *rq) -{ + return rq; } -#endif - /** * schedule_tail - first thing a freshly forked thread must call. * @prev: the thread we just switched away from. */ -asmlinkage void schedule_tail(struct task_struct *prev) +asmlinkage __visible void schedule_tail(struct task_struct *prev) __releases(rq->lock) { - struct rq *rq = this_rq(); - - finish_task_switch(rq, prev); - /* - * FIXME: do we need to worry about rq being invalidated by the - * task_switch? + * New tasks start with FORK_PREEMPT_COUNT, see there and + * finish_task_switch() for details. + * + * finish_task_switch() will drop rq->lock() and lower preempt_count + * and the preempt_enable() will end up enabling preemption (on + * PREEMPT_COUNT kernels). */ - post_schedule(rq); -#ifdef __ARCH_WANT_UNLOCKED_CTXSW - /* In this case, finish_task_switch does not reenable preemption */ + 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(); -#endif + if (current->set_child_tid) put_user(task_pid_vnr(current), current->set_child_tid); + + calculate_sigpending(); } /* - * context_switch - switch to the new MM and the new - * thread's register state. + * context_switch - switch to the new MM and the new thread's register state. */ -static inline void +static __always_inline struct rq * context_switch(struct rq *rq, struct task_struct *prev, - struct task_struct *next) + struct task_struct *next, struct rq_flags *rf) { - struct mm_struct *mm, *oldmm; - prepare_task_switch(rq, prev, next); - mm = next->mm; - oldmm = prev->active_mm; /* * For paravirt, this is coupled with an exit in switch_to to * combine the page table reload and the switch backend into @@ -1989,38 +5207,56 @@ context_switch(struct rq *rq, struct task_struct *prev, */ arch_start_context_switch(prev); - if (!mm) { - next->active_mm = oldmm; - atomic_inc(&oldmm->mm_count); - enter_lazy_tlb(oldmm, next); - } else - switch_mm(oldmm, mm, next); + /* + * kernel -> kernel lazy + transfer active + * user -> kernel lazy + mmgrab_lazy_tlb() active + * + * kernel -> user switch + mmdrop_lazy_tlb() active + * user -> user switch + */ + if (!next->mm) { // to kernel + enter_lazy_tlb(prev->active_mm, next); + + next->active_mm = prev->active_mm; + if (prev->mm) // from user + mmgrab_lazy_tlb(prev->active_mm); + else + prev->active_mm = NULL; + } else { // to user + membarrier_switch_mm(rq, prev->active_mm, next->mm); + /* + * sys_membarrier() requires an smp_mb() between setting + * rq->curr / membarrier_switch_mm() and returning to userspace. + * + * The below provides this either through switch_mm(), or in + * case 'prev->active_mm == next->mm' through + * finish_task_switch()'s mmdrop(). + */ + switch_mm_irqs_off(prev->active_mm, next->mm, next); + lru_gen_use_mm(next->mm); - if (!prev->mm) { - prev->active_mm = NULL; - rq->prev_mm = oldmm; + if (!prev->mm) { // from kernel + /* will mmdrop_lazy_tlb() in finish_task_switch(). */ + rq->prev_mm = prev->active_mm; + prev->active_mm = NULL; + } } + + mm_cid_switch_to(prev, next); + /* - * Since the runqueue lock will be released by the next - * task (which is an invalid locking op but in the case - * of the scheduler it's an obvious special-case), so we - * do an early lockdep release here: + * Tell rseq that the task was scheduled in. Must be after + * switch_mm_cid() to get the TIF flag set. */ -#ifndef __ARCH_WANT_UNLOCKED_CTXSW - spin_release(&rq->lock.dep_map, 1, _THIS_IP_); -#endif + rseq_sched_switch_event(next); + + prepare_lock_switch(rq, next, rf); - context_tracking_task_switch(prev, next); /* Here we just switch the register state and the stack. */ switch_to(prev, next, prev); - barrier(); - /* - * this_rq must be evaluated again because prev may have moved - * CPUs since it called schedule(), thus the 'rq' on its stack - * frame will be invalid. - */ - finish_task_switch(this_rq(), prev); + + return finish_task_switch(prev); } /* @@ -2029,9 +5265,9 @@ context_switch(struct rq *rq, struct task_struct *prev, * externally visible scheduler statistics: current number of runnable * threads, total number of context switches performed since bootup. */ -unsigned long nr_running(void) +unsigned int nr_running(void) { - unsigned long i, sum = 0; + unsigned int i, sum = 0; for_each_online_cpu(i) sum += cpu_rq(i)->nr_running; @@ -2039,6 +5275,30 @@ unsigned long nr_running(void) return sum; } +/* + * Check if only the current task is running on the CPU. + * + * Caution: this function does not check that the caller has disabled + * preemption, thus the result might have a time-of-check-to-time-of-use + * race. The caller is responsible to use it correctly, for example: + * + * - from a non-preemptible section (of course) + * + * - from a thread that is bound to a single CPU + * + * - in a loop with very short iterations (e.g. a polling loop) + */ +bool single_task_running(void) +{ + return raw_rq()->nr_running == 1; +} +EXPORT_SYMBOL(single_task_running); + +unsigned long long nr_context_switches_cpu(int cpu) +{ + return cpu_rq(cpu)->nr_switches; +} + unsigned long long nr_context_switches(void) { int i; @@ -2050,24 +5310,58 @@ unsigned long long nr_context_switches(void) return sum; } -unsigned long nr_iowait(void) +/* + * Consumers of these two interfaces, like for example the cpuidle menu + * governor, are using nonsensical data. Preferring shallow idle state selection + * for a CPU that has IO-wait which might not even end up running the task when + * it does become runnable. + */ + +unsigned int nr_iowait_cpu(int cpu) { - unsigned long i, sum = 0; + return atomic_read(&cpu_rq(cpu)->nr_iowait); +} + +/* + * IO-wait accounting, and how it's mostly bollocks (on SMP). + * + * The idea behind IO-wait account is to account the idle time that we could + * have spend running if it were not for IO. That is, if we were to improve the + * storage performance, we'd have a proportional reduction in IO-wait time. + * + * This all works nicely on UP, where, when a task blocks on IO, we account + * idle time as IO-wait, because if the storage were faster, it could've been + * running and we'd not be idle. + * + * This has been extended to SMP, by doing the same for each CPU. This however + * is broken. + * + * Imagine for instance the case where two tasks block on one CPU, only the one + * CPU will have IO-wait accounted, while the other has regular idle. Even + * though, if the storage were faster, both could've ran at the same time, + * utilising both CPUs. + * + * This means, that when looking globally, the current IO-wait accounting on + * SMP is a lower bound, by reason of under accounting. + * + * Worse, since the numbers are provided per CPU, they are sometimes + * interpreted per CPU, and that is nonsensical. A blocked task isn't strictly + * associated with any one particular CPU, it can wake to another CPU than it + * blocked on. This means the per CPU IO-wait number is meaningless. + * + * Task CPU affinities can make all that even more 'interesting'. + */ + +unsigned int nr_iowait(void) +{ + unsigned int i, sum = 0; for_each_possible_cpu(i) - sum += atomic_read(&cpu_rq(i)->nr_iowait); + sum += nr_iowait_cpu(i); return sum; } -unsigned long nr_iowait_cpu(int cpu) -{ - struct rq *this = cpu_rq(cpu); - return atomic_read(&this->nr_iowait); -} - -#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. @@ -2075,27 +5369,22 @@ unsigned long nr_iowait_cpu(int cpu) void sched_exec(void) { struct task_struct *p = current; - unsigned long flags; + struct migration_arg arg; int dest_cpu; - raw_spin_lock_irqsave(&p->pi_lock, flags); - dest_cpu = p->sched_class->select_task_rq(p, SD_BALANCE_EXEC, 0); - if (dest_cpu == smp_processor_id()) - goto unlock; + scoped_guard (raw_spinlock_irqsave, &p->pi_lock) { + dest_cpu = p->sched_class->select_task_rq(p, task_cpu(p), WF_EXEC); + if (dest_cpu == smp_processor_id()) + return; - if (likely(cpu_active(dest_cpu))) { - struct migration_arg arg = { p, dest_cpu }; + if (unlikely(!cpu_active(dest_cpu))) + return; - raw_spin_unlock_irqrestore(&p->pi_lock, flags); - stop_one_cpu(task_cpu(p), migration_cpu_stop, &arg); - return; + arg = (struct migration_arg){ p, dest_cpu }; } -unlock: - raw_spin_unlock_irqrestore(&p->pi_lock, flags); + 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); @@ -2103,123 +5392,325 @@ EXPORT_PER_CPU_SYMBOL(kstat); EXPORT_PER_CPU_SYMBOL(kernel_cpustat); /* - * Return any ns on the sched_clock that have not yet been accounted in - * @p in case that task is currently running. - * - * Called with task_rq_lock() held on @rq. + * The function fair_sched_class.update_curr accesses the struct curr + * and its field curr->exec_start; when called from task_sched_runtime(), + * we observe a high rate of cache misses in practice. + * Prefetching this data results in improved performance. */ -static u64 do_task_delta_exec(struct task_struct *p, struct rq *rq) +static inline void prefetch_curr_exec_start(struct task_struct *p) { - u64 ns = 0; +#ifdef CONFIG_FAIR_GROUP_SCHED + struct sched_entity *curr = p->se.cfs_rq->curr; +#else + struct sched_entity *curr = task_rq(p)->cfs.curr; +#endif + prefetch(curr); + prefetch(&curr->exec_start); +} - if (task_current(rq, p)) { +/* + * Return accounted runtime for the task. + * In case the task is currently running, return the runtime plus current's + * pending runtime that have not been accounted yet. + */ +unsigned long long task_sched_runtime(struct task_struct *p) +{ + struct rq_flags rf; + struct rq *rq; + u64 ns; + +#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. + * Reading ->on_cpu is racy, but this is OK. + * + * If we race with it leaving CPU, we'll take a lock. So we're correct. + * If we race with it entering CPU, unaccounted time is 0. This is + * indistinguishable from the read occurring a few cycles earlier. + * If we see ->on_cpu without ->on_rq, the task is leaving, and has + * been accounted, so we're correct here as well. + */ + if (!p->on_cpu || !task_on_rq_queued(p)) + return p->se.sum_exec_runtime; +#endif + + rq = task_rq_lock(p, &rf); + /* + * Must be ->curr _and_ ->on_rq. If dequeued, we would + * project cycles that may never be accounted to this + * thread, breaking clock_gettime(). + */ + if (task_current_donor(rq, p) && task_on_rq_queued(p)) { + prefetch_curr_exec_start(p); update_rq_clock(rq); - ns = rq_clock_task(rq) - p->se.exec_start; - if ((s64)ns < 0) - ns = 0; + p->sched_class->update_curr(rq); } + ns = p->se.sum_exec_runtime; + task_rq_unlock(rq, p, &rf); return ns; } -unsigned long long task_delta_exec(struct task_struct *p) +static u64 cpu_resched_latency(struct rq *rq) { - unsigned long flags; - struct rq *rq; - u64 ns = 0; + int latency_warn_ms = READ_ONCE(sysctl_resched_latency_warn_ms); + u64 resched_latency, now = rq_clock(rq); + static bool warned_once; - rq = task_rq_lock(p, &flags); - ns = do_task_delta_exec(p, rq); - task_rq_unlock(rq, p, &flags); + if (sysctl_resched_latency_warn_once && warned_once) + return 0; - return ns; + if (!need_resched() || !latency_warn_ms) + return 0; + + if (system_state == SYSTEM_BOOTING) + return 0; + + if (!rq->last_seen_need_resched_ns) { + rq->last_seen_need_resched_ns = now; + rq->ticks_without_resched = 0; + return 0; + } + + rq->ticks_without_resched++; + resched_latency = now - rq->last_seen_need_resched_ns; + if (resched_latency <= latency_warn_ms * NSEC_PER_MSEC) + return 0; + + warned_once = true; + + return resched_latency; } -/* - * Return accounted runtime for the task. - * In case the task is currently running, return the runtime plus current's - * pending runtime that have not been accounted yet. - */ -unsigned long long task_sched_runtime(struct task_struct *p) +static int __init setup_resched_latency_warn_ms(char *str) { - unsigned long flags; - struct rq *rq; - u64 ns = 0; + long val; - rq = task_rq_lock(p, &flags); - ns = p->se.sum_exec_runtime + do_task_delta_exec(p, rq); - task_rq_unlock(rq, p, &flags); + if ((kstrtol(str, 0, &val))) { + pr_warn("Unable to set resched_latency_warn_ms\n"); + return 1; + } - return ns; + sysctl_resched_latency_warn_ms = val; + return 1; } +__setup("resched_latency_warn_ms=", setup_resched_latency_warn_ms); /* * This function gets called by the timer code, with HZ frequency. * We call it with interrupts disabled. */ -void scheduler_tick(void) +void sched_tick(void) { int cpu = smp_processor_id(); struct rq *rq = cpu_rq(cpu); - struct task_struct *curr = rq->curr; + /* accounting goes to the donor task */ + struct task_struct *donor; + struct rq_flags rf; + unsigned long hw_pressure; + u64 resched_latency; + + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) + arch_scale_freq_tick(); sched_clock_tick(); - raw_spin_lock(&rq->lock); + rq_lock(rq, &rf); + donor = rq->donor; + + psi_account_irqtime(rq, donor, NULL); + update_rq_clock(rq); - curr->sched_class->task_tick(rq, curr, 0); - update_cpu_load_active(rq); - raw_spin_unlock(&rq->lock); + hw_pressure = arch_scale_hw_pressure(cpu_of(rq)); + update_hw_load_avg(rq_clock_task(rq), rq, hw_pressure); + + if (dynamic_preempt_lazy() && tif_test_bit(TIF_NEED_RESCHED_LAZY)) + resched_curr(rq); + + donor->sched_class->task_tick(rq, donor, 0); + if (sched_feat(LATENCY_WARN)) + resched_latency = cpu_resched_latency(rq); + calc_global_load_tick(rq); + sched_core_tick(rq); + scx_tick(rq); + + rq_unlock(rq, &rf); + + if (sched_feat(LATENCY_WARN) && resched_latency) + resched_latency_warn(cpu, resched_latency); perf_event_task_tick(); -#ifdef CONFIG_SMP - rq->idle_balance = idle_cpu(cpu); - trigger_load_balance(rq, cpu); -#endif - rq_last_tick_reset(rq); + if (donor->flags & PF_WQ_WORKER) + wq_worker_tick(donor); + + if (!scx_switched_all()) { + rq->idle_balance = idle_cpu(cpu); + sched_balance_trigger(rq); + } } #ifdef CONFIG_NO_HZ_FULL -/** - * scheduler_tick_max_deferment + +struct tick_work { + int cpu; + atomic_t state; + struct delayed_work work; +}; +/* Values for ->state, see diagram below. */ +#define TICK_SCHED_REMOTE_OFFLINE 0 +#define TICK_SCHED_REMOTE_OFFLINING 1 +#define TICK_SCHED_REMOTE_RUNNING 2 + +/* + * State diagram for ->state: + * * - * Keep at least one tick per second when a single - * active task is running because the scheduler doesn't - * yet completely support full dynticks environment. + * TICK_SCHED_REMOTE_OFFLINE + * | ^ + * | | + * | | sched_tick_remote() + * | | + * | | + * +--TICK_SCHED_REMOTE_OFFLINING + * | ^ + * | | + * sched_tick_start() | | sched_tick_stop() + * | | + * V | + * TICK_SCHED_REMOTE_RUNNING * - * This makes sure that uptime, CFS vruntime, load - * balancing, etc... continue to move forward, even - * with a very low granularity. + * + * Other transitions get WARN_ON_ONCE(), except that sched_tick_remote() + * and sched_tick_start() are happy to leave the state in RUNNING. */ -u64 scheduler_tick_max_deferment(void) + +static struct tick_work __percpu *tick_work_cpu; + +static void sched_tick_remote(struct work_struct *work) { - struct rq *rq = this_rq(); - unsigned long next, now = ACCESS_ONCE(jiffies); + struct delayed_work *dwork = to_delayed_work(work); + struct tick_work *twork = container_of(dwork, struct tick_work, work); + int cpu = twork->cpu; + struct rq *rq = cpu_rq(cpu); + int os; + + /* + * Handle the tick only if it appears the remote CPU is running in full + * dynticks mode. The check is racy by nature, but missing a tick or + * having one too much is no big deal because the scheduler tick updates + * statistics and checks timeslices in a time-independent way, regardless + * of when exactly it is running. + */ + if (tick_nohz_tick_stopped_cpu(cpu)) { + guard(rq_lock_irq)(rq); + struct task_struct *curr = rq->curr; - next = rq->last_sched_tick + HZ; + if (cpu_online(cpu)) { + /* + * Since this is a remote tick for full dynticks mode, + * we are always sure that there is no proxy (only a + * single task is running). + */ + WARN_ON_ONCE(rq->curr != rq->donor); + update_rq_clock(rq); - if (time_before_eq(next, now)) - return 0; + if (!is_idle_task(curr)) { + /* + * Make sure the next tick runs within a + * reasonable amount of time. + */ + u64 delta = rq_clock_task(rq) - curr->se.exec_start; + WARN_ON_ONCE(delta > (u64)NSEC_PER_SEC * 30); + } + curr->sched_class->task_tick(rq, curr, 0); - return jiffies_to_usecs(next - now) * NSEC_PER_USEC; + calc_load_nohz_remote(rq); + } + } + + /* + * Run the remote tick once per second (1Hz). This arbitrary + * frequency is large enough to avoid overload but short enough + * to keep scheduler internal stats reasonably up to date. But + * first update state to reflect hotplug activity if required. + */ + 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); } -#endif -notrace unsigned long get_parent_ip(unsigned long addr) +static void sched_tick_start(int cpu) { - if (in_lock_functions(addr)) { - addr = CALLER_ADDR2; - if (in_lock_functions(addr)) - addr = CALLER_ADDR3; + int os; + struct tick_work *twork; + + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) + return; + + WARN_ON_ONCE(!tick_work_cpu); + + twork = per_cpu_ptr(tick_work_cpu, cpu); + os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_RUNNING); + WARN_ON_ONCE(os == TICK_SCHED_REMOTE_RUNNING); + 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); } - return addr; } -#if defined(CONFIG_PREEMPT) && (defined(CONFIG_DEBUG_PREEMPT) || \ - defined(CONFIG_PREEMPT_TRACER)) +#ifdef CONFIG_HOTPLUG_CPU +static void sched_tick_stop(int cpu) +{ + struct tick_work *twork; + int os; + + if (housekeeping_cpu(cpu, HK_TYPE_KERNEL_NOISE)) + return; + + WARN_ON_ONCE(!tick_work_cpu); + + twork = per_cpu_ptr(tick_work_cpu, cpu); + /* There cannot be competing actions, but don't rely on stop-machine. */ + os = atomic_xchg(&twork->state, TICK_SCHED_REMOTE_OFFLINING); + WARN_ON_ONCE(os != TICK_SCHED_REMOTE_RUNNING); + /* Don't cancel, as this would mess up the state machine. */ +} +#endif /* CONFIG_HOTPLUG_CPU */ + +int __init sched_tick_offload_init(void) +{ + tick_work_cpu = alloc_percpu(struct tick_work); + BUG_ON(!tick_work_cpu); + return 0; +} + +#else /* !CONFIG_NO_HZ_FULL: */ +static inline void sched_tick_start(int cpu) { } +static inline void sched_tick_stop(int cpu) { } +#endif /* !CONFIG_NO_HZ_FULL */ + +#if defined(CONFIG_PREEMPTION) && (defined(CONFIG_DEBUG_PREEMPT) || \ + defined(CONFIG_TRACE_PREEMPT_TOGGLE)) +/* + * If the value passed in is equal to the current preempt count + * then we just disabled preemption. Start timing the latency. + */ +static inline void preempt_latency_start(int val) +{ + if (preempt_count() == val) { + unsigned long ip = get_lock_parent_ip(); +#ifdef CONFIG_DEBUG_PREEMPT + current->preempt_disable_ip = ip; +#endif + trace_preempt_off(CALLER_ADDR0, ip); + } +} -void __kprobes add_preempt_count(int val) +void preempt_count_add(int val) { #ifdef CONFIG_DEBUG_PREEMPT /* @@ -2228,7 +5719,7 @@ void __kprobes add_preempt_count(int val) if (DEBUG_LOCKS_WARN_ON((preempt_count() < 0))) return; #endif - preempt_count() += val; + __preempt_count_add(val); #ifdef CONFIG_DEBUG_PREEMPT /* * Spinlock count overflowing soon? @@ -2236,12 +5727,22 @@ void __kprobes add_preempt_count(int val) DEBUG_LOCKS_WARN_ON((preempt_count() & PREEMPT_MASK) >= PREEMPT_MASK - 10); #endif + preempt_latency_start(val); +} +EXPORT_SYMBOL(preempt_count_add); +NOKPROBE_SYMBOL(preempt_count_add); + +/* + * If the value passed in equals to the current preempt count + * then we just enabled preemption. Stop timing the latency. + */ +static inline void preempt_latency_stop(int val) +{ if (preempt_count() == val) - trace_preempt_off(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1)); + trace_preempt_on(CALLER_ADDR0, get_lock_parent_ip()); } -EXPORT_SYMBOL(add_preempt_count); -void __kprobes sub_preempt_count(int val) +void preempt_count_sub(int val) { #ifdef CONFIG_DEBUG_PREEMPT /* @@ -2257,19 +5758,34 @@ void __kprobes sub_preempt_count(int val) return; #endif - if (preempt_count() == val) - trace_preempt_on(CALLER_ADDR0, get_parent_ip(CALLER_ADDR1)); - preempt_count() -= val; + preempt_latency_stop(val); + __preempt_count_sub(val); } -EXPORT_SYMBOL(sub_preempt_count); +EXPORT_SYMBOL(preempt_count_sub); +NOKPROBE_SYMBOL(preempt_count_sub); +#else +static inline void preempt_latency_start(int val) { } +static inline void preempt_latency_stop(int val) { } #endif +static inline unsigned long get_preempt_disable_ip(struct task_struct *p) +{ +#ifdef CONFIG_DEBUG_PREEMPT + return p->preempt_disable_ip; +#else + return 0; +#endif +} + /* * Print scheduling while atomic bug: */ static noinline void __schedule_bug(struct task_struct *prev) { + /* Save this before calling printk(), since that will clobber it */ + unsigned long preempt_disable_ip = get_preempt_disable_ip(current); + if (oops_in_progress) return; @@ -2280,6 +5796,12 @@ static noinline void __schedule_bug(struct task_struct *prev) print_modules(); if (irqs_disabled()) print_irqtrace_events(prev); + if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)) { + pr_err("Preemption disabled at:"); + print_ip_sym(KERN_ERR, preempt_disable_ip); + } + check_panic_on_warn("scheduling while atomic"); + dump_stack(); add_taint(TAINT_WARN, LOCKDEP_STILL_OK); } @@ -2287,1527 +5809,1644 @@ static noinline void __schedule_bug(struct task_struct *prev) /* * Various schedule()-time debugging checks and statistics: */ -static inline void schedule_debug(struct task_struct *prev) +static inline void schedule_debug(struct task_struct *prev, bool preempt) { - /* - * Test if we are atomic. Since do_exit() needs to call into - * schedule() atomically, we ignore that path for now. - * Otherwise, whine if we are scheduling when we should not be. - */ - if (unlikely(in_atomic_preempt_off() && !prev->exit_state)) +#ifdef CONFIG_SCHED_STACK_END_CHECK + if (task_stack_end_corrupted(prev)) + panic("corrupted stack end detected inside scheduler\n"); + + if (task_scs_end_corrupted(prev)) + panic("corrupted shadow stack detected inside scheduler\n"); +#endif + +#ifdef CONFIG_DEBUG_ATOMIC_SLEEP + if (!preempt && READ_ONCE(prev->__state) && prev->non_block_count) { + printk(KERN_ERR "BUG: scheduling in a non-blocking section: %s/%d/%i\n", + prev->comm, prev->pid, prev->non_block_count); + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); + } +#endif + + if (unlikely(in_atomic_preempt_off())) { __schedule_bug(prev); + preempt_count_set(PREEMPT_DISABLED); + } rcu_sleep_check(); + WARN_ON_ONCE(ct_state() == CT_STATE_USER); profile_hit(SCHED_PROFILING, __builtin_return_address(0)); - schedstat_inc(this_rq(), sched_count); + schedstat_inc(this_rq()->sched_count); } -static void put_prev_task(struct rq *rq, struct task_struct *prev) +static void prev_balance(struct rq *rq, struct task_struct *prev, + struct rq_flags *rf) { - if (prev->on_rq || rq->skip_clock_update < 0) - update_rq_clock(rq); - prev->sched_class->put_prev_task(rq, prev); + const struct sched_class *start_class = prev->sched_class; + const struct sched_class *class; + + /* + * We must do the balancing pass before put_prev_task(), such + * that when we release the rq->lock the task is in the same + * state as before we took rq->lock. + * + * We can terminate the balance pass as soon as we know there is + * a runnable task of @class priority or higher. + */ + for_active_class_range(class, start_class, &idle_sched_class) { + if (class->balance && class->balance(rq, prev, rf)) + break; + } } /* * Pick up the highest-prio task: */ static inline struct task_struct * -pick_next_task(struct rq *rq) +__pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) { const struct sched_class *class; struct task_struct *p; + rq->dl_server = NULL; + + if (scx_enabled()) + goto restart; + /* - * Optimization: we know that if all tasks are in - * the fair class we can call that function directly: + * Optimization: we know that if all tasks are in the fair class we can + * call that function directly, but only if the @prev task wasn't of a + * higher scheduling class, because otherwise those lose the + * opportunity to pull in more work from other CPUs. */ - if (likely(rq->nr_running == rq->cfs.h_nr_running)) { - p = fair_sched_class.pick_next_task(rq); - if (likely(p)) - return p; + if (likely(!sched_class_above(prev->sched_class, &fair_sched_class) && + rq->nr_running == rq->cfs.h_nr_queued)) { + + p = pick_next_task_fair(rq, prev, rf); + if (unlikely(p == RETRY_TASK)) + goto restart; + + /* Assume the next prioritized class is idle_sched_class */ + if (!p) { + p = pick_task_idle(rq, rf); + put_prev_set_next_task(rq, prev, p); + } + + return p; } - for_each_class(class) { - p = class->pick_next_task(rq); - if (p) - return p; +restart: + prev_balance(rq, prev, rf); + + for_each_active_class(class) { + if (class->pick_next_task) { + 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, rf); + if (unlikely(p == RETRY_TASK)) + goto restart; + if (p) { + put_prev_set_next_task(rq, prev, p); + return p; + } + } } - BUG(); /* the idle class will always have a runnable task */ + BUG(); /* The idle class should always have a runnable task. */ +} + +#ifdef CONFIG_SCHED_CORE +static inline bool is_task_rq_idle(struct task_struct *t) +{ + return (task_rq(t)->idle == t); +} + +static inline bool cookie_equals(struct task_struct *a, unsigned long cookie) +{ + return is_task_rq_idle(a) || (a->core_cookie == cookie); +} + +static inline bool cookie_match(struct task_struct *a, struct task_struct *b) +{ + if (is_task_rq_idle(a) || is_task_rq_idle(b)) + return true; + + return a->core_cookie == b->core_cookie; } /* - * __schedule() is the main scheduler function. - * - * The main means of driving the scheduler and thus entering this function are: - * - * 1. Explicit blocking: mutex, semaphore, waitqueue, etc. - * - * 2. TIF_NEED_RESCHED flag is checked on interrupt and userspace return - * paths. For example, see arch/x86/entry_64.S. - * - * To drive preemption between tasks, the scheduler sets the flag in timer - * interrupt handler scheduler_tick(). - * - * 3. Wakeups don't really cause entry into schedule(). They add a - * task to the run-queue and that's it. - * - * Now, if the new task added to the run-queue preempts the current - * task, then the wakeup sets TIF_NEED_RESCHED and schedule() gets - * called on the nearest possible occasion: - * - * - If the kernel is preemptible (CONFIG_PREEMPT=y): - * - * - in syscall or exception context, at the next outmost - * preempt_enable(). (this might be as soon as the wake_up()'s - * spin_unlock()!) - * - * - in IRQ context, return from interrupt-handler to - * preemptible context - * - * - If the kernel is not preemptible (CONFIG_PREEMPT is not set) - * then at the next: - * - * - cond_resched() call - * - explicit schedule() call - * - return from syscall or exception to user-space - * - return from interrupt-handler to user-space + * Careful; this can return RETRY_TASK, it does not include the retry-loop + * itself due to the whole SMT pick retry thing below. */ -static void __sched __schedule(void) +static inline struct task_struct *pick_task(struct rq *rq, struct rq_flags *rf) { - struct task_struct *prev, *next; - unsigned long *switch_count; - struct rq *rq; - int cpu; + const struct sched_class *class; + struct task_struct *p; -need_resched: - preempt_disable(); - cpu = smp_processor_id(); - rq = cpu_rq(cpu); - rcu_note_context_switch(cpu); - prev = rq->curr; + rq->dl_server = NULL; - schedule_debug(prev); + for_each_active_class(class) { + p = class->pick_task(rq, rf); + if (p) + return p; + } - if (sched_feat(HRTICK)) - hrtick_clear(rq); + BUG(); /* The idle class should always have a runnable task. */ +} - raw_spin_lock_irq(&rq->lock); +extern void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi); - switch_count = &prev->nivcsw; - if (prev->state && !(preempt_count() & PREEMPT_ACTIVE)) { - if (unlikely(signal_pending_state(prev->state, prev))) { - prev->state = TASK_RUNNING; - } else { - deactivate_task(rq, prev, DEQUEUE_SLEEP); - prev->on_rq = 0; +static void queue_core_balance(struct rq *rq); - /* - * If a worker went to sleep, notify and ask workqueue - * whether it wants to wake up a task to maintain - * concurrency. - */ - if (prev->flags & PF_WQ_WORKER) { - struct task_struct *to_wakeup; +static struct task_struct * +pick_next_task(struct rq *rq, struct task_struct *prev, struct rq_flags *rf) +{ + struct task_struct *next, *p, *max; + const struct cpumask *smt_mask; + bool fi_before = false; + bool core_clock_updated = (rq == rq->core); + unsigned long cookie; + int i, cpu, occ = 0; + struct rq *rq_i; + bool need_sync; - to_wakeup = wq_worker_sleeping(prev, cpu); - if (to_wakeup) - try_to_wake_up_local(to_wakeup); - } - } - switch_count = &prev->nvcsw; + if (!sched_core_enabled(rq)) + return __pick_next_task(rq, prev, rf); + + cpu = cpu_of(rq); + + /* Stopper task is switching into idle, no need core-wide selection. */ + if (cpu_is_offline(cpu)) { + /* + * Reset core_pick so that we don't enter the fastpath when + * coming online. core_pick would already be migrated to + * another cpu during offline. + */ + rq->core_pick = NULL; + rq->core_dl_server = NULL; + return __pick_next_task(rq, prev, rf); } - pre_schedule(rq, prev); + /* + * If there were no {en,de}queues since we picked (IOW, the task + * pointers are all still valid), and we haven't scheduled the last + * pick yet, do so now. + * + * rq->core_pick can be NULL if no selection was made for a CPU because + * it was either offline or went offline during a sibling's core-wide + * selection. In this case, do a core-wide selection. + */ + if (rq->core->core_pick_seq == rq->core->core_task_seq && + rq->core->core_pick_seq != rq->core_sched_seq && + rq->core_pick) { + WRITE_ONCE(rq->core_sched_seq, rq->core->core_pick_seq); - if (unlikely(!rq->nr_running)) - idle_balance(cpu, rq); + next = rq->core_pick; + rq->dl_server = rq->core_dl_server; + rq->core_pick = NULL; + rq->core_dl_server = NULL; + goto out_set_next; + } - put_prev_task(rq, prev); - next = pick_next_task(rq); - clear_tsk_need_resched(prev); - rq->skip_clock_update = 0; + prev_balance(rq, prev, rf); - if (likely(prev != next)) { - rq->nr_switches++; - rq->curr = next; - ++*switch_count; + smt_mask = cpu_smt_mask(cpu); + need_sync = !!rq->core->core_cookie; - context_switch(rq, prev, next); /* unlocks the rq */ - /* - * The context switch have flipped the stack from under us - * and restored the local variables which were saved when - * this task called schedule() in the past. prev == current - * is still correct, but it can be moved to another cpu/rq. - */ - cpu = smp_processor_id(); - rq = cpu_rq(cpu); - } else - raw_spin_unlock_irq(&rq->lock); + /* reset state */ + rq->core->core_cookie = 0UL; + if (rq->core->core_forceidle_count) { + if (!core_clock_updated) { + update_rq_clock(rq->core); + core_clock_updated = true; + } + sched_core_account_forceidle(rq); + /* reset after accounting force idle */ + rq->core->core_forceidle_start = 0; + rq->core->core_forceidle_count = 0; + rq->core->core_forceidle_occupation = 0; + need_sync = true; + fi_before = true; + } - post_schedule(rq); + /* + * core->core_task_seq, core->core_pick_seq, rq->core_sched_seq + * + * @task_seq guards the task state ({en,de}queues) + * @pick_seq is the @task_seq we did a selection on + * @sched_seq is the @pick_seq we scheduled + * + * However, preemptions can cause multiple picks on the same task set. + * 'Fix' this by also increasing @task_seq for every pick. + */ + rq->core->core_task_seq++; - sched_preempt_enable_no_resched(); - if (need_resched()) - goto need_resched; -} + /* + * Optimize for common case where this CPU has no cookies + * and there are no cookied tasks running on siblings. + */ + if (!need_sync) { +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; + /* + * For robustness, update the min_vruntime_fi for + * unconstrained picks as well. + */ + WARN_ON_ONCE(fi_before); + task_vruntime_update(rq, next, false); + goto out_set_next; + } + } -static inline void sched_submit_work(struct task_struct *tsk) -{ - if (!tsk->state || tsk_is_pi_blocked(tsk)) - return; /* - * If we are going to sleep and we have plugged IO queued, - * make sure to submit it to avoid deadlocks. + * For each thread: do the regular task pick and find the max prio task + * amongst them. + * + * Tie-break prio towards the current CPU */ - if (blk_needs_flush_plug(tsk)) - blk_schedule_flush_plug(tsk); -} +restart_multi: + max = NULL; + for_each_cpu_wrap(i, smt_mask, cpu) { + rq_i = cpu_rq(i); -asmlinkage void __sched schedule(void) -{ - struct task_struct *tsk = current; + /* + * Current cpu always has its clock updated on entrance to + * pick_next_task(). If the current cpu is not the core, + * the core may also have been updated above. + */ + if (i != cpu && (rq_i != rq->core || !core_clock_updated)) + update_rq_clock(rq_i); - sched_submit_work(tsk); - __schedule(); -} -EXPORT_SYMBOL(schedule); + 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)) + max = p; + } + + cookie = rq->core->core_cookie = max->core_cookie; -#ifdef CONFIG_CONTEXT_TRACKING -asmlinkage void __sched schedule_user(void) -{ /* - * If we come here after a random call to set_need_resched(), - * or we have been woken up remotely but the IPI has not yet arrived, - * we haven't yet exited the RCU idle mode. Do it here manually until - * we find a better solution. + * For each thread: try and find a runnable task that matches @max or + * force idle. */ - user_exit(); - schedule(); - user_enter(); -} -#endif + for_each_cpu(i, smt_mask) { + rq_i = cpu_rq(i); + p = rq_i->core_pick; + + if (!cookie_equals(p, cookie)) { + p = NULL; + if (cookie) + p = sched_core_find(rq_i, cookie); + if (!p) + p = idle_sched_class.pick_task(rq_i, rf); + } -/** - * schedule_preempt_disabled - called with preemption disabled - * - * Returns with preemption disabled. Note: preempt_count must be 1 - */ -void __sched schedule_preempt_disabled(void) -{ - sched_preempt_enable_no_resched(); - schedule(); - preempt_disable(); -} + rq_i->core_pick = p; + rq_i->core_dl_server = NULL; -#ifdef CONFIG_PREEMPT -/* - * this is the entry point to schedule() from in-kernel preemption - * off of preempt_enable. Kernel preemptions off return from interrupt - * occur there and call schedule directly. - */ -asmlinkage void __sched notrace preempt_schedule(void) -{ - struct thread_info *ti = current_thread_info(); + if (p == rq_i->idle) { + if (rq_i->nr_running) { + rq->core->core_forceidle_count++; + if (!fi_before) + rq->core->core_forceidle_seq++; + } + } else { + occ++; + } + } + + if (schedstat_enabled() && rq->core->core_forceidle_count) { + rq->core->core_forceidle_start = rq_clock(rq->core); + rq->core->core_forceidle_occupation = occ; + } + + rq->core->core_pick_seq = rq->core->core_task_seq; + next = rq->core_pick; + rq->core_sched_seq = rq->core->core_pick_seq; + + /* Something should have been selected for current CPU */ + WARN_ON_ONCE(!next); /* - * If there is a non-zero preempt_count or interrupts are disabled, - * we do not want to preempt the current task. Just return.. + * Reschedule siblings + * + * NOTE: L1TF -- at this point we're no longer running the old task and + * sending an IPI (below) ensures the sibling will no longer be running + * their task. This ensures there is no inter-sibling overlap between + * non-matching user state. */ - if (likely(ti->preempt_count || irqs_disabled())) - return; + for_each_cpu(i, smt_mask) { + rq_i = cpu_rq(i); - do { - add_preempt_count_notrace(PREEMPT_ACTIVE); - __schedule(); - sub_preempt_count_notrace(PREEMPT_ACTIVE); + /* + * An online sibling might have gone offline before a task + * could be picked for it, or it might be offline but later + * happen to come online, but its too late and nothing was + * picked for it. That's Ok - it will pick tasks for itself, + * so ignore it. + */ + if (!rq_i->core_pick) + continue; /* - * Check again in case we missed a preemption opportunity - * between schedule and now. + * Update for new !FI->FI transitions, or if continuing to be in !FI: + * fi_before fi update? + * 0 0 1 + * 0 1 1 + * 1 0 1 + * 1 1 0 */ - barrier(); - } while (need_resched()); + if (!(fi_before && rq->core->core_forceidle_count)) + task_vruntime_update(rq_i, rq_i->core_pick, !!rq->core->core_forceidle_count); + + rq_i->core_pick->core_occupation = occ; + + if (i == cpu) { + rq_i->core_pick = NULL; + rq_i->core_dl_server = NULL; + continue; + } + + /* Did we break L1TF mitigation requirements? */ + WARN_ON_ONCE(!cookie_match(next, rq_i->core_pick)); + + if (rq_i->curr == rq_i->core_pick) { + rq_i->core_pick = NULL; + rq_i->core_dl_server = NULL; + continue; + } + + resched_curr(rq_i); + } + +out_set_next: + put_prev_set_next_task(rq, prev, next); + if (rq->core->core_forceidle_count && next == rq->idle) + queue_core_balance(rq); + + return next; } -EXPORT_SYMBOL(preempt_schedule); -/* - * this is the entry point to schedule() from kernel preemption - * off of irq context. - * Note, that this is called and return with irqs disabled. This will - * protect us against recursive calling from irq. - */ -asmlinkage void __sched preempt_schedule_irq(void) +static bool try_steal_cookie(int this, int that) { - struct thread_info *ti = current_thread_info(); - enum ctx_state prev_state; + struct rq *dst = cpu_rq(this), *src = cpu_rq(that); + struct task_struct *p; + unsigned long cookie; + bool success = false; - /* Catch callers which need to be fixed */ - BUG_ON(ti->preempt_count || !irqs_disabled()); + guard(irq)(); + guard(double_rq_lock)(dst, src); - prev_state = exception_enter(); + cookie = dst->core->core_cookie; + if (!cookie) + return false; + + if (dst->curr != dst->idle) + return false; + + p = sched_core_find(src, cookie); + if (!p) + return false; do { - add_preempt_count(PREEMPT_ACTIVE); - local_irq_enable(); - __schedule(); - local_irq_disable(); - sub_preempt_count(PREEMPT_ACTIVE); + if (p == src->core_pick || p == src->curr) + goto next; + if (!is_cpu_allowed(p, this)) + goto next; + + if (p->core_occupation > dst->idle->core_occupation) + goto next; /* - * Check again in case we missed a preemption opportunity - * between schedule and now. + * sched_core_find() and sched_core_next() will ensure + * that task @p is not throttled now, we also need to + * check whether the runqueue of the destination CPU is + * being throttled. */ - barrier(); - } while (need_resched()); + if (sched_task_is_throttled(p, this)) + goto next; - exception_exit(prev_state); -} + move_queued_task_locked(src, dst, p); + resched_curr(dst); + + success = true; + break; -#endif /* CONFIG_PREEMPT */ +next: + p = sched_core_next(p, cookie); + } while (p); -int default_wake_function(wait_queue_t *curr, unsigned mode, int wake_flags, - void *key) -{ - return try_to_wake_up(curr->private, mode, wake_flags); + return success; } -EXPORT_SYMBOL(default_wake_function); -/* - * The core wakeup function. Non-exclusive wakeups (nr_exclusive == 0) just - * wake everything up. If it's an exclusive wakeup (nr_exclusive == small +ve - * number) then we wake all the non-exclusive tasks and one 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 - * zero in this (rare) case, and we handle it by continuing to scan the queue. - */ -static void __wake_up_common(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, int wake_flags, void *key) +static bool steal_cookie_task(int cpu, struct sched_domain *sd) { - wait_queue_t *curr, *next; + int i; - list_for_each_entry_safe(curr, next, &q->task_list, task_list) { - unsigned flags = curr->flags; + for_each_cpu_wrap(i, sched_domain_span(sd), cpu + 1) { + if (i == cpu) + continue; - if (curr->func(curr, mode, wake_flags, key) && - (flags & WQ_FLAG_EXCLUSIVE) && !--nr_exclusive) + if (need_resched()) break; + + if (try_steal_cookie(cpu, i)) + return true; } + + return false; } -/** - * __wake_up - wake up threads blocked on a waitqueue. - * @q: the waitqueue - * @mode: which threads - * @nr_exclusive: how many wake-one or wake-many threads to wake up - * @key: is directly passed to the wakeup function - * - * It may be assumed that this function implies a write memory barrier before - * changing the task state if and only if any tasks are woken up. - */ -void __wake_up(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, void *key) +static void sched_core_balance(struct rq *rq) { - unsigned long flags; + struct sched_domain *sd; + int cpu = cpu_of(rq); - spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, 0, key); - spin_unlock_irqrestore(&q->lock, flags); -} -EXPORT_SYMBOL(__wake_up); + guard(preempt)(); + guard(rcu)(); -/* - * Same as __wake_up but called with the spinlock in wait_queue_head_t held. - */ -void __wake_up_locked(wait_queue_head_t *q, unsigned int mode, int nr) -{ - __wake_up_common(q, mode, nr, 0, NULL); -} -EXPORT_SYMBOL_GPL(__wake_up_locked); + raw_spin_rq_unlock_irq(rq); + for_each_domain(cpu, sd) { + if (need_resched()) + break; -void __wake_up_locked_key(wait_queue_head_t *q, unsigned int mode, void *key) -{ - __wake_up_common(q, mode, 1, 0, key); + if (steal_cookie_task(cpu, sd)) + break; + } + raw_spin_rq_lock_irq(rq); } -EXPORT_SYMBOL_GPL(__wake_up_locked_key); -/** - * __wake_up_sync_key - wake up threads blocked on a waitqueue. - * @q: the waitqueue - * @mode: which threads - * @nr_exclusive: how many wake-one or wake-many threads to wake up - * @key: opaque value to be passed to wakeup targets - * - * The sync wakeup differs that the waker knows that it will schedule - * away soon, so while the target thread will be woken up, it will not - * be migrated to another CPU - ie. the two threads are 'synchronized' - * with each other. This can prevent needless bouncing between CPUs. - * - * On UP it can prevent extra preemption. - * - * It may be assumed that this function implies a write memory barrier before - * changing the task state if and only if any tasks are woken up. - */ -void __wake_up_sync_key(wait_queue_head_t *q, unsigned int mode, - int nr_exclusive, void *key) +static DEFINE_PER_CPU(struct balance_callback, core_balance_head); + +static void queue_core_balance(struct rq *rq) { - unsigned long flags; - int wake_flags = WF_SYNC; + if (!sched_core_enabled(rq)) + return; - if (unlikely(!q)) + if (!rq->core->core_cookie) return; - if (unlikely(!nr_exclusive)) - wake_flags = 0; + if (!rq->nr_running) /* not forced idle */ + return; - spin_lock_irqsave(&q->lock, flags); - __wake_up_common(q, mode, nr_exclusive, wake_flags, key); - spin_unlock_irqrestore(&q->lock, flags); + queue_balance_callback(rq, &per_cpu(core_balance_head, rq->cpu), sched_core_balance); } -EXPORT_SYMBOL_GPL(__wake_up_sync_key); -/* - * __wake_up_sync - see __wake_up_sync_key() - */ -void __wake_up_sync(wait_queue_head_t *q, unsigned int mode, int nr_exclusive) -{ - __wake_up_sync_key(q, mode, nr_exclusive, NULL); -} -EXPORT_SYMBOL_GPL(__wake_up_sync); /* For internal use only */ +DEFINE_LOCK_GUARD_1(core_lock, int, + sched_core_lock(*_T->lock, &_T->flags), + sched_core_unlock(*_T->lock, &_T->flags), + unsigned long flags) -/** - * complete: - signals a single thread waiting on this completion - * @x: holds the state of this particular completion - * - * This will wake up a single thread waiting on this completion. Threads will be - * awakened in the same order in which they were queued. - * - * See also complete_all(), wait_for_completion() and related routines. - * - * It may be assumed that this function implies a write memory barrier before - * changing the task state if and only if any tasks are woken up. - */ -void complete(struct completion *x) +static void sched_core_cpu_starting(unsigned int cpu) { - unsigned long flags; + const struct cpumask *smt_mask = cpu_smt_mask(cpu); + struct rq *rq = cpu_rq(cpu), *core_rq = NULL; + int t; - spin_lock_irqsave(&x->wait.lock, flags); - x->done++; - __wake_up_common(&x->wait, TASK_NORMAL, 1, 0, NULL); - spin_unlock_irqrestore(&x->wait.lock, flags); -} -EXPORT_SYMBOL(complete); + guard(core_lock)(&cpu); -/** - * complete_all: - signals all threads waiting on this completion - * @x: holds the state of this particular completion - * - * This will wake up all threads waiting on this particular completion event. - * - * It may be assumed that this function implies a write memory barrier before - * changing the task state if and only if any tasks are woken up. - */ -void complete_all(struct completion *x) -{ - unsigned long flags; + WARN_ON_ONCE(rq->core != rq); - spin_lock_irqsave(&x->wait.lock, flags); - x->done += UINT_MAX/2; - __wake_up_common(&x->wait, TASK_NORMAL, 0, 0, NULL); - spin_unlock_irqrestore(&x->wait.lock, flags); -} -EXPORT_SYMBOL(complete_all); + /* if we're the first, we'll be our own leader */ + if (cpumask_weight(smt_mask) == 1) + return; -static inline long __sched -do_wait_for_common(struct completion *x, - long (*action)(long), long timeout, int state) -{ - if (!x->done) { - DECLARE_WAITQUEUE(wait, current); + /* find the leader */ + for_each_cpu(t, smt_mask) { + if (t == cpu) + continue; + rq = cpu_rq(t); + if (rq->core == rq) { + core_rq = rq; + break; + } + } - __add_wait_queue_tail_exclusive(&x->wait, &wait); - do { - if (signal_pending_state(state, current)) { - timeout = -ERESTARTSYS; - break; - } - __set_current_state(state); - spin_unlock_irq(&x->wait.lock); - timeout = action(timeout); - spin_lock_irq(&x->wait.lock); - } while (!x->done && timeout); - __remove_wait_queue(&x->wait, &wait); - if (!x->done) - return timeout; + if (WARN_ON_ONCE(!core_rq)) /* whoopsie */ + return; + + /* install and validate core_rq */ + for_each_cpu(t, smt_mask) { + rq = cpu_rq(t); + + if (t == cpu) + rq->core = core_rq; + + WARN_ON_ONCE(rq->core != core_rq); } - x->done--; - return timeout ?: 1; } -static inline long __sched -__wait_for_common(struct completion *x, - long (*action)(long), long timeout, int state) +static void sched_core_cpu_deactivate(unsigned int cpu) { - might_sleep(); + const struct cpumask *smt_mask = cpu_smt_mask(cpu); + struct rq *rq = cpu_rq(cpu), *core_rq = NULL; + int t; - spin_lock_irq(&x->wait.lock); - timeout = do_wait_for_common(x, action, timeout, state); - spin_unlock_irq(&x->wait.lock); - return timeout; -} + guard(core_lock)(&cpu); -static long __sched -wait_for_common(struct completion *x, long timeout, int state) -{ - return __wait_for_common(x, schedule_timeout, timeout, state); -} + /* if we're the last man standing, nothing to do */ + if (cpumask_weight(smt_mask) == 1) { + WARN_ON_ONCE(rq->core != rq); + return; + } -static long __sched -wait_for_common_io(struct completion *x, long timeout, int state) -{ - return __wait_for_common(x, io_schedule_timeout, timeout, state); -} + /* if we're not the leader, nothing to do */ + if (rq->core != rq) + return; -/** - * wait_for_completion: - waits for completion of a task - * @x: holds the state of this particular completion - * - * This waits to be signaled for completion of a specific task. It is NOT - * interruptible and there is no timeout. - * - * See also similar routines (i.e. wait_for_completion_timeout()) with timeout - * and interrupt capability. Also see complete(). - */ -void __sched wait_for_completion(struct completion *x) -{ - wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE); -} -EXPORT_SYMBOL(wait_for_completion); + /* find a new leader */ + for_each_cpu(t, smt_mask) { + if (t == cpu) + continue; + core_rq = cpu_rq(t); + break; + } -/** - * wait_for_completion_timeout: - waits for completion of a task (w/timeout) - * @x: holds the state of this particular completion - * @timeout: timeout value in jiffies - * - * This waits for either a completion of a specific task to be signaled or for a - * specified timeout to expire. The timeout is in jiffies. It is not - * interruptible. - * - * The return value is 0 if timed out, and positive (at least 1, or number of - * jiffies left till timeout) if completed. - */ -unsigned long __sched -wait_for_completion_timeout(struct completion *x, unsigned long timeout) -{ - return wait_for_common(x, timeout, TASK_UNINTERRUPTIBLE); -} -EXPORT_SYMBOL(wait_for_completion_timeout); + if (WARN_ON_ONCE(!core_rq)) /* impossible */ + return; -/** - * wait_for_completion_io: - waits for completion of a task - * @x: holds the state of this particular completion - * - * This waits to be signaled for completion of a specific task. It is NOT - * interruptible and there is no timeout. The caller is accounted as waiting - * for IO. - */ -void __sched wait_for_completion_io(struct completion *x) -{ - wait_for_common_io(x, MAX_SCHEDULE_TIMEOUT, TASK_UNINTERRUPTIBLE); -} -EXPORT_SYMBOL(wait_for_completion_io); + /* copy the shared state to the new leader */ + core_rq->core_task_seq = rq->core_task_seq; + core_rq->core_pick_seq = rq->core_pick_seq; + core_rq->core_cookie = rq->core_cookie; + core_rq->core_forceidle_count = rq->core_forceidle_count; + core_rq->core_forceidle_seq = rq->core_forceidle_seq; + core_rq->core_forceidle_occupation = rq->core_forceidle_occupation; -/** - * wait_for_completion_io_timeout: - waits for completion of a task (w/timeout) - * @x: holds the state of this particular completion - * @timeout: timeout value in jiffies - * - * This waits for either a completion of a specific task to be signaled or for a - * specified timeout to expire. The timeout is in jiffies. It is not - * interruptible. The caller is accounted as waiting for IO. - * - * The return value is 0 if timed out, and positive (at least 1, or number of - * jiffies left till timeout) if completed. - */ -unsigned long __sched -wait_for_completion_io_timeout(struct completion *x, unsigned long timeout) -{ - return wait_for_common_io(x, timeout, TASK_UNINTERRUPTIBLE); -} -EXPORT_SYMBOL(wait_for_completion_io_timeout); + /* + * Accounting edge for forced idle is handled in pick_next_task(). + * Don't need another one here, since the hotplug thread shouldn't + * have a cookie. + */ + core_rq->core_forceidle_start = 0; -/** - * wait_for_completion_interruptible: - waits for completion of a task (w/intr) - * @x: holds the state of this particular completion - * - * This waits for completion of a specific task to be signaled. It is - * interruptible. - * - * The return value is -ERESTARTSYS if interrupted, 0 if completed. - */ -int __sched wait_for_completion_interruptible(struct completion *x) -{ - long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_INTERRUPTIBLE); - if (t == -ERESTARTSYS) - return t; - return 0; + /* install new leader */ + for_each_cpu(t, smt_mask) { + rq = cpu_rq(t); + rq->core = core_rq; + } } -EXPORT_SYMBOL(wait_for_completion_interruptible); -/** - * wait_for_completion_interruptible_timeout: - waits for completion (w/(to,intr)) - * @x: holds the state of this particular completion - * @timeout: timeout value in jiffies - * - * This waits for either a completion of a specific task to be signaled or for a - * specified timeout to expire. It is interruptible. The timeout is in jiffies. - * - * The return value is -ERESTARTSYS if interrupted, 0 if timed out, - * positive (at least 1, or number of jiffies left till timeout) if completed. - */ -long __sched -wait_for_completion_interruptible_timeout(struct completion *x, - unsigned long timeout) +static inline void sched_core_cpu_dying(unsigned int cpu) { - return wait_for_common(x, timeout, TASK_INTERRUPTIBLE); -} -EXPORT_SYMBOL(wait_for_completion_interruptible_timeout); + struct rq *rq = cpu_rq(cpu); -/** - * wait_for_completion_killable: - waits for completion of a task (killable) - * @x: holds the state of this particular completion - * - * This waits to be signaled for completion of a specific task. It can be - * interrupted by a kill signal. - * - * The return value is -ERESTARTSYS if interrupted, 0 if completed. - */ -int __sched wait_for_completion_killable(struct completion *x) -{ - long t = wait_for_common(x, MAX_SCHEDULE_TIMEOUT, TASK_KILLABLE); - if (t == -ERESTARTSYS) - return t; - return 0; + if (rq->core != rq) + rq->core = rq; } -EXPORT_SYMBOL(wait_for_completion_killable); -/** - * wait_for_completion_killable_timeout: - waits for completion of a task (w/(to,killable)) - * @x: holds the state of this particular completion - * @timeout: timeout value in jiffies - * - * This waits for either a completion of a specific task to be - * signaled or for a specified timeout to expire. It can be - * interrupted by a kill signal. The timeout is in jiffies. - * - * The return value is -ERESTARTSYS if interrupted, 0 if timed out, - * positive (at least 1, or number of jiffies left till timeout) if completed. - */ -long __sched -wait_for_completion_killable_timeout(struct completion *x, - unsigned long timeout) +#else /* !CONFIG_SCHED_CORE: */ + +static inline void sched_core_cpu_starting(unsigned int cpu) {} +static inline void sched_core_cpu_deactivate(unsigned int cpu) {} +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) { - return wait_for_common(x, timeout, TASK_KILLABLE); + return __pick_next_task(rq, prev, rf); } -EXPORT_SYMBOL(wait_for_completion_killable_timeout); -/** - * try_wait_for_completion - try to decrement a completion without blocking - * @x: completion structure - * - * Returns: 0 if a decrement cannot be done without blocking - * 1 if a decrement succeeded. +#endif /* !CONFIG_SCHED_CORE */ + +/* + * Constants for the sched_mode argument of __schedule(). * - * If a completion is being used as a counting completion, - * attempt to decrement the counter without blocking. This - * enables us to avoid waiting if the resource the completion - * is protecting is not available. + * The mode argument allows RT enabled kernels to differentiate a + * preemption from blocking on an 'sleeping' spin/rwlock. */ -bool try_wait_for_completion(struct completion *x) -{ - unsigned long flags; - int ret = 1; - - spin_lock_irqsave(&x->wait.lock, flags); - if (!x->done) - ret = 0; - else - x->done--; - spin_unlock_irqrestore(&x->wait.lock, flags); - return ret; -} -EXPORT_SYMBOL(try_wait_for_completion); +#define SM_IDLE (-1) +#define SM_NONE 0 +#define SM_PREEMPT 1 +#define SM_RTLOCK_WAIT 2 -/** - * completion_done - Test to see if a completion has any waiters - * @x: completion structure - * - * Returns: 0 if there are waiters (wait_for_completion() in progress) - * 1 if there are no waiters. +/* + * Helper function for __schedule() * + * 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). */ -bool completion_done(struct completion *x) +static bool try_to_block_task(struct rq *rq, struct task_struct *p, + unsigned long *task_state_p, bool should_block) { - unsigned long flags; - int ret = 1; - - spin_lock_irqsave(&x->wait.lock, flags); - if (!x->done) - ret = 0; - spin_unlock_irqrestore(&x->wait.lock, flags); - return ret; -} -EXPORT_SYMBOL(completion_done); + unsigned long task_state = *task_state_p; + int flags = DEQUEUE_NOCLOCK; -static long __sched -sleep_on_common(wait_queue_head_t *q, int state, long timeout) -{ - unsigned long flags; - wait_queue_t wait; + if (signal_pending_state(task_state, p)) { + WRITE_ONCE(p->__state, TASK_RUNNING); + *task_state_p = TASK_RUNNING; + return false; + } - init_waitqueue_entry(&wait, current); + /* + * 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; - __set_current_state(state); + p->sched_contributes_to_load = + (task_state & TASK_UNINTERRUPTIBLE) && + !(task_state & TASK_NOLOAD) && + !(task_state & TASK_FROZEN); - spin_lock_irqsave(&q->lock, flags); - __add_wait_queue(q, &wait); - spin_unlock(&q->lock); - timeout = schedule_timeout(timeout); - spin_lock_irq(&q->lock); - __remove_wait_queue(q, &wait); - spin_unlock_irqrestore(&q->lock, flags); + if (unlikely(is_special_task_state(task_state))) + flags |= DEQUEUE_SPECIAL; - return timeout; + /* + * __schedule() ttwu() + * prev_state = prev->state; if (p->on_rq && ...) + * if (prev_state) goto out; + * p->on_rq = 0; smp_acquire__after_ctrl_dep(); + * p->state = TASK_WAKING + * + * Where __schedule() and ttwu() have matching control dependencies. + * + * After this, schedule() must not care about p->state any more. + */ + block_task(rq, p, flags); + return true; } -void __sched interruptible_sleep_on(wait_queue_head_t *q) +#ifdef CONFIG_SCHED_PROXY_EXEC +static inline struct task_struct *proxy_resched_idle(struct rq *rq) { - sleep_on_common(q, TASK_INTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT); + 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; } -EXPORT_SYMBOL(interruptible_sleep_on); -long __sched -interruptible_sleep_on_timeout(wait_queue_head_t *q, long timeout) +static bool __proxy_deactivate(struct rq *rq, struct task_struct *donor) { - return sleep_on_common(q, TASK_INTERRUPTIBLE, timeout); -} -EXPORT_SYMBOL(interruptible_sleep_on_timeout); + unsigned long state = READ_ONCE(donor->__state); -void __sched sleep_on(wait_queue_head_t *q) -{ - sleep_on_common(q, TASK_UNINTERRUPTIBLE, MAX_SCHEDULE_TIMEOUT); + /* 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); } -EXPORT_SYMBOL(sleep_on); -long __sched sleep_on_timeout(wait_queue_head_t *q, long timeout) +static struct task_struct *proxy_deactivate(struct rq *rq, struct task_struct *donor) { - return sleep_on_common(q, TASK_UNINTERRUPTIBLE, timeout); + if (!__proxy_deactivate(rq, donor)) { + /* + * XXX: For now, if deactivation failed, set donor + * as unblocked, as we aren't doing proxy-migrations + * yet (more logic will be needed then). + */ + donor->blocked_on = NULL; + } + return NULL; } -EXPORT_SYMBOL(sleep_on_timeout); - -#ifdef CONFIG_RT_MUTEXES /* - * rt_mutex_setprio - set the current priority of a task - * @p: task - * @prio: prio value (kernel-internal form) + * Find runnable lock owner to proxy for mutex blocked donor * - * This function changes the 'effective' priority of a task. It does - * not touch ->normal_prio like __setscheduler(). + * Follow the blocked-on relation: + * task->blocked_on -> mutex->owner -> task... + * + * Lock order: * - * Used by the rt_mutex code to implement priority inheritance logic. + * p->pi_lock + * rq->lock + * mutex->wait_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)). */ -void rt_mutex_setprio(struct task_struct *p, int prio) +static struct task_struct * +find_proxy_task(struct rq *rq, struct task_struct *donor, struct rq_flags *rf) { - int oldprio, on_rq, running; - struct rq *rq; - const struct sched_class *prev_class; - - BUG_ON(prio < 0 || prio > MAX_PRIO); - - rq = __task_rq_lock(p); - - /* - * Idle task boosting is a nono in general. There is one - * exception, when PREEMPT_RT and NOHZ is active: - * - * The idle task calls get_next_timer_interrupt() and holds - * the timer wheel base->lock on the CPU and another CPU wants - * to access the timer (probably to cancel it). We can safely - * ignore the boosting request, as the idle CPU runs this code - * with interrupts disabled and will complete the lock - * protected section without being interrupted. So there is no - * real need to boost. - */ - if (unlikely(p == rq->idle)) { - WARN_ON(p != rq->curr); - WARN_ON(p->pi_blocked_on); - goto out_unlock; - } + struct task_struct *owner = NULL; + int this_cpu = cpu_of(rq); + struct task_struct *p; + struct mutex *mutex; - trace_sched_pi_setprio(p, prio); - oldprio = p->prio; - prev_class = p->sched_class; - on_rq = p->on_rq; - running = task_current(rq, p); - if (on_rq) - dequeue_task(rq, p, 0); - if (running) - p->sched_class->put_prev_task(rq, p); + /* Follow blocked_on chain. */ + for (p = donor; task_is_blocked(p); p = owner) { + mutex = p->blocked_on; + /* Something changed in the chain, so pick again */ + if (!mutex) + return NULL; + /* + * By taking mutex->wait_lock we hold off concurrent mutex_unlock() + * and ensure @owner sticks around. + */ + guard(raw_spinlock)(&mutex->wait_lock); - if (rt_prio(prio)) - p->sched_class = &rt_sched_class; - else - p->sched_class = &fair_sched_class; + /* Check again that p is blocked with wait_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; + } - p->prio = prio; + owner = __mutex_owner(mutex); + if (!owner) { + __clear_task_blocked_on(p, mutex); + return p; + } - if (running) - p->sched_class->set_curr_task(rq); - if (on_rq) - enqueue_task(rq, p, oldprio < prio ? ENQUEUE_HEAD : 0); + if (!READ_ONCE(owner->on_rq) || owner->se.sched_delayed) { + /* XXX Don't handle blocked owners/delayed dequeue yet */ + return proxy_deactivate(rq, donor); + } - check_class_changed(rq, p, prev_class, oldprio); -out_unlock: - __task_rq_unlock(rq); -} -#endif -void set_user_nice(struct task_struct *p, long nice) -{ - int old_prio, delta, on_rq; - unsigned long flags; - struct rq *rq; + if (task_cpu(owner) != this_cpu) { + /* XXX Don't handle migrations yet */ + return proxy_deactivate(rq, donor); + } - if (TASK_NICE(p) == nice || nice < -20 || nice > 19) - return; - /* - * We have to be careful, if called from sys_setpriority(), - * the task might be in the middle of scheduling on another CPU. - */ - rq = task_rq_lock(p, &flags); - /* - * The RT priorities are set via sched_setscheduler(), but we still - * allow the 'normal' nice value to be set - but as expected - * it wont have any effect on scheduling until the task is - * SCHED_FIFO/SCHED_RR: - */ - if (task_has_rt_policy(p)) { - p->static_prio = NICE_TO_PRIO(nice); - goto out_unlock; - } - on_rq = p->on_rq; - if (on_rq) - dequeue_task(rq, p, 0); + 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); + } - p->static_prio = NICE_TO_PRIO(nice); - set_load_weight(p); - old_prio = p->prio; - p->prio = effective_prio(p); - delta = p->prio - old_prio; + /* + * 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 (on_rq) { - enqueue_task(rq, p, 0); + 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); + } /* - * If the task increased its priority or is running and - * lowered its priority, then reschedule its CPU: + * 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(). */ - if (delta < 0 || (delta > 0 && task_running(rq, p))) - resched_task(rq->curr); } -out_unlock: - task_rq_unlock(rq, p, &flags); -} -EXPORT_SYMBOL(set_user_nice); -/* - * can_nice - check if a task can reduce its nice value - * @p: task - * @nice: nice value - */ -int can_nice(const struct task_struct *p, const int nice) + WARN_ON_ONCE(owner && !owner->on_rq); + return owner; +} +#else /* SCHED_PROXY_EXEC */ +static struct task_struct * +find_proxy_task(struct rq *rq, struct task_struct *donor, struct rq_flags *rf) { - /* convert nice value [19,-20] to rlimit style value [1,40] */ - int nice_rlim = 20 - nice; - - return (nice_rlim <= task_rlimit(p, RLIMIT_NICE) || - capable(CAP_SYS_NICE)); + WARN_ONCE(1, "This should never be called in the !SCHED_PROXY_EXEC case\n"); + return donor; } +#endif /* SCHED_PROXY_EXEC */ -#ifdef __ARCH_WANT_SYS_NICE +static inline void proxy_tag_curr(struct rq *rq, struct task_struct *owner) +{ + if (!sched_proxy_exec()) + return; + /* + * pick_next_task() calls set_next_task() on the chosen task + * at some point, which ensures it is not push/pullable. + * However, the chosen/donor task *and* the mutex owner form an + * atomic pair wrt push/pull. + * + * Make sure owner we run is not pushable. Unfortunately we can + * only deal with that by means of a dequeue/enqueue cycle. :-/ + */ + dequeue_task(rq, owner, DEQUEUE_NOCLOCK | DEQUEUE_SAVE); + enqueue_task(rq, owner, ENQUEUE_NOCLOCK | ENQUEUE_RESTORE); +} /* - * sys_nice - change the priority of the current process. - * @increment: priority increment + * __schedule() is the main scheduler function. + * + * The main means of driving the scheduler and thus entering this function are: + * + * 1. Explicit blocking: mutex, semaphore, waitqueue, etc. + * + * 2. TIF_NEED_RESCHED flag is checked on interrupt and userspace return + * paths. For example, see arch/x86/entry_64.S. + * + * To drive preemption between tasks, the scheduler sets the flag in timer + * interrupt handler sched_tick(). + * + * 3. Wakeups don't really cause entry into schedule(). They add a + * task to the run-queue and that's it. + * + * Now, if the new task added to the run-queue preempts the current + * task, then the wakeup sets TIF_NEED_RESCHED and schedule() gets + * called on the nearest possible occasion: + * + * - If the kernel is preemptible (CONFIG_PREEMPTION=y): + * + * - in syscall or exception context, at the next outmost + * preempt_enable(). (this might be as soon as the wake_up()'s + * spin_unlock()!) + * + * - in IRQ context, return from interrupt-handler to + * preemptible context * - * sys_setpriority is a more generic, but much slower function that - * does similar things. + * - If the kernel is not preemptible (CONFIG_PREEMPTION is not set) + * then at the next: + * + * - cond_resched() call + * - explicit schedule() call + * - return from syscall or exception to user-space + * - return from interrupt-handler to user-space + * + * WARNING: must be called with preemption disabled! */ -SYSCALL_DEFINE1(nice, int, increment) +static void __sched notrace __schedule(int sched_mode) { - long nice, retval; - + struct task_struct *prev, *next; /* - * Setpriority might change our priority at the same moment. - * We don't have to worry. Conceptually one call occurs first - * and we have a single winner. + * On PREEMPT_RT kernel, SM_RTLOCK_WAIT is noted + * as a preemption by schedule_debug() and RCU. */ - if (increment < -40) - increment = -40; - if (increment > 40) - increment = 40; + 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; - nice = TASK_NICE(current) + increment; - if (nice < -20) - nice = -20; - if (nice > 19) - nice = 19; + /* Trace preemptions consistently with task switches */ + trace_sched_entry_tp(sched_mode == SM_PREEMPT); - if (increment < 0 && !can_nice(current, nice)) - return -EPERM; + cpu = smp_processor_id(); + rq = cpu_rq(cpu); + prev = rq->curr; - retval = security_task_setnice(current, nice); - if (retval) - return retval; + schedule_debug(prev, preempt); - set_user_nice(current, nice); - return 0; -} + if (sched_feat(HRTICK) || sched_feat(HRTICK_DL)) + hrtick_clear(rq); -#endif + klp_sched_try_switch(prev); -/** - * task_prio - return the priority value of a given task. - * @p: the task in question. - * - * This is the priority value as seen by users in /proc. - * RT tasks are offset by -200. Normal tasks are centered - * around 0, value goes from -16 to +15. - */ -int task_prio(const struct task_struct *p) -{ - return p->prio - MAX_RT_PRIO; -} + local_irq_disable(); + rcu_note_context_switch(preempt); + migrate_disable_switch(rq, prev); -/** - * task_nice - return the nice value of a given task. - * @p: the task in question. - */ -int task_nice(const struct task_struct *p) -{ - return TASK_NICE(p); -} -EXPORT_SYMBOL(task_nice); + /* + * Make sure that signal_pending_state()->signal_pending() below + * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) + * done by the caller to avoid the race with signal_wake_up(): + * + * __set_current_state(@state) signal_wake_up() + * schedule() set_tsk_thread_flag(p, TIF_SIGPENDING) + * wake_up_state(p, state) + * LOCK rq->lock LOCK p->pi_state + * smp_mb__after_spinlock() smp_mb__after_spinlock() + * if (signal_pending_state()) if (p->state & @state) + * + * Also, the membarrier system call requires a full memory barrier + * after coming from user-space, before storing to rq->curr; this + * barrier matches a full barrier in the proximity of the membarrier + * system call exit. + */ + rq_lock(rq, &rf); + smp_mb__after_spinlock(); -/** - * idle_cpu - is a given cpu idle currently? - * @cpu: the processor in question. - */ -int idle_cpu(int cpu) -{ - struct rq *rq = cpu_rq(cpu); + /* Promote REQ to ACT */ + rq->clock_update_flags <<= 1; + update_rq_clock(rq); + rq->clock_update_flags = RQCF_UPDATED; - if (rq->curr != rq->idle) - return 0; + switch_count = &prev->nivcsw; - if (rq->nr_running) - return 0; + /* Task state changes only considers SM_PREEMPT as preemption */ + preempt = sched_mode == SM_PREEMPT; -#ifdef CONFIG_SMP - if (!llist_empty(&rq->wake_list)) - return 0; -#endif + /* + * We must load prev->state once (task_struct::state is volatile), such + * that we form a control dependency vs deactivate_task() below. + */ + prev_state = READ_ONCE(prev->__state); + if (sched_mode == SM_IDLE) { + /* SCX must consult the BPF scheduler to tell if rq is empty */ + if (!rq->nr_running && !scx_enabled()) { + next = prev; + goto picked; + } + } else if (!preempt && 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; + } - return 1; -} +pick_again: + next = pick_next_task(rq, rq->donor, &rf); + rq_set_donor(rq, next); + if (unlikely(task_is_blocked(next))) { + next = find_proxy_task(rq, next, &rf); + if (!next) + goto pick_again; + if (next == rq->idle) + goto keep_resched; + } +picked: + clear_tsk_need_resched(prev); + clear_preempt_need_resched(); +keep_resched: + rq->last_seen_need_resched_ns = 0; -/** - * idle_task - return the idle task for a given cpu. - * @cpu: the processor in question. - */ -struct task_struct *idle_task(int cpu) -{ - return cpu_rq(cpu)->idle; -} + 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); -/** - * find_process_by_pid - find a process with a matching PID value. - * @pid: the pid in question. - */ -static struct task_struct *find_process_by_pid(pid_t pid) -{ - return pid ? find_task_by_vpid(pid) : current; -} + if (!task_current_donor(rq, next)) + proxy_tag_curr(rq, next); -/* Actually do priority change: must hold rq lock. */ -static void -__setscheduler(struct rq *rq, struct task_struct *p, int policy, int prio) -{ - p->policy = policy; - p->rt_priority = prio; - p->normal_prio = normal_prio(p); - /* we are holding p->pi_lock already */ - p->prio = rt_mutex_getprio(p); - if (rt_prio(p->prio)) - p->sched_class = &rt_sched_class; - else - p->sched_class = &fair_sched_class; - set_load_weight(p); -} + /* + * The membarrier system call requires each architecture + * to have a full memory barrier after updating + * rq->curr, before returning to user-space. + * + * Here are the schemes providing that barrier on the + * various architectures: + * - mm ? switch_mm() : mmdrop() for x86, s390, sparc, PowerPC, + * RISC-V. switch_mm() relies on membarrier_arch_switch_mm() + * on PowerPC and on RISC-V. + * - finish_lock_switch() for weakly-ordered + * architectures where spin_unlock is a full barrier, + * - switch_to() for arm64 (weakly-ordered, spin_unlock + * is a RELEASE barrier), + * + * The barrier matches a full barrier in the proximity of + * the membarrier system call entry. + * + * On RISC-V, this barrier pairing is also needed for the + * SYNC_CORE command when switching between processes, cf. + * the inline comments in membarrier_arch_switch_mm(). + */ + ++*switch_count; -/* - * check the target process has a UID that matches the current process's - */ -static bool check_same_owner(struct task_struct *p) -{ - const struct cred *cred = current_cred(), *pcred; - bool match; + psi_account_irqtime(rq, prev, next); + psi_sched_switch(prev, next, !task_on_rq_queued(prev) || + prev->se.sched_delayed); - rcu_read_lock(); - pcred = __task_cred(p); - match = (uid_eq(cred->euid, pcred->euid) || - uid_eq(cred->euid, pcred->uid)); - rcu_read_unlock(); - return match; -} + trace_sched_switch(preempt, prev, next, prev_state); -static int __sched_setscheduler(struct task_struct *p, int policy, - const struct sched_param *param, bool user) -{ - int retval, oldprio, oldpolicy = -1, on_rq, running; - unsigned long flags; - const struct sched_class *prev_class; - struct rq *rq; - int reset_on_fork; - - /* may grab non-irq protected spin_locks */ - BUG_ON(in_interrupt()); -recheck: - /* double check policy once rq lock held */ - if (policy < 0) { - reset_on_fork = p->sched_reset_on_fork; - policy = oldpolicy = p->policy; + /* Also unlocks the rq: */ + rq = context_switch(rq, prev, next, &rf); } else { - reset_on_fork = !!(policy & SCHED_RESET_ON_FORK); - policy &= ~SCHED_RESET_ON_FORK; + /* In case next was already curr but just got blocked_donor */ + if (!task_current_donor(rq, next)) + proxy_tag_curr(rq, next); - if (policy != SCHED_FIFO && policy != SCHED_RR && - policy != SCHED_NORMAL && policy != SCHED_BATCH && - policy != SCHED_IDLE) - return -EINVAL; + rq_unpin_lock(rq, &rf); + __balance_callbacks(rq); + raw_spin_rq_unlock_irq(rq); } + trace_sched_exit_tp(is_switch); +} - /* - * Valid priorities for SCHED_FIFO and SCHED_RR are - * 1..MAX_USER_RT_PRIO-1, valid priority for SCHED_NORMAL, - * SCHED_BATCH and SCHED_IDLE is 0. - */ - if (param->sched_priority < 0 || - (p->mm && param->sched_priority > MAX_USER_RT_PRIO-1) || - (!p->mm && param->sched_priority > MAX_RT_PRIO-1)) - return -EINVAL; - if (rt_policy(policy) != (param->sched_priority != 0)) - return -EINVAL; - - /* - * Allow unprivileged RT tasks to decrease priority: - */ - if (user && !capable(CAP_SYS_NICE)) { - if (rt_policy(policy)) { - unsigned long rlim_rtprio = - task_rlimit(p, RLIMIT_RTPRIO); +void __noreturn do_task_dead(void) +{ + /* Causes final put_task_struct in finish_task_switch(): */ + set_special_state(TASK_DEAD); - /* can't set/change the rt policy */ - if (policy != p->policy && !rlim_rtprio) - return -EPERM; + /* Tell freezer to ignore us: */ + current->flags |= PF_NOFREEZE; - /* can't increase priority */ - if (param->sched_priority > p->rt_priority && - param->sched_priority > rlim_rtprio) - return -EPERM; - } + __schedule(SM_NONE); + BUG(); - /* - * Treat SCHED_IDLE as nice 20. Only allow a switch to - * SCHED_NORMAL if the RLIMIT_NICE would normally permit it. - */ - if (p->policy == SCHED_IDLE && policy != SCHED_IDLE) { - if (!can_nice(p, TASK_NICE(p))) - return -EPERM; - } - - /* can't change other user's priorities */ - if (!check_same_owner(p)) - return -EPERM; + /* Avoid "noreturn function does return" - but don't continue if BUG() is a NOP: */ + for (;;) + cpu_relax(); +} - /* Normal users shall not reset the sched_reset_on_fork flag */ - if (p->sched_reset_on_fork && !reset_on_fork) - return -EPERM; - } +static inline void sched_submit_work(struct task_struct *tsk) +{ + static DEFINE_WAIT_OVERRIDE_MAP(sched_map, LD_WAIT_CONFIG); + unsigned int task_flags; - if (user) { - retval = security_task_setscheduler(p); - if (retval) - return retval; - } + /* + * Establish LD_WAIT_CONFIG context to ensure none of the code called + * will use a blocking primitive -- which would lead to recursion. + */ + lock_map_acquire_try(&sched_map); + task_flags = tsk->flags; /* - * make sure no PI-waiters arrive (or leave) while we are - * changing the priority of the task: - * - * To be able to change p->policy safely, the appropriate - * runqueue lock must be held. + * If a worker goes to sleep, notify and ask workqueue whether it + * wants to wake up a task to maintain concurrency. */ - rq = task_rq_lock(p, &flags); + if (task_flags & PF_WQ_WORKER) + wq_worker_sleeping(tsk); + else if (task_flags & PF_IO_WORKER) + io_wq_worker_sleeping(tsk); /* - * Changing the policy of the stop threads its a very bad idea + * spinlock and rwlock must not flush block requests. This will + * deadlock if the callback attempts to acquire a lock which is + * already acquired. */ - if (p == rq->stop) { - task_rq_unlock(rq, p, &flags); - return -EINVAL; - } + WARN_ON_ONCE(current->__state & TASK_RTLOCK_WAIT); /* - * If not changing anything there's no need to proceed further: + * If we are going to sleep and we have plugged IO queued, + * make sure to submit it to avoid deadlocks. */ - if (unlikely(policy == p->policy && (!rt_policy(policy) || - param->sched_priority == p->rt_priority))) { - task_rq_unlock(rq, p, &flags); - return 0; - } + blk_flush_plug(tsk->plug, true); -#ifdef CONFIG_RT_GROUP_SCHED - if (user) { - /* - * Do not allow realtime tasks into groups that have no runtime - * assigned. - */ - if (rt_bandwidth_enabled() && rt_policy(policy) && - task_group(p)->rt_bandwidth.rt_runtime == 0 && - !task_group_is_autogroup(task_group(p))) { - task_rq_unlock(rq, p, &flags); - return -EPERM; - } - } -#endif + lock_map_release(&sched_map); +} - /* recheck policy now with rq lock held */ - if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { - policy = oldpolicy = -1; - task_rq_unlock(rq, p, &flags); - goto recheck; +static void sched_update_worker(struct task_struct *tsk) +{ + if (tsk->flags & (PF_WQ_WORKER | PF_IO_WORKER | PF_BLOCK_TS)) { + if (tsk->flags & PF_BLOCK_TS) + blk_plug_invalidate_ts(tsk); + if (tsk->flags & PF_WQ_WORKER) + wq_worker_running(tsk); + else if (tsk->flags & PF_IO_WORKER) + io_wq_worker_running(tsk); } - on_rq = p->on_rq; - running = task_current(rq, p); - if (on_rq) - dequeue_task(rq, p, 0); - if (running) - p->sched_class->put_prev_task(rq, p); - - p->sched_reset_on_fork = reset_on_fork; - - oldprio = p->prio; - prev_class = p->sched_class; - __setscheduler(rq, p, policy, param->sched_priority); +} - if (running) - p->sched_class->set_curr_task(rq); - if (on_rq) - enqueue_task(rq, p, 0); +static __always_inline void __schedule_loop(int sched_mode) +{ + do { + preempt_disable(); + __schedule(sched_mode); + sched_preempt_enable_no_resched(); + } while (need_resched()); +} - check_class_changed(rq, p, prev_class, oldprio); - task_rq_unlock(rq, p, &flags); +asmlinkage __visible void __sched schedule(void) +{ + struct task_struct *tsk = current; - rt_mutex_adjust_pi(p); +#ifdef CONFIG_RT_MUTEXES + lockdep_assert(!tsk->sched_rt_mutex); +#endif - return 0; + if (!task_is_running(tsk)) + sched_submit_work(tsk); + __schedule_loop(SM_NONE); + sched_update_worker(tsk); } +EXPORT_SYMBOL(schedule); -/** - * sched_setscheduler - change the scheduling policy and/or RT priority of a thread. - * @p: the task in question. - * @policy: new policy. - * @param: structure containing the new RT priority. +/* + * synchronize_rcu_tasks() makes sure that no task is stuck in preempted + * state (have scheduled out non-voluntarily) by making sure that all + * tasks have either left the run queue or have gone into user space. + * As idle tasks do not do either, they must not ever be preempted + * (schedule out non-voluntarily). * - * NOTE that the task may be already dead. + * schedule_idle() is similar to schedule_preempt_disable() except that it + * never enables preemption because it does not call sched_submit_work(). */ -int sched_setscheduler(struct task_struct *p, int policy, - const struct sched_param *param) +void __sched schedule_idle(void) { - return __sched_setscheduler(p, policy, param, true); + /* + * As this skips calling sched_submit_work(), which the idle task does + * regardless because that function is a NOP when the task is in a + * TASK_RUNNING state, make sure this isn't used someplace that the + * current task can be in any other state. Note, idle is always in the + * TASK_RUNNING state. + */ + WARN_ON_ONCE(current->__state); + do { + __schedule(SM_IDLE); + } while (need_resched()); } -EXPORT_SYMBOL_GPL(sched_setscheduler); + +#if defined(CONFIG_CONTEXT_TRACKING_USER) && !defined(CONFIG_HAVE_CONTEXT_TRACKING_USER_OFFSTACK) +asmlinkage __visible void __sched schedule_user(void) +{ + /* + * If we come here after a random call to set_need_resched(), + * or we have been woken up remotely but the IPI has not yet arrived, + * we haven't yet exited the RCU idle mode. Do it here manually until + * we find a better solution. + * + * NB: There are buggy callers of this function. Ideally we + * should warn if prev_state != CT_STATE_USER, but that will trigger + * too frequently to make sense yet. + */ + enum ctx_state prev_state = exception_enter(); + schedule(); + exception_exit(prev_state); +} +#endif /** - * sched_setscheduler_nocheck - change the scheduling policy and/or RT priority of a thread from kernelspace. - * @p: the task in question. - * @policy: new policy. - * @param: structure containing the new RT priority. + * schedule_preempt_disabled - called with preemption disabled * - * Just like sched_setscheduler, only don't bother checking if the - * current context has permission. For example, this is needed in - * stop_machine(): we create temporary high priority worker threads, - * but our caller might not have that capability. + * Returns with preemption disabled. Note: preempt_count must be 1 */ -int sched_setscheduler_nocheck(struct task_struct *p, int policy, - const struct sched_param *param) +void __sched schedule_preempt_disabled(void) { - return __sched_setscheduler(p, policy, param, false); + sched_preempt_enable_no_resched(); + schedule(); + preempt_disable(); } -static int -do_sched_setscheduler(pid_t pid, int policy, struct sched_param __user *param) +#ifdef CONFIG_PREEMPT_RT +void __sched notrace schedule_rtlock(void) { - struct sched_param lparam; - struct task_struct *p; - int retval; - - if (!param || pid < 0) - return -EINVAL; - if (copy_from_user(&lparam, param, sizeof(struct sched_param))) - return -EFAULT; - - rcu_read_lock(); - retval = -ESRCH; - p = find_process_by_pid(pid); - if (p != NULL) - retval = sched_setscheduler(p, policy, &lparam); - rcu_read_unlock(); - - return retval; + __schedule_loop(SM_RTLOCK_WAIT); } +NOKPROBE_SYMBOL(schedule_rtlock); +#endif -/** - * sys_sched_setscheduler - set/change the scheduler policy and RT priority - * @pid: the pid in question. - * @policy: new policy. - * @param: structure containing the new RT priority. - */ -SYSCALL_DEFINE3(sched_setscheduler, pid_t, pid, int, policy, - struct sched_param __user *, param) +static void __sched notrace preempt_schedule_common(void) { - /* negative values for policy are not valid */ - if (policy < 0) - return -EINVAL; + do { + /* + * Because the function tracer can trace preempt_count_sub() + * and it also uses preempt_enable/disable_notrace(), if + * NEED_RESCHED is set, the preempt_enable_notrace() called + * by the function tracer will call this function again and + * cause infinite recursion. + * + * Preemption must be disabled here before the function + * tracer can trace. Break up preempt_disable() into two + * calls. One to disable preemption without fear of being + * traced. The other to still record the preemption latency, + * which can also be traced by the function tracer. + */ + preempt_disable_notrace(); + preempt_latency_start(1); + __schedule(SM_PREEMPT); + preempt_latency_stop(1); + preempt_enable_no_resched_notrace(); - return do_sched_setscheduler(pid, policy, param); + /* + * Check again in case we missed a preemption opportunity + * between schedule and now. + */ + } while (need_resched()); } -/** - * sys_sched_setparam - set/change the RT priority of a thread - * @pid: the pid in question. - * @param: structure containing the new RT priority. +#ifdef CONFIG_PREEMPTION +/* + * This is the entry point to schedule() from in-kernel preemption + * off of preempt_enable. */ -SYSCALL_DEFINE2(sched_setparam, pid_t, pid, struct sched_param __user *, param) +asmlinkage __visible void __sched notrace preempt_schedule(void) { - return do_sched_setscheduler(pid, -1, param); + /* + * If there is a non-zero preempt_count or interrupts are disabled, + * we do not want to preempt the current task. Just return.. + */ + if (likely(!preemptible())) + return; + preempt_schedule_common(); +} +NOKPROBE_SYMBOL(preempt_schedule); +EXPORT_SYMBOL(preempt_schedule); + +#ifdef CONFIG_PREEMPT_DYNAMIC +# 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) +static DEFINE_STATIC_KEY_TRUE(sk_dynamic_preempt_schedule); +void __sched notrace dynamic_preempt_schedule(void) +{ + if (!static_branch_unlikely(&sk_dynamic_preempt_schedule)) + return; + preempt_schedule(); } +NOKPROBE_SYMBOL(dynamic_preempt_schedule); +EXPORT_SYMBOL(dynamic_preempt_schedule); +# endif +#endif /* CONFIG_PREEMPT_DYNAMIC */ /** - * sys_sched_getscheduler - get the policy (scheduling class) of a thread - * @pid: the pid in question. + * preempt_schedule_notrace - preempt_schedule called by tracing + * + * The tracing infrastructure uses preempt_enable_notrace to prevent + * recursion and tracing preempt enabling caused by the tracing + * infrastructure itself. But as tracing can happen in areas coming + * from userspace or just about to enter userspace, a preempt enable + * can occur before user_exit() is called. This will cause the scheduler + * to be called when the system is still in usermode. + * + * To prevent this, the preempt_enable_notrace will use this function + * instead of preempt_schedule() to exit user context if needed before + * calling the scheduler. */ -SYSCALL_DEFINE1(sched_getscheduler, pid_t, pid) +asmlinkage __visible void __sched notrace preempt_schedule_notrace(void) { - struct task_struct *p; - int retval; + enum ctx_state prev_ctx; - if (pid < 0) - return -EINVAL; + if (likely(!preemptible())) + return; - retval = -ESRCH; - rcu_read_lock(); - p = find_process_by_pid(pid); - if (p) { - retval = security_task_getscheduler(p); - if (!retval) - retval = p->policy - | (p->sched_reset_on_fork ? SCHED_RESET_ON_FORK : 0); - } - rcu_read_unlock(); - return retval; + do { + /* + * Because the function tracer can trace preempt_count_sub() + * and it also uses preempt_enable/disable_notrace(), if + * NEED_RESCHED is set, the preempt_enable_notrace() called + * by the function tracer will call this function again and + * cause infinite recursion. + * + * Preemption must be disabled here before the function + * tracer can trace. Break up preempt_disable() into two + * calls. One to disable preemption without fear of being + * traced. The other to still record the preemption latency, + * which can also be traced by the function tracer. + */ + preempt_disable_notrace(); + preempt_latency_start(1); + /* + * Needs preempt disabled in case user_exit() is traced + * and the tracer calls preempt_enable_notrace() causing + * an infinite recursion. + */ + prev_ctx = exception_enter(); + __schedule(SM_PREEMPT); + exception_exit(prev_ctx); + + preempt_latency_stop(1); + preempt_enable_no_resched_notrace(); + } while (need_resched()); +} +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 +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) +static DEFINE_STATIC_KEY_TRUE(sk_dynamic_preempt_schedule_notrace); +void __sched notrace dynamic_preempt_schedule_notrace(void) +{ + if (!static_branch_unlikely(&sk_dynamic_preempt_schedule_notrace)) + return; + preempt_schedule_notrace(); } +NOKPROBE_SYMBOL(dynamic_preempt_schedule_notrace); +EXPORT_SYMBOL(dynamic_preempt_schedule_notrace); +# endif +#endif -/** - * sys_sched_getparam - get the RT priority of a thread - * @pid: the pid in question. - * @param: structure containing the RT priority. +#endif /* CONFIG_PREEMPTION */ + +/* + * This is the entry point to schedule() from kernel preemption + * off of IRQ context. + * Note, that this is called and return with IRQs disabled. This will + * protect us against recursive calling from IRQ contexts. */ -SYSCALL_DEFINE2(sched_getparam, pid_t, pid, struct sched_param __user *, param) +asmlinkage __visible void __sched preempt_schedule_irq(void) { - struct sched_param lp; - struct task_struct *p; - int retval; - - if (!param || pid < 0) - return -EINVAL; - - rcu_read_lock(); - p = find_process_by_pid(pid); - retval = -ESRCH; - if (!p) - goto out_unlock; - - retval = security_task_getscheduler(p); - if (retval) - goto out_unlock; + enum ctx_state prev_state; - lp.sched_priority = p->rt_priority; - rcu_read_unlock(); + /* Catch callers which need to be fixed */ + BUG_ON(preempt_count() || !irqs_disabled()); - /* - * This one might sleep, we cannot do it with a spinlock held ... - */ - retval = copy_to_user(param, &lp, sizeof(*param)) ? -EFAULT : 0; + prev_state = exception_enter(); - return retval; + do { + preempt_disable(); + local_irq_enable(); + __schedule(SM_PREEMPT); + local_irq_disable(); + sched_preempt_enable_no_resched(); + } while (need_resched()); -out_unlock: - rcu_read_unlock(); - return retval; + exception_exit(prev_state); } -long sched_setaffinity(pid_t pid, const struct cpumask *in_mask) +int default_wake_function(wait_queue_entry_t *curr, unsigned mode, int wake_flags, + void *key) { - cpumask_var_t cpus_allowed, new_mask; - struct task_struct *p; - int retval; + 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); - get_online_cpus(); - rcu_read_lock(); +const struct sched_class *__setscheduler_class(int policy, int prio) +{ + if (dl_prio(prio)) + return &dl_sched_class; - p = find_process_by_pid(pid); - if (!p) { - rcu_read_unlock(); - put_online_cpus(); - return -ESRCH; - } + if (rt_prio(prio)) + return &rt_sched_class; - /* Prevent p going away */ - get_task_struct(p); - rcu_read_unlock(); +#ifdef CONFIG_SCHED_CLASS_EXT + if (task_should_scx(policy)) + return &ext_sched_class; +#endif - if (p->flags & PF_NO_SETAFFINITY) { - retval = -EINVAL; - goto out_put_task; - } - if (!alloc_cpumask_var(&cpus_allowed, GFP_KERNEL)) { - retval = -ENOMEM; - goto out_put_task; - } - if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) { - retval = -ENOMEM; - goto out_free_cpus_allowed; - } - retval = -EPERM; - if (!check_same_owner(p)) { - rcu_read_lock(); - if (!ns_capable(__task_cred(p)->user_ns, CAP_SYS_NICE)) { - rcu_read_unlock(); - goto out_unlock; - } - rcu_read_unlock(); - } + return &fair_sched_class; +} - retval = security_task_setscheduler(p); - if (retval) - goto out_unlock; +#ifdef CONFIG_RT_MUTEXES - cpuset_cpus_allowed(p, cpus_allowed); - cpumask_and(new_mask, in_mask, cpus_allowed); -again: - retval = set_cpus_allowed_ptr(p, new_mask); +/* + * Would be more useful with typeof()/auto_type but they don't mix with + * bit-fields. Since it's a local thing, use int. Keep the generic sounding + * name such that if someone were to implement this function we get to compare + * notes. + */ +#define fetch_and_set(x, v) ({ int _x = (x); (x) = (v); _x; }) - if (!retval) { - cpuset_cpus_allowed(p, cpus_allowed); - if (!cpumask_subset(new_mask, cpus_allowed)) { - /* - * We must have raced with a concurrent cpuset - * update. Just reset the cpus_allowed to the - * cpuset's cpus_allowed - */ - cpumask_copy(new_mask, cpus_allowed); - goto again; - } - } -out_unlock: - free_cpumask_var(new_mask); -out_free_cpus_allowed: - free_cpumask_var(cpus_allowed); -out_put_task: - put_task_struct(p); - put_online_cpus(); - return retval; +void rt_mutex_pre_schedule(void) +{ + lockdep_assert(!fetch_and_set(current->sched_rt_mutex, 1)); + sched_submit_work(current); } -static int get_user_cpu_mask(unsigned long __user *user_mask_ptr, unsigned len, - struct cpumask *new_mask) +void rt_mutex_schedule(void) { - if (len < cpumask_size()) - cpumask_clear(new_mask); - else if (len > cpumask_size()) - len = cpumask_size(); + lockdep_assert(current->sched_rt_mutex); + __schedule_loop(SM_NONE); +} - return copy_from_user(new_mask, user_mask_ptr, len) ? -EFAULT : 0; +void rt_mutex_post_schedule(void) +{ + sched_update_worker(current); + lockdep_assert(fetch_and_set(current->sched_rt_mutex, 0)); } -/** - * sys_sched_setaffinity - set the cpu affinity of a process - * @pid: pid of the process - * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to the new cpu mask +/* + * rt_mutex_setprio - set the current priority of a task + * @p: task to boost + * @pi_task: donor task + * + * This function changes the 'effective' priority of a task. It does + * not touch ->normal_prio like __setscheduler(). + * + * Used by the rt_mutex code to implement priority inheritance + * logic. Call site only calls if the priority of the task changed. */ -SYSCALL_DEFINE3(sched_setaffinity, pid_t, pid, unsigned int, len, - unsigned long __user *, user_mask_ptr) +void rt_mutex_setprio(struct task_struct *p, struct task_struct *pi_task) { - cpumask_var_t new_mask; - int retval; - - if (!alloc_cpumask_var(&new_mask, GFP_KERNEL)) - return -ENOMEM; + int prio, oldprio, queue_flag = + DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; + const struct sched_class *prev_class, *next_class; + struct rq_flags rf; + struct rq *rq; - retval = get_user_cpu_mask(user_mask_ptr, len, new_mask); - if (retval == 0) - retval = sched_setaffinity(pid, new_mask); - free_cpumask_var(new_mask); - return retval; -} + /* XXX used to be waiter->prio, not waiter->task->prio */ + prio = __rt_effective_prio(pi_task, p->normal_prio); -long sched_getaffinity(pid_t pid, struct cpumask *mask) -{ - struct task_struct *p; - unsigned long flags; - int retval; + /* + * If nothing changed; bail early. + */ + if (p->pi_top_task == pi_task && prio == p->prio && !dl_prio(prio)) + return; - get_online_cpus(); - rcu_read_lock(); + rq = __task_rq_lock(p, &rf); + update_rq_clock(rq); + /* + * Set under pi_lock && rq->lock, such that the value can be used under + * either lock. + * + * Note that there is loads of tricky to make this pointer cache work + * right. rt_mutex_slowunlock()+rt_mutex_postunlock() work together to + * ensure a task is de-boosted (pi_task is set to NULL) before the + * task is allowed to run again (and can exit). This ensures the pointer + * points to a blocked task -- which guarantees the task is present. + */ + p->pi_top_task = pi_task; - retval = -ESRCH; - p = find_process_by_pid(pid); - if (!p) + /* + * For FIFO/RR we only need to set prio, if that matches we're done. + */ + if (prio == p->prio && !dl_prio(prio)) goto out_unlock; - retval = security_task_getscheduler(p); - if (retval) + /* + * Idle task boosting is a no-no in general. There is one + * exception, when PREEMPT_RT and NOHZ is active: + * + * The idle task calls get_next_timer_interrupt() and holds + * the timer wheel base->lock on the CPU and another CPU wants + * to access the timer (probably to cancel it). We can safely + * ignore the boosting request, as the idle CPU runs this code + * with interrupts disabled and will complete the lock + * protected section without being interrupted. So there is no + * real need to boost. + */ + if (unlikely(p == rq->idle)) { + WARN_ON(p != rq->curr); + WARN_ON(p->pi_blocked_on); goto out_unlock; + } - raw_spin_lock_irqsave(&p->pi_lock, flags); - cpumask_and(mask, &p->cpus_allowed, cpu_online_mask); - raw_spin_unlock_irqrestore(&p->pi_lock, flags); - -out_unlock: - rcu_read_unlock(); - put_online_cpus(); - - return retval; -} + trace_sched_pi_setprio(p, pi_task); + oldprio = p->prio; -/** - * sys_sched_getaffinity - get the cpu affinity of a process - * @pid: pid of the process - * @len: length in bytes of the bitmask pointed to by user_mask_ptr - * @user_mask_ptr: user-space pointer to hold the current cpu mask - */ -SYSCALL_DEFINE3(sched_getaffinity, pid_t, pid, unsigned int, len, - unsigned long __user *, user_mask_ptr) -{ - int ret; - cpumask_var_t mask; + if (oldprio == prio) + queue_flag &= ~DEQUEUE_MOVE; - if ((len * BITS_PER_BYTE) < nr_cpu_ids) - return -EINVAL; - if (len & (sizeof(unsigned long)-1)) - return -EINVAL; + prev_class = p->sched_class; + next_class = __setscheduler_class(p->policy, prio); - if (!alloc_cpumask_var(&mask, GFP_KERNEL)) - return -ENOMEM; + if (prev_class != next_class) + queue_flag |= DEQUEUE_CLASS; - ret = sched_getaffinity(pid, mask); - if (ret == 0) { - size_t retlen = min_t(size_t, len, cpumask_size()); + 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 { + if (dl_prio(oldprio)) + p->dl.pi_se = &p->dl; + if (rt_prio(oldprio)) + p->rt.timeout = 0; + } - if (copy_to_user(user_mask_ptr, mask, retlen)) - ret = -EFAULT; - else - ret = retlen; + p->sched_class = next_class; + p->prio = prio; } - free_cpumask_var(mask); +out_unlock: + /* Caller holds task_struct::pi_lock, IRQs are still disabled */ - return ret; + rq_unpin_lock(rq, &rf); + __balance_callbacks(rq); + rq_repin_lock(rq, &rf); + __task_rq_unlock(rq, p, &rf); } +#endif /* CONFIG_RT_MUTEXES */ -/** - * sys_sched_yield - yield the current processor to other threads. - * - * This function yields the current CPU to other tasks. If there are no - * other threads running on this CPU then this function will return. - */ -SYSCALL_DEFINE0(sched_yield) +#if !defined(CONFIG_PREEMPTION) || defined(CONFIG_PREEMPT_DYNAMIC) +int __sched __cond_resched(void) { - struct rq *rq = this_rq_lock(); - - schedstat_inc(rq, yld_count); - current->sched_class->yield_task(rq); - + if (should_resched(0) && !irqs_disabled()) { + preempt_schedule_common(); + return 1; + } /* - * Since we are going to call schedule() anyway, there's - * no need to preempt or enable interrupts: + * 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, + * RCU readers leave no in-memory hints, which means that CPU-bound + * processes executing in kernel context might never report an + * 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(). */ - __release(rq->lock); - spin_release(&rq->lock.dep_map, 1, _THIS_IP_); - do_raw_spin_unlock(&rq->lock); - sched_preempt_enable_no_resched(); - - schedule(); - +#ifndef CONFIG_PREEMPT_RCU + rcu_all_qs(); +#endif return 0; } +EXPORT_SYMBOL(__cond_resched); +#endif -static inline int should_resched(void) -{ - return need_resched() && !(preempt_count() & PREEMPT_ACTIVE); -} - -static void __cond_resched(void) -{ - add_preempt_count(PREEMPT_ACTIVE); - __schedule(); - sub_preempt_count(PREEMPT_ACTIVE); +#ifdef CONFIG_PREEMPT_DYNAMIC +# 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_STATIC_CALL_RET0(might_resched, __cond_resched); +EXPORT_STATIC_CALL_TRAMP(might_resched); +# elif defined(CONFIG_HAVE_PREEMPT_DYNAMIC_KEY) +static DEFINE_STATIC_KEY_FALSE(sk_dynamic_cond_resched); +int __sched dynamic_cond_resched(void) +{ + if (!static_branch_unlikely(&sk_dynamic_cond_resched)) + return 0; + return __cond_resched(); } +EXPORT_SYMBOL(dynamic_cond_resched); -int __sched _cond_resched(void) +static DEFINE_STATIC_KEY_FALSE(sk_dynamic_might_resched); +int __sched dynamic_might_resched(void) { - if (should_resched()) { - __cond_resched(); - return 1; - } - return 0; + if (!static_branch_unlikely(&sk_dynamic_might_resched)) + return 0; + return __cond_resched(); } -EXPORT_SYMBOL(_cond_resched); +EXPORT_SYMBOL(dynamic_might_resched); +# endif +#endif /* CONFIG_PREEMPT_DYNAMIC */ /* * __cond_resched_lock() - if a reschedule is pending, drop the given lock, * call schedule, and on return reacquire the lock. * - * This works OK both with and without CONFIG_PREEMPT. We do strange low-level + * This works OK both with and without CONFIG_PREEMPTION. We do strange low-level * operations here to prevent schedule() from being called twice (once via * spin_unlock(), once by hand). */ int __cond_resched_lock(spinlock_t *lock) { - int resched = should_resched(); + int resched = should_resched(PREEMPT_LOCK_OFFSET); int ret = 0; lockdep_assert_held(lock); if (spin_needbreak(lock) || resched) { spin_unlock(lock); - if (resched) - __cond_resched(); - else + if (!_cond_resched()) cpu_relax(); ret = 1; spin_lock(lock); @@ -3816,315 +7455,418 @@ int __cond_resched_lock(spinlock_t *lock) } EXPORT_SYMBOL(__cond_resched_lock); -int __sched __cond_resched_softirq(void) +int __cond_resched_rwlock_read(rwlock_t *lock) { - BUG_ON(!in_softirq()); + int resched = should_resched(PREEMPT_LOCK_OFFSET); + int ret = 0; - if (should_resched()) { - local_bh_enable(); - __cond_resched(); - local_bh_disable(); - return 1; + lockdep_assert_held_read(lock); + + if (rwlock_needbreak(lock) || resched) { + read_unlock(lock); + if (!_cond_resched()) + cpu_relax(); + ret = 1; + read_lock(lock); } - return 0; + return ret; } -EXPORT_SYMBOL(__cond_resched_softirq); +EXPORT_SYMBOL(__cond_resched_rwlock_read); -/** - * yield - yield the current processor to other threads. - * - * Do not ever use this function, there's a 99% chance you're doing it wrong. +int __cond_resched_rwlock_write(rwlock_t *lock) +{ + int resched = should_resched(PREEMPT_LOCK_OFFSET); + int ret = 0; + + lockdep_assert_held_write(lock); + + if (rwlock_needbreak(lock) || resched) { + write_unlock(lock); + if (!_cond_resched()) + cpu_relax(); + ret = 1; + write_lock(lock); + } + return ret; +} +EXPORT_SYMBOL(__cond_resched_rwlock_write); + +#ifdef CONFIG_PREEMPT_DYNAMIC + +# ifdef CONFIG_GENERIC_IRQ_ENTRY +# include <linux/irq-entry-common.h> +# endif + +/* + * SC:cond_resched + * SC:might_resched + * SC:preempt_schedule + * SC:preempt_schedule_notrace + * SC:irqentry_exit_cond_resched * - * The scheduler is at all times free to pick the calling task as the most - * eligible task to run, if removing the yield() call from your code breaks - * it, its already broken. * - * Typical broken usage is: + * NONE: + * cond_resched <- __cond_resched + * might_resched <- RET0 + * preempt_schedule <- NOP + * preempt_schedule_notrace <- NOP + * irqentry_exit_cond_resched <- NOP + * dynamic_preempt_lazy <- false * - * while (!event) - * yield(); + * VOLUNTARY: + * cond_resched <- __cond_resched + * might_resched <- __cond_resched + * preempt_schedule <- NOP + * preempt_schedule_notrace <- NOP + * irqentry_exit_cond_resched <- NOP + * dynamic_preempt_lazy <- false * - * where one assumes that yield() will let 'the other' process run that will - * make event true. If the current task is a SCHED_FIFO task that will never - * happen. Never use yield() as a progress guarantee!! + * FULL: + * cond_resched <- RET0 + * might_resched <- RET0 + * preempt_schedule <- preempt_schedule + * preempt_schedule_notrace <- preempt_schedule_notrace + * irqentry_exit_cond_resched <- irqentry_exit_cond_resched + * dynamic_preempt_lazy <- false * - * If you want to use yield() to wait for something, use wait_event(). - * If you want to use yield() to be 'nice' for others, use cond_resched(). - * If you still want to use yield(), do not! - */ -void __sched yield(void) + * LAZY: + * cond_resched <- RET0 + * might_resched <- RET0 + * preempt_schedule <- preempt_schedule + * preempt_schedule_notrace <- preempt_schedule_notrace + * irqentry_exit_cond_resched <- irqentry_exit_cond_resched + * dynamic_preempt_lazy <- true + */ + +enum { + preempt_dynamic_undefined = -1, + preempt_dynamic_none, + preempt_dynamic_voluntary, + preempt_dynamic_full, + preempt_dynamic_lazy, +}; + +int preempt_dynamic_mode = preempt_dynamic_undefined; + +int sched_dynamic_mode(const char *str) { - set_current_state(TASK_RUNNING); - sys_sched_yield(); +# ifndef CONFIG_PREEMPT_RT + if (!strcmp(str, "none")) + return preempt_dynamic_none; + + if (!strcmp(str, "voluntary")) + return preempt_dynamic_voluntary; +# endif + + if (!strcmp(str, "full")) + return preempt_dynamic_full; + +# ifdef CONFIG_ARCH_HAS_PREEMPT_LAZY + if (!strcmp(str, "lazy")) + return preempt_dynamic_lazy; +# endif + + return -EINVAL; } -EXPORT_SYMBOL(yield); -/** - * yield_to - yield the current processor to another thread in - * your thread group, or accelerate that thread toward the - * processor it's on. - * @p: target task - * @preempt: whether task preemption is allowed or not - * - * It's the caller's job to ensure that the target task struct - * can't go away on us before we can do any checks. - * - * Returns: - * true (>0) if we indeed boosted the target task. - * false (0) if we failed to boost the target. - * -ESRCH if there's no task to yield to. - */ -bool __sched yield_to(struct task_struct *p, bool preempt) -{ - struct task_struct *curr = current; - struct rq *rq, *p_rq; - unsigned long flags; - int yielded = 0; +# 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) - local_irq_save(flags); - rq = this_rq(); +# 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 -again: - p_rq = task_rq(p); +static DEFINE_MUTEX(sched_dynamic_mutex); + +static void __sched_dynamic_update(int mode) +{ /* - * If we're the only runnable task on the rq and target rq also - * has only one task, there's absolutely no point in yielding. + * Avoid {NONE,VOLUNTARY} -> FULL transitions from ever ending up in + * the ZERO state, which is invalid. */ - if (rq->nr_running == 1 && p_rq->nr_running == 1) { - yielded = -ESRCH; - goto out_irq; - } + preempt_dynamic_enable(cond_resched); + preempt_dynamic_enable(might_resched); + preempt_dynamic_enable(preempt_schedule); + preempt_dynamic_enable(preempt_schedule_notrace); + preempt_dynamic_enable(irqentry_exit_cond_resched); + preempt_dynamic_key_disable(preempt_lazy); + + switch (mode) { + case preempt_dynamic_none: + preempt_dynamic_enable(cond_resched); + preempt_dynamic_disable(might_resched); + preempt_dynamic_disable(preempt_schedule); + preempt_dynamic_disable(preempt_schedule_notrace); + preempt_dynamic_disable(irqentry_exit_cond_resched); + preempt_dynamic_key_disable(preempt_lazy); + if (mode != preempt_dynamic_mode) + pr_info("Dynamic Preempt: none\n"); + break; - double_rq_lock(rq, p_rq); - while (task_rq(p) != p_rq) { - double_rq_unlock(rq, p_rq); - goto again; + case preempt_dynamic_voluntary: + preempt_dynamic_enable(cond_resched); + preempt_dynamic_enable(might_resched); + preempt_dynamic_disable(preempt_schedule); + preempt_dynamic_disable(preempt_schedule_notrace); + preempt_dynamic_disable(irqentry_exit_cond_resched); + preempt_dynamic_key_disable(preempt_lazy); + if (mode != preempt_dynamic_mode) + pr_info("Dynamic Preempt: voluntary\n"); + break; + + case preempt_dynamic_full: + preempt_dynamic_disable(cond_resched); + preempt_dynamic_disable(might_resched); + preempt_dynamic_enable(preempt_schedule); + preempt_dynamic_enable(preempt_schedule_notrace); + preempt_dynamic_enable(irqentry_exit_cond_resched); + preempt_dynamic_key_disable(preempt_lazy); + if (mode != preempt_dynamic_mode) + pr_info("Dynamic Preempt: full\n"); + break; + + case preempt_dynamic_lazy: + preempt_dynamic_disable(cond_resched); + preempt_dynamic_disable(might_resched); + preempt_dynamic_enable(preempt_schedule); + preempt_dynamic_enable(preempt_schedule_notrace); + preempt_dynamic_enable(irqentry_exit_cond_resched); + preempt_dynamic_key_enable(preempt_lazy); + if (mode != preempt_dynamic_mode) + pr_info("Dynamic Preempt: lazy\n"); + break; } - if (!curr->sched_class->yield_to_task) - goto out_unlock; + preempt_dynamic_mode = mode; +} - if (curr->sched_class != p->sched_class) - goto out_unlock; +void sched_dynamic_update(int mode) +{ + mutex_lock(&sched_dynamic_mutex); + __sched_dynamic_update(mode); + mutex_unlock(&sched_dynamic_mutex); +} - if (task_running(p_rq, p) || p->state) - goto out_unlock; +static int __init setup_preempt_mode(char *str) +{ + int mode = sched_dynamic_mode(str); + if (mode < 0) { + pr_warn("Dynamic Preempt: unsupported mode: %s\n", str); + return 0; + } - yielded = curr->sched_class->yield_to_task(rq, p, preempt); - if (yielded) { - schedstat_inc(rq, yld_count); - /* - * Make p's CPU reschedule; pick_next_entity takes care of - * fairness. - */ - if (preempt && rq != p_rq) - resched_task(p_rq->curr); + sched_dynamic_update(mode); + return 1; +} +__setup("preempt=", setup_preempt_mode); + +static void __init preempt_dynamic_init(void) +{ + if (preempt_dynamic_mode == preempt_dynamic_undefined) { + if (IS_ENABLED(CONFIG_PREEMPT_NONE)) { + sched_dynamic_update(preempt_dynamic_none); + } else if (IS_ENABLED(CONFIG_PREEMPT_VOLUNTARY)) { + sched_dynamic_update(preempt_dynamic_voluntary); + } else if (IS_ENABLED(CONFIG_PREEMPT_LAZY)) { + sched_dynamic_update(preempt_dynamic_lazy); + } else { + /* Default static call setting, nothing to do */ + WARN_ON_ONCE(!IS_ENABLED(CONFIG_PREEMPT)); + preempt_dynamic_mode = preempt_dynamic_full; + pr_info("Dynamic Preempt: full\n"); + } } +} -out_unlock: - double_rq_unlock(rq, p_rq); -out_irq: - local_irq_restore(flags); +# define PREEMPT_MODEL_ACCESSOR(mode) \ + bool preempt_model_##mode(void) \ + { \ + WARN_ON_ONCE(preempt_dynamic_mode == preempt_dynamic_undefined); \ + return preempt_dynamic_mode == preempt_dynamic_##mode; \ + } \ + EXPORT_SYMBOL_GPL(preempt_model_##mode) - if (yielded > 0) - schedule(); +PREEMPT_MODEL_ACCESSOR(none); +PREEMPT_MODEL_ACCESSOR(voluntary); +PREEMPT_MODEL_ACCESSOR(full); +PREEMPT_MODEL_ACCESSOR(lazy); - return yielded; -} -EXPORT_SYMBOL_GPL(yield_to); +#else /* !CONFIG_PREEMPT_DYNAMIC: */ -/* - * This task is about to go to sleep on IO. Increment rq->nr_iowait so - * that process accounting knows that this is a task in IO wait state. - */ -void __sched io_schedule(void) +#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) { - struct rq *rq = raw_rq(); + bool brace = IS_ENABLED(CONFIG_PREEMPT_RT) && + (IS_ENABLED(CONFIG_PREEMPT_DYNAMIC) || + IS_ENABLED(CONFIG_PREEMPT_LAZY)); + static char buf[128]; - delayacct_blkio_start(); - atomic_inc(&rq->nr_iowait); - blk_flush_plug(current); - current->in_iowait = 1; - schedule(); - current->in_iowait = 0; - atomic_dec(&rq->nr_iowait); - delayacct_blkio_end(); + 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"; } -EXPORT_SYMBOL(io_schedule); -long __sched io_schedule_timeout(long timeout) +int io_schedule_prepare(void) { - struct rq *rq = raw_rq(); - long ret; + int old_iowait = current->in_iowait; - delayacct_blkio_start(); - atomic_inc(&rq->nr_iowait); - blk_flush_plug(current); current->in_iowait = 1; - ret = schedule_timeout(timeout); - current->in_iowait = 0; - atomic_dec(&rq->nr_iowait); - delayacct_blkio_end(); - return ret; + blk_flush_plug(current->plug, true); + return old_iowait; } -/** - * sys_sched_get_priority_max - return maximum RT priority. - * @policy: scheduling class. - * - * this syscall returns the maximum rt_priority that can be used - * by a given scheduling class. - */ -SYSCALL_DEFINE1(sched_get_priority_max, int, policy) +void io_schedule_finish(int token) { - int ret = -EINVAL; - - switch (policy) { - case SCHED_FIFO: - case SCHED_RR: - ret = MAX_USER_RT_PRIO-1; - break; - case SCHED_NORMAL: - case SCHED_BATCH: - case SCHED_IDLE: - ret = 0; - break; - } - return ret; + current->in_iowait = token; } -/** - * sys_sched_get_priority_min - return minimum RT priority. - * @policy: scheduling class. - * - * this syscall returns the minimum rt_priority that can be used - * by a given scheduling class. +/* + * This task is about to go to sleep on IO. Increment rq->nr_iowait so + * that process accounting knows that this is a task in IO wait state. */ -SYSCALL_DEFINE1(sched_get_priority_min, int, policy) +long __sched io_schedule_timeout(long timeout) { - int ret = -EINVAL; + int token; + long ret; + + token = io_schedule_prepare(); + ret = schedule_timeout(timeout); + io_schedule_finish(token); - switch (policy) { - case SCHED_FIFO: - case SCHED_RR: - ret = 1; - break; - case SCHED_NORMAL: - case SCHED_BATCH: - case SCHED_IDLE: - ret = 0; - } return ret; } +EXPORT_SYMBOL(io_schedule_timeout); -/** - * sys_sched_rr_get_interval - return the default timeslice of a process. - * @pid: pid of the process. - * @interval: userspace pointer to the timeslice value. - * - * this syscall writes the default timeslice value of a given process - * into the user-space timespec buffer. A value of '0' means infinity. - */ -SYSCALL_DEFINE2(sched_rr_get_interval, pid_t, pid, - struct timespec __user *, interval) +void __sched io_schedule(void) { - struct task_struct *p; - unsigned int time_slice; - unsigned long flags; - struct rq *rq; - int retval; - struct timespec t; - - if (pid < 0) - return -EINVAL; - - retval = -ESRCH; - rcu_read_lock(); - p = find_process_by_pid(pid); - if (!p) - goto out_unlock; + int token; - retval = security_task_getscheduler(p); - if (retval) - goto out_unlock; - - rq = task_rq_lock(p, &flags); - time_slice = p->sched_class->get_rr_interval(rq, p); - task_rq_unlock(rq, p, &flags); - - rcu_read_unlock(); - jiffies_to_timespec(time_slice, &t); - retval = copy_to_user(interval, &t, sizeof(t)) ? -EFAULT : 0; - return retval; - -out_unlock: - rcu_read_unlock(); - return retval; + token = io_schedule_prepare(); + schedule(); + io_schedule_finish(token); } - -static const char stat_nam[] = TASK_STATE_TO_CHAR_STR; +EXPORT_SYMBOL(io_schedule); void sched_show_task(struct task_struct *p) { - unsigned long free = 0; + unsigned long free; int ppid; - unsigned state; - state = p->state ? __ffs(p->state) + 1 : 0; - printk(KERN_INFO "%-15.15s %c", p->comm, - state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'); -#if BITS_PER_LONG == 32 - if (state == TASK_RUNNING) - printk(KERN_CONT " running "); - else - printk(KERN_CONT " %08lx ", thread_saved_pc(p)); -#else - if (state == TASK_RUNNING) - printk(KERN_CONT " running task "); - else - printk(KERN_CONT " %016lx ", thread_saved_pc(p)); -#endif -#ifdef CONFIG_DEBUG_STACK_USAGE + if (!try_get_task_stack(p)) + return; + + pr_info("task:%-15.15s state:%c", p->comm, task_state_to_char(p)); + + if (task_is_running(p)) + pr_cont(" running task "); free = stack_not_used(p); -#endif + ppid = 0; rcu_read_lock(); - ppid = task_pid_nr(rcu_dereference(p->real_parent)); + if (pid_alive(p)) + ppid = task_pid_nr(rcu_dereference(p->real_parent)); rcu_read_unlock(); - printk(KERN_CONT "%5lu %5d %6d 0x%08lx\n", free, - task_pid_nr(p), ppid, - (unsigned long)task_thread_info(p)->flags); + pr_cont(" stack:%-5lu pid:%-5d tgid:%-5d ppid:%-6d task_flags:0x%04x flags:0x%08lx\n", + free, task_pid_nr(p), task_tgid_nr(p), + ppid, p->flags, read_task_thread_flags(p)); print_worker_info(KERN_INFO, p); - show_stack(p, NULL); + print_stop_info(KERN_INFO, p); + print_scx_info(KERN_INFO, p); + show_stack(p, NULL, KERN_INFO); + put_task_stack(p); +} +EXPORT_SYMBOL_GPL(sched_show_task); + +static inline bool +state_filter_match(unsigned long state_filter, struct task_struct *p) +{ + unsigned int state = READ_ONCE(p->__state); + + /* no filter, everything matches */ + if (!state_filter) + return true; + + /* filter, but doesn't match */ + if (!(state & state_filter)) + return false; + + /* + * When looking for TASK_UNINTERRUPTIBLE skip TASK_IDLE (allows + * TASK_KILLABLE). + */ + if (state_filter == TASK_UNINTERRUPTIBLE && (state & TASK_NOLOAD)) + return false; + + return true; } -void show_state_filter(unsigned long state_filter) + +void show_state_filter(unsigned int state_filter) { struct task_struct *g, *p; -#if BITS_PER_LONG == 32 - printk(KERN_INFO - " task PC stack pid father\n"); -#else - printk(KERN_INFO - " task PC stack pid father\n"); -#endif rcu_read_lock(); - do_each_thread(g, p) { + for_each_process_thread(g, p) { /* * reset the NMI-timeout, listing all files on a slow * console might take a lot of time: + * Also, reset softlockup watchdogs on all CPUs, because + * another CPU might be blocked waiting for us to process + * an IPI. */ touch_nmi_watchdog(); - if (!state_filter || (p->state & state_filter)) + touch_all_softlockup_watchdogs(); + if (state_filter_match(state_filter, p)) sched_show_task(p); - } while_each_thread(g, p); + } - touch_all_softlockup_watchdogs(); + if (!state_filter) + sysrq_sched_debug_show(); -#ifdef CONFIG_SCHED_DEBUG - sysrq_sched_debug_show(); -#endif rcu_read_unlock(); /* * Only show locks if all tasks are dumped: @@ -4133,34 +7875,43 @@ void show_state_filter(unsigned long state_filter) debug_show_all_locks(); } -void __cpuinit init_idle_bootup_task(struct task_struct *idle) -{ - idle->sched_class = &idle_sched_class; -} - /** * init_idle - set up an idle thread for a given CPU * @idle: task in question - * @cpu: cpu the idle task belongs to + * @cpu: CPU the idle task belongs to * * NOTE: this function does not set the idle thread's NEED_RESCHED * flag, to make booting more robust. */ -void __cpuinit init_idle(struct task_struct *idle, int cpu) +void __init init_idle(struct task_struct *idle, int cpu) { + struct affinity_context ac = (struct affinity_context) { + .new_mask = cpumask_of(cpu), + .flags = 0, + }; struct rq *rq = cpu_rq(cpu); unsigned long flags; - raw_spin_lock_irqsave(&rq->lock, flags); + raw_spin_lock_irqsave(&idle->pi_lock, flags); + raw_spin_rq_lock(rq); - __sched_fork(idle); - idle->state = TASK_RUNNING; + idle->__state = TASK_RUNNING; idle->se.exec_start = sched_clock(); + /* + * PF_KTHREAD should already be set at this point; regardless, make it + * look like a proper per-CPU kthread. + */ + idle->flags |= PF_KTHREAD | PF_NO_SETAFFINITY; + kthread_set_per_cpu(idle, cpu); - do_set_cpus_allowed(idle, cpumask_of(cpu)); + /* + * 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); /* * 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 + * holding rq->lock, the CPU isn't yet set to this CPU so the * lockdep check in task_group() will fail. * * Similar case to sched_fork(). / Alternatively we could @@ -4172,14 +7923,16 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu) __set_task_cpu(idle, cpu); rcu_read_unlock(); - rq->curr = rq->idle = idle; -#if defined(CONFIG_SMP) + rq->idle = idle; + rq_set_donor(rq, idle); + rcu_assign_pointer(rq->curr, idle); + idle->on_rq = TASK_ON_RQ_QUEUED; idle->on_cpu = 1; -#endif - raw_spin_unlock_irqrestore(&rq->lock, flags); + raw_spin_rq_unlock(rq); + raw_spin_unlock_irqrestore(&idle->pi_lock, flags); /* Set the preempt count _outside_ the spinlocks! */ - task_thread_info(idle)->preempt_count = 0; + init_idle_preempt_count(idle, cpu); /* * The idle tasks have their own, simple scheduling class: @@ -4187,416 +7940,241 @@ void __cpuinit 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); -#if defined(CONFIG_SMP) sprintf(idle->comm, "%s/%d", INIT_TASK_COMM, cpu); -#endif } -#ifdef CONFIG_SMP -void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask) +int cpuset_cpumask_can_shrink(const struct cpumask *cur, + const struct cpumask *trial) { - if (p->sched_class && p->sched_class->set_cpus_allowed) - p->sched_class->set_cpus_allowed(p, new_mask); + int ret = 1; - cpumask_copy(&p->cpus_allowed, new_mask); - p->nr_cpus_allowed = cpumask_weight(new_mask); -} + if (cpumask_empty(cur)) + return ret; -/* - * This is how migration works: - * - * 1) we invoke migration_cpu_stop() on the target CPU using - * stop_one_cpu(). - * 2) stopper starts to run (implicitly forcing the migrated thread - * off the CPU) - * 3) it checks whether the migrated task is still in the wrong runqueue. - * 4) if it's in the wrong runqueue then the migration thread removes - * it and puts it into the right queue. - * 5) stopper completes and stop_one_cpu() returns and the migration - * is done. - */ + ret = dl_cpuset_cpumask_can_shrink(cur, trial); -/* - * Change a given task's CPU affinity. Migrate the thread to a - * proper CPU and schedule it away if the CPU it's executing on - * is removed from the allowed bitmask. - * - * NOTE: the caller must have a valid reference to the task, the - * task must not exit() & deallocate itself prematurely. The - * call is not atomic; no spinlocks may be held. - */ -int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask) + return ret; +} + +int task_can_attach(struct task_struct *p) { - unsigned long flags; - struct rq *rq; - unsigned int dest_cpu; int ret = 0; - rq = task_rq_lock(p, &flags); - - if (cpumask_equal(&p->cpus_allowed, new_mask)) - goto out; - - if (!cpumask_intersects(new_mask, cpu_active_mask)) { + /* + * Kthreads which disallow setaffinity shouldn't be moved + * to a new cpuset; we don't want to change their CPU + * affinity and isolating such threads by their set of + * allowed nodes is unnecessary. Thus, cpusets are not + * applicable for such threads. This prevents checking for + * success of set_cpus_allowed_ptr() on all attached tasks + * before cpus_mask may be changed. + */ + if (p->flags & PF_NO_SETAFFINITY) ret = -EINVAL; - goto out; - } - - do_set_cpus_allowed(p, new_mask); - - /* Can the task run on the task's current CPU? If so, we're done */ - if (cpumask_test_cpu(task_cpu(p), new_mask)) - goto out; - - dest_cpu = cpumask_any_and(cpu_active_mask, new_mask); - if (p->on_rq) { - struct migration_arg arg = { p, dest_cpu }; - /* Need help from migration thread: drop lock and wait. */ - task_rq_unlock(rq, p, &flags); - stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg); - tlb_migrate_finish(p->mm); - return 0; - } -out: - task_rq_unlock(rq, p, &flags); return ret; } -EXPORT_SYMBOL_GPL(set_cpus_allowed_ptr); -/* - * Move (not current) task off this cpu, onto dest cpu. We're doing - * this because either it can't run here any more (set_cpus_allowed() - * away from this CPU, or CPU going down), or because we're - * attempting to rebalance this task on exec (sched_exec). - * - * So we race with normal scheduler movements, but that's OK, as long - * as the task is no longer on this CPU. - * - * Returns non-zero if task was successfully migrated. - */ -static int __migrate_task(struct task_struct *p, int src_cpu, int dest_cpu) +bool sched_smp_initialized __read_mostly; + +#ifdef CONFIG_NUMA_BALANCING +/* Migrate current task p to target_cpu */ +int migrate_task_to(struct task_struct *p, int target_cpu) { - struct rq *rq_dest, *rq_src; - int ret = 0; + struct migration_arg arg = { p, target_cpu }; + int curr_cpu = task_cpu(p); - if (unlikely(!cpu_active(dest_cpu))) - return ret; + if (curr_cpu == target_cpu) + return 0; - rq_src = cpu_rq(src_cpu); - rq_dest = cpu_rq(dest_cpu); + if (!cpumask_test_cpu(target_cpu, p->cpus_ptr)) + return -EINVAL; - raw_spin_lock(&p->pi_lock); - double_rq_lock(rq_src, rq_dest); - /* Already moved. */ - if (task_cpu(p) != src_cpu) - goto done; - /* Affinity changed (again). */ - if (!cpumask_test_cpu(dest_cpu, tsk_cpus_allowed(p))) - goto fail; + /* TODO: This is not properly updating schedstats */ - /* - * If we're not on a rq, the next wake-up will ensure we're - * placed properly. - */ - if (p->on_rq) { - dequeue_task(rq_src, p, 0); - set_task_cpu(p, dest_cpu); - enqueue_task(rq_dest, p, 0); - check_preempt_curr(rq_dest, p, 0); - } -done: - ret = 1; -fail: - double_rq_unlock(rq_src, rq_dest); - raw_spin_unlock(&p->pi_lock); - return ret; + trace_sched_move_numa(p, curr_cpu, target_cpu); + return stop_one_cpu(curr_cpu, migration_cpu_stop, &arg); } /* - * migration_cpu_stop - this will be executed by a highprio stopper thread - * and performs thread migration by bumping thread off CPU then - * 'pushing' onto another runqueue. + * Requeue a task on a given node and accurately track the number of NUMA + * tasks on the runqueues */ -static int migration_cpu_stop(void *data) +void sched_setnuma(struct task_struct *p, int nid) { - struct migration_arg *arg = data; - - /* - * The original target cpu might have gone down and we might - * be on another cpu but it doesn't matter. - */ - local_irq_disable(); - __migrate_task(arg->task, raw_smp_processor_id(), arg->dest_cpu); - local_irq_enable(); - return 0; + guard(task_rq_lock)(p); + scoped_guard (sched_change, p, DEQUEUE_SAVE) + p->numa_preferred_nid = nid; } +#endif /* CONFIG_NUMA_BALANCING */ #ifdef CONFIG_HOTPLUG_CPU - /* - * Ensures that the idle task is using init_mm right before its cpu goes - * offline. + * Invoked on the outgoing CPU in context of the CPU hotplug thread + * after ensuring that there are no user space tasks left on the CPU. + * + * If there is a lazy mm in use on the hotplug thread, drop it and + * switch to init_mm. + * + * The reference count on init_mm is dropped in finish_cpu(). */ -void idle_task_exit(void) +static void sched_force_init_mm(void) { struct mm_struct *mm = current->active_mm; - BUG_ON(cpu_online(smp_processor_id())); + if (mm != &init_mm) { + mmgrab_lazy_tlb(&init_mm); + local_irq_disable(); + current->active_mm = &init_mm; + switch_mm_irqs_off(mm, &init_mm, current); + local_irq_enable(); + finish_arch_post_lock_switch(); + mmdrop_lazy_tlb(mm); + } - if (mm != &init_mm) - switch_mm(mm, &init_mm, current); - mmdrop(mm); + /* finish_cpu(), as ran on the BP, will clean up the active_mm state */ } -/* - * Since this CPU is going 'away' for a while, fold any nr_active delta - * we might have. Assumes we're called after migrate_tasks() so that the - * nr_active count is stable. - * - * Also see the comment "Global load-average calculations". - */ -static void calc_load_migrate(struct rq *rq) +static int __balance_push_cpu_stop(void *arg) { - long delta = calc_load_fold_active(rq); - if (delta) - atomic_long_add(delta, &calc_load_tasks); + struct task_struct *p = arg; + struct rq *rq = this_rq(); + struct rq_flags rf; + int cpu; + + scoped_guard (raw_spinlock_irq, &p->pi_lock) { + cpu = select_fallback_rq(rq->cpu, p); + + 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); + + return 0; } +static DEFINE_PER_CPU(struct cpu_stop_work, push_work); + /* - * Migrate all tasks from the rq, sleeping tasks will be migrated by - * try_to_wake_up()->select_task_rq(). + * Ensure we only run per-cpu kthreads once the CPU goes !active. * - * Called with rq->lock held even though we'er in stop_machine() and - * there's no concurrency possible, we hold the required locks anyway - * because of lock validation efforts. + * This is enabled below SCHED_AP_ACTIVE; when !cpu_active(), but only + * effective when the hotplug motion is down. */ -static void migrate_tasks(unsigned int dead_cpu) +static void balance_push(struct rq *rq) { - struct rq *rq = cpu_rq(dead_cpu); - struct task_struct *next, *stop = rq->stop; - int dest_cpu; + struct task_struct *push_task = rq->curr; + + lockdep_assert_rq_held(rq); /* - * Fudge the rq selection such that the below task selection loop - * doesn't get stuck on the currently eligible stop task. - * - * We're currently inside stop_machine() and the rq is either stuck - * in the stop_machine_cpu_stop() loop, or we're executing this code, - * either way we should never end up calling schedule() until we're - * done here. + * Ensure the thing is persistent until balance_push_set(.on = false); */ - rq->stop = NULL; + rq->balance_callback = &balance_push_callback; /* - * put_prev_task() and pick_next_task() sched - * class method both need to have an up-to-date - * value of rq->clock[_task] + * Only active while going offline and when invoked on the outgoing + * CPU. */ - update_rq_clock(rq); + if (!cpu_dying(rq->cpu) || rq != this_rq()) + return; + + /* + * Both the cpu-hotplug and stop task are in this case and are + * required to complete the hotplug process. + */ + if (kthread_is_per_cpu(push_task) || + is_migration_disabled(push_task)) { - for ( ; ; ) { /* - * There's this thread running, bail when that's the only - * remaining thread. + * If this is the idle task on the outgoing CPU try to wake + * up the hotplug control thread which might wait for the + * last task to vanish. The rcuwait_active() check is + * accurate here because the waiter is pinned on this CPU + * and can't obviously be running in parallel. + * + * On RT kernels this also has to check whether there are + * pinned and scheduled out tasks on the runqueue. They + * need to leave the migrate disabled section first. */ - if (rq->nr_running == 1) - break; - - next = pick_next_task(rq); - BUG_ON(!next); - next->sched_class->put_prev_task(rq, next); - - /* Find suitable destination for @next, with force if needed. */ - dest_cpu = select_fallback_rq(dead_cpu, next); - raw_spin_unlock(&rq->lock); - - __migrate_task(next, dead_cpu, dest_cpu); - - raw_spin_lock(&rq->lock); + if (!rq->nr_running && !rq_has_pinned_tasks(rq) && + rcuwait_active(&rq->hotplug_wait)) { + raw_spin_rq_unlock(rq); + rcuwait_wake_up(&rq->hotplug_wait); + raw_spin_rq_lock(rq); + } + return; } - rq->stop = stop; -} - -#endif /* CONFIG_HOTPLUG_CPU */ - -#if defined(CONFIG_SCHED_DEBUG) && defined(CONFIG_SYSCTL) - -static struct ctl_table sd_ctl_dir[] = { - { - .procname = "sched_domain", - .mode = 0555, - }, - {} -}; - -static struct ctl_table sd_ctl_root[] = { - { - .procname = "kernel", - .mode = 0555, - .child = sd_ctl_dir, - }, - {} -}; - -static struct ctl_table *sd_alloc_ctl_entry(int n) -{ - struct ctl_table *entry = - kcalloc(n, sizeof(struct ctl_table), GFP_KERNEL); - - return entry; -} - -static void sd_free_ctl_entry(struct ctl_table **tablep) -{ - struct ctl_table *entry; - + get_task_struct(push_task); /* - * In the intermediate directories, both the child directory and - * procname are dynamically allocated and could fail but the mode - * will always be set. In the lowest directory the names are - * static strings and all have proc handlers. + * Temporarily drop rq->lock such that we can wake-up the stop task. + * Both preemption and IRQs are still disabled. */ - for (entry = *tablep; entry->mode; entry++) { - if (entry->child) - sd_free_ctl_entry(&entry->child); - if (entry->proc_handler == NULL) - kfree(entry->procname); - } - - kfree(*tablep); - *tablep = NULL; + preempt_disable(); + raw_spin_rq_unlock(rq); + stop_one_cpu_nowait(rq->cpu, __balance_push_cpu_stop, push_task, + this_cpu_ptr(&push_work)); + preempt_enable(); + /* + * At this point need_resched() is true and we'll take the loop in + * schedule(). The next pick is obviously going to be the stop task + * which kthread_is_per_cpu() and will push this task away. + */ + raw_spin_rq_lock(rq); } -static int min_load_idx = 0; -static int max_load_idx = CPU_LOAD_IDX_MAX-1; - -static void -set_table_entry(struct ctl_table *entry, - const char *procname, void *data, int maxlen, - umode_t mode, proc_handler *proc_handler, - bool load_idx) +static void balance_push_set(int cpu, bool on) { - entry->procname = procname; - entry->data = data; - entry->maxlen = maxlen; - entry->mode = mode; - entry->proc_handler = proc_handler; + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; - if (load_idx) { - entry->extra1 = &min_load_idx; - entry->extra2 = &max_load_idx; + rq_lock_irqsave(rq, &rf); + if (on) { + WARN_ON_ONCE(rq->balance_callback); + rq->balance_callback = &balance_push_callback; + } else if (rq->balance_callback == &balance_push_callback) { + rq->balance_callback = NULL; } + rq_unlock_irqrestore(rq, &rf); } -static struct ctl_table * -sd_alloc_ctl_domain_table(struct sched_domain *sd) +/* + * Invoked from a CPUs hotplug control thread after the CPU has been marked + * inactive. All tasks which are not per CPU kernel threads are either + * pushed off this CPU now via balance_push() or placed on a different CPU + * during wakeup. Wait until the CPU is quiescent. + */ +static void balance_hotplug_wait(void) { - struct ctl_table *table = sd_alloc_ctl_entry(13); - - if (table == NULL) - return NULL; - - set_table_entry(&table[0], "min_interval", &sd->min_interval, - sizeof(long), 0644, proc_doulongvec_minmax, false); - set_table_entry(&table[1], "max_interval", &sd->max_interval, - sizeof(long), 0644, proc_doulongvec_minmax, false); - set_table_entry(&table[2], "busy_idx", &sd->busy_idx, - sizeof(int), 0644, proc_dointvec_minmax, true); - set_table_entry(&table[3], "idle_idx", &sd->idle_idx, - sizeof(int), 0644, proc_dointvec_minmax, true); - set_table_entry(&table[4], "newidle_idx", &sd->newidle_idx, - sizeof(int), 0644, proc_dointvec_minmax, true); - set_table_entry(&table[5], "wake_idx", &sd->wake_idx, - sizeof(int), 0644, proc_dointvec_minmax, true); - set_table_entry(&table[6], "forkexec_idx", &sd->forkexec_idx, - sizeof(int), 0644, proc_dointvec_minmax, true); - set_table_entry(&table[7], "busy_factor", &sd->busy_factor, - sizeof(int), 0644, proc_dointvec_minmax, false); - set_table_entry(&table[8], "imbalance_pct", &sd->imbalance_pct, - sizeof(int), 0644, proc_dointvec_minmax, false); - set_table_entry(&table[9], "cache_nice_tries", - &sd->cache_nice_tries, - sizeof(int), 0644, proc_dointvec_minmax, false); - set_table_entry(&table[10], "flags", &sd->flags, - sizeof(int), 0644, proc_dointvec_minmax, false); - set_table_entry(&table[11], "name", sd->name, - CORENAME_MAX_SIZE, 0444, proc_dostring, false); - /* &table[12] is terminator */ - - return table; -} - -static struct ctl_table *sd_alloc_ctl_cpu_table(int cpu) -{ - struct ctl_table *entry, *table; - struct sched_domain *sd; - int domain_num = 0, i; - char buf[32]; - - for_each_domain(cpu, sd) - domain_num++; - entry = table = sd_alloc_ctl_entry(domain_num + 1); - if (table == NULL) - return NULL; + struct rq *rq = this_rq(); - i = 0; - for_each_domain(cpu, sd) { - snprintf(buf, 32, "domain%d", i); - entry->procname = kstrdup(buf, GFP_KERNEL); - entry->mode = 0555; - entry->child = sd_alloc_ctl_domain_table(sd); - entry++; - i++; - } - return table; + rcuwait_wait_event(&rq->hotplug_wait, + rq->nr_running == 1 && !rq_has_pinned_tasks(rq), + TASK_UNINTERRUPTIBLE); } -static struct ctl_table_header *sd_sysctl_header; -static void register_sched_domain_sysctl(void) -{ - int i, cpu_num = num_possible_cpus(); - struct ctl_table *entry = sd_alloc_ctl_entry(cpu_num + 1); - char buf[32]; - - WARN_ON(sd_ctl_dir[0].child); - sd_ctl_dir[0].child = entry; +#else /* !CONFIG_HOTPLUG_CPU: */ - if (entry == NULL) - return; - - for_each_possible_cpu(i) { - snprintf(buf, 32, "cpu%d", i); - entry->procname = kstrdup(buf, GFP_KERNEL); - entry->mode = 0555; - entry->child = sd_alloc_ctl_cpu_table(i); - entry++; - } - - WARN_ON(sd_sysctl_header); - sd_sysctl_header = register_sysctl_table(sd_ctl_root); -} - -/* may be called multiple times per register */ -static void unregister_sched_domain_sysctl(void) +static inline void balance_push(struct rq *rq) { - if (sd_sysctl_header) - unregister_sysctl_table(sd_sysctl_header); - sd_sysctl_header = NULL; - if (sd_ctl_dir[0].child) - sd_free_ctl_entry(&sd_ctl_dir[0].child); } -#else -static void register_sched_domain_sysctl(void) + +static inline void balance_push_set(int cpu, bool on) { } -static void unregister_sched_domain_sysctl(void) + +static inline void balance_hotplug_wait(void) { } -#endif -static void set_rq_online(struct rq *rq) +#endif /* !CONFIG_HOTPLUG_CPU */ + +void set_rq_online(struct rq *rq) { if (!rq->online) { const struct sched_class *class; @@ -4611,11 +8189,12 @@ static void set_rq_online(struct rq *rq) } } -static void set_rq_offline(struct rq *rq) +void set_rq_offline(struct rq *rq) { if (rq->online) { const struct sched_class *class; + update_rq_clock(rq); for_each_class(class) { if (class->rq_offline) class->rq_offline(rq); @@ -4626,1699 +8205,321 @@ static void set_rq_offline(struct rq *rq) } } -/* - * migration_call - callback that gets triggered when a CPU is added. - * Here we can start up the necessary migration thread for the new CPU. - */ -static int __cpuinit -migration_call(struct notifier_block *nfb, unsigned long action, void *hcpu) -{ - int cpu = (long)hcpu; - unsigned long flags; - struct rq *rq = cpu_rq(cpu); - - switch (action & ~CPU_TASKS_FROZEN) { - - case CPU_UP_PREPARE: - rq->calc_load_update = calc_load_update; - break; - - case CPU_ONLINE: - /* Update our root-domain */ - raw_spin_lock_irqsave(&rq->lock, flags); - if (rq->rd) { - BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); - - set_rq_online(rq); - } - raw_spin_unlock_irqrestore(&rq->lock, flags); - break; - -#ifdef CONFIG_HOTPLUG_CPU - case CPU_DYING: - sched_ttwu_pending(); - /* Update our root-domain */ - raw_spin_lock_irqsave(&rq->lock, flags); - if (rq->rd) { - BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); - set_rq_offline(rq); - } - migrate_tasks(cpu); - BUG_ON(rq->nr_running != 1); /* the migration thread */ - raw_spin_unlock_irqrestore(&rq->lock, flags); - break; - - case CPU_DEAD: - calc_load_migrate(rq); - break; -#endif - } - - update_max_interval(); - - return NOTIFY_OK; -} - -/* - * Register at high priority so that task migration (migrate_all_tasks) - * happens before everything else. This has to be lower priority than - * the notifier in the perf_event subsystem, though. - */ -static struct notifier_block __cpuinitdata migration_notifier = { - .notifier_call = migration_call, - .priority = CPU_PRI_MIGRATION, -}; - -static int __cpuinit sched_cpu_active(struct notifier_block *nfb, - unsigned long action, void *hcpu) -{ - switch (action & ~CPU_TASKS_FROZEN) { - case CPU_STARTING: - case CPU_DOWN_FAILED: - set_cpu_active((long)hcpu, true); - return NOTIFY_OK; - default: - return NOTIFY_DONE; - } -} - -static int __cpuinit sched_cpu_inactive(struct notifier_block *nfb, - unsigned long action, void *hcpu) -{ - switch (action & ~CPU_TASKS_FROZEN) { - case CPU_DOWN_PREPARE: - set_cpu_active((long)hcpu, false); - return NOTIFY_OK; - default: - return NOTIFY_DONE; - } -} - -static int __init migration_init(void) -{ - void *cpu = (void *)(long)smp_processor_id(); - int err; - - /* Initialize migration for the boot CPU */ - err = migration_call(&migration_notifier, CPU_UP_PREPARE, cpu); - BUG_ON(err == NOTIFY_BAD); - migration_call(&migration_notifier, CPU_ONLINE, cpu); - register_cpu_notifier(&migration_notifier); - - /* Register cpu active notifiers */ - cpu_notifier(sched_cpu_active, CPU_PRI_SCHED_ACTIVE); - cpu_notifier(sched_cpu_inactive, CPU_PRI_SCHED_INACTIVE); - - return 0; -} -early_initcall(migration_init); -#endif - -#ifdef CONFIG_SMP - -static cpumask_var_t sched_domains_tmpmask; /* sched_domains_mutex */ - -#ifdef CONFIG_SCHED_DEBUG - -static __read_mostly int sched_debug_enabled; - -static int __init sched_debug_setup(char *str) -{ - sched_debug_enabled = 1; - - return 0; -} -early_param("sched_debug", sched_debug_setup); - -static inline bool sched_debug(void) -{ - return sched_debug_enabled; -} - -static int sched_domain_debug_one(struct sched_domain *sd, int cpu, int level, - struct cpumask *groupmask) -{ - struct sched_group *group = sd->groups; - char str[256]; - - cpulist_scnprintf(str, sizeof(str), sched_domain_span(sd)); - cpumask_clear(groupmask); - - printk(KERN_DEBUG "%*s domain %d: ", level, "", level); - - if (!(sd->flags & SD_LOAD_BALANCE)) { - printk("does not load-balance\n"); - if (sd->parent) - printk(KERN_ERR "ERROR: !SD_LOAD_BALANCE domain" - " has parent"); - return -1; - } - - printk(KERN_CONT "span %s level %s\n", str, sd->name); - - if (!cpumask_test_cpu(cpu, sched_domain_span(sd))) { - printk(KERN_ERR "ERROR: domain->span does not contain " - "CPU%d\n", cpu); - } - if (!cpumask_test_cpu(cpu, sched_group_cpus(group))) { - printk(KERN_ERR "ERROR: domain->groups does not contain" - " CPU%d\n", cpu); - } - - printk(KERN_DEBUG "%*s groups:", level + 1, ""); - do { - if (!group) { - printk("\n"); - printk(KERN_ERR "ERROR: group is NULL\n"); - break; - } - - /* - * Even though we initialize ->power to something semi-sane, - * we leave power_orig unset. This allows us to detect if - * domain iteration is still funny without causing /0 traps. - */ - if (!group->sgp->power_orig) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: domain->cpu_power not " - "set\n"); - break; - } - - if (!cpumask_weight(sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: empty group\n"); - break; - } - - if (!(sd->flags & SD_OVERLAP) && - cpumask_intersects(groupmask, sched_group_cpus(group))) { - printk(KERN_CONT "\n"); - printk(KERN_ERR "ERROR: repeated CPUs\n"); - break; - } - - cpumask_or(groupmask, groupmask, sched_group_cpus(group)); - - cpulist_scnprintf(str, sizeof(str), sched_group_cpus(group)); - - printk(KERN_CONT " %s", str); - if (group->sgp->power != SCHED_POWER_SCALE) { - printk(KERN_CONT " (cpu_power = %d)", - group->sgp->power); - } - - group = group->next; - } while (group != sd->groups); - printk(KERN_CONT "\n"); - - if (!cpumask_equal(sched_domain_span(sd), groupmask)) - printk(KERN_ERR "ERROR: groups don't span domain->span\n"); - - if (sd->parent && - !cpumask_subset(groupmask, sched_domain_span(sd->parent))) - printk(KERN_ERR "ERROR: parent span is not a superset " - "of domain->span\n"); - return 0; -} - -static void sched_domain_debug(struct sched_domain *sd, int cpu) -{ - int level = 0; - - if (!sched_debug_enabled) - return; - - if (!sd) { - printk(KERN_DEBUG "CPU%d attaching NULL sched-domain.\n", cpu); - return; - } - - printk(KERN_DEBUG "CPU%d attaching sched-domain:\n", cpu); - - for (;;) { - if (sched_domain_debug_one(sd, cpu, level, sched_domains_tmpmask)) - break; - level++; - sd = sd->parent; - if (!sd) - break; - } -} -#else /* !CONFIG_SCHED_DEBUG */ -# define sched_domain_debug(sd, cpu) do { } while (0) -static inline bool sched_debug(void) -{ - return false; -} -#endif /* CONFIG_SCHED_DEBUG */ - -static int sd_degenerate(struct sched_domain *sd) -{ - if (cpumask_weight(sched_domain_span(sd)) == 1) - return 1; - - /* Following flags need at least 2 groups */ - if (sd->flags & (SD_LOAD_BALANCE | - SD_BALANCE_NEWIDLE | - SD_BALANCE_FORK | - SD_BALANCE_EXEC | - SD_SHARE_CPUPOWER | - SD_SHARE_PKG_RESOURCES)) { - if (sd->groups != sd->groups->next) - return 0; - } - - /* Following flags don't use groups */ - if (sd->flags & (SD_WAKE_AFFINE)) - return 0; - - return 1; -} - -static int -sd_parent_degenerate(struct sched_domain *sd, struct sched_domain *parent) -{ - unsigned long cflags = sd->flags, pflags = parent->flags; - - if (sd_degenerate(parent)) - return 1; - - if (!cpumask_equal(sched_domain_span(sd), sched_domain_span(parent))) - return 0; - - /* Flags needing groups don't count if only 1 group in parent */ - if (parent->groups == parent->groups->next) { - pflags &= ~(SD_LOAD_BALANCE | - SD_BALANCE_NEWIDLE | - SD_BALANCE_FORK | - SD_BALANCE_EXEC | - SD_SHARE_CPUPOWER | - SD_SHARE_PKG_RESOURCES); - if (nr_node_ids == 1) - pflags &= ~SD_SERIALIZE; - } - if (~cflags & pflags) - return 0; - - return 1; -} - -static void free_rootdomain(struct rcu_head *rcu) +static inline void sched_set_rq_online(struct rq *rq, int cpu) { - struct root_domain *rd = container_of(rcu, struct root_domain, rcu); - - cpupri_cleanup(&rd->cpupri); - free_cpumask_var(rd->rto_mask); - free_cpumask_var(rd->online); - free_cpumask_var(rd->span); - kfree(rd); -} - -static void rq_attach_root(struct rq *rq, struct root_domain *rd) -{ - struct root_domain *old_rd = NULL; - unsigned long flags; - - raw_spin_lock_irqsave(&rq->lock, flags); + struct rq_flags rf; + rq_lock_irqsave(rq, &rf); if (rq->rd) { - old_rd = rq->rd; - - if (cpumask_test_cpu(rq->cpu, old_rd->online)) - set_rq_offline(rq); - - cpumask_clear_cpu(rq->cpu, old_rd->span); - - /* - * If we dont want to free the old_rt yet then - * set old_rd to NULL to skip the freeing later - * in this function: - */ - if (!atomic_dec_and_test(&old_rd->refcount)) - old_rd = NULL; - } - - atomic_inc(&rd->refcount); - rq->rd = rd; - - cpumask_set_cpu(rq->cpu, rd->span); - if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_online(rq); - - raw_spin_unlock_irqrestore(&rq->lock, flags); - - if (old_rd) - call_rcu_sched(&old_rd->rcu, free_rootdomain); -} - -static int init_rootdomain(struct root_domain *rd) -{ - memset(rd, 0, sizeof(*rd)); - - if (!alloc_cpumask_var(&rd->span, GFP_KERNEL)) - goto out; - if (!alloc_cpumask_var(&rd->online, GFP_KERNEL)) - goto free_span; - if (!alloc_cpumask_var(&rd->rto_mask, GFP_KERNEL)) - goto free_online; - - if (cpupri_init(&rd->cpupri) != 0) - goto free_rto_mask; - return 0; - -free_rto_mask: - free_cpumask_var(rd->rto_mask); -free_online: - free_cpumask_var(rd->online); -free_span: - free_cpumask_var(rd->span); -out: - return -ENOMEM; -} - -/* - * By default the system creates a single root-domain with all cpus as - * members (mimicking the global state we have today). - */ -struct root_domain def_root_domain; - -static void init_defrootdomain(void) -{ - init_rootdomain(&def_root_domain); - - atomic_set(&def_root_domain.refcount, 1); -} - -static struct root_domain *alloc_rootdomain(void) -{ - struct root_domain *rd; - - rd = kmalloc(sizeof(*rd), GFP_KERNEL); - if (!rd) - return NULL; - - if (init_rootdomain(rd) != 0) { - kfree(rd); - return NULL; } - - return rd; + rq_unlock_irqrestore(rq, &rf); } -static void free_sched_groups(struct sched_group *sg, int free_sgp) +static inline void sched_set_rq_offline(struct rq *rq, int cpu) { - struct sched_group *tmp, *first; - - if (!sg) - return; - - first = sg; - do { - tmp = sg->next; + struct rq_flags rf; - if (free_sgp && atomic_dec_and_test(&sg->sgp->ref)) - kfree(sg->sgp); - - kfree(sg); - sg = tmp; - } while (sg != first); -} - -static void free_sched_domain(struct rcu_head *rcu) -{ - struct sched_domain *sd = container_of(rcu, struct sched_domain, rcu); - - /* - * If its an overlapping domain it has private groups, iterate and - * nuke them all. - */ - if (sd->flags & SD_OVERLAP) { - free_sched_groups(sd->groups, 1); - } else if (atomic_dec_and_test(&sd->groups->ref)) { - kfree(sd->groups->sgp); - kfree(sd->groups); + rq_lock_irqsave(rq, &rf); + if (rq->rd) { + BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); + set_rq_offline(rq); } - kfree(sd); -} - -static void destroy_sched_domain(struct sched_domain *sd, int cpu) -{ - call_rcu(&sd->rcu, free_sched_domain); -} - -static void destroy_sched_domains(struct sched_domain *sd, int cpu) -{ - for (; sd; sd = sd->parent) - destroy_sched_domain(sd, cpu); + rq_unlock_irqrestore(rq, &rf); } /* - * Keep a special pointer to the highest sched_domain that has - * SD_SHARE_PKG_RESOURCE set (Last Level Cache Domain) for this - * allows us to avoid some pointer chasing select_idle_sibling(). - * - * Also keep a unique ID per domain (we use the first cpu number in - * the cpumask of the domain), this allows us to quickly tell if - * two cpus are in the same cache domain, see cpus_share_cache(). + * used to mark begin/end of suspend/resume: */ -DEFINE_PER_CPU(struct sched_domain *, sd_llc); -DEFINE_PER_CPU(int, sd_llc_id); - -static void update_top_cache_domain(int cpu) -{ - struct sched_domain *sd; - int id = cpu; - - sd = highest_flag_domain(cpu, SD_SHARE_PKG_RESOURCES); - if (sd) - id = cpumask_first(sched_domain_span(sd)); - - rcu_assign_pointer(per_cpu(sd_llc, cpu), sd); - per_cpu(sd_llc_id, cpu) = id; -} +static int num_cpus_frozen; /* - * Attach the domain 'sd' to 'cpu' as its base domain. Callers must - * hold the hotplug lock. - */ -static void -cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) -{ - struct rq *rq = cpu_rq(cpu); - struct sched_domain *tmp; - - /* Remove the sched domains which do not contribute to scheduling. */ - for (tmp = sd; tmp; ) { - struct sched_domain *parent = tmp->parent; - if (!parent) - break; - - if (sd_parent_degenerate(tmp, parent)) { - tmp->parent = parent->parent; - if (parent->parent) - parent->parent->child = tmp; - destroy_sched_domain(parent, cpu); - } else - tmp = tmp->parent; - } - - if (sd && sd_degenerate(sd)) { - tmp = sd; - sd = sd->parent; - destroy_sched_domain(tmp, cpu); - if (sd) - sd->child = NULL; - } - - sched_domain_debug(sd, cpu); - - rq_attach_root(rq, rd); - tmp = rq->sd; - rcu_assign_pointer(rq->sd, sd); - destroy_sched_domains(tmp, cpu); - - update_top_cache_domain(cpu); -} - -/* cpus with isolated domains */ -static cpumask_var_t cpu_isolated_map; - -/* Setup the mask of cpus configured for isolated domains */ -static int __init isolated_cpu_setup(char *str) -{ - alloc_bootmem_cpumask_var(&cpu_isolated_map); - cpulist_parse(str, cpu_isolated_map); - return 1; -} - -__setup("isolcpus=", isolated_cpu_setup); - -static const struct cpumask *cpu_cpu_mask(int cpu) -{ - return cpumask_of_node(cpu_to_node(cpu)); -} - -struct sd_data { - struct sched_domain **__percpu sd; - struct sched_group **__percpu sg; - struct sched_group_power **__percpu sgp; -}; - -struct s_data { - struct sched_domain ** __percpu sd; - struct root_domain *rd; -}; - -enum s_alloc { - sa_rootdomain, - sa_sd, - sa_sd_storage, - sa_none, -}; - -struct sched_domain_topology_level; - -typedef struct sched_domain *(*sched_domain_init_f)(struct sched_domain_topology_level *tl, int cpu); -typedef const struct cpumask *(*sched_domain_mask_f)(int cpu); - -#define SDTL_OVERLAP 0x01 - -struct sched_domain_topology_level { - sched_domain_init_f init; - sched_domain_mask_f mask; - int flags; - int numa_level; - struct sd_data data; -}; - -/* - * Build an iteration mask that can exclude certain CPUs from the upwards - * domain traversal. - * - * Asymmetric node setups can result in situations where the domain tree is of - * unequal depth, make sure to skip domains that already cover the entire - * range. - * - * In that case build_sched_domains() will have terminated the iteration early - * and our sibling sd spans will be empty. Domains should always include the - * cpu they're built on, so check that. + * Update cpusets according to cpu_active mask. If cpusets are + * disabled, cpuset_update_active_cpus() becomes a simple wrapper + * around partition_sched_domains(). * + * If we come here as part of a suspend/resume, don't touch cpusets because we + * want to restore it back to its original state upon resume anyway. */ -static void build_group_mask(struct sched_domain *sd, struct sched_group *sg) -{ - const struct cpumask *span = sched_domain_span(sd); - struct sd_data *sdd = sd->private; - struct sched_domain *sibling; - int i; - - for_each_cpu(i, span) { - sibling = *per_cpu_ptr(sdd->sd, i); - if (!cpumask_test_cpu(i, sched_domain_span(sibling))) - continue; - - cpumask_set_cpu(i, sched_group_mask(sg)); - } -} - -/* - * Return the canonical balance cpu for this group, this is the first cpu - * of this group that's also in the iteration mask. - */ -int group_balance_cpu(struct sched_group *sg) +static void cpuset_cpu_active(void) { - return cpumask_first_and(sched_group_cpus(sg), sched_group_mask(sg)); -} - -static int -build_overlap_sched_groups(struct sched_domain *sd, int cpu) -{ - struct sched_group *first = NULL, *last = NULL, *groups = NULL, *sg; - const struct cpumask *span = sched_domain_span(sd); - struct cpumask *covered = sched_domains_tmpmask; - struct sd_data *sdd = sd->private; - struct sched_domain *child; - int i; - - cpumask_clear(covered); - - for_each_cpu(i, span) { - struct cpumask *sg_span; - - if (cpumask_test_cpu(i, covered)) - continue; - - child = *per_cpu_ptr(sdd->sd, i); - - /* See the comment near build_group_mask(). */ - if (!cpumask_test_cpu(i, sched_domain_span(child))) - continue; - - sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), - GFP_KERNEL, cpu_to_node(cpu)); - - if (!sg) - goto fail; - - sg_span = sched_group_cpus(sg); - if (child->child) { - child = child->child; - cpumask_copy(sg_span, sched_domain_span(child)); - } else - cpumask_set_cpu(i, sg_span); - - cpumask_or(covered, covered, sg_span); - - sg->sgp = *per_cpu_ptr(sdd->sgp, i); - if (atomic_inc_return(&sg->sgp->ref) == 1) - build_group_mask(sd, sg); - + if (cpuhp_tasks_frozen) { /* - * Initialize sgp->power such that even if we mess up the - * domains and no possible iteration will get us here, we won't - * die on a /0 trap. + * num_cpus_frozen tracks how many CPUs are involved in suspend + * resume sequence. As long as this is not the last online + * operation in the resume sequence, just build a single sched + * domain, ignoring cpusets. */ - sg->sgp->power = SCHED_POWER_SCALE * cpumask_weight(sg_span); - + cpuset_reset_sched_domains(); + if (--num_cpus_frozen) + return; /* - * Make sure the first group of this domain contains the - * canonical balance cpu. Otherwise the sched_domain iteration - * breaks. See update_sg_lb_stats(). + * This is the last CPU online operation. So fall through and + * restore the original sched domains by considering the + * cpuset configurations. */ - if ((!groups && cpumask_test_cpu(cpu, sg_span)) || - group_balance_cpu(sg) == cpu) - groups = sg; - - if (!first) - first = sg; - if (last) - last->next = sg; - last = sg; - last->next = first; - } - sd->groups = groups; - - return 0; - -fail: - free_sched_groups(first, 0); - - return -ENOMEM; -} - -static int get_group(int cpu, struct sd_data *sdd, struct sched_group **sg) -{ - struct sched_domain *sd = *per_cpu_ptr(sdd->sd, cpu); - struct sched_domain *child = sd->child; - - if (child) - cpu = cpumask_first(sched_domain_span(child)); - - if (sg) { - *sg = *per_cpu_ptr(sdd->sg, cpu); - (*sg)->sgp = *per_cpu_ptr(sdd->sgp, cpu); - atomic_set(&(*sg)->sgp->ref, 1); /* for claim_allocations */ + cpuset_force_rebuild(); } - - return cpu; + cpuset_update_active_cpus(); } -/* - * build_sched_groups will build a circular linked list of the groups - * covered by the given span, and will set each group's ->cpumask correctly, - * and ->cpu_power to 0. - * - * Assumes the sched_domain tree is fully constructed - */ -static int -build_sched_groups(struct sched_domain *sd, int cpu) +static void cpuset_cpu_inactive(unsigned int cpu) { - struct sched_group *first = NULL, *last = NULL; - struct sd_data *sdd = sd->private; - const struct cpumask *span = sched_domain_span(sd); - struct cpumask *covered; - int i; - - get_group(cpu, sdd, &sd->groups); - atomic_inc(&sd->groups->ref); - - if (cpu != cpumask_first(span)) - return 0; - - lockdep_assert_held(&sched_domains_mutex); - covered = sched_domains_tmpmask; - - cpumask_clear(covered); - - for_each_cpu(i, span) { - struct sched_group *sg; - int group, j; - - if (cpumask_test_cpu(i, covered)) - continue; - - group = get_group(i, sdd, &sg); - cpumask_clear(sched_group_cpus(sg)); - sg->sgp->power = 0; - cpumask_setall(sched_group_mask(sg)); - - for_each_cpu(j, span) { - if (get_group(j, sdd, NULL) != group) - continue; - - cpumask_set_cpu(j, covered); - cpumask_set_cpu(j, sched_group_cpus(sg)); - } - - if (!first) - first = sg; - if (last) - last->next = sg; - last = sg; + if (!cpuhp_tasks_frozen) { + cpuset_update_active_cpus(); + } else { + num_cpus_frozen++; + cpuset_reset_sched_domains(); } - last->next = first; - - return 0; } -/* - * Initialize sched groups cpu_power. - * - * cpu_power indicates the capacity of sched group, which is used while - * distributing the load between different sched groups in a sched domain. - * Typically cpu_power for all the groups in a sched domain will be same unless - * there are asymmetries in the topology. If there are asymmetries, group - * having more cpu_power will pickup more load compared to the group having - * less cpu_power. - */ -static void init_sched_groups_power(int cpu, struct sched_domain *sd) +static inline void sched_smt_present_inc(int cpu) { - struct sched_group *sg = sd->groups; - - WARN_ON(!sg); - - do { - sg->group_weight = cpumask_weight(sched_group_cpus(sg)); - sg = sg->next; - } while (sg != sd->groups); - - if (cpu != group_balance_cpu(sg)) - return; - - update_group_power(sd, cpu); - atomic_set(&sg->sgp->nr_busy_cpus, sg->group_weight); -} - -int __weak arch_sd_sibling_asym_packing(void) -{ - return 0*SD_ASYM_PACKING; -} - -/* - * Initializers for schedule domains - * Non-inlined to reduce accumulated stack pressure in build_sched_domains() - */ - -#ifdef CONFIG_SCHED_DEBUG -# define SD_INIT_NAME(sd, type) sd->name = #type -#else -# define SD_INIT_NAME(sd, type) do { } while (0) -#endif - -#define SD_INIT_FUNC(type) \ -static noinline struct sched_domain * \ -sd_init_##type(struct sched_domain_topology_level *tl, int cpu) \ -{ \ - struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu); \ - *sd = SD_##type##_INIT; \ - SD_INIT_NAME(sd, type); \ - sd->private = &tl->data; \ - return sd; \ -} - -SD_INIT_FUNC(CPU) #ifdef CONFIG_SCHED_SMT - SD_INIT_FUNC(SIBLING) -#endif -#ifdef CONFIG_SCHED_MC - SD_INIT_FUNC(MC) -#endif -#ifdef CONFIG_SCHED_BOOK - SD_INIT_FUNC(BOOK) + if (cpumask_weight(cpu_smt_mask(cpu)) == 2) + static_branch_inc_cpuslocked(&sched_smt_present); #endif - -static int default_relax_domain_level = -1; -int sched_domain_level_max; - -static int __init setup_relax_domain_level(char *str) -{ - if (kstrtoint(str, 0, &default_relax_domain_level)) - pr_warn("Unable to set relax_domain_level\n"); - - return 1; -} -__setup("relax_domain_level=", setup_relax_domain_level); - -static void set_domain_attribute(struct sched_domain *sd, - struct sched_domain_attr *attr) -{ - int request; - - if (!attr || attr->relax_domain_level < 0) { - if (default_relax_domain_level < 0) - return; - else - request = default_relax_domain_level; - } else - request = attr->relax_domain_level; - if (request < sd->level) { - /* turn off idle balance on this domain */ - sd->flags &= ~(SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); - } else { - /* turn on idle balance on this domain */ - sd->flags |= (SD_BALANCE_WAKE|SD_BALANCE_NEWIDLE); - } -} - -static void __sdt_free(const struct cpumask *cpu_map); -static int __sdt_alloc(const struct cpumask *cpu_map); - -static void __free_domain_allocs(struct s_data *d, enum s_alloc what, - const struct cpumask *cpu_map) -{ - switch (what) { - case sa_rootdomain: - if (!atomic_read(&d->rd->refcount)) - free_rootdomain(&d->rd->rcu); /* fall through */ - case sa_sd: - free_percpu(d->sd); /* fall through */ - case sa_sd_storage: - __sdt_free(cpu_map); /* fall through */ - case sa_none: - break; - } -} - -static enum s_alloc __visit_domain_allocation_hell(struct s_data *d, - const struct cpumask *cpu_map) -{ - memset(d, 0, sizeof(*d)); - - if (__sdt_alloc(cpu_map)) - return sa_sd_storage; - d->sd = alloc_percpu(struct sched_domain *); - if (!d->sd) - return sa_sd_storage; - d->rd = alloc_rootdomain(); - if (!d->rd) - return sa_sd; - return sa_rootdomain; } -/* - * NULL the sd_data elements we've used to build the sched_domain and - * 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) -{ - struct sd_data *sdd = sd->private; - - WARN_ON_ONCE(*per_cpu_ptr(sdd->sd, cpu) != sd); - *per_cpu_ptr(sdd->sd, 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->sgp, cpu))->ref)) - *per_cpu_ptr(sdd->sgp, cpu) = NULL; -} - -#ifdef CONFIG_SCHED_SMT -static const struct cpumask *cpu_smt_mask(int cpu) +static inline void sched_smt_present_dec(int cpu) { - return topology_thread_cpumask(cpu); -} -#endif - -/* - * Topology list, bottom-up. - */ -static struct sched_domain_topology_level default_topology[] = { #ifdef CONFIG_SCHED_SMT - { sd_init_SIBLING, cpu_smt_mask, }, -#endif -#ifdef CONFIG_SCHED_MC - { sd_init_MC, cpu_coregroup_mask, }, + if (cpumask_weight(cpu_smt_mask(cpu)) == 2) + static_branch_dec_cpuslocked(&sched_smt_present); #endif -#ifdef CONFIG_SCHED_BOOK - { sd_init_BOOK, cpu_book_mask, }, -#endif - { sd_init_CPU, cpu_cpu_mask, }, - { NULL, }, -}; - -static struct sched_domain_topology_level *sched_domain_topology = default_topology; - -#define for_each_sd_topology(tl) \ - for (tl = sched_domain_topology; tl->init; tl++) - -#ifdef CONFIG_NUMA - -static int sched_domains_numa_levels; -static int *sched_domains_numa_distance; -static struct cpumask ***sched_domains_numa_masks; -static int sched_domains_curr_level; +} -static inline int sd_local_flags(int level) +int sched_cpu_activate(unsigned int cpu) { - if (sched_domains_numa_distance[level] > RECLAIM_DISTANCE) - return 0; - - return SD_BALANCE_EXEC | SD_BALANCE_FORK | SD_WAKE_AFFINE; -} - -static struct sched_domain * -sd_numa_init(struct sched_domain_topology_level *tl, int cpu) -{ - struct sched_domain *sd = *per_cpu_ptr(tl->data.sd, cpu); - int level = tl->numa_level; - int sd_weight = cpumask_weight( - sched_domains_numa_masks[level][cpu_to_node(cpu)]); - - *sd = (struct sched_domain){ - .min_interval = sd_weight, - .max_interval = 2*sd_weight, - .busy_factor = 32, - .imbalance_pct = 125, - .cache_nice_tries = 2, - .busy_idx = 3, - .idle_idx = 2, - .newidle_idx = 0, - .wake_idx = 0, - .forkexec_idx = 0, - - .flags = 1*SD_LOAD_BALANCE - | 1*SD_BALANCE_NEWIDLE - | 0*SD_BALANCE_EXEC - | 0*SD_BALANCE_FORK - | 0*SD_BALANCE_WAKE - | 0*SD_WAKE_AFFINE - | 0*SD_SHARE_CPUPOWER - | 0*SD_SHARE_PKG_RESOURCES - | 1*SD_SERIALIZE - | 0*SD_PREFER_SIBLING - | sd_local_flags(level) - , - .last_balance = jiffies, - .balance_interval = sd_weight, - }; - SD_INIT_NAME(sd, NUMA); - sd->private = &tl->data; + struct rq *rq = cpu_rq(cpu); /* - * Ugly hack to pass state to sd_numa_mask()... + * Clear the balance_push callback and prepare to schedule + * regular tasks. */ - sched_domains_curr_level = tl->numa_level; - - return sd; -} + balance_push_set(cpu, false); -static const struct cpumask *sd_numa_mask(int cpu) -{ - return sched_domains_numa_masks[sched_domains_curr_level][cpu_to_node(cpu)]; -} - -static void sched_numa_warn(const char *str) -{ - static int done = false; - int i,j; - - if (done) - return; - - done = true; - - printk(KERN_WARNING "ERROR: %s\n\n", str); + /* + * When going up, increment the number of cores with SMT present. + */ + sched_smt_present_inc(cpu); + set_cpu_active(cpu, true); - for (i = 0; i < nr_node_ids; i++) { - printk(KERN_WARNING " "); - for (j = 0; j < nr_node_ids; j++) - printk(KERN_CONT "%02d ", node_distance(i,j)); - printk(KERN_CONT "\n"); + if (sched_smp_initialized) { + sched_update_numa(cpu, true); + sched_domains_numa_masks_set(cpu); + cpuset_cpu_active(); } - printk(KERN_WARNING "\n"); -} -static bool find_numa_distance(int distance) -{ - int i; - - if (distance == node_distance(0, 0)) - return true; + scx_rq_activate(rq); - for (i = 0; i < sched_domains_numa_levels; i++) { - if (sched_domains_numa_distance[i] == distance) - return true; - } + /* + * Put the rq online, if not already. This happens: + * + * 1) In the early boot process, because we build the real domains + * after all CPUs have been brought up. + * + * 2) At runtime, if cpuset_cpu_active() fails to rebuild the + * domains. + */ + sched_set_rq_online(rq, cpu); - return false; + return 0; } -static void sched_init_numa(void) +int sched_cpu_deactivate(unsigned int cpu) { - int next_distance, curr_distance = node_distance(0, 0); - struct sched_domain_topology_level *tl; - int level = 0; - int i, j, k; + struct rq *rq = cpu_rq(cpu); + int ret; - sched_domains_numa_distance = kzalloc(sizeof(int) * nr_node_ids, GFP_KERNEL); - if (!sched_domains_numa_distance) - return; + ret = dl_bw_deactivate(cpu); + + if (ret) + return ret; /* - * O(nr_nodes^2) deduplicating selection sort -- in order to find the - * unique distances in the node_distance() table. - * - * Assumes node_distance(0,j) includes all distances in - * node_distance(i,j) in order to avoid cubic time. + * Remove CPU from nohz.idle_cpus_mask to prevent participating in + * load balancing when not active */ - next_distance = curr_distance; - for (i = 0; i < nr_node_ids; i++) { - for (j = 0; j < nr_node_ids; j++) { - for (k = 0; k < nr_node_ids; k++) { - int distance = node_distance(i, k); - - if (distance > curr_distance && - (distance < next_distance || - next_distance == curr_distance)) - next_distance = distance; - - /* - * While not a strong assumption it would be nice to know - * about cases where if node A is connected to B, B is not - * equally connected to A. - */ - if (sched_debug() && node_distance(k, i) != distance) - sched_numa_warn("Node-distance not symmetric"); + nohz_balance_exit_idle(rq); - if (sched_debug() && i && !find_numa_distance(distance)) - sched_numa_warn("Node-0 not representative"); - } - if (next_distance != curr_distance) { - sched_domains_numa_distance[level++] = next_distance; - sched_domains_numa_levels = level; - curr_distance = next_distance; - } else break; - } + set_cpu_active(cpu, false); - /* - * In case of sched_debug() we verify the above assumption. - */ - if (!sched_debug()) - break; - } /* - * 'level' contains the number of unique distances, excluding the - * identity distance node_distance(i,i). - * - * The sched_domains_numa_distance[] array includes the actual distance - * numbers. + * From this point forward, this CPU will refuse to run any task that + * is not: migrate_disable() or KTHREAD_IS_PER_CPU, and will actively + * push those tasks away until this gets cleared, see + * sched_cpu_dying(). */ + balance_push_set(cpu, true); /* - * Here, we should temporarily reset sched_domains_numa_levels to 0. - * If it fails to allocate memory for array sched_domains_numa_masks[][], - * the array will contain less then 'level' members. This could be - * dangerous when we use it to iterate array sched_domains_numa_masks[][] - * in other functions. + * We've cleared cpu_active_mask / set balance_push, wait for all + * preempt-disabled and RCU users of this state to go away such that + * all new such users will observe it. + * + * Specifically, we rely on ttwu to no longer target this CPU, see + * ttwu_queue_cond() and is_cpu_allowed(). * - * We reset it to 'level' at the end of this function. + * Do sync before park smpboot threads to take care the RCU boost case. */ - sched_domains_numa_levels = 0; + synchronize_rcu(); - sched_domains_numa_masks = kzalloc(sizeof(void *) * level, GFP_KERNEL); - if (!sched_domains_numa_masks) - return; + sched_set_rq_offline(rq, cpu); - /* - * Now for each level, construct a mask per node which contains all - * cpus of nodes that are that many hops away from us. - */ - for (i = 0; i < level; i++) { - sched_domains_numa_masks[i] = - kzalloc(nr_node_ids * sizeof(void *), GFP_KERNEL); - if (!sched_domains_numa_masks[i]) - return; - - for (j = 0; j < nr_node_ids; j++) { - struct cpumask *mask = kzalloc(cpumask_size(), GFP_KERNEL); - if (!mask) - return; - - sched_domains_numa_masks[i][j] = mask; - - for (k = 0; k < nr_node_ids; k++) { - if (node_distance(j, k) > sched_domains_numa_distance[i]) - continue; - - cpumask_or(mask, mask, cpumask_of_node(k)); - } - } - } - - tl = kzalloc((ARRAY_SIZE(default_topology) + level) * - sizeof(struct sched_domain_topology_level), GFP_KERNEL); - if (!tl) - return; - - /* - * Copy the default topology bits.. - */ - for (i = 0; default_topology[i].init; i++) - tl[i] = default_topology[i]; + scx_rq_deactivate(rq); /* - * .. and append 'j' levels of NUMA goodness. + * When going down, decrement the number of cores with SMT present. */ - for (j = 0; j < level; i++, j++) { - tl[i] = (struct sched_domain_topology_level){ - .init = sd_numa_init, - .mask = sd_numa_mask, - .flags = SDTL_OVERLAP, - .numa_level = j, - }; - } - - sched_domain_topology = tl; - - sched_domains_numa_levels = level; -} - -static void sched_domains_numa_masks_set(int cpu) -{ - int i, j; - int node = cpu_to_node(cpu); + sched_smt_present_dec(cpu); - for (i = 0; i < sched_domains_numa_levels; i++) { - for (j = 0; j < nr_node_ids; j++) { - if (node_distance(j, node) <= sched_domains_numa_distance[i]) - cpumask_set_cpu(cpu, sched_domains_numa_masks[i][j]); - } - } -} - -static void sched_domains_numa_masks_clear(int cpu) -{ - int i, j; - for (i = 0; i < sched_domains_numa_levels; i++) { - for (j = 0; j < nr_node_ids; j++) - cpumask_clear_cpu(cpu, sched_domains_numa_masks[i][j]); - } -} - -/* - * Update sched_domains_numa_masks[level][node] array when new cpus - * are onlined. - */ -static int sched_domains_numa_masks_update(struct notifier_block *nfb, - unsigned long action, - void *hcpu) -{ - int cpu = (long)hcpu; - - switch (action & ~CPU_TASKS_FROZEN) { - case CPU_ONLINE: - sched_domains_numa_masks_set(cpu); - break; - - case CPU_DEAD: - sched_domains_numa_masks_clear(cpu); - break; - - default: - return NOTIFY_DONE; - } - - return NOTIFY_OK; -} -#else -static inline void sched_init_numa(void) -{ -} - -static int sched_domains_numa_masks_update(struct notifier_block *nfb, - unsigned long action, - void *hcpu) -{ - return 0; -} -#endif /* CONFIG_NUMA */ - -static int __sdt_alloc(const struct cpumask *cpu_map) -{ - struct sched_domain_topology_level *tl; - int j; - - for_each_sd_topology(tl) { - struct sd_data *sdd = &tl->data; - - sdd->sd = alloc_percpu(struct sched_domain *); - if (!sdd->sd) - return -ENOMEM; - - sdd->sg = alloc_percpu(struct sched_group *); - if (!sdd->sg) - return -ENOMEM; - - sdd->sgp = alloc_percpu(struct sched_group_power *); - if (!sdd->sgp) - return -ENOMEM; - - for_each_cpu(j, cpu_map) { - struct sched_domain *sd; - struct sched_group *sg; - struct sched_group_power *sgp; - - sd = kzalloc_node(sizeof(struct sched_domain) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sd) - return -ENOMEM; - - *per_cpu_ptr(sdd->sd, j) = sd; - - sg = kzalloc_node(sizeof(struct sched_group) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sg) - return -ENOMEM; - - sg->next = sg; - - *per_cpu_ptr(sdd->sg, j) = sg; - - sgp = kzalloc_node(sizeof(struct sched_group_power) + cpumask_size(), - GFP_KERNEL, cpu_to_node(j)); - if (!sgp) - return -ENOMEM; +#ifdef CONFIG_SCHED_SMT + sched_core_cpu_deactivate(cpu); +#endif - *per_cpu_ptr(sdd->sgp, j) = sgp; - } - } + if (!sched_smp_initialized) + return 0; + sched_update_numa(cpu, false); + cpuset_cpu_inactive(cpu); + sched_domains_numa_masks_clear(cpu); return 0; } -static void __sdt_free(const struct cpumask *cpu_map) +static void sched_rq_cpu_starting(unsigned int cpu) { - struct sched_domain_topology_level *tl; - int j; - - for_each_sd_topology(tl) { - struct sd_data *sdd = &tl->data; - - for_each_cpu(j, cpu_map) { - struct sched_domain *sd; - - if (sdd->sd) { - sd = *per_cpu_ptr(sdd->sd, j); - if (sd && (sd->flags & SD_OVERLAP)) - free_sched_groups(sd->groups, 0); - kfree(*per_cpu_ptr(sdd->sd, j)); - } - - if (sdd->sg) - kfree(*per_cpu_ptr(sdd->sg, j)); - if (sdd->sgp) - kfree(*per_cpu_ptr(sdd->sgp, j)); - } - free_percpu(sdd->sd); - sdd->sd = NULL; - free_percpu(sdd->sg); - sdd->sg = NULL; - free_percpu(sdd->sgp); - sdd->sgp = NULL; - } -} - -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) -{ - struct sched_domain *sd = tl->init(tl, cpu); - if (!sd) - return child; - - cpumask_and(sched_domain_span(sd), cpu_map, tl->mask(cpu)); - if (child) { - sd->level = child->level + 1; - sched_domain_level_max = max(sched_domain_level_max, sd->level); - child->parent = sd; - sd->child = child; - } - set_domain_attribute(sd, attr); - - return sd; -} - -/* - * Build sched domains for a given set of cpus and attach the sched domains - * to the individual cpus - */ -static int build_sched_domains(const struct cpumask *cpu_map, - struct sched_domain_attr *attr) -{ - enum s_alloc alloc_state; - struct sched_domain *sd; - struct s_data d; - int i, ret = -ENOMEM; - - alloc_state = __visit_domain_allocation_hell(&d, cpu_map); - if (alloc_state != sa_rootdomain) - goto error; - - /* Set up domains for cpus specified by the cpu_map. */ - for_each_cpu(i, cpu_map) { - struct sched_domain_topology_level *tl; - - sd = NULL; - for_each_sd_topology(tl) { - sd = build_sched_domain(tl, cpu_map, attr, sd, i); - if (tl == sched_domain_topology) - *per_cpu_ptr(d.sd, i) = sd; - if (tl->flags & SDTL_OVERLAP || sched_feat(FORCE_SD_OVERLAP)) - sd->flags |= SD_OVERLAP; - if (cpumask_equal(cpu_map, sched_domain_span(sd))) - break; - } - } - - /* 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 (build_overlap_sched_groups(sd, i)) - goto error; - } else { - if (build_sched_groups(sd, i)) - goto error; - } - } - } - - /* Calculate CPU power for physical packages and nodes */ - for (i = nr_cpumask_bits-1; i >= 0; i--) { - if (!cpumask_test_cpu(i, cpu_map)) - continue; - - for (sd = *per_cpu_ptr(d.sd, i); sd; sd = sd->parent) { - claim_allocations(i, sd); - init_sched_groups_power(i, sd); - } - } - - /* Attach the domains */ - rcu_read_lock(); - for_each_cpu(i, cpu_map) { - sd = *per_cpu_ptr(d.sd, i); - cpu_attach_domain(sd, d.rd, i); - } - rcu_read_unlock(); + struct rq *rq = cpu_rq(cpu); - ret = 0; -error: - __free_domain_allocs(&d, alloc_state, cpu_map); - return ret; + rq->calc_load_update = calc_load_update; + update_max_interval(); } -static cpumask_var_t *doms_cur; /* current sched domains */ -static int ndoms_cur; /* number of sched domains in 'doms_cur' */ -static struct sched_domain_attr *dattr_cur; - /* attribues of custom domains in 'doms_cur' */ - -/* - * Special case: If a kmalloc of a doms_cur partition (array of - * cpumask) fails, then fallback to a single sched domain, - * as determined by the single cpumask fallback_doms. - */ -static cpumask_var_t fallback_doms; - -/* - * arch_update_cpu_topology lets virtualized architectures update the - * cpu core maps. It is supposed to return 1 if the topology changed - * or 0 if it stayed the same. - */ -int __attribute__((weak)) arch_update_cpu_topology(void) +int sched_cpu_starting(unsigned int cpu) { + sched_core_cpu_starting(cpu); + sched_rq_cpu_starting(cpu); + sched_tick_start(cpu); return 0; } -cpumask_var_t *alloc_sched_domains(unsigned int ndoms) -{ - int i; - cpumask_var_t *doms; - - doms = kmalloc(sizeof(*doms) * ndoms, GFP_KERNEL); - if (!doms) - return NULL; - for (i = 0; i < ndoms; i++) { - if (!alloc_cpumask_var(&doms[i], GFP_KERNEL)) { - free_sched_domains(doms, i); - return NULL; - } - } - return doms; -} - -void free_sched_domains(cpumask_var_t doms[], unsigned int ndoms) -{ - unsigned int i; - for (i = 0; i < ndoms; i++) - free_cpumask_var(doms[i]); - kfree(doms); -} +#ifdef CONFIG_HOTPLUG_CPU /* - * Set up scheduler domains and groups. Callers must hold the hotplug lock. - * For now this just excludes isolated cpus, but could be used to - * exclude other special cases in the future. + * Invoked immediately before the stopper thread is invoked to bring the + * CPU down completely. At this point all per CPU kthreads except the + * hotplug thread (current) and the stopper thread (inactive) have been + * either parked or have been unbound from the outgoing CPU. Ensure that + * any of those which might be on the way out are gone. + * + * If after this point a bound task is being woken on this CPU then the + * responsible hotplug callback has failed to do it's job. + * sched_cpu_dying() will catch it with the appropriate fireworks. */ -static int init_sched_domains(const struct cpumask *cpu_map) +int sched_cpu_wait_empty(unsigned int cpu) { - int err; - - arch_update_cpu_topology(); - ndoms_cur = 1; - doms_cur = alloc_sched_domains(ndoms_cur); - if (!doms_cur) - doms_cur = &fallback_doms; - cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); - err = build_sched_domains(doms_cur[0], NULL); - register_sched_domain_sysctl(); - - return err; + balance_hotplug_wait(); + sched_force_init_mm(); + return 0; } /* - * Detach sched domains from a group of cpus specified in cpu_map - * These cpus will now be attached to the NULL domain + * Since this CPU is going 'away' for a while, fold any nr_active delta we + * might have. Called from the CPU stopper task after ensuring that the + * stopper is the last running task on the CPU, so nr_active count is + * stable. We need to take the tear-down thread which is calling this into + * account, so we hand in adjust = 1 to the load calculation. + * + * Also see the comment "Global load-average calculations". */ -static void detach_destroy_domains(const struct cpumask *cpu_map) -{ - int i; - - rcu_read_lock(); - for_each_cpu(i, cpu_map) - cpu_attach_domain(NULL, &def_root_domain, i); - rcu_read_unlock(); -} - -/* handle null as "default" */ -static int dattrs_equal(struct sched_domain_attr *cur, int idx_cur, - struct sched_domain_attr *new, int idx_new) +static void calc_load_migrate(struct rq *rq) { - struct sched_domain_attr tmp; + long delta = calc_load_fold_active(rq, 1); - /* fast path */ - if (!new && !cur) - return 1; - - tmp = SD_ATTR_INIT; - return !memcmp(cur ? (cur + idx_cur) : &tmp, - new ? (new + idx_new) : &tmp, - sizeof(struct sched_domain_attr)); + if (delta) + atomic_long_add(delta, &calc_load_tasks); } -/* - * Partition sched domains as specified by the 'ndoms_new' - * cpumasks in the array doms_new[] of cpumasks. This compares - * doms_new[] to the current sched domain partitioning, doms_cur[]. - * It destroys each deleted domain and builds each new domain. - * - * 'doms_new' is an array of cpumask_var_t's of length 'ndoms_new'. - * The masks don't intersect (don't overlap.) We should setup one - * sched domain for each mask. CPUs not in any of the cpumasks will - * not be load balanced. If the same cpumask appears both in the - * current 'doms_cur' domains and in the new 'doms_new', we can leave - * it as it is. - * - * The passed in 'doms_new' should be allocated using - * alloc_sched_domains. This routine takes ownership of it and will - * free_sched_domains it when done with it. If the caller failed the - * alloc call, then it can pass in doms_new == NULL && ndoms_new == 1, - * and partition_sched_domains() will fallback to the single partition - * 'fallback_doms', it also forces the domains to be rebuilt. - * - * If doms_new == NULL it will be replaced with cpu_online_mask. - * ndoms_new == 0 is a special case for destroying existing domains, - * and it will not create the default domain. - * - * Call with hotplug lock held - */ -void partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], - struct sched_domain_attr *dattr_new) +static void dump_rq_tasks(struct rq *rq, const char *loglvl) { - int i, j, n; - int new_topology; - - mutex_lock(&sched_domains_mutex); - - /* always unregister in case we don't destroy any domains */ - unregister_sched_domain_sysctl(); - - /* Let architecture update cpu core mappings. */ - new_topology = arch_update_cpu_topology(); + struct task_struct *g, *p; + int cpu = cpu_of(rq); - n = doms_new ? ndoms_new : 0; + lockdep_assert_rq_held(rq); - /* Destroy deleted domains */ - 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)) - goto match1; - } - /* no match - a current sched domain not in new doms_new[] */ - detach_destroy_domains(doms_cur[i]); -match1: - ; - } + printk("%sCPU%d enqueued tasks (%u total):\n", loglvl, cpu, rq->nr_running); + for_each_process_thread(g, p) { + if (task_cpu(p) != cpu) + continue; - if (doms_new == NULL) { - ndoms_cur = 0; - doms_new = &fallback_doms; - cpumask_andnot(doms_new[0], cpu_active_mask, cpu_isolated_map); - WARN_ON_ONCE(dattr_new); - } + if (!task_on_rq_queued(p)) + continue; - /* Build new domains */ - for (i = 0; i < ndoms_new; i++) { - for (j = 0; j < ndoms_cur && !new_topology; j++) { - if (cpumask_equal(doms_new[i], doms_cur[j]) - && dattrs_equal(dattr_new, i, dattr_cur, j)) - goto match2; - } - /* no match - add a new doms_new */ - build_sched_domains(doms_new[i], dattr_new ? dattr_new + i : NULL); -match2: - ; + printk("%s\tpid: %d, name: %s\n", loglvl, p->pid, p->comm); } - - /* Remember the new sched domains */ - if (doms_cur != &fallback_doms) - free_sched_domains(doms_cur, ndoms_cur); - kfree(dattr_cur); /* kfree(NULL) is safe */ - doms_cur = doms_new; - dattr_cur = dattr_new; - ndoms_cur = ndoms_new; - - register_sched_domain_sysctl(); - - mutex_unlock(&sched_domains_mutex); } -static int num_cpus_frozen; /* used to mark begin/end of suspend/resume */ - -/* - * Update cpusets according to cpu_active mask. If cpusets are - * disabled, cpuset_update_active_cpus() becomes a simple wrapper - * around partition_sched_domains(). - * - * If we come here as part of a suspend/resume, don't touch cpusets because we - * want to restore it back to its original state upon resume anyway. - */ -static int cpuset_cpu_active(struct notifier_block *nfb, unsigned long action, - void *hcpu) +int sched_cpu_dying(unsigned int cpu) { - switch (action) { - case CPU_ONLINE_FROZEN: - case CPU_DOWN_FAILED_FROZEN: - - /* - * num_cpus_frozen tracks how many CPUs are involved in suspend - * resume sequence. As long as this is not the last online - * operation in the resume sequence, just build a single sched - * domain, ignoring cpusets. - */ - num_cpus_frozen--; - if (likely(num_cpus_frozen)) { - partition_sched_domains(1, NULL, NULL); - break; - } + struct rq *rq = cpu_rq(cpu); + struct rq_flags rf; - /* - * This is the last CPU online operation. So fall through and - * restore the original sched domains by considering the - * cpuset configurations. - */ + /* Handle pending wakeups and then migrate everything off */ + sched_tick_stop(cpu); - case CPU_ONLINE: - case CPU_DOWN_FAILED: - cpuset_update_active_cpus(true); - break; - default: - return NOTIFY_DONE; + 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); } - return NOTIFY_OK; -} + dl_server_stop(&rq->fair_server); + rq_unlock_irqrestore(rq, &rf); -static int cpuset_cpu_inactive(struct notifier_block *nfb, unsigned long action, - void *hcpu) -{ - switch (action) { - case CPU_DOWN_PREPARE: - cpuset_update_active_cpus(false); - break; - case CPU_DOWN_PREPARE_FROZEN: - num_cpus_frozen++; - partition_sched_domains(1, NULL, NULL); - break; - default: - return NOTIFY_DONE; - } - return NOTIFY_OK; + calc_load_migrate(rq); + update_max_interval(); + hrtick_clear(rq); + sched_core_cpu_dying(cpu); + return 0; } +#endif /* CONFIG_HOTPLUG_CPU */ void __init sched_init_smp(void) { - cpumask_var_t non_isolated_cpus; - - alloc_cpumask_var(&non_isolated_cpus, GFP_KERNEL); - alloc_cpumask_var(&fallback_doms, GFP_KERNEL); - - sched_init_numa(); + sched_init_numa(NUMA_NO_NODE); - get_online_cpus(); - mutex_lock(&sched_domains_mutex); - init_sched_domains(cpu_active_mask); - cpumask_andnot(non_isolated_cpus, cpu_possible_mask, cpu_isolated_map); - if (cpumask_empty(non_isolated_cpus)) - cpumask_set_cpu(smp_processor_id(), non_isolated_cpus); - mutex_unlock(&sched_domains_mutex); - put_online_cpus(); + prandom_init_once(&sched_rnd_state); - hotcpu_notifier(sched_domains_numa_masks_update, CPU_PRI_SCHED_ACTIVE); - hotcpu_notifier(cpuset_cpu_active, CPU_PRI_CPUSET_ACTIVE); - hotcpu_notifier(cpuset_cpu_inactive, CPU_PRI_CPUSET_INACTIVE); - - init_hrtick(); + /* + * 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. + */ + sched_domains_mutex_lock(); + sched_init_domains(cpu_active_mask); + sched_domains_mutex_unlock(); /* Move init over to a non-isolated CPU */ - if (set_cpus_allowed_ptr(current, non_isolated_cpus) < 0) + if (set_cpus_allowed_ptr(current, housekeeping_cpumask(HK_TYPE_DOMAIN)) < 0) BUG(); + current->flags &= ~PF_NO_SETAFFINITY; sched_init_granularity(); - free_cpumask_var(non_isolated_cpus); init_sched_rt_class(); + init_sched_dl_class(); + + sched_init_dl_servers(); + + sched_smp_initialized = true; } -#else -void __init sched_init_smp(void) + +static int __init migration_init(void) { - sched_init_granularity(); + sched_cpu_starting(smp_processor_id()); + return 0; } -#endif /* CONFIG_SMP */ - -const_debug unsigned int sysctl_timer_migration = 1; +early_initcall(migration_init); int in_sched_functions(unsigned long addr) { @@ -6334,26 +8535,36 @@ int in_sched_functions(unsigned long addr) */ struct task_group root_task_group; LIST_HEAD(task_groups); -#endif -DECLARE_PER_CPU(cpumask_var_t, load_balance_mask); +/* Cacheline aligned slab cache for task_group */ +static struct kmem_cache *task_group_cache __ro_after_init; +#endif void __init sched_init(void) { - int i, j; - unsigned long alloc_size = 0, ptr; + unsigned long ptr = 0; + int i; + + /* Make sure the linker didn't screw up */ + BUG_ON(!sched_class_above(&stop_sched_class, &dl_sched_class)); + 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)); +#ifdef CONFIG_SCHED_CLASS_EXT + BUG_ON(!sched_class_above(&fair_sched_class, &ext_sched_class)); + BUG_ON(!sched_class_above(&ext_sched_class, &idle_sched_class)); +#endif + + wait_bit_init(); #ifdef CONFIG_FAIR_GROUP_SCHED - alloc_size += 2 * nr_cpu_ids * sizeof(void **); + ptr += 2 * nr_cpu_ids * sizeof(void **); #endif #ifdef CONFIG_RT_GROUP_SCHED - alloc_size += 2 * nr_cpu_ids * sizeof(void **); -#endif -#ifdef CONFIG_CPUMASK_OFFSTACK - alloc_size += num_possible_cpus() * cpumask_size(); + ptr += 2 * nr_cpu_ids * sizeof(void **); #endif - if (alloc_size) { - ptr = (unsigned long)kzalloc(alloc_size, GFP_NOWAIT); + if (ptr) { + ptr = (unsigned long)kzalloc(ptr, GFP_NOWAIT); #ifdef CONFIG_FAIR_GROUP_SCHED root_task_group.se = (struct sched_entity **)ptr; @@ -6362,7 +8573,12 @@ void __init sched_init(void) root_task_group.cfs_rq = (struct cfs_rq **)ptr; ptr += nr_cpu_ids * sizeof(void **); + root_task_group.shares = ROOT_TASK_GROUP_LOAD; + init_cfs_bandwidth(&root_task_group.cfs_bandwidth, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ +#ifdef CONFIG_EXT_GROUP_SCHED + 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; ptr += nr_cpu_ids * sizeof(void **); @@ -6371,20 +8587,9 @@ void __init sched_init(void) ptr += nr_cpu_ids * sizeof(void **); #endif /* CONFIG_RT_GROUP_SCHED */ -#ifdef CONFIG_CPUMASK_OFFSTACK - for_each_possible_cpu(i) { - per_cpu(load_balance_mask, i) = (void *)ptr; - ptr += cpumask_size(); - } -#endif /* CONFIG_CPUMASK_OFFSTACK */ } -#ifdef CONFIG_SMP init_defrootdomain(); -#endif - - init_rt_bandwidth(&def_rt_bandwidth, - global_rt_period(), global_rt_runtime()); #ifdef CONFIG_RT_GROUP_SCHED init_rt_bandwidth(&root_task_group.rt_bandwidth, @@ -6392,65 +8597,63 @@ void __init sched_init(void) #endif /* CONFIG_RT_GROUP_SCHED */ #ifdef CONFIG_CGROUP_SCHED + task_group_cache = KMEM_CACHE(task_group, 0); + list_add(&root_task_group.list, &task_groups); INIT_LIST_HEAD(&root_task_group.children); INIT_LIST_HEAD(&root_task_group.siblings); autogroup_init(&init_task); - #endif /* CONFIG_CGROUP_SCHED */ for_each_possible_cpu(i) { struct rq *rq; rq = cpu_rq(i); - raw_spin_lock_init(&rq->lock); + raw_spin_lock_init(&rq->__lock); rq->nr_running = 0; rq->calc_load_active = 0; rq->calc_load_update = jiffies + LOAD_FREQ; init_cfs_rq(&rq->cfs); - init_rt_rq(&rq->rt, rq); + init_rt_rq(&rq->rt); + init_dl_rq(&rq->dl); #ifdef CONFIG_FAIR_GROUP_SCHED - root_task_group.shares = ROOT_TASK_GROUP_LOAD; INIT_LIST_HEAD(&rq->leaf_cfs_rq_list); + rq->tmp_alone_branch = &rq->leaf_cfs_rq_list; /* - * How much cpu bandwidth does root_task_group get? + * How much CPU bandwidth does root_task_group get? * - * In case of task-groups formed thr' the cgroup filesystem, it - * gets 100% of the cpu resources in the system. This overall - * system cpu resource is divided among the tasks of + * In case of task-groups formed through the cgroup filesystem, it + * gets 100% of the CPU resources in the system. This overall + * system CPU resource is divided among the tasks of * root_task_group and its child task-groups in a fair manner, * based on each entity's (task or task-group's) weight * (se->load.weight). * * In other words, if root_task_group has 10 tasks of weight * 1024) and two child groups A0 and A1 (of weight 1024 each), - * then A0's share of the cpu resource is: + * then A0's share of the CPU resource is: * * A0's bandwidth = 1024 / (10*1024 + 1024 + 1024) = 8.33% * * We achieve this by letting root_task_group's tasks sit * directly in rq->cfs (i.e root_task_group->se[] = NULL). */ - init_cfs_bandwidth(&root_task_group.cfs_bandwidth); init_tg_cfs_entry(&root_task_group, &rq->cfs, NULL, i, NULL); #endif /* CONFIG_FAIR_GROUP_SCHED */ - rq->rt.rt_runtime = def_rt_bandwidth.rt_runtime; #ifdef CONFIG_RT_GROUP_SCHED - INIT_LIST_HEAD(&rq->leaf_rt_rq_list); + /* + * This is required for init cpu because rt.c:__enable_runtime() + * starts working after scheduler_running, which is not the case + * yet. + */ + rq->rt.rt_runtime = global_rt_runtime(); init_tg_rt_entry(&root_task_group, &rq->rt, NULL, i, NULL); #endif - - for (j = 0; j < CPU_LOAD_IDX_MAX; j++) - rq->cpu_load[j] = 0; - - rq->last_load_update_tick = jiffies; - -#ifdef CONFIG_SMP rq->sd = NULL; rq->rd = NULL; - rq->cpu_power = SCHED_POWER_SCALE; - rq->post_schedule = 0; + rq->cpu_capacity = SCHED_CAPACITY_SCALE; + rq->balance_callback = &balance_push_callback; rq->active_balance = 0; rq->next_balance = jiffies; rq->push_cpu = 0; @@ -6458,167 +8661,277 @@ void __init sched_init(void) rq->online = 0; rq->idle_stamp = 0; rq->avg_idle = 2*sysctl_sched_migration_cost; + rq->max_idle_balance_cost = sysctl_sched_migration_cost; INIT_LIST_HEAD(&rq->cfs_tasks); rq_attach_root(rq, &def_root_domain); #ifdef CONFIG_NO_HZ_COMMON - rq->nohz_flags = 0; -#endif -#ifdef CONFIG_NO_HZ_FULL - rq->last_sched_tick = 0; + rq->last_blocked_load_update_tick = jiffies; + atomic_set(&rq->nohz_flags, 0); + + INIT_CSD(&rq->nohz_csd, nohz_csd_func, rq); #endif +#ifdef CONFIG_HOTPLUG_CPU + rcuwait_init(&rq->hotplug_wait); #endif - init_rq_hrtick(rq); + hrtick_rq_init(rq); atomic_set(&rq->nr_iowait, 0); - } - - set_load_weight(&init_task); - -#ifdef CONFIG_PREEMPT_NOTIFIERS - INIT_HLIST_HEAD(&init_task.preempt_notifiers); + fair_server_init(rq); + +#ifdef CONFIG_SCHED_CORE + rq->core = rq; + rq->core_pick = NULL; + rq->core_dl_server = NULL; + rq->core_enabled = 0; + rq->core_tree = RB_ROOT; + rq->core_forceidle_count = 0; + rq->core_forceidle_occupation = 0; + rq->core_forceidle_start = 0; + + rq->core_cookie = 0UL; #endif + zalloc_cpumask_var_node(&rq->scratch_mask, GFP_KERNEL, cpu_to_node(i)); + } -#ifdef CONFIG_RT_MUTEXES - plist_head_init(&init_task.pi_waiters); -#endif + set_load_weight(&init_task, false); + init_task.se.slice = sysctl_sched_base_slice, /* * The boot idle thread does lazy MMU switching as well: */ - atomic_inc(&init_mm.mm_count); + mmgrab_lazy_tlb(&init_mm); enter_lazy_tlb(&init_mm, current); /* + * The idle task doesn't need the kthread struct to function, but it + * is dressed up as a per-CPU kthread and thus needs to play the part + * if we want to avoid special-casing it in code that deals with per-CPU + * kthreads. + */ + WARN_ON(!set_kthread_struct(current)); + + /* * Make us the idle thread. Technically, schedule() should not be * called from this thread, however somewhere below it might be, * but because we are the idle thread, we just pick up running again * when this runqueue becomes "idle". */ + __sched_fork(0, current); init_idle(current, smp_processor_id()); calc_load_update = jiffies + LOAD_FREQ; - /* - * During early bootup we pretend to be a normal task: - */ - current->sched_class = &fair_sched_class; - -#ifdef CONFIG_SMP - zalloc_cpumask_var(&sched_domains_tmpmask, GFP_NOWAIT); - /* May be allocated at isolcpus cmdline parse time */ - if (cpu_isolated_map == NULL) - zalloc_cpumask_var(&cpu_isolated_map, GFP_NOWAIT); idle_thread_set_boot_cpu(); -#endif + + balance_push_set(smp_processor_id(), false); init_sched_fair_class(); + init_sched_ext_class(); + + psi_init(); + + init_uclamp(); + + preempt_dynamic_init(); scheduler_running = 1; } #ifdef CONFIG_DEBUG_ATOMIC_SLEEP -static inline int preempt_count_equals(int preempt_offset) + +void __might_sleep(const char *file, int line) +{ + unsigned int state = get_current_state(); + /* + * Blocking primitives will set (and therefore destroy) current->state, + * since we will exit with TASK_RUNNING make sure we enter with it, + * otherwise we will destroy state. + */ + WARN_ONCE(state != TASK_RUNNING && current->task_state_change, + "do not call blocking ops when !TASK_RUNNING; " + "state=%x set at [<%p>] %pS\n", state, + (void *)current->task_state_change, + (void *)current->task_state_change); + + __might_resched(file, line, 0); +} +EXPORT_SYMBOL(__might_sleep); + +static void print_preempt_disable_ip(int preempt_offset, unsigned long ip) +{ + if (!IS_ENABLED(CONFIG_DEBUG_PREEMPT)) + return; + + if (preempt_count() == preempt_offset) + return; + + pr_err("Preemption disabled at:"); + print_ip_sym(KERN_ERR, ip); +} + +static inline bool resched_offsets_ok(unsigned int offsets) +{ + unsigned int nested = preempt_count(); + + nested += rcu_preempt_depth() << MIGHT_RESCHED_RCU_SHIFT; + + return nested == offsets; +} + +void __might_resched(const char *file, int line, unsigned int offsets) { - int nested = (preempt_count() & ~PREEMPT_ACTIVE) + rcu_preempt_depth(); + /* Ratelimiting timestamp: */ + static unsigned long prev_jiffy; + + unsigned long preempt_disable_ip; + + /* WARN_ON_ONCE() by default, no rate limit required: */ + rcu_sleep_check(); + + if ((resched_offsets_ok(offsets) && !irqs_disabled() && + !is_idle_task(current) && !current->non_block_count) || + system_state == SYSTEM_BOOTING || system_state > SYSTEM_RUNNING || + oops_in_progress) + return; + + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) + return; + prev_jiffy = jiffies; + + /* Save this before calling printk(), since that will clobber it: */ + preempt_disable_ip = get_preempt_disable_ip(current); + + pr_err("BUG: sleeping function called from invalid context at %s:%d\n", + file, line); + pr_err("in_atomic(): %d, irqs_disabled(): %d, non_block: %d, pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), current->non_block_count, + current->pid, current->comm); + pr_err("preempt_count: %x, expected: %x\n", preempt_count(), + offsets & MIGHT_RESCHED_PREEMPT_MASK); + + if (IS_ENABLED(CONFIG_PREEMPT_RCU)) { + pr_err("RCU nest depth: %d, expected: %u\n", + rcu_preempt_depth(), offsets >> MIGHT_RESCHED_RCU_SHIFT); + } + + if (task_stack_end_corrupted(current)) + pr_emerg("Thread overran stack, or stack corrupted\n"); + + debug_show_held_locks(current); + if (irqs_disabled()) + print_irqtrace_events(current); - return (nested == preempt_offset); + print_preempt_disable_ip(offsets & MIGHT_RESCHED_PREEMPT_MASK, + preempt_disable_ip); + + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); } +EXPORT_SYMBOL(__might_resched); -void __might_sleep(const char *file, int line, int preempt_offset) +void __cant_sleep(const char *file, int line, int preempt_offset) { - static unsigned long prev_jiffy; /* ratelimiting */ + static unsigned long prev_jiffy; - rcu_sleep_check(); /* WARN_ON_ONCE() by default, no rate limit reqd. */ - if ((preempt_count_equals(preempt_offset) && !irqs_disabled()) || - system_state != SYSTEM_RUNNING || oops_in_progress) + if (irqs_disabled()) return; + + if (!IS_ENABLED(CONFIG_PREEMPT_COUNT)) + return; + + if (preempt_count() > preempt_offset) + return; + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) return; prev_jiffy = jiffies; - printk(KERN_ERR - "BUG: sleeping function called from invalid context at %s:%d\n", - file, line); - printk(KERN_ERR - "in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", + printk(KERN_ERR "BUG: assuming atomic context at %s:%d\n", file, line); + printk(KERN_ERR "in_atomic(): %d, irqs_disabled(): %d, pid: %d, name: %s\n", in_atomic(), irqs_disabled(), current->pid, current->comm); debug_show_held_locks(current); - if (irqs_disabled()) - print_irqtrace_events(current); dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); } -EXPORT_SYMBOL(__might_sleep); -#endif +EXPORT_SYMBOL_GPL(__cant_sleep); -#ifdef CONFIG_MAGIC_SYSRQ -static void normalize_task(struct rq *rq, struct task_struct *p) +# ifdef CONFIG_SMP +void __cant_migrate(const char *file, int line) { - const struct sched_class *prev_class = p->sched_class; - int old_prio = p->prio; - int on_rq; + static unsigned long prev_jiffy; - on_rq = p->on_rq; - if (on_rq) - dequeue_task(rq, p, 0); - __setscheduler(rq, p, SCHED_NORMAL, 0); - if (on_rq) { - enqueue_task(rq, p, 0); - resched_task(rq->curr); - } + if (irqs_disabled()) + return; + + if (is_migration_disabled(current)) + return; + + if (!IS_ENABLED(CONFIG_PREEMPT_COUNT)) + return; + + if (preempt_count() > 0) + return; + + if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy) + return; + prev_jiffy = jiffies; + + pr_err("BUG: assuming non migratable context at %s:%d\n", file, line); + pr_err("in_atomic(): %d, irqs_disabled(): %d, migration_disabled() %u pid: %d, name: %s\n", + in_atomic(), irqs_disabled(), is_migration_disabled(current), + current->pid, current->comm); - check_class_changed(rq, p, prev_class, old_prio); + debug_show_held_locks(current); + dump_stack(); + add_taint(TAINT_WARN, LOCKDEP_STILL_OK); } +EXPORT_SYMBOL_GPL(__cant_migrate); +# endif /* CONFIG_SMP */ +#endif /* CONFIG_DEBUG_ATOMIC_SLEEP */ +#ifdef CONFIG_MAGIC_SYSRQ void normalize_rt_tasks(void) { struct task_struct *g, *p; - unsigned long flags; - struct rq *rq; + struct sched_attr attr = { + .sched_policy = SCHED_NORMAL, + }; - read_lock_irqsave(&tasklist_lock, flags); - do_each_thread(g, p) { + read_lock(&tasklist_lock); + for_each_process_thread(g, p) { /* * Only normalize user tasks: */ - if (!p->mm) + if (p->flags & PF_KTHREAD) continue; - p->se.exec_start = 0; -#ifdef CONFIG_SCHEDSTATS - p->se.statistics.wait_start = 0; - p->se.statistics.sleep_start = 0; - p->se.statistics.block_start = 0; -#endif + p->se.exec_start = 0; + schedstat_set(p->stats.wait_start, 0); + schedstat_set(p->stats.sleep_start, 0); + schedstat_set(p->stats.block_start, 0); - if (!rt_task(p)) { + if (!rt_or_dl_task(p)) { /* * Renice negative nice level userspace * tasks back to 0: */ - if (TASK_NICE(p) < 0 && p->mm) + if (task_nice(p) < 0) set_user_nice(p, 0); continue; } - raw_spin_lock(&p->pi_lock); - rq = __task_rq_lock(p); - - normalize_task(rq, p); - - __task_rq_unlock(rq); - raw_spin_unlock(&p->pi_lock); - } while_each_thread(g, p); - - read_unlock_irqrestore(&tasklist_lock, flags); + __sched_setscheduler(p, &attr, false, false); + } + read_unlock(&tasklist_lock); } #endif /* CONFIG_MAGIC_SYSRQ */ -#if defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) +#ifdef CONFIG_KGDB_KDB /* - * These functions are only useful for the IA64 MCA handling, or kdb. + * These functions are only useful for KDB. * * They can only be called when the whole system has been * stopped - every CPU needs to be quiescent, and no scheduling @@ -6628,51 +8941,60 @@ void normalize_rt_tasks(void) */ /** - * curr_task - return the current task for a given cpu. + * curr_task - return the current task for a given CPU. * @cpu: the processor in question. * * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! + * + * Return: The current task for @cpu. */ struct task_struct *curr_task(int cpu) { return cpu_curr(cpu); } -#endif /* defined(CONFIG_IA64) || defined(CONFIG_KGDB_KDB) */ - -#ifdef CONFIG_IA64 -/** - * set_curr_task - set the current task for a given cpu. - * @cpu: the processor in question. - * @p: the task pointer to set. - * - * Description: This function must only be used when non-maskable interrupts - * are serviced on a separate stack. It allows the architecture to switch the - * notion of the current task on a cpu in a non-blocking manner. This function - * must be called with all CPU's synchronized, and interrupts disabled, the - * and caller must save the original value of the current task (see - * curr_task() above) and restore that value before reenabling interrupts and - * re-starting the system. - * - * ONLY VALID WHEN THE WHOLE SYSTEM IS STOPPED! - */ -void set_curr_task(int cpu, struct task_struct *p) -{ - cpu_curr(cpu) = p; -} - -#endif +#endif /* CONFIG_KGDB_KDB */ #ifdef CONFIG_CGROUP_SCHED /* task_group_lock serializes the addition/removal of task groups */ static DEFINE_SPINLOCK(task_group_lock); -static void free_sched_group(struct task_group *tg) +static inline void alloc_uclamp_sched_group(struct task_group *tg, + struct task_group *parent) +{ +#ifdef CONFIG_UCLAMP_TASK_GROUP + enum uclamp_id clamp_id; + + for_each_clamp_id(clamp_id) { + uclamp_se_set(&tg->uclamp_req[clamp_id], + uclamp_none(clamp_id), false); + tg->uclamp[clamp_id] = parent->uclamp[clamp_id]; + } +#endif +} + +static void sched_free_group(struct task_group *tg) { free_fair_sched_group(tg); free_rt_sched_group(tg); autogroup_free(tg); - kfree(tg); + kmem_cache_free(task_group_cache, tg); +} + +static void sched_free_group_rcu(struct rcu_head *rcu) +{ + sched_free_group(container_of(rcu, struct task_group, rcu)); +} + +static void sched_unregister_group(struct task_group *tg) +{ + unregister_fair_sched_group(tg); + unregister_rt_sched_group(tg); + /* + * We have to wait for yet another RCU grace period to expire, as + * print_cfs_stats() might run concurrently. + */ + call_rcu(&tg->rcu, sched_free_group_rcu); } /* allocate runqueue etc for a new task group */ @@ -6680,7 +9002,7 @@ struct task_group *sched_create_group(struct task_group *parent) { struct task_group *tg; - tg = kzalloc(sizeof(*tg), GFP_KERNEL); + tg = kmem_cache_alloc(task_group_cache, GFP_KERNEL | __GFP_ZERO); if (!tg) return ERR_PTR(-ENOMEM); @@ -6690,10 +9012,13 @@ struct task_group *sched_create_group(struct task_group *parent) if (!alloc_rt_sched_group(tg, parent)) goto err; + scx_tg_init(tg); + alloc_uclamp_sched_group(tg, parent); + return tg; err: - free_sched_group(tg); + sched_free_group(tg); return ERR_PTR(-ENOMEM); } @@ -6702,586 +9027,498 @@ 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); - WARN_ON(!parent); /* root should already exist */ + /* Root should already exist: */ + WARN_ON(!parent); tg->parent = parent; INIT_LIST_HEAD(&tg->children); list_add_rcu(&tg->siblings, &parent->children); spin_unlock_irqrestore(&task_group_lock, flags); + + online_fair_sched_group(tg); } -/* rcu callback to free various structures associated with a task group */ -static void free_sched_group_rcu(struct rcu_head *rhp) +/* RCU callback to free various structures associated with a task group */ +static void sched_unregister_group_rcu(struct rcu_head *rhp) { - /* now it should be safe to free those cfs_rqs */ - free_sched_group(container_of(rhp, struct task_group, rcu)); + /* Now it should be safe to free those cfs_rqs: */ + sched_unregister_group(container_of(rhp, struct task_group, rcu)); } -/* Destroy runqueue etc associated with a task group */ void sched_destroy_group(struct task_group *tg) { - /* wait for possible concurrent references to cfs_rqs complete */ - call_rcu(&tg->rcu, free_sched_group_rcu); + /* Wait for possible concurrent references to cfs_rqs complete: */ + call_rcu(&tg->rcu, sched_unregister_group_rcu); } -void sched_offline_group(struct task_group *tg) +void sched_release_group(struct task_group *tg) { unsigned long flags; - int i; - - /* end participation in shares distribution */ - for_each_possible_cpu(i) - unregister_fair_sched_group(tg, i); + /* + * Unlink first, to avoid walk_tg_tree_from() from finding us (via + * sched_cfs_period_timer()). + * + * For this to be effective, we have to wait for all pending users of + * this task group to leave their RCU critical section to ensure no new + * user will see our dying task group any more. Specifically ensure + * that tg_unthrottle_up() won't add decayed cfs_rq's to it. + * + * We therefore defer calling unregister_fair_sched_group() to + * sched_unregister_group() which is guarantied to get called only after the + * current RCU grace period has expired. + */ spin_lock_irqsave(&task_group_lock, flags); list_del_rcu(&tg->list); list_del_rcu(&tg->siblings); spin_unlock_irqrestore(&task_group_lock, flags); } -/* change task's runqueue when it moves between groups. - * The caller of this function should have put the task in its new group - * by now. This function just updates tsk->se.cfs_rq and tsk->se.parent to - * reflect its new group. - */ -void sched_move_task(struct task_struct *tsk) +static void sched_change_group(struct task_struct *tsk) { struct task_group *tg; - int on_rq, running; - unsigned long flags; - struct rq *rq; - - rq = task_rq_lock(tsk, &flags); - - running = task_current(rq, tsk); - on_rq = tsk->on_rq; - if (on_rq) - dequeue_task(rq, tsk, 0); - if (unlikely(running)) - tsk->sched_class->put_prev_task(rq, tsk); - - tg = container_of(task_subsys_state_check(tsk, cpu_cgroup_subsys_id, - lockdep_is_held(&tsk->sighand->siglock)), + /* + * All callers are synchronized by task_rq_lock(); we do not use RCU + * which is pointless here. Thus, we pass "true" to task_css_check() + * to prevent lockdep warnings. + */ + tg = container_of(task_css_check(tsk, cpu_cgrp_id, true), struct task_group, css); tg = autogroup_task_group(tsk, tg); tsk->sched_task_group = tg; #ifdef CONFIG_FAIR_GROUP_SCHED - if (tsk->sched_class->task_move_group) - tsk->sched_class->task_move_group(tsk, on_rq); + if (tsk->sched_class->task_change_group) + tsk->sched_class->task_change_group(tsk); else #endif set_task_rq(tsk, task_cpu(tsk)); - - if (unlikely(running)) - tsk->sched_class->set_curr_task(rq); - if (on_rq) - enqueue_task(rq, tsk, 0); - - task_rq_unlock(rq, tsk, &flags); } -#endif /* CONFIG_CGROUP_SCHED */ - -#if defined(CONFIG_RT_GROUP_SCHED) || defined(CONFIG_CFS_BANDWIDTH) -static unsigned long to_ratio(u64 period, u64 runtime) -{ - if (runtime == RUNTIME_INF) - return 1ULL << 20; - return div64_u64(runtime << 20, period); -} -#endif - -#ifdef CONFIG_RT_GROUP_SCHED /* - * Ensure that the real time constraints are schedulable. + * Change task's runqueue when it moves between groups. + * + * The caller of this function should have put the task in its new group by + * now. This function just updates tsk->se.cfs_rq and tsk->se.parent to reflect + * its new group. */ -static DEFINE_MUTEX(rt_constraints_mutex); - -/* Must be called with tasklist_lock held */ -static inline int tg_has_rt_tasks(struct task_group *tg) +void sched_move_task(struct task_struct *tsk, bool for_autogroup) { - struct task_struct *g, *p; + unsigned int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE; + bool resched = false; + struct rq *rq; - do_each_thread(g, p) { - if (rt_task(p) && task_rq(p)->rt.tg == tg) - return 1; - } while_each_thread(g, p); + CLASS(task_rq_lock, rq_guard)(tsk); + rq = rq_guard.rq; - return 0; -} + scoped_guard (sched_change, tsk, queue_flags) { + sched_change_group(tsk); + if (!for_autogroup) + scx_cgroup_move_task(tsk); + if (scope->running) + resched = true; + } -struct rt_schedulable_data { - struct task_group *tg; - u64 rt_period; - u64 rt_runtime; -}; + if (resched) + resched_curr(rq); +} -static int tg_rt_schedulable(struct task_group *tg, void *data) +static struct cgroup_subsys_state * +cpu_cgroup_css_alloc(struct cgroup_subsys_state *parent_css) { - struct rt_schedulable_data *d = data; - struct task_group *child; - unsigned long total, sum = 0; - u64 period, runtime; - - period = ktime_to_ns(tg->rt_bandwidth.rt_period); - runtime = tg->rt_bandwidth.rt_runtime; + struct task_group *parent = css_tg(parent_css); + struct task_group *tg; - if (tg == d->tg) { - period = d->rt_period; - runtime = d->rt_runtime; + if (!parent) { + /* This is early initialization for the top cgroup */ + return &root_task_group.css; } - /* - * Cannot have more runtime than the period. - */ - if (runtime > period && runtime != RUNTIME_INF) - return -EINVAL; - - /* - * Ensure we don't starve existing RT tasks. - */ - if (rt_bandwidth_enabled() && !runtime && tg_has_rt_tasks(tg)) - return -EBUSY; - - total = to_ratio(period, runtime); + tg = sched_create_group(parent); + if (IS_ERR(tg)) + return ERR_PTR(-ENOMEM); - /* - * Nobody can have more than the global setting allows. - */ - if (total > to_ratio(global_rt_period(), global_rt_runtime())) - return -EINVAL; + return &tg->css; +} - /* - * The sum of our children's runtime should not exceed our own. - */ - list_for_each_entry_rcu(child, &tg->children, siblings) { - period = ktime_to_ns(child->rt_bandwidth.rt_period); - runtime = child->rt_bandwidth.rt_runtime; +/* Expose task group only after completing cgroup initialization */ +static int cpu_cgroup_css_online(struct cgroup_subsys_state *css) +{ + struct task_group *tg = css_tg(css); + struct task_group *parent = css_tg(css->parent); + int ret; - if (child == d->tg) { - period = d->rt_period; - runtime = d->rt_runtime; - } + ret = scx_tg_online(tg); + if (ret) + return ret; - sum += to_ratio(period, runtime); - } + if (parent) + sched_online_group(tg, parent); - if (sum > total) - return -EINVAL; +#ifdef CONFIG_UCLAMP_TASK_GROUP + /* Propagate the effective uclamp value for the new group */ + guard(mutex)(&uclamp_mutex); + guard(rcu)(); + cpu_util_update_eff(css); +#endif return 0; } -static int __rt_schedulable(struct task_group *tg, u64 period, u64 runtime) +static void cpu_cgroup_css_offline(struct cgroup_subsys_state *css) { - int ret; + struct task_group *tg = css_tg(css); - struct rt_schedulable_data data = { - .tg = tg, - .rt_period = period, - .rt_runtime = runtime, - }; - - rcu_read_lock(); - ret = walk_tg_tree(tg_rt_schedulable, tg_nop, &data); - rcu_read_unlock(); - - return ret; + scx_tg_offline(tg); } -static int tg_set_rt_bandwidth(struct task_group *tg, - u64 rt_period, u64 rt_runtime) +static void cpu_cgroup_css_released(struct cgroup_subsys_state *css) { - int i, err = 0; - - mutex_lock(&rt_constraints_mutex); - read_lock(&tasklist_lock); - err = __rt_schedulable(tg, rt_period, rt_runtime); - if (err) - goto unlock; - - raw_spin_lock_irq(&tg->rt_bandwidth.rt_runtime_lock); - tg->rt_bandwidth.rt_period = ns_to_ktime(rt_period); - tg->rt_bandwidth.rt_runtime = rt_runtime; - - for_each_possible_cpu(i) { - struct rt_rq *rt_rq = tg->rt_rq[i]; + struct task_group *tg = css_tg(css); - raw_spin_lock(&rt_rq->rt_runtime_lock); - rt_rq->rt_runtime = rt_runtime; - raw_spin_unlock(&rt_rq->rt_runtime_lock); - } - raw_spin_unlock_irq(&tg->rt_bandwidth.rt_runtime_lock); -unlock: - read_unlock(&tasklist_lock); - mutex_unlock(&rt_constraints_mutex); - - return err; + sched_release_group(tg); } -static int sched_group_set_rt_runtime(struct task_group *tg, long rt_runtime_us) +static void cpu_cgroup_css_free(struct cgroup_subsys_state *css) { - u64 rt_runtime, rt_period; + struct task_group *tg = css_tg(css); - rt_period = ktime_to_ns(tg->rt_bandwidth.rt_period); - rt_runtime = (u64)rt_runtime_us * NSEC_PER_USEC; - if (rt_runtime_us < 0) - rt_runtime = RUNTIME_INF; - - return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); + /* + * Relies on the RCU grace period between css_released() and this. + */ + sched_unregister_group(tg); } -static long sched_group_rt_runtime(struct task_group *tg) +static int cpu_cgroup_can_attach(struct cgroup_taskset *tset) { - u64 rt_runtime_us; +#ifdef CONFIG_RT_GROUP_SCHED + struct task_struct *task; + struct cgroup_subsys_state *css; - if (tg->rt_bandwidth.rt_runtime == RUNTIME_INF) - return -1; + if (!rt_group_sched_enabled()) + goto scx_check; - rt_runtime_us = tg->rt_bandwidth.rt_runtime; - do_div(rt_runtime_us, NSEC_PER_USEC); - return rt_runtime_us; + cgroup_taskset_for_each(task, css, tset) { + if (!sched_rt_can_attach(css_tg(css), task)) + return -EINVAL; + } +scx_check: +#endif /* CONFIG_RT_GROUP_SCHED */ + return scx_cgroup_can_attach(tset); } -static int sched_group_set_rt_period(struct task_group *tg, long rt_period_us) +static void cpu_cgroup_attach(struct cgroup_taskset *tset) { - u64 rt_runtime, rt_period; - - rt_period = (u64)rt_period_us * NSEC_PER_USEC; - rt_runtime = tg->rt_bandwidth.rt_runtime; - - if (rt_period == 0) - return -EINVAL; + struct task_struct *task; + struct cgroup_subsys_state *css; - return tg_set_rt_bandwidth(tg, rt_period, rt_runtime); + cgroup_taskset_for_each(task, css, tset) + sched_move_task(task, false); } -static long sched_group_rt_period(struct task_group *tg) +static void cpu_cgroup_cancel_attach(struct cgroup_taskset *tset) { - u64 rt_period_us; - - rt_period_us = ktime_to_ns(tg->rt_bandwidth.rt_period); - do_div(rt_period_us, NSEC_PER_USEC); - return rt_period_us; + scx_cgroup_cancel_attach(tset); } -static int sched_rt_global_constraints(void) +#ifdef CONFIG_UCLAMP_TASK_GROUP +static void cpu_util_update_eff(struct cgroup_subsys_state *css) { - u64 runtime, period; - int ret = 0; - - if (sysctl_sched_rt_period <= 0) - return -EINVAL; + struct cgroup_subsys_state *top_css = css; + struct uclamp_se *uc_parent = NULL; + struct uclamp_se *uc_se = NULL; + unsigned int eff[UCLAMP_CNT]; + enum uclamp_id clamp_id; + unsigned int clamps; - runtime = global_rt_runtime(); - period = global_rt_period(); + lockdep_assert_held(&uclamp_mutex); + WARN_ON_ONCE(!rcu_read_lock_held()); - /* - * Sanity check on the sysctl variables. - */ - if (runtime > period && runtime != RUNTIME_INF) - return -EINVAL; + css_for_each_descendant_pre(css, top_css) { + uc_parent = css_tg(css)->parent + ? css_tg(css)->parent->uclamp : NULL; - mutex_lock(&rt_constraints_mutex); - read_lock(&tasklist_lock); - ret = __rt_schedulable(NULL, 0, 0); - read_unlock(&tasklist_lock); - mutex_unlock(&rt_constraints_mutex); + for_each_clamp_id(clamp_id) { + /* Assume effective clamps matches requested clamps */ + eff[clamp_id] = css_tg(css)->uclamp_req[clamp_id].value; + /* Cap effective clamps with parent's effective clamps */ + if (uc_parent && + eff[clamp_id] > uc_parent[clamp_id].value) { + eff[clamp_id] = uc_parent[clamp_id].value; + } + } + /* Ensure protection is always capped by limit */ + eff[UCLAMP_MIN] = min(eff[UCLAMP_MIN], eff[UCLAMP_MAX]); + + /* Propagate most restrictive effective clamps */ + clamps = 0x0; + uc_se = css_tg(css)->uclamp; + for_each_clamp_id(clamp_id) { + if (eff[clamp_id] == uc_se[clamp_id].value) + continue; + uc_se[clamp_id].value = eff[clamp_id]; + uc_se[clamp_id].bucket_id = uclamp_bucket_id(eff[clamp_id]); + clamps |= (0x1 << clamp_id); + } + if (!clamps) { + css = css_rightmost_descendant(css); + continue; + } - return ret; + /* Immediately update descendants RUNNABLE tasks */ + uclamp_update_active_tasks(css); + } } -static int sched_rt_can_attach(struct task_group *tg, struct task_struct *tsk) -{ - /* Don't accept realtime tasks when there is no way for them to run */ - if (rt_task(tsk) && tg->rt_bandwidth.rt_runtime == 0) - return 0; - - return 1; -} +/* + * Integer 10^N with a given N exponent by casting to integer the literal "1eN" + * C expression. Since there is no way to convert a macro argument (N) into a + * character constant, use two levels of macros. + */ +#define _POW10(exp) ((unsigned int)1e##exp) +#define POW10(exp) _POW10(exp) + +struct uclamp_request { +#define UCLAMP_PERCENT_SHIFT 2 +#define UCLAMP_PERCENT_SCALE (100 * POW10(UCLAMP_PERCENT_SHIFT)) + s64 percent; + u64 util; + int ret; +}; -#else /* !CONFIG_RT_GROUP_SCHED */ -static int sched_rt_global_constraints(void) +static inline struct uclamp_request +capacity_from_percent(char *buf) { - unsigned long flags; - int i; - - if (sysctl_sched_rt_period <= 0) - return -EINVAL; - - /* - * There's always some RT tasks in the root group - * -- migration, kstopmachine etc.. - */ - if (sysctl_sched_rt_runtime == 0) - return -EBUSY; + struct uclamp_request req = { + .percent = UCLAMP_PERCENT_SCALE, + .util = SCHED_CAPACITY_SCALE, + .ret = 0, + }; - raw_spin_lock_irqsave(&def_rt_bandwidth.rt_runtime_lock, flags); - for_each_possible_cpu(i) { - struct rt_rq *rt_rq = &cpu_rq(i)->rt; + buf = strim(buf); + if (strcmp(buf, "max")) { + req.ret = cgroup_parse_float(buf, UCLAMP_PERCENT_SHIFT, + &req.percent); + if (req.ret) + return req; + if ((u64)req.percent > UCLAMP_PERCENT_SCALE) { + req.ret = -ERANGE; + return req; + } - raw_spin_lock(&rt_rq->rt_runtime_lock); - rt_rq->rt_runtime = global_rt_runtime(); - raw_spin_unlock(&rt_rq->rt_runtime_lock); + req.util = req.percent << SCHED_CAPACITY_SHIFT; + req.util = DIV_ROUND_CLOSEST_ULL(req.util, UCLAMP_PERCENT_SCALE); } - raw_spin_unlock_irqrestore(&def_rt_bandwidth.rt_runtime_lock, flags); - return 0; + return req; } -#endif /* CONFIG_RT_GROUP_SCHED */ -int sched_rr_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) +static ssize_t cpu_uclamp_write(struct kernfs_open_file *of, char *buf, + size_t nbytes, loff_t off, + enum uclamp_id clamp_id) { - int ret; - static DEFINE_MUTEX(mutex); + struct uclamp_request req; + struct task_group *tg; - mutex_lock(&mutex); - ret = proc_dointvec(table, write, buffer, lenp, ppos); - /* make sure that internally we keep jiffies */ - /* also, writing zero resets timeslice to default */ - if (!ret && write) { - sched_rr_timeslice = sched_rr_timeslice <= 0 ? - RR_TIMESLICE : msecs_to_jiffies(sched_rr_timeslice); - } - mutex_unlock(&mutex); - return ret; -} + req = capacity_from_percent(buf); + if (req.ret) + return req.ret; -int sched_rt_handler(struct ctl_table *table, int write, - void __user *buffer, size_t *lenp, - loff_t *ppos) -{ - int ret; - int old_period, old_runtime; - static DEFINE_MUTEX(mutex); + sched_uclamp_enable(); - mutex_lock(&mutex); - old_period = sysctl_sched_rt_period; - old_runtime = sysctl_sched_rt_runtime; + guard(mutex)(&uclamp_mutex); + guard(rcu)(); - ret = proc_dointvec(table, write, buffer, lenp, ppos); + tg = css_tg(of_css(of)); + if (tg->uclamp_req[clamp_id].value != req.util) + uclamp_se_set(&tg->uclamp_req[clamp_id], req.util, false); - if (!ret && write) { - ret = sched_rt_global_constraints(); - if (ret) { - sysctl_sched_rt_period = old_period; - sysctl_sched_rt_runtime = old_runtime; - } else { - def_rt_bandwidth.rt_runtime = global_rt_runtime(); - def_rt_bandwidth.rt_period = - ns_to_ktime(global_rt_period()); - } - } - mutex_unlock(&mutex); + /* + * Because of not recoverable conversion rounding we keep track of the + * exact requested value + */ + tg->uclamp_pct[clamp_id] = req.percent; - return ret; + /* Update effective clamps to track the most restrictive value */ + cpu_util_update_eff(of_css(of)); + + return nbytes; } -#ifdef CONFIG_CGROUP_SCHED +static ssize_t cpu_uclamp_min_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, + loff_t off) +{ + return cpu_uclamp_write(of, buf, nbytes, off, UCLAMP_MIN); +} -/* return corresponding task_group object of a cgroup */ -static inline struct task_group *cgroup_tg(struct cgroup *cgrp) +static ssize_t cpu_uclamp_max_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, + loff_t off) { - return container_of(cgroup_subsys_state(cgrp, cpu_cgroup_subsys_id), - struct task_group, css); + return cpu_uclamp_write(of, buf, nbytes, off, UCLAMP_MAX); } -static struct cgroup_subsys_state *cpu_cgroup_css_alloc(struct cgroup *cgrp) +static inline void cpu_uclamp_print(struct seq_file *sf, + enum uclamp_id clamp_id) { - struct task_group *tg, *parent; + struct task_group *tg; + u64 util_clamp; + u64 percent; + u32 rem; - if (!cgrp->parent) { - /* This is early initialization for the top cgroup */ - return &root_task_group.css; + scoped_guard (rcu) { + tg = css_tg(seq_css(sf)); + util_clamp = tg->uclamp_req[clamp_id].value; } - parent = cgroup_tg(cgrp->parent); - tg = sched_create_group(parent); - if (IS_ERR(tg)) - return ERR_PTR(-ENOMEM); + if (util_clamp == SCHED_CAPACITY_SCALE) { + seq_puts(sf, "max\n"); + return; + } - return &tg->css; + percent = tg->uclamp_pct[clamp_id]; + percent = div_u64_rem(percent, POW10(UCLAMP_PERCENT_SHIFT), &rem); + seq_printf(sf, "%llu.%0*u\n", percent, UCLAMP_PERCENT_SHIFT, rem); } -static int cpu_cgroup_css_online(struct cgroup *cgrp) +static int cpu_uclamp_min_show(struct seq_file *sf, void *v) { - struct task_group *tg = cgroup_tg(cgrp); - struct task_group *parent; - - if (!cgrp->parent) - return 0; - - parent = cgroup_tg(cgrp->parent); - sched_online_group(tg, parent); + cpu_uclamp_print(sf, UCLAMP_MIN); return 0; } -static void cpu_cgroup_css_free(struct cgroup *cgrp) +static int cpu_uclamp_max_show(struct seq_file *sf, void *v) { - struct task_group *tg = cgroup_tg(cgrp); - - sched_destroy_group(tg); -} - -static void cpu_cgroup_css_offline(struct cgroup *cgrp) -{ - struct task_group *tg = cgroup_tg(cgrp); - - sched_offline_group(tg); + cpu_uclamp_print(sf, UCLAMP_MAX); + return 0; } +#endif /* CONFIG_UCLAMP_TASK_GROUP */ -static int cpu_cgroup_can_attach(struct cgroup *cgrp, - struct cgroup_taskset *tset) +#ifdef CONFIG_GROUP_SCHED_WEIGHT +static unsigned long tg_weight(struct task_group *tg) { - struct task_struct *task; - - cgroup_taskset_for_each(task, cgrp, tset) { -#ifdef CONFIG_RT_GROUP_SCHED - if (!sched_rt_can_attach(cgroup_tg(cgrp), task)) - return -EINVAL; +#ifdef CONFIG_FAIR_GROUP_SCHED + return scale_load_down(tg->shares); #else - /* We don't support RT-tasks being in separate groups */ - if (task->sched_class != &fair_sched_class) - return -EINVAL; + return sched_weight_from_cgroup(tg->scx.weight); #endif - } - return 0; -} - -static void cpu_cgroup_attach(struct cgroup *cgrp, - struct cgroup_taskset *tset) -{ - struct task_struct *task; - - cgroup_taskset_for_each(task, cgrp, tset) - sched_move_task(task); } -static void -cpu_cgroup_exit(struct cgroup *cgrp, struct cgroup *old_cgrp, - struct task_struct *task) +static int cpu_shares_write_u64(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 shareval) { - /* - * cgroup_exit() is called in the copy_process() failure path. - * Ignore this case since the task hasn't ran yet, this avoids - * trying to poke a half freed task state from generic code. - */ - if (!(task->flags & PF_EXITING)) - return; - - sched_move_task(task); -} + int ret; -#ifdef CONFIG_FAIR_GROUP_SCHED -static int cpu_shares_write_u64(struct cgroup *cgrp, struct cftype *cftype, - u64 shareval) -{ - return sched_group_set_shares(cgroup_tg(cgrp), scale_load(shareval)); + if (shareval > scale_load_down(ULONG_MAX)) + shareval = MAX_SHARES; + ret = sched_group_set_shares(css_tg(css), scale_load(shareval)); + if (!ret) + scx_group_set_weight(css_tg(css), + sched_weight_to_cgroup(shareval)); + return ret; } -static u64 cpu_shares_read_u64(struct cgroup *cgrp, struct cftype *cft) +static u64 cpu_shares_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) { - struct task_group *tg = cgroup_tg(cgrp); - - return (u64) scale_load_down(tg->shares); + return tg_weight(css_tg(css)); } +#endif /* CONFIG_GROUP_SCHED_WEIGHT */ #ifdef CONFIG_CFS_BANDWIDTH static DEFINE_MUTEX(cfs_constraints_mutex); -const u64 max_cfs_quota_period = 1 * NSEC_PER_SEC; /* 1s */ -const u64 min_cfs_quota_period = 1 * NSEC_PER_MSEC; /* 1ms */ - 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) +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; + period = (u64)period_us * NSEC_PER_USEC; - /* - * 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; + if (quota_us == RUNTIME_INF) + quota = RUNTIME_INF; + else + quota = (u64)quota_us * NSEC_PER_USEC; + + burst = (u64)burst_us * NSEC_PER_USEC; /* - * Likewise, bound things on the otherside by preventing insane quota - * periods. This also allows us to normalize in computing quota - * feasibility. + * Prevent race between setting of cfs_rq->runtime_enabled and + * unthrottle_offline_cfs_rqs(). */ - if (period > max_cfs_quota_period) - return -EINVAL; + guard(cpus_read_lock)(); + guard(mutex)(&cfs_constraints_mutex); - mutex_lock(&cfs_constraints_mutex); ret = __cfs_schedulable(tg, period, quota); if (ret) - goto out_unlock; + return ret; runtime_enabled = quota != RUNTIME_INF; runtime_was_enabled = cfs_b->quota != RUNTIME_INF; - account_cfs_bandwidth_used(runtime_enabled, runtime_was_enabled); - raw_spin_lock_irq(&cfs_b->lock); - cfs_b->period = ns_to_ktime(period); - cfs_b->quota = quota; + /* + * If we need to toggle cfs_bandwidth_used, off->on must occur + * before making related changes, and on->off must occur afterwards + */ + if (runtime_enabled && !runtime_was_enabled) + cfs_bandwidth_usage_inc(); + + scoped_guard (raw_spinlock_irq, &cfs_b->lock) { + cfs_b->period = ns_to_ktime(period); + cfs_b->quota = quota; + cfs_b->burst = burst; - __refill_cfs_bandwidth_runtime(cfs_b); - /* restart the period timer (if active) to handle new period expiry */ - if (runtime_enabled && cfs_b->timer_active) { - /* force a reprogram */ - cfs_b->timer_active = 0; - __start_cfs_bandwidth(cfs_b); + __refill_cfs_bandwidth_runtime(cfs_b); + + /* + * Restart the period timer (if active) to handle new + * period expiry: + */ + if (runtime_enabled) + start_cfs_bandwidth(cfs_b); } - raw_spin_unlock_irq(&cfs_b->lock); - for_each_possible_cpu(i) { + for_each_online_cpu(i) { struct cfs_rq *cfs_rq = tg->cfs_rq[i]; struct rq *rq = cfs_rq->rq; - raw_spin_lock_irq(&rq->lock); + 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); - raw_spin_unlock_irq(&rq->lock); } -out_unlock: - mutex_unlock(&cfs_constraints_mutex); - return ret; + if (runtime_was_enabled && !runtime_enabled) + cfs_bandwidth_usage_dec(); + + return 0; } -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; + u64 cfs_period_us; - period = ktime_to_ns(tg->cfs_bandwidth.period); - if (cfs_quota_us < 0) - quota = RUNTIME_INF; - else - quota = (u64)cfs_quota_us * NSEC_PER_USEC; + 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); + return cfs_period_us; } -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); @@ -7289,46 +9526,14 @@ long tg_get_cfs_quota(struct task_group *tg) return quota_us; } -int tg_set_cfs_period(struct task_group *tg, long cfs_period_us) -{ - u64 quota, period; - - period = (u64)cfs_period_us * NSEC_PER_USEC; - quota = tg->cfs_bandwidth.quota; - - return tg_set_cfs_bandwidth(tg, period, quota); -} - -long tg_get_cfs_period(struct task_group *tg) +static u64 tg_get_cfs_burst(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); + u64 burst_us; - return cfs_period_us; -} + burst_us = tg->cfs_bandwidth.burst; + do_div(burst_us, NSEC_PER_USEC); -static s64 cpu_cfs_quota_read_s64(struct cgroup *cgrp, struct cftype *cft) -{ - return tg_get_cfs_quota(cgroup_tg(cgrp)); -} - -static int cpu_cfs_quota_write_s64(struct cgroup *cgrp, struct cftype *cftype, - s64 cfs_quota_us) -{ - return tg_set_cfs_quota(cgroup_tg(cgrp), cfs_quota_us); -} - -static u64 cpu_cfs_period_read_u64(struct cgroup *cgrp, struct cftype *cft) -{ - return tg_get_cfs_period(cgroup_tg(cgrp)); -} - -static int cpu_cfs_period_write_u64(struct cgroup *cgrp, struct cftype *cftype, - u64 cfs_period_us) -{ - return tg_set_cfs_period(cgroup_tg(cgrp), cfs_period_us); + return burst_us; } struct cfs_schedulable_data { @@ -7372,25 +9577,34 @@ static int tg_cfs_schedulable_down(struct task_group *tg, void *data) struct cfs_bandwidth *parent_b = &tg->parent->cfs_bandwidth; quota = normalize_cfs_quota(tg, d); - parent_quota = parent_b->hierarchal_quota; + parent_quota = parent_b->hierarchical_quota; /* - * ensure max(child_quota) <= parent_quota, inherit when no - * limit is set + * Ensure max(child_quota) <= parent_quota. On cgroup2, + * always take the non-RUNTIME_INF min. On cgroup1, only + * inherit when no limit is set. In both cases this is used + * by the scheduler to determine if a given CFS task has a + * bandwidth constraint at some higher level. */ - if (quota == RUNTIME_INF) - quota = parent_quota; - else if (parent_quota != RUNTIME_INF && quota > parent_quota) - return -EINVAL; + if (cgroup_subsys_on_dfl(cpu_cgrp_subsys)) { + if (quota == RUNTIME_INF) + quota = parent_quota; + else if (parent_quota != RUNTIME_INF) + quota = min(quota, parent_quota); + } else { + if (quota == RUNTIME_INF) + quota = parent_quota; + else if (parent_quota != RUNTIME_INF && quota > parent_quota) + return -EINVAL; + } } - cfs_b->hierarchal_quota = quota; + cfs_b->hierarchical_quota = quota; return 0; } static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) { - int ret; struct cfs_schedulable_data data = { .tg = tg, .period = period, @@ -7402,77 +9616,301 @@ static int __cfs_schedulable(struct task_group *tg, u64 period, u64 quota) do_div(data.quota, NSEC_PER_USEC); } - rcu_read_lock(); - ret = walk_tg_tree(tg_cfs_schedulable_down, tg_nop, &data); - rcu_read_unlock(); - - return ret; + guard(rcu)(); + return walk_tg_tree(tg_cfs_schedulable_down, tg_nop, &data); } -static int cpu_stats_show(struct cgroup *cgrp, struct cftype *cft, - struct cgroup_map_cb *cb) +static int cpu_cfs_stat_show(struct seq_file *sf, void *v) { - struct task_group *tg = cgroup_tg(cgrp); + struct task_group *tg = css_tg(seq_css(sf)); struct cfs_bandwidth *cfs_b = &tg->cfs_bandwidth; - cb->fill(cb, "nr_periods", cfs_b->nr_periods); - cb->fill(cb, "nr_throttled", cfs_b->nr_throttled); - cb->fill(cb, "throttled_time", cfs_b->throttled_time); + seq_printf(sf, "nr_periods %d\n", cfs_b->nr_periods); + seq_printf(sf, "nr_throttled %d\n", cfs_b->nr_throttled); + seq_printf(sf, "throttled_time %llu\n", cfs_b->throttled_time); + + if (schedstat_enabled() && tg != &root_task_group) { + struct sched_statistics *stats; + u64 ws = 0; + int i; + + for_each_possible_cpu(i) { + stats = __schedstats_from_se(tg->se[i]); + ws += schedstat_val(stats->wait_sum); + } + + seq_printf(sf, "wait_sum %llu\n", ws); + } + + seq_printf(sf, "nr_bursts %d\n", cfs_b->nr_burst); + seq_printf(sf, "burst_time %llu\n", cfs_b->burst_time); + + return 0; +} + +static u64 throttled_time_self(struct task_group *tg) +{ + int i; + u64 total = 0; + + for_each_possible_cpu(i) { + total += READ_ONCE(tg->cfs_rq[i]->throttled_clock_self_time); + } + + return total; +} + +static int cpu_cfs_local_stat_show(struct seq_file *sf, void *v) +{ + struct task_group *tg = css_tg(seq_css(sf)); + + seq_printf(sf, "throttled_time %llu\n", throttled_time_self(tg)); return 0; } #endif /* CONFIG_CFS_BANDWIDTH */ -#endif /* CONFIG_FAIR_GROUP_SCHED */ + +#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 *cgrp, struct cftype *cft, - s64 val) +static int cpu_rt_runtime_write(struct cgroup_subsys_state *css, + struct cftype *cft, s64 val) { - return sched_group_set_rt_runtime(cgroup_tg(cgrp), val); + return sched_group_set_rt_runtime(css_tg(css), val); } -static s64 cpu_rt_runtime_read(struct cgroup *cgrp, struct cftype *cft) +static s64 cpu_rt_runtime_read(struct cgroup_subsys_state *css, + struct cftype *cft) { - return sched_group_rt_runtime(cgroup_tg(cgrp)); + return sched_group_rt_runtime(css_tg(css)); } -static int cpu_rt_period_write_uint(struct cgroup *cgrp, struct cftype *cftype, - u64 rt_period_us) +static int cpu_rt_period_write_uint(struct cgroup_subsys_state *css, + struct cftype *cftype, u64 rt_period_us) { - return sched_group_set_rt_period(cgroup_tg(cgrp), rt_period_us); + return sched_group_set_rt_period(css_tg(css), rt_period_us); } -static u64 cpu_rt_period_read_uint(struct cgroup *cgrp, struct cftype *cft) +static u64 cpu_rt_period_read_uint(struct cgroup_subsys_state *css, + struct cftype *cft) { - return sched_group_rt_period(cgroup_tg(cgrp)); + return sched_group_rt_period(css_tg(css)); } #endif /* CONFIG_RT_GROUP_SCHED */ -static struct cftype cpu_files[] = { -#ifdef CONFIG_FAIR_GROUP_SCHED +#ifdef CONFIG_GROUP_SCHED_WEIGHT +static s64 cpu_idle_read_s64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + return css_tg(css)->idle; +} + +static int cpu_idle_write_s64(struct cgroup_subsys_state *css, + struct cftype *cft, s64 idle) +{ + int ret; + + ret = sched_group_set_idle(css_tg(css), idle); + if (!ret) + scx_group_set_idle(css_tg(css), idle); + return ret; +} +#endif /* CONFIG_GROUP_SCHED_WEIGHT */ + +static struct cftype cpu_legacy_files[] = { +#ifdef CONFIG_GROUP_SCHED_WEIGHT { .name = "shares", .read_u64 = cpu_shares_read_u64, .write_u64 = cpu_shares_write_u64, }, + { + .name = "idle", + .read_s64 = cpu_idle_read_s64, + .write_s64 = cpu_idle_write_s64, + }, #endif -#ifdef CONFIG_CFS_BANDWIDTH +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH + { + .name = "cfs_period_us", + .read_u64 = cpu_period_read_u64, + .write_u64 = cpu_period_write_u64, + }, { .name = "cfs_quota_us", - .read_s64 = cpu_cfs_quota_read_s64, - .write_s64 = cpu_cfs_quota_write_s64, + .read_s64 = cpu_quota_read_s64, + .write_s64 = cpu_quota_write_s64, }, { - .name = "cfs_period_us", - .read_u64 = cpu_cfs_period_read_u64, - .write_u64 = cpu_cfs_period_write_u64, + .name = "cfs_burst_us", + .read_u64 = cpu_burst_read_u64, + .write_u64 = cpu_burst_write_u64, }, +#endif +#ifdef CONFIG_CFS_BANDWIDTH { .name = "stat", - .read_map = cpu_stats_show, + .seq_show = cpu_cfs_stat_show, + }, + { + .name = "stat.local", + .seq_show = cpu_cfs_local_stat_show, + }, +#endif +#ifdef CONFIG_UCLAMP_TASK_GROUP + { + .name = "uclamp.min", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cpu_uclamp_min_show, + .write = cpu_uclamp_min_write, + }, + { + .name = "uclamp.max", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cpu_uclamp_max_show, + .write = cpu_uclamp_max_write, }, #endif + { } /* Terminate */ +}; + #ifdef CONFIG_RT_GROUP_SCHED +static struct cftype rt_group_files[] = { { .name = "rt_runtime_us", .read_s64 = cpu_rt_runtime_read, @@ -7483,28 +9921,909 @@ static struct cftype cpu_files[] = { .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) +{ +#ifdef CONFIG_CFS_BANDWIDTH + { + struct task_group *tg = css_tg(css); + struct cfs_bandwidth *cfs_b = &tg->cfs_bandwidth; + u64 throttled_usec, burst_usec; + + throttled_usec = cfs_b->throttled_time; + do_div(throttled_usec, NSEC_PER_USEC); + burst_usec = cfs_b->burst_time; + do_div(burst_usec, NSEC_PER_USEC); + + seq_printf(sf, "nr_periods %d\n" + "nr_throttled %d\n" + "throttled_usec %llu\n" + "nr_bursts %d\n" + "burst_usec %llu\n", + cfs_b->nr_periods, cfs_b->nr_throttled, + throttled_usec, cfs_b->nr_burst, burst_usec); + } +#endif /* CONFIG_CFS_BANDWIDTH */ + return 0; +} + +static int cpu_local_stat_show(struct seq_file *sf, + struct cgroup_subsys_state *css) +{ +#ifdef CONFIG_CFS_BANDWIDTH + { + struct task_group *tg = css_tg(css); + u64 throttled_self_usec; + + throttled_self_usec = throttled_time_self(tg); + do_div(throttled_self_usec, NSEC_PER_USEC); + + seq_printf(sf, "throttled_usec %llu\n", + throttled_self_usec); + } +#endif + return 0; +} + +#ifdef CONFIG_GROUP_SCHED_WEIGHT + +static u64 cpu_weight_read_u64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + return sched_weight_to_cgroup(tg_weight(css_tg(css))); +} + +static int cpu_weight_write_u64(struct cgroup_subsys_state *css, + struct cftype *cft, u64 cgrp_weight) +{ + unsigned long weight; + int ret; + + if (cgrp_weight < CGROUP_WEIGHT_MIN || cgrp_weight > CGROUP_WEIGHT_MAX) + return -ERANGE; + + weight = sched_weight_from_cgroup(cgrp_weight); + + ret = sched_group_set_shares(css_tg(css), scale_load(weight)); + if (!ret) + scx_group_set_weight(css_tg(css), cgrp_weight); + return ret; +} + +static s64 cpu_weight_nice_read_s64(struct cgroup_subsys_state *css, + struct cftype *cft) +{ + unsigned long weight = tg_weight(css_tg(css)); + int last_delta = INT_MAX; + int prio, delta; + + /* find the closest nice value to the current weight */ + for (prio = 0; prio < ARRAY_SIZE(sched_prio_to_weight); prio++) { + delta = abs(sched_prio_to_weight[prio] - weight); + if (delta >= last_delta) + break; + last_delta = delta; + } + + return PRIO_TO_NICE(prio - 1 + MAX_RT_PRIO); +} + +static int cpu_weight_nice_write_s64(struct cgroup_subsys_state *css, + struct cftype *cft, s64 nice) +{ + unsigned long weight; + int idx, ret; + + if (nice < MIN_NICE || nice > MAX_NICE) + return -ERANGE; + + idx = NICE_TO_PRIO(nice) - MAX_RT_PRIO; + idx = array_index_nospec(idx, 40); + weight = sched_prio_to_weight[idx]; + + ret = sched_group_set_shares(css_tg(css), scale_load(weight)); + if (!ret) + scx_group_set_weight(css_tg(css), + sched_weight_to_cgroup(weight)); + return ret; +} +#endif /* CONFIG_GROUP_SCHED_WEIGHT */ + +static void __maybe_unused cpu_period_quota_print(struct seq_file *sf, + long period, long quota) +{ + if (quota < 0) + seq_puts(sf, "max"); + else + seq_printf(sf, "%ld", quota); + + seq_printf(sf, " %ld\n", period); +} + +/* caller should put the current value in *@periodp before calling */ +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, period_us_p) < 1) + 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_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; + + tg_bandwidth(tg, &period_us, "a_us, NULL); + cpu_period_quota_print(sf, period_us, quota_us); + return 0; +} + +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_us, quota_us, burst_us; + int ret; + + tg_bandwidth(tg, &period_us, NULL, &burst_us); + ret = cpu_period_quota_parse(buf, &period_us, "a_us); + if (!ret) + ret = tg_set_bandwidth(tg, period_us, quota_us, burst_us); + return ret ?: nbytes; +} +#endif /* CONFIG_CFS_BANDWIDTH */ + +static struct cftype cpu_files[] = { +#ifdef CONFIG_GROUP_SCHED_WEIGHT + { + .name = "weight", + .flags = CFTYPE_NOT_ON_ROOT, + .read_u64 = cpu_weight_read_u64, + .write_u64 = cpu_weight_write_u64, + }, + { + .name = "weight.nice", + .flags = CFTYPE_NOT_ON_ROOT, + .read_s64 = cpu_weight_nice_read_s64, + .write_s64 = cpu_weight_nice_write_s64, + }, + { + .name = "idle", + .flags = CFTYPE_NOT_ON_ROOT, + .read_s64 = cpu_idle_read_s64, + .write_s64 = cpu_idle_write_s64, + }, #endif +#ifdef CONFIG_GROUP_SCHED_BANDWIDTH + { + .name = "max", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cpu_max_show, + .write = cpu_max_write, + }, + { + .name = "max.burst", + .flags = CFTYPE_NOT_ON_ROOT, + .read_u64 = cpu_burst_read_u64, + .write_u64 = cpu_burst_write_u64, + }, +#endif /* CONFIG_CFS_BANDWIDTH */ +#ifdef CONFIG_UCLAMP_TASK_GROUP + { + .name = "uclamp.min", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cpu_uclamp_min_show, + .write = cpu_uclamp_min_write, + }, + { + .name = "uclamp.max", + .flags = CFTYPE_NOT_ON_ROOT, + .seq_show = cpu_uclamp_max_show, + .write = cpu_uclamp_max_write, + }, +#endif /* CONFIG_UCLAMP_TASK_GROUP */ { } /* terminate */ }; -struct cgroup_subsys cpu_cgroup_subsys = { - .name = "cpu", +struct cgroup_subsys cpu_cgrp_subsys = { .css_alloc = cpu_cgroup_css_alloc, - .css_free = cpu_cgroup_css_free, .css_online = cpu_cgroup_css_online, .css_offline = cpu_cgroup_css_offline, + .css_released = cpu_cgroup_css_released, + .css_free = cpu_cgroup_css_free, + .css_extra_stat_show = cpu_extra_stat_show, + .css_local_stat_show = cpu_local_stat_show, .can_attach = cpu_cgroup_can_attach, .attach = cpu_cgroup_attach, - .exit = cpu_cgroup_exit, - .subsys_id = cpu_cgroup_subsys_id, - .base_cftypes = cpu_files, - .early_init = 1, + .cancel_attach = cpu_cgroup_cancel_attach, + .legacy_cftypes = cpu_legacy_files, + .dfl_cftypes = cpu_files, + .early_init = true, + .threaded = true, }; -#endif /* CONFIG_CGROUP_SCHED */ +#endif /* CONFIG_CGROUP_SCHED */ void dump_cpu_task(int cpu) { + if (in_hardirq() && cpu == smp_processor_id()) { + struct pt_regs *regs; + + regs = get_irq_regs(); + if (regs) { + show_regs(regs); + return; + } + } + + if (trigger_single_cpu_backtrace(cpu)) + return; + pr_info("Task dump for CPU %d:\n", cpu); sched_show_task(cpu_curr(cpu)); } + +/* + * Nice levels are multiplicative, with a gentle 10% change for every + * nice level changed. I.e. when a CPU-bound task goes from nice 0 to + * nice 1, it will get ~10% less CPU time than another CPU-bound task + * that remained on nice 0. + * + * The "10% effect" is relative and cumulative: from _any_ nice level, + * if you go up 1 level, it's -10% CPU usage, if you go down 1 level + * it's +10% CPU usage. (to achieve that we use a multiplier of 1.25. + * If a task goes up by ~10% and another task goes down by ~10% then + * the relative distance between them is ~25%.) + */ +const int sched_prio_to_weight[40] = { + /* -20 */ 88761, 71755, 56483, 46273, 36291, + /* -15 */ 29154, 23254, 18705, 14949, 11916, + /* -10 */ 9548, 7620, 6100, 4904, 3906, + /* -5 */ 3121, 2501, 1991, 1586, 1277, + /* 0 */ 1024, 820, 655, 526, 423, + /* 5 */ 335, 272, 215, 172, 137, + /* 10 */ 110, 87, 70, 56, 45, + /* 15 */ 36, 29, 23, 18, 15, +}; + +/* + * Inverse (2^32/x) values of the sched_prio_to_weight[] array, pre-calculated. + * + * In cases where the weight does not change often, we can use the + * pre-calculated inverse to speed up arithmetics by turning divisions + * into multiplications: + */ +const u32 sched_prio_to_wmult[40] = { + /* -20 */ 48388, 59856, 76040, 92818, 118348, + /* -15 */ 147320, 184698, 229616, 287308, 360437, + /* -10 */ 449829, 563644, 704093, 875809, 1099582, + /* -5 */ 1376151, 1717300, 2157191, 2708050, 3363326, + /* 0 */ 4194304, 5237765, 6557202, 8165337, 10153587, + /* 5 */ 12820798, 15790321, 19976592, 24970740, 31350126, + /* 10 */ 39045157, 49367440, 61356676, 76695844, 95443717, + /* 15 */ 119304647, 148102320, 186737708, 238609294, 286331153, +}; + +void call_trace_sched_update_nr_running(struct rq *rq, int count) +{ + trace_sched_update_nr_running_tp(rq, count); +} + +#ifdef CONFIG_SCHED_MM_CID +/* + * Concurrency IDentifier management + * + * Serialization rules: + * + * mm::mm_cid::mutex: Serializes fork() and exit() and therefore + * protects mm::mm_cid::users. + * + * 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. + * + * CID ownership: + * + * 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. During transition from CPU to task ownership mode, + * MM_CID_TRANSIT is set on the per task CIDs. When this bit is set the + * task needs to drop the CID into the pool when scheduling out. 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. + * + * Mode switching: + * + * Switching to per CPU mode happens when the user count becomes greater + * than the maximum number of CIDs, which is calculated by: + * + * opt_cids = min(mm_cid::nr_cpus_allowed, mm_cid::users); + * max_cids = min(1.25 * opt_cids, num_possible_cpus()); + * + * 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. + * + * 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 transfers each tasks owned CID to the CPU the task runs on or + * drops it into the CID pool if a task is not on a CPU at that point in + * time. 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's guaranteed that no task related to that MM owns a CID anymore. + * + * 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 transfers the CPU owned CIDs to a related task which + * runs on the CPU or drops it into the pool. Tasks which schedule in on a + * CPU which the walk did not cover yet do the handover themself. + * + * This transition from CPU to per task ownership happens in two phases: + * + * 1) mm:mm_cid.transit contains MM_CID_TRANSIT This is OR'ed on the task + * CID and denotes that the CID is only temporarily owned by the + * task. When it schedules out the task drops the CID back into the + * pool if this bit is set. + * + * 2) The initiating context walks the per CPU space and after completion + * clears mm:mm_cid.transit. So after that point the CIDs are strictly + * task owned again. + * + * This two phase transition is required to prevent CID space exhaustion + * during the transition as a direct transfer of ownership would fail if + * two tasks are scheduled in on the same CPU before the fixup freed per + * CPU CIDs. + * + * When mm_cid_fixup_cpus_to_tasks() completes it's guaranteed that no CID + * related to that MM is owned by a CPU anymore. + */ + +/* + * 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) +{ + unsigned int opt_cids, max_cids; + + /* Calculate the new optimal constraint */ + opt_cids = min(mc->nr_cpus_allowed, mc->users); + + /* 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); +} + +static inline unsigned int mm_cid_calc_pcpu_thrs(struct mm_mm_cid *mc) +{ + unsigned int opt_cids; + + 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 bool mm_update_max_cids(struct mm_struct *mm) +{ + struct mm_mm_cid *mc = &mm->mm_cid; + + lockdep_assert_held(&mm->mm_cid.lock); + + /* Clear deferred mode switch flag. A change is handled by the caller */ + mc->update_deferred = false; + __mm_update_max_cids(mc); + + /* Check whether owner mode must be changed */ + if (!mc->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 (!!mc->percpu == !!mc->pcpu_thrs) + return false; + /* When switching back to per TASK mode, set the transition flag */ + if (!mc->pcpu_thrs) + WRITE_ONCE(mc->transit, MM_CID_TRANSIT); + WRITE_ONCE(mc->percpu, !!mc->pcpu_thrs); + return true; +} + +static inline void mm_update_cpus_allowed(struct mm_struct *mm, const struct cpumask *affmsk) +{ + struct cpumask *mm_allowed; + struct mm_mm_cid *mc; + unsigned int weight; + + if (!mm || !READ_ONCE(mm->mm_cid.users)) + return; + /* + * 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; + + WRITE_ONCE(mc->nr_cpus_allowed, weight); + __mm_update_max_cids(mc); + if (!mc->percpu) + return; + + /* 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; + + /* Don't queue twice */ + if (mc->update_deferred) + return; + + /* Queue the irq work, which schedules the real work */ + mc->update_deferred = true; + irq_work_queue(&mc->irq_work); +} + +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 mm_cid_fixup_cpus_to_tasks(struct mm_struct *mm) +{ + unsigned int 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); + + /* 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; + } + } + } + /* Clear the transition bit */ + WRITE_ONCE(mm->mm_cid.transit, 0); +} + +static inline void mm_cid_transfer_to_cpu(struct task_struct *t, struct mm_cid_pcpu *pcp) +{ + if (cid_on_task(t->mm_cid.cid)) { + t->mm_cid.cid = cid_to_cpu_cid(t->mm_cid.cid); + pcp->cid = t->mm_cid.cid; + } +} + +static bool mm_cid_fixup_task_to_cpu(struct task_struct *t, struct mm_struct *mm) +{ + /* Remote access to mm::mm_cid::pcpu requires rq_lock */ + guard(task_rq_lock)(t); + /* If the task is not active it is not in the users count */ + if (!t->mm_cid.active) + return false; + if (cid_on_task(t->mm_cid.cid)) { + /* If running on the CPU, transfer the CID, otherwise drop it */ + if (task_rq(t)->curr == t) + mm_cid_transfer_to_cpu(t, per_cpu_ptr(mm->mm_cid.pcpu, task_cpu(t))); + else + mm_unset_cid_on_task(t); + } + return true; +} + +static void mm_cid_fixup_tasks_to_cpus(void) +{ + struct mm_struct *mm = current->mm; + struct task_struct *p, *t; + unsigned int users; + + /* + * This can obviously race with a concurrent affinity change, which + * increases the number of allowed CPUs for this mm, but that does + * not affect the mode and only changes the CID constraints. A + * possible switch back to per task mode happens either in the + * deferred handler function or in the next fork()/exit(). + * + * The caller has already transferred. The newly incoming task is + * already accounted for, but not yet visible. + */ + users = mm->mm_cid.users - 2; + if (!users) + return; + + guard(rcu)(); + for_other_threads(current, t) { + if (mm_cid_fixup_task_to_cpu(t, mm)) + users--; + } + + if (!users) + return; + + /* Happens only for VM_CLONE processes. */ + for_each_process_thread(p, t) { + if (t == current || t->mm != mm) + continue; + if (mm_cid_fixup_task_to_cpu(t, mm)) { + if (--users == 0) + return; + } + } +} + +static bool sched_mm_cid_add_user(struct task_struct *t, struct mm_struct *mm) +{ + t->mm_cid.active = 1; + mm->mm_cid.users++; + return mm_update_max_cids(mm); +} + +void sched_mm_cid_fork(struct task_struct *t) +{ + struct mm_struct *mm = t->mm; + bool percpu; + + WARN_ON_ONCE(!mm || 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 (!mm->mm_cid.percpu) + t->mm_cid.cid = mm_get_cid(mm); + return; + } + + /* Handle the mode change and transfer current's CID */ + percpu = !!mm->mm_cid.percpu; + if (!percpu) + mm_cid_transit_to_task(current, pcp); + else + mm_cid_transfer_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) +{ + t->mm_cid.active = 0; + scoped_guard(preempt) { + /* Clear the transition bit */ + t->mm_cid.cid = cid_from_transit_cid(t->mm_cid.cid); + mm_unset_cid_on_task(t); + } + 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; + /* + * 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. + */ + if (WARN_ON_ONCE(mm->mm_cid.percpu)) + 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; +} + +/* + * 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; + + 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 required. Transfer currents CID */ + mm_cid_transit_to_task(current, 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); + } + } + + /* + * 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. + */ + 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 successful execve() */ +void sched_mm_cid_after_execve(struct task_struct *t) +{ + sched_mm_cid_fork(t); +} + +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; + + 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(mm->mm_cid.percpu)) + return; + } + mm_cid_fixup_cpus_to_tasks(mm); +} + +static void mm_cid_irq_work(struct irq_work *work) +{ + 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); +} + +void mm_init_cid(struct mm_struct *mm, struct task_struct *p) +{ + mm->mm_cid.max_cids = 0; + mm->mm_cid.percpu = 0; + mm->mm_cid.transit = 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); + 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) { } +#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); + + if (!(flags & DEQUEUE_NOCLOCK)) { + update_rq_clock(rq); + flags |= DEQUEUE_NOCLOCK; + } + + if (flags & DEQUEUE_CLASS) { + if (p->sched_class->switching_from) + p->sched_class->switching_from(rq, p); + } + + *ctx = (struct sched_change_ctx){ + .p = p, + .flags = flags, + .queued = task_on_rq_queued(p), + .running = task_current_donor(rq, p), + }; + + 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, 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_change_end(struct sched_change_ctx *ctx) +{ + struct task_struct *p = ctx->p; + struct rq *rq = task_rq(p); + + lockdep_assert_rq_held(rq); + + if ((ctx->flags & ENQUEUE_CLASS) && p->sched_class->switching_to) + p->sched_class->switching_to(rq, p); + + if (ctx->queued) + enqueue_task(rq, p, ctx->flags); + if (ctx->running) + set_next_task(rq, p); + + if (ctx->flags & ENQUEUE_CLASS) { + if (p->sched_class->switched_to) + p->sched_class->switched_to(rq, p); + } else { + p->sched_class->prio_changed(rq, p, ctx->prio); + } +} |