diff options
Diffstat (limited to 'kernel/sched/psi.c')
| -rw-r--r-- | kernel/sched/psi.c | 745 |
1 files changed, 411 insertions, 334 deletions
diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 8ac8b81bfee6..59fdb7ebbf22 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -41,7 +41,7 @@ * What it means for a task to be productive is defined differently * for each resource. For IO, productive means a running task. For * memory, productive means a running task that isn't a reclaimer. For - * CPU, productive means an oncpu task. + * CPU, productive means an on-CPU task. * * Naturally, the FULL state doesn't exist for the CPU resource at the * system level, but exist at the cgroup level. At the cgroup level, @@ -49,7 +49,7 @@ * resource which is being used by others outside of the cgroup or * throttled by the cgroup cpu.max configuration. * - * The percentage of wallclock time spent in those compound stall + * The percentage of wall clock time spent in those compound stall * states gives pressure numbers between 0 and 100 for each resource, * where the SOME percentage indicates workload slowdowns and the FULL * percentage indicates reduced CPU utilization: @@ -136,11 +136,15 @@ * cost-wise, yet way more sensitive and accurate than periodic * sampling of the aggregate task states would be. */ +#include <linux/sched/clock.h> +#include <linux/workqueue.h> +#include <linux/psi.h> +#include "sched.h" static int psi_bug __read_mostly; DEFINE_STATIC_KEY_FALSE(psi_disabled); -DEFINE_STATIC_KEY_TRUE(psi_cgroups_enabled); +static DEFINE_STATIC_KEY_TRUE(psi_cgroups_enabled); #ifdef CONFIG_PSI_DEFAULT_DISABLED static bool psi_enable; @@ -160,7 +164,6 @@ __setup("psi=", setup_psi); #define EXP_300s 2034 /* 1/exp(2s/300s) */ /* PSI trigger definitions */ -#define WINDOW_MIN_US 500000 /* Min window size is 500ms */ #define WINDOW_MAX_US 10000000 /* Max window size is 10s */ #define UPDATES_PER_WINDOW 10 /* 10 updates per window */ @@ -173,30 +176,53 @@ struct psi_group psi_system = { .pcpu = &system_group_pcpu, }; +static DEFINE_PER_CPU(seqcount_t, psi_seq) = SEQCNT_ZERO(psi_seq); + +static inline void psi_write_begin(int cpu) +{ + write_seqcount_begin(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline void psi_write_end(int cpu) +{ + write_seqcount_end(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline u32 psi_read_begin(int cpu) +{ + return read_seqcount_begin(per_cpu_ptr(&psi_seq, cpu)); +} + +static inline bool psi_read_retry(int cpu, u32 seq) +{ + return read_seqcount_retry(per_cpu_ptr(&psi_seq, cpu), seq); +} + static void psi_avgs_work(struct work_struct *work); static void poll_timer_fn(struct timer_list *t); static void group_init(struct psi_group *group) { - int cpu; - group->enabled = true; - for_each_possible_cpu(cpu) - seqcount_init(&per_cpu_ptr(group->pcpu, cpu)->seq); group->avg_last_update = sched_clock(); group->avg_next_update = group->avg_last_update + psi_period; - INIT_DELAYED_WORK(&group->avgs_work, psi_avgs_work); mutex_init(&group->avgs_lock); - /* Init trigger-related members */ - atomic_set(&group->poll_scheduled, 0); - mutex_init(&group->trigger_lock); - INIT_LIST_HEAD(&group->triggers); - group->poll_min_period = U32_MAX; - group->polling_next_update = ULLONG_MAX; - init_waitqueue_head(&group->poll_wait); - timer_setup(&group->poll_timer, poll_timer_fn, 0); - rcu_assign_pointer(group->poll_task, NULL); + + /* Init avg trigger-related members */ + INIT_LIST_HEAD(&group->avg_triggers); + memset(group->avg_nr_triggers, 0, sizeof(group->avg_nr_triggers)); + INIT_DELAYED_WORK(&group->avgs_work, psi_avgs_work); + + /* Init rtpoll trigger-related members */ + atomic_set(&group->rtpoll_scheduled, 0); + mutex_init(&group->rtpoll_trigger_lock); + INIT_LIST_HEAD(&group->rtpoll_triggers); + group->rtpoll_min_period = U32_MAX; + group->rtpoll_next_update = ULLONG_MAX; + init_waitqueue_head(&group->rtpoll_wait); + timer_setup(&group->rtpoll_timer, poll_timer_fn, 0); + rcu_assign_pointer(group->rtpoll_task, NULL); } void __init psi_init(void) @@ -214,28 +240,32 @@ void __init psi_init(void) group_init(&psi_system); } -static bool test_state(unsigned int *tasks, enum psi_states state, bool oncpu) +static u32 test_states(unsigned int *tasks, u32 state_mask) { - switch (state) { - case PSI_IO_SOME: - return unlikely(tasks[NR_IOWAIT]); - case PSI_IO_FULL: - return unlikely(tasks[NR_IOWAIT] && !tasks[NR_RUNNING]); - case PSI_MEM_SOME: - return unlikely(tasks[NR_MEMSTALL]); - case PSI_MEM_FULL: - return unlikely(tasks[NR_MEMSTALL] && - tasks[NR_RUNNING] == tasks[NR_MEMSTALL_RUNNING]); - case PSI_CPU_SOME: - return unlikely(tasks[NR_RUNNING] > oncpu); - case PSI_CPU_FULL: - return unlikely(tasks[NR_RUNNING] && !oncpu); - case PSI_NONIDLE: - return tasks[NR_IOWAIT] || tasks[NR_MEMSTALL] || - tasks[NR_RUNNING]; - default: - return false; + const bool oncpu = state_mask & PSI_ONCPU; + + if (tasks[NR_IOWAIT]) { + state_mask |= BIT(PSI_IO_SOME); + if (!tasks[NR_RUNNING]) + state_mask |= BIT(PSI_IO_FULL); } + + if (tasks[NR_MEMSTALL]) { + state_mask |= BIT(PSI_MEM_SOME); + if (tasks[NR_RUNNING] == tasks[NR_MEMSTALL_RUNNING]) + state_mask |= BIT(PSI_MEM_FULL); + } + + if (tasks[NR_RUNNING] > oncpu) + state_mask |= BIT(PSI_CPU_SOME); + + if (tasks[NR_RUNNING] && !oncpu) + state_mask |= BIT(PSI_CPU_FULL); + + if (tasks[NR_IOWAIT] || tasks[NR_MEMSTALL] || tasks[NR_RUNNING]) + state_mask |= BIT(PSI_NONIDLE); + + return state_mask; } static void get_recent_times(struct psi_group *group, int cpu, @@ -254,14 +284,14 @@ static void get_recent_times(struct psi_group *group, int cpu, /* Snapshot a coherent view of the CPU state */ do { - seq = read_seqcount_begin(&groupc->seq); + seq = psi_read_begin(cpu); now = cpu_clock(cpu); memcpy(times, groupc->times, sizeof(groupc->times)); state_mask = groupc->state_mask; state_start = groupc->state_start; if (cpu == current_cpu) memcpy(tasks, groupc->tasks, sizeof(groupc->tasks)); - } while (read_seqcount_retry(&groupc->seq, seq)); + } while (psi_read_retry(cpu, seq)); /* Calculate state time deltas against the previous snapshot */ for (s = 0; s < NR_PSI_STATES; s++) { @@ -341,7 +371,7 @@ static void collect_percpu_times(struct psi_group *group, /* * Collect the per-cpu time buckets and average them into a - * single time sample that is normalized to wallclock time. + * single time sample that is normalized to wall clock time. * * For averaging, each CPU is weighted by its non-idle time in * the sampling period. This eliminates artifacts from uneven @@ -384,6 +414,114 @@ static void collect_percpu_times(struct psi_group *group, *pchanged_states = changed_states; } +/* Trigger tracking window manipulations */ +static void window_reset(struct psi_window *win, u64 now, u64 value, + u64 prev_growth) +{ + win->start_time = now; + win->start_value = value; + win->prev_growth = prev_growth; +} + +/* + * PSI growth tracking window update and growth calculation routine. + * + * This approximates a sliding tracking window by interpolating + * partially elapsed windows using historical growth data from the + * previous intervals. This minimizes memory requirements (by not storing + * all the intermediate values in the previous window) and simplifies + * the calculations. It works well because PSI signal changes only in + * positive direction and over relatively small window sizes the growth + * is close to linear. + */ +static u64 window_update(struct psi_window *win, u64 now, u64 value) +{ + u64 elapsed; + u64 growth; + + elapsed = now - win->start_time; + growth = value - win->start_value; + /* + * After each tracking window passes win->start_value and + * win->start_time get reset and win->prev_growth stores + * the average per-window growth of the previous window. + * win->prev_growth is then used to interpolate additional + * growth from the previous window assuming it was linear. + */ + if (elapsed > win->size) + window_reset(win, now, value, growth); + else { + u32 remaining; + + remaining = win->size - elapsed; + growth += div64_u64(win->prev_growth * remaining, win->size); + } + + return growth; +} + +static void update_triggers(struct psi_group *group, u64 now, + enum psi_aggregators aggregator) +{ + struct psi_trigger *t; + u64 *total = group->total[aggregator]; + struct list_head *triggers; + u64 *aggregator_total; + + if (aggregator == PSI_AVGS) { + triggers = &group->avg_triggers; + aggregator_total = group->avg_total; + } else { + triggers = &group->rtpoll_triggers; + aggregator_total = group->rtpoll_total; + } + + /* + * On subsequent updates, calculate growth deltas and let + * watchers know when their specified thresholds are exceeded. + */ + list_for_each_entry(t, triggers, node) { + u64 growth; + bool new_stall; + + new_stall = aggregator_total[t->state] != total[t->state]; + + /* Check for stall activity or a previous threshold breach */ + if (!new_stall && !t->pending_event) + continue; + /* + * Check for new stall activity, as well as deferred + * events that occurred in the last window after the + * trigger had already fired (we want to ratelimit + * events without dropping any). + */ + if (new_stall) { + /* Calculate growth since last update */ + growth = window_update(&t->win, now, total[t->state]); + if (!t->pending_event) { + if (growth < t->threshold) + continue; + + t->pending_event = true; + } + } + /* Limit event signaling to once per window */ + if (now < t->last_event_time + t->win.size) + continue; + + /* Generate an event */ + if (cmpxchg(&t->event, 0, 1) == 0) { + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); + } + t->last_event_time = now; + /* Reset threshold breach flag once event got generated */ + t->pending_event = false; + } +} + static u64 update_averages(struct psi_group *group, u64 now) { unsigned long missed_periods = 0; @@ -459,8 +597,10 @@ static void psi_avgs_work(struct work_struct *work) * Once restarted, we'll catch up the running averages in one * go - see calc_avgs() and missed_periods. */ - if (now >= group->avg_next_update) + if (now >= group->avg_next_update) { + update_triggers(group, now, PSI_AVGS); group->avg_next_update = update_averages(group, now); + } if (changed_states & PSI_STATE_RESCHEDULE) { schedule_delayed_work(dwork, nsecs_to_jiffies( @@ -470,257 +610,155 @@ static void psi_avgs_work(struct work_struct *work) mutex_unlock(&group->avgs_lock); } -/* Trigger tracking window manipulations */ -static void window_reset(struct psi_window *win, u64 now, u64 value, - u64 prev_growth) -{ - win->start_time = now; - win->start_value = value; - win->prev_growth = prev_growth; -} - -/* - * PSI growth tracking window update and growth calculation routine. - * - * This approximates a sliding tracking window by interpolating - * partially elapsed windows using historical growth data from the - * previous intervals. This minimizes memory requirements (by not storing - * all the intermediate values in the previous window) and simplifies - * the calculations. It works well because PSI signal changes only in - * positive direction and over relatively small window sizes the growth - * is close to linear. - */ -static u64 window_update(struct psi_window *win, u64 now, u64 value) -{ - u64 elapsed; - u64 growth; - - elapsed = now - win->start_time; - growth = value - win->start_value; - /* - * After each tracking window passes win->start_value and - * win->start_time get reset and win->prev_growth stores - * the average per-window growth of the previous window. - * win->prev_growth is then used to interpolate additional - * growth from the previous window assuming it was linear. - */ - if (elapsed > win->size) - window_reset(win, now, value, growth); - else { - u32 remaining; - - remaining = win->size - elapsed; - growth += div64_u64(win->prev_growth * remaining, win->size); - } - - return growth; -} - -static void init_triggers(struct psi_group *group, u64 now) +static void init_rtpoll_triggers(struct psi_group *group, u64 now) { struct psi_trigger *t; - list_for_each_entry(t, &group->triggers, node) + list_for_each_entry(t, &group->rtpoll_triggers, node) window_reset(&t->win, now, group->total[PSI_POLL][t->state], 0); - memcpy(group->polling_total, group->total[PSI_POLL], - sizeof(group->polling_total)); - group->polling_next_update = now + group->poll_min_period; + memcpy(group->rtpoll_total, group->total[PSI_POLL], + sizeof(group->rtpoll_total)); + group->rtpoll_next_update = now + group->rtpoll_min_period; } -static u64 update_triggers(struct psi_group *group, u64 now) -{ - struct psi_trigger *t; - bool update_total = false; - u64 *total = group->total[PSI_POLL]; - - /* - * On subsequent updates, calculate growth deltas and let - * watchers know when their specified thresholds are exceeded. - */ - list_for_each_entry(t, &group->triggers, node) { - u64 growth; - bool new_stall; - - new_stall = group->polling_total[t->state] != total[t->state]; - - /* Check for stall activity or a previous threshold breach */ - if (!new_stall && !t->pending_event) - continue; - /* - * Check for new stall activity, as well as deferred - * events that occurred in the last window after the - * trigger had already fired (we want to ratelimit - * events without dropping any). - */ - if (new_stall) { - /* - * Multiple triggers might be looking at the same state, - * remember to update group->polling_total[] once we've - * been through all of them. Also remember to extend the - * polling time if we see new stall activity. - */ - update_total = true; - - /* Calculate growth since last update */ - growth = window_update(&t->win, now, total[t->state]); - if (!t->pending_event) { - if (growth < t->threshold) - continue; - - t->pending_event = true; - } - } - /* Limit event signaling to once per window */ - if (now < t->last_event_time + t->win.size) - continue; - - /* Generate an event */ - if (cmpxchg(&t->event, 0, 1) == 0) - wake_up_interruptible(&t->event_wait); - t->last_event_time = now; - /* Reset threshold breach flag once event got generated */ - t->pending_event = false; - } - - if (update_total) - memcpy(group->polling_total, total, - sizeof(group->polling_total)); - - return now + group->poll_min_period; -} - -/* Schedule polling if it's not already scheduled or forced. */ -static void psi_schedule_poll_work(struct psi_group *group, unsigned long delay, +/* Schedule rtpolling if it's not already scheduled or forced. */ +static void psi_schedule_rtpoll_work(struct psi_group *group, unsigned long delay, bool force) { struct task_struct *task; /* * atomic_xchg should be called even when !force to provide a - * full memory barrier (see the comment inside psi_poll_work). + * full memory barrier (see the comment inside psi_rtpoll_work). */ - if (atomic_xchg(&group->poll_scheduled, 1) && !force) + if (atomic_xchg(&group->rtpoll_scheduled, 1) && !force) return; rcu_read_lock(); - task = rcu_dereference(group->poll_task); + task = rcu_dereference(group->rtpoll_task); /* * kworker might be NULL in case psi_trigger_destroy races with * psi_task_change (hotpath) which can't use locks */ if (likely(task)) - mod_timer(&group->poll_timer, jiffies + delay); + mod_timer(&group->rtpoll_timer, jiffies + delay); else - atomic_set(&group->poll_scheduled, 0); + atomic_set(&group->rtpoll_scheduled, 0); rcu_read_unlock(); } -static void psi_poll_work(struct psi_group *group) +static void psi_rtpoll_work(struct psi_group *group) { bool force_reschedule = false; u32 changed_states; u64 now; - mutex_lock(&group->trigger_lock); + mutex_lock(&group->rtpoll_trigger_lock); now = sched_clock(); - if (now > group->polling_until) { + if (now > group->rtpoll_until) { /* - * We are either about to start or might stop polling if no - * state change was recorded. Resetting poll_scheduled leaves + * We are either about to start or might stop rtpolling if no + * state change was recorded. Resetting rtpoll_scheduled leaves * a small window for psi_group_change to sneak in and schedule - * an immediate poll_work before we get to rescheduling. One - * potential extra wakeup at the end of the polling window - * should be negligible and polling_next_update still keeps + * an immediate rtpoll_work before we get to rescheduling. One + * potential extra wakeup at the end of the rtpolling window + * should be negligible and rtpoll_next_update still keeps * updates correctly on schedule. */ - atomic_set(&group->poll_scheduled, 0); + atomic_set(&group->rtpoll_scheduled, 0); /* - * A task change can race with the poll worker that is supposed to + * A task change can race with the rtpoll worker that is supposed to * report on it. To avoid missing events, ensure ordering between - * poll_scheduled and the task state accesses, such that if the poll - * worker misses the state update, the task change is guaranteed to - * reschedule the poll worker: + * rtpoll_scheduled and the task state accesses, such that if the + * rtpoll worker misses the state update, the task change is + * guaranteed to reschedule the rtpoll worker: * - * poll worker: - * atomic_set(poll_scheduled, 0) + * rtpoll worker: + * atomic_set(rtpoll_scheduled, 0) * smp_mb() * LOAD states * * task change: * STORE states - * if atomic_xchg(poll_scheduled, 1) == 0: - * schedule poll worker + * if atomic_xchg(rtpoll_scheduled, 1) == 0: + * schedule rtpoll worker * * The atomic_xchg() implies a full barrier. */ smp_mb(); } else { - /* Polling window is not over, keep rescheduling */ + /* The rtpolling window is not over, keep rescheduling */ force_reschedule = true; } collect_percpu_times(group, PSI_POLL, &changed_states); - if (changed_states & group->poll_states) { - /* Initialize trigger windows when entering polling mode */ - if (now > group->polling_until) - init_triggers(group, now); + if (changed_states & group->rtpoll_states) { + /* Initialize trigger windows when entering rtpolling mode */ + if (now > group->rtpoll_until) + init_rtpoll_triggers(group, now); /* * Keep the monitor active for at least the duration of the * minimum tracking window as long as monitor states are * changing. */ - group->polling_until = now + - group->poll_min_period * UPDATES_PER_WINDOW; + group->rtpoll_until = now + + group->rtpoll_min_period * UPDATES_PER_WINDOW; } - if (now > group->polling_until) { - group->polling_next_update = ULLONG_MAX; + if (now > group->rtpoll_until) { + group->rtpoll_next_update = ULLONG_MAX; goto out; } - if (now >= group->polling_next_update) - group->polling_next_update = update_triggers(group, now); + if (now >= group->rtpoll_next_update) { + if (changed_states & group->rtpoll_states) { + update_triggers(group, now, PSI_POLL); + memcpy(group->rtpoll_total, group->total[PSI_POLL], + sizeof(group->rtpoll_total)); + } + group->rtpoll_next_update = now + group->rtpoll_min_period; + } - psi_schedule_poll_work(group, - nsecs_to_jiffies(group->polling_next_update - now) + 1, + psi_schedule_rtpoll_work(group, + nsecs_to_jiffies(group->rtpoll_next_update - now) + 1, force_reschedule); out: - mutex_unlock(&group->trigger_lock); + mutex_unlock(&group->rtpoll_trigger_lock); } -static int psi_poll_worker(void *data) +static int psi_rtpoll_worker(void *data) { struct psi_group *group = (struct psi_group *)data; sched_set_fifo_low(current); while (true) { - wait_event_interruptible(group->poll_wait, - atomic_cmpxchg(&group->poll_wakeup, 1, 0) || + wait_event_interruptible(group->rtpoll_wait, + atomic_cmpxchg(&group->rtpoll_wakeup, 1, 0) || kthread_should_stop()); if (kthread_should_stop()) break; - psi_poll_work(group); + psi_rtpoll_work(group); } return 0; } static void poll_timer_fn(struct timer_list *t) { - struct psi_group *group = from_timer(group, t, poll_timer); + struct psi_group *group = timer_container_of(group, t, rtpoll_timer); - atomic_set(&group->poll_wakeup, 1); - wake_up_interruptible(&group->poll_wait); + atomic_set(&group->rtpoll_wakeup, 1); + wake_up_interruptible(&group->rtpoll_wait); } static void record_times(struct psi_group_cpu *groupc, u64 now) @@ -752,29 +790,21 @@ static void record_times(struct psi_group_cpu *groupc, u64 now) groupc->times[PSI_NONIDLE] += delta; } +#define for_each_group(iter, group) \ + for (typeof(group) iter = group; iter; iter = iter->parent) + static void psi_group_change(struct psi_group *group, int cpu, - unsigned int clear, unsigned int set, u64 now, - bool wake_clock) + unsigned int clear, unsigned int set, + u64 now, bool wake_clock) { struct psi_group_cpu *groupc; unsigned int t, m; - enum psi_states s; u32 state_mask; + lockdep_assert_rq_held(cpu_rq(cpu)); groupc = per_cpu_ptr(group->pcpu, cpu); /* - * First we update the task counts according to the state - * change requested through the @clear and @set bits. - * - * Then if the cgroup PSI stats accounting enabled, we - * assess the aggregate resource states this CPU's tasks - * have been in since the last change, and account any - * SOME and FULL time these may have resulted in. - */ - write_seqcount_begin(&groupc->seq); - - /* * Start with TSK_ONCPU, which doesn't have a corresponding * task count - it's just a boolean flag directly encoded in * the state mask. Clear, set, or carry the current state if @@ -825,14 +855,10 @@ static void psi_group_change(struct psi_group *group, int cpu, groupc->state_mask = state_mask; - write_seqcount_end(&groupc->seq); return; } - for (s = 0; s < NR_PSI_STATES; s++) { - if (test_state(groupc->tasks, s, state_mask & PSI_ONCPU)) - state_mask |= (1 << s); - } + state_mask = test_states(groupc->tasks, state_mask); /* * Since we care about lost potential, a memstall is FULL @@ -849,10 +875,8 @@ static void psi_group_change(struct psi_group *group, int cpu, groupc->state_mask = state_mask; - write_seqcount_end(&groupc->seq); - - if (state_mask & group->poll_states) - psi_schedule_poll_work(group, 1, false); + if (state_mask & group->rtpoll_states) + psi_schedule_rtpoll_work(group, 1, false); if (wake_clock && !delayed_work_pending(&group->avgs_work)) schedule_delayed_work(&group->avgs_work, PSI_FREQ); @@ -885,7 +909,6 @@ static void psi_flags_change(struct task_struct *task, int clear, int set) void psi_task_change(struct task_struct *task, int clear, int set) { int cpu = task_cpu(task); - struct psi_group *group; u64 now; if (!task->pid) @@ -893,20 +916,22 @@ void psi_task_change(struct task_struct *task, int clear, int set) psi_flags_change(task, clear, set); + psi_write_begin(cpu); now = cpu_clock(cpu); - - group = task_psi_group(task); - do { + for_each_group(group, task_psi_group(task)) psi_group_change(group, cpu, clear, set, now, true); - } while ((group = group->parent)); + psi_write_end(cpu); } void psi_task_switch(struct task_struct *prev, struct task_struct *next, bool sleep) { - struct psi_group *group, *common = NULL; + struct psi_group *common = NULL; int cpu = task_cpu(prev); - u64 now = cpu_clock(cpu); + u64 now; + + psi_write_begin(cpu); + now = cpu_clock(cpu); if (next->pid) { psi_flags_change(next, 0, TSK_ONCPU); @@ -915,16 +940,15 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, * ancestors with @prev, those will already have @prev's * TSK_ONCPU bit set, and we can stop the iteration there. */ - group = task_psi_group(next); - do { - if (per_cpu_ptr(group->pcpu, cpu)->state_mask & - PSI_ONCPU) { + for_each_group(group, task_psi_group(next)) { + struct psi_group_cpu *groupc = per_cpu_ptr(group->pcpu, cpu); + + if (groupc->state_mask & PSI_ONCPU) { common = group; break; } - psi_group_change(group, cpu, 0, TSK_ONCPU, now, true); - } while ((group = group->parent)); + } } if (prev->pid) { @@ -957,12 +981,11 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, psi_flags_change(prev, clear, set); - group = task_psi_group(prev); - do { + for_each_group(group, task_psi_group(prev)) { if (group == common) break; psi_group_change(group, cpu, clear, set, now, wake_clock); - } while ((group = group->parent)); + } /* * TSK_ONCPU is handled up to the common ancestor. If there are @@ -972,44 +995,56 @@ void psi_task_switch(struct task_struct *prev, struct task_struct *next, */ if ((prev->psi_flags ^ next->psi_flags) & ~TSK_ONCPU) { clear &= ~TSK_ONCPU; - for (; group; group = group->parent) + for_each_group(group, common) psi_group_change(group, cpu, clear, set, now, wake_clock); } } + psi_write_end(cpu); } #ifdef CONFIG_IRQ_TIME_ACCOUNTING -void psi_account_irqtime(struct task_struct *task, u32 delta) +void psi_account_irqtime(struct rq *rq, struct task_struct *curr, struct task_struct *prev) { - int cpu = task_cpu(task); - struct psi_group *group; + int cpu = task_cpu(curr); struct psi_group_cpu *groupc; + s64 delta; + u64 irq; u64 now; - if (!task->pid) + if (static_branch_likely(&psi_disabled) || !irqtime_enabled()) return; + if (!curr->pid) + return; + + lockdep_assert_rq_held(rq); + if (prev && task_psi_group(prev) == task_psi_group(curr)) + return; + + irq = irq_time_read(cpu); + delta = (s64)(irq - rq->psi_irq_time); + if (delta < 0) + return; + rq->psi_irq_time = irq; + + psi_write_begin(cpu); now = cpu_clock(cpu); - group = task_psi_group(task); - do { + for_each_group(group, task_psi_group(curr)) { if (!group->enabled) continue; groupc = per_cpu_ptr(group->pcpu, cpu); - write_seqcount_begin(&groupc->seq); - record_times(groupc, now); groupc->times[PSI_IRQ_FULL] += delta; - write_seqcount_end(&groupc->seq); - - if (group->poll_states & (1 << PSI_IRQ_FULL)) - psi_schedule_poll_work(group, 1, false); - } while ((group = group->parent)); + if (group->rtpoll_states & (1 << PSI_IRQ_FULL)) + psi_schedule_rtpoll_work(group, 1, false); + } + psi_write_end(cpu); } -#endif +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ /** * psi_memstall_enter - mark the beginning of a memory stall section @@ -1101,7 +1136,7 @@ void psi_cgroup_free(struct cgroup *cgroup) cancel_delayed_work_sync(&cgroup->psi->avgs_work); free_percpu(cgroup->psi->pcpu); /* All triggers must be removed by now */ - WARN_ONCE(cgroup->psi->poll_states, "psi: trigger leak\n"); + WARN_ONCE(cgroup->psi->rtpoll_states, "psi: trigger leak\n"); kfree(cgroup->psi); } @@ -1179,7 +1214,7 @@ void psi_cgroup_restart(struct psi_group *group) /* * After we disable psi_group->enabled, we don't actually * stop percpu tasks accounting in each psi_group_cpu, - * instead only stop test_state() loop, record_times() + * instead only stop test_states() loop, record_times() * and averaging worker, see psi_group_change() for details. * * When disable cgroup PSI, this function has nothing to sync @@ -1187,7 +1222,7 @@ void psi_cgroup_restart(struct psi_group *group) * would see !psi_group->enabled and only do task accounting. * * When re-enable cgroup PSI, this function use psi_group_change() - * to get correct state mask from test_state() loop on tasks[], + * to get correct state mask from test_states() loop on tasks[], * and restart groupc->state_start from now, use .clear = .set = 0 * here since no task status really changed. */ @@ -1195,14 +1230,14 @@ void psi_cgroup_restart(struct psi_group *group) return; for_each_possible_cpu(cpu) { - struct rq *rq = cpu_rq(cpu); - struct rq_flags rf; u64 now; - rq_lock_irq(rq, &rf); + guard(rq_lock_irq)(cpu_rq(cpu)); + + psi_write_begin(cpu); now = cpu_clock(cpu); psi_group_change(group, cpu, 0, 0, now, true); - rq_unlock_irq(rq, &rf); + psi_write_end(cpu); } } #endif /* CONFIG_CGROUPS */ @@ -1216,6 +1251,11 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res) if (static_branch_likely(&psi_disabled)) return -EOPNOTSUPP; +#ifdef CONFIG_IRQ_TIME_ACCOUNTING + if (!irqtime_enabled() && res == PSI_IRQ) + return -EOPNOTSUPP; +#endif + /* Update averages before reporting them */ mutex_lock(&group->avgs_lock); now = sched_clock(); @@ -1252,17 +1292,25 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res) return 0; } -struct psi_trigger *psi_trigger_create(struct psi_group *group, - char *buf, enum psi_res res) +struct psi_trigger *psi_trigger_create(struct psi_group *group, char *buf, + enum psi_res res, struct file *file, + struct kernfs_open_file *of) { struct psi_trigger *t; enum psi_states state; u32 threshold_us; + bool privileged; u32 window_us; if (static_branch_likely(&psi_disabled)) return ERR_PTR(-EOPNOTSUPP); + /* + * Checking the privilege here on file->f_cred implies that a privileged user + * could open the file and delegate the write to an unprivileged one. + */ + privileged = cap_raised(file->f_cred->cap_effective, CAP_SYS_RESOURCE); + if (sscanf(buf, "some %u %u", &threshold_us, &window_us) == 2) state = PSI_IO_SOME + res * 2; else if (sscanf(buf, "full %u %u", &threshold_us, &window_us) == 2) @@ -1278,8 +1326,14 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, if (state >= PSI_NONIDLE) return ERR_PTR(-EINVAL); - if (window_us < WINDOW_MIN_US || - window_us > WINDOW_MAX_US) + if (window_us == 0 || window_us > WINDOW_MAX_US) + return ERR_PTR(-EINVAL); + + /* + * Unprivileged users can only use 2s windows so that averages aggregation + * work is used, and no RT threads need to be spawned. + */ + if (!privileged && window_us % 2000000) return ERR_PTR(-EINVAL); /* Check threshold */ @@ -1299,33 +1353,44 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, t->event = 0; t->last_event_time = 0; - init_waitqueue_head(&t->event_wait); + t->of = of; + if (!of) + init_waitqueue_head(&t->event_wait); t->pending_event = false; + t->aggregator = privileged ? PSI_POLL : PSI_AVGS; - mutex_lock(&group->trigger_lock); + if (privileged) { + mutex_lock(&group->rtpoll_trigger_lock); - if (!rcu_access_pointer(group->poll_task)) { - struct task_struct *task; + if (!rcu_access_pointer(group->rtpoll_task)) { + struct task_struct *task; - task = kthread_create(psi_poll_worker, group, "psimon"); - if (IS_ERR(task)) { - kfree(t); - mutex_unlock(&group->trigger_lock); - return ERR_CAST(task); + task = kthread_create(psi_rtpoll_worker, group, "psimon"); + if (IS_ERR(task)) { + kfree(t); + mutex_unlock(&group->rtpoll_trigger_lock); + return ERR_CAST(task); + } + atomic_set(&group->rtpoll_wakeup, 0); + wake_up_process(task); + rcu_assign_pointer(group->rtpoll_task, task); } - atomic_set(&group->poll_wakeup, 0); - wake_up_process(task); - rcu_assign_pointer(group->poll_task, task); - } - list_add(&t->node, &group->triggers); - group->poll_min_period = min(group->poll_min_period, - div_u64(t->win.size, UPDATES_PER_WINDOW)); - group->nr_triggers[t->state]++; - group->poll_states |= (1 << t->state); + list_add(&t->node, &group->rtpoll_triggers); + group->rtpoll_min_period = min(group->rtpoll_min_period, + div_u64(t->win.size, UPDATES_PER_WINDOW)); + group->rtpoll_nr_triggers[t->state]++; + group->rtpoll_states |= (1 << t->state); + + mutex_unlock(&group->rtpoll_trigger_lock); + } else { + mutex_lock(&group->avgs_lock); - mutex_unlock(&group->trigger_lock); + list_add(&t->node, &group->avg_triggers); + group->avg_nr_triggers[t->state]++; + mutex_unlock(&group->avgs_lock); + } return t; } @@ -1343,56 +1408,73 @@ void psi_trigger_destroy(struct psi_trigger *t) group = t->group; /* - * Wakeup waiters to stop polling. Can happen if cgroup is deleted - * from under a polling process. + * Wakeup waiters to stop polling and clear the queue to prevent it from + * being accessed later. Can happen if cgroup is deleted from under a + * polling process. */ - wake_up_interruptible(&t->event_wait); - - mutex_lock(&group->trigger_lock); - - if (!list_empty(&t->node)) { - struct psi_trigger *tmp; - u64 period = ULLONG_MAX; - - list_del(&t->node); - group->nr_triggers[t->state]--; - if (!group->nr_triggers[t->state]) - group->poll_states &= ~(1 << t->state); - /* reset min update period for the remaining triggers */ - list_for_each_entry(tmp, &group->triggers, node) - period = min(period, div_u64(tmp->win.size, - UPDATES_PER_WINDOW)); - group->poll_min_period = period; - /* Destroy poll_task when the last trigger is destroyed */ - if (group->poll_states == 0) { - group->polling_until = 0; - task_to_destroy = rcu_dereference_protected( - group->poll_task, - lockdep_is_held(&group->trigger_lock)); - rcu_assign_pointer(group->poll_task, NULL); - del_timer(&group->poll_timer); + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); + + if (t->aggregator == PSI_AVGS) { + mutex_lock(&group->avgs_lock); + if (!list_empty(&t->node)) { + list_del(&t->node); + group->avg_nr_triggers[t->state]--; } + mutex_unlock(&group->avgs_lock); + } else { + mutex_lock(&group->rtpoll_trigger_lock); + if (!list_empty(&t->node)) { + struct psi_trigger *tmp; + u64 period = ULLONG_MAX; + + list_del(&t->node); + group->rtpoll_nr_triggers[t->state]--; + if (!group->rtpoll_nr_triggers[t->state]) + group->rtpoll_states &= ~(1 << t->state); + /* + * Reset min update period for the remaining triggers + * iff the destroying trigger had the min window size. + */ + if (group->rtpoll_min_period == div_u64(t->win.size, UPDATES_PER_WINDOW)) { + list_for_each_entry(tmp, &group->rtpoll_triggers, node) + period = min(period, div_u64(tmp->win.size, + UPDATES_PER_WINDOW)); + group->rtpoll_min_period = period; + } + /* Destroy rtpoll_task when the last trigger is destroyed */ + if (group->rtpoll_states == 0) { + group->rtpoll_until = 0; + task_to_destroy = rcu_dereference_protected( + group->rtpoll_task, + lockdep_is_held(&group->rtpoll_trigger_lock)); + rcu_assign_pointer(group->rtpoll_task, NULL); + timer_delete(&group->rtpoll_timer); + } + } + mutex_unlock(&group->rtpoll_trigger_lock); } - mutex_unlock(&group->trigger_lock); - /* - * Wait for psi_schedule_poll_work RCU to complete its read-side + * Wait for psi_schedule_rtpoll_work RCU to complete its read-side * critical section before destroying the trigger and optionally the - * poll_task. + * rtpoll_task. */ synchronize_rcu(); /* - * Stop kthread 'psimon' after releasing trigger_lock to prevent a - * deadlock while waiting for psi_poll_work to acquire trigger_lock + * Stop kthread 'psimon' after releasing rtpoll_trigger_lock to prevent + * a deadlock while waiting for psi_rtpoll_work to acquire + * rtpoll_trigger_lock */ if (task_to_destroy) { /* * After the RCU grace period has expired, the worker - * can no longer be found through group->poll_task. + * can no longer be found through group->rtpoll_task. */ kthread_stop(task_to_destroy); - atomic_set(&group->poll_scheduled, 0); + atomic_set(&group->rtpoll_scheduled, 0); } kfree(t); } @@ -1410,7 +1492,10 @@ __poll_t psi_trigger_poll(void **trigger_ptr, if (!t) return DEFAULT_POLLMASK | EPOLLERR | EPOLLPRI; - poll_wait(file, &t->event_wait, wait); + if (t->of) + kernfs_generic_poll(t->of, wait); + else + poll_wait(file, &t->event_wait, wait); if (cmpxchg(&t->event, 1, 0) == 1) ret |= EPOLLPRI; @@ -1434,27 +1519,19 @@ static int psi_cpu_show(struct seq_file *m, void *v) return psi_show(m, &psi_system, PSI_CPU); } -static int psi_open(struct file *file, int (*psi_show)(struct seq_file *, void *)) -{ - if (file->f_mode & FMODE_WRITE && !capable(CAP_SYS_RESOURCE)) - return -EPERM; - - return single_open(file, psi_show, NULL); -} - static int psi_io_open(struct inode *inode, struct file *file) { - return psi_open(file, psi_io_show); + return single_open(file, psi_io_show, NULL); } static int psi_memory_open(struct inode *inode, struct file *file) { - return psi_open(file, psi_memory_show); + return single_open(file, psi_memory_show, NULL); } static int psi_cpu_open(struct inode *inode, struct file *file) { - return psi_open(file, psi_cpu_show); + return single_open(file, psi_cpu_show, NULL); } static ssize_t psi_write(struct file *file, const char __user *user_buf, @@ -1488,7 +1565,7 @@ static ssize_t psi_write(struct file *file, const char __user *user_buf, return -EBUSY; } - new = psi_trigger_create(&psi_system, buf, res); + new = psi_trigger_create(&psi_system, buf, res, file, NULL); if (IS_ERR(new)) { mutex_unlock(&seq->lock); return PTR_ERR(new); @@ -1568,7 +1645,7 @@ static int psi_irq_show(struct seq_file *m, void *v) static int psi_irq_open(struct inode *inode, struct file *file) { - return psi_open(file, psi_irq_show); + return single_open(file, psi_irq_show, NULL); } static ssize_t psi_irq_write(struct file *file, const char __user *user_buf, @@ -1585,7 +1662,7 @@ static const struct proc_ops psi_irq_proc_ops = { .proc_poll = psi_fop_poll, .proc_release = psi_fop_release, }; -#endif +#endif /* CONFIG_IRQ_TIME_ACCOUNTING */ static int __init psi_proc_init(void) { |