From 967fc04671feea4dbf780c9e55a0bc8fcf68a14e Mon Sep 17 00:00:00 2001 From: Gregory Haskins Date: Mon, 29 Dec 2008 09:39:52 -0500 Subject: sched: add sched_class->needs_post_schedule() member We currently run class->post_schedule() outside of the rq->lock, which means that we need to test for the need to post_schedule outside of the lock to avoid a forced reacquistion. This is currently not a problem as we only look at rq->rt.overloaded. However, we want to enhance this going forward to look at more state to reduce the need to post_schedule to a bare minimum set. Therefore, we introduce a new member-func called needs_post_schedule() which tests for the post_schedule condtion without actually performing the work. Therefore it is safe to call this function before the rq->lock is released, because we are guaranteed not to drop the lock at an intermediate point (such as what post_schedule() may do). We will use this later in the series [ rostedt: removed paranoid BUG_ON ] Signed-off-by: Gregory Haskins --- include/linux/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index e5f928a079e..836a86c32a6 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1012,6 +1012,7 @@ struct sched_class { struct rq *busiest, struct sched_domain *sd, enum cpu_idle_type idle); void (*pre_schedule) (struct rq *this_rq, struct task_struct *task); + int (*needs_post_schedule) (struct rq *this_rq); void (*post_schedule) (struct rq *this_rq); void (*task_wake_up) (struct rq *this_rq, struct task_struct *task); -- cgit v1.2.3-70-g09d2 From 4075134e40804821f90866d7de56802e4dcecb1e Mon Sep 17 00:00:00 2001 From: Gregory Haskins Date: Mon, 29 Dec 2008 09:39:53 -0500 Subject: plist: fix PLIST_NODE_INIT to work with debug enabled It seems that PLIST_NODE_INIT breaks if used and DEBUG_PI_LIST is defined. Since there are no current users of PLIST_NODE_INIT, this has gone undetected. This patch fixes the build issue that enables the DEBUG_PI_LIST later in the series when we use it in init_task.h Signed-off-by: Gregory Haskins --- include/linux/plist.h | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) (limited to 'include') diff --git a/include/linux/plist.h b/include/linux/plist.h index 85de2f05587..45926d77d6a 100644 --- a/include/linux/plist.h +++ b/include/linux/plist.h @@ -96,6 +96,10 @@ struct plist_node { # define PLIST_HEAD_LOCK_INIT(_lock) #endif +#define _PLIST_HEAD_INIT(head) \ + .prio_list = LIST_HEAD_INIT((head).prio_list), \ + .node_list = LIST_HEAD_INIT((head).node_list) + /** * PLIST_HEAD_INIT - static struct plist_head initializer * @head: struct plist_head variable name @@ -103,8 +107,7 @@ struct plist_node { */ #define PLIST_HEAD_INIT(head, _lock) \ { \ - .prio_list = LIST_HEAD_INIT((head).prio_list), \ - .node_list = LIST_HEAD_INIT((head).node_list), \ + _PLIST_HEAD_INIT(head), \ PLIST_HEAD_LOCK_INIT(&(_lock)) \ } @@ -116,7 +119,7 @@ struct plist_node { #define PLIST_NODE_INIT(node, __prio) \ { \ .prio = (__prio), \ - .plist = PLIST_HEAD_INIT((node).plist, NULL), \ + .plist = { _PLIST_HEAD_INIT((node).plist) }, \ } /** -- cgit v1.2.3-70-g09d2 From 917b627d4d981dc614519d7b34ea31a976b14e12 Mon Sep 17 00:00:00 2001 From: Gregory Haskins Date: Mon, 29 Dec 2008 09:39:53 -0500 Subject: sched: create "pushable_tasks" list to limit pushing to one attempt The RT scheduler employs a "push/pull" design to actively balance tasks within the system (on a per disjoint cpuset basis). When a task is awoken, it is immediately determined if there are any lower priority cpus which should be preempted. This is opposed to the way normal SCHED_OTHER tasks behave, which will wait for a periodic rebalancing operation to occur before spreading out load. When a particular RQ has more than 1 active RT task, it is said to be in an "overloaded" state. Once this occurs, the system enters the active balancing mode, where it will try to push the task away, or persuade a different cpu to pull it over. The system will stay in this state until the system falls back below the <= 1 queued RT task per RQ. However, the current implementation suffers from a limitation in the push logic. Once overloaded, all tasks (other than current) on the RQ are analyzed on every push operation, even if it was previously unpushable (due to affinity, etc). Whats more, the operation stops at the first task that is unpushable and will not look at items lower in the queue. This causes two problems: 1) We can have the same tasks analyzed over and over again during each push, which extends out the fast path in the scheduler for no gain. Consider a RQ that has dozens of tasks that are bound to a core. Each one of those tasks will be encountered and skipped for each push operation while they are queued. 2) There may be lower-priority tasks under the unpushable task that could have been successfully pushed, but will never be considered until either the unpushable task is cleared, or a pull operation succeeds. The net result is a potential latency source for mid priority tasks. This patch aims to rectify these two conditions by introducing a new priority sorted list: "pushable_tasks". A task is added to the list each time a task is activated or preempted. It is removed from the list any time it is deactivated, made current, or fails to push. This works because a task only needs to be attempted to push once. After an initial failure to push, the other cpus will eventually try to pull the task when the conditions are proper. This also solves the problem that we don't completely analyze all tasks due to encountering an unpushable tasks. Now every task will have a push attempted (when appropriate). This reduces latency both by shorting the critical section of the rq->lock for certain workloads, and by making sure the algorithm considers all eligible tasks in the system. [ rostedt: added a couple more BUG_ONs ] Signed-off-by: Gregory Haskins Acked-by: Steven Rostedt --- include/linux/init_task.h | 1 + include/linux/sched.h | 1 + kernel/sched.c | 4 ++ kernel/sched_rt.c | 119 +++++++++++++++++++++++++++++++++++++++------- 4 files changed, 107 insertions(+), 18 deletions(-) (limited to 'include') diff --git a/include/linux/init_task.h b/include/linux/init_task.h index 23fd8909b9e..6851225f44a 100644 --- a/include/linux/init_task.h +++ b/include/linux/init_task.h @@ -140,6 +140,7 @@ extern struct group_info init_groups; .nr_cpus_allowed = NR_CPUS, \ }, \ .tasks = LIST_HEAD_INIT(tsk.tasks), \ + .pushable_tasks = PLIST_NODE_INIT(tsk.pushable_tasks, MAX_PRIO), \ .ptraced = LIST_HEAD_INIT(tsk.ptraced), \ .ptrace_entry = LIST_HEAD_INIT(tsk.ptrace_entry), \ .real_parent = &tsk, \ diff --git a/include/linux/sched.h b/include/linux/sched.h index 836a86c32a6..440cabb2d43 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1179,6 +1179,7 @@ struct task_struct { #endif struct list_head tasks; + struct plist_node pushable_tasks; struct mm_struct *mm, *active_mm; diff --git a/kernel/sched.c b/kernel/sched.c index 3acbad8991a..24ab80c2876 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -471,6 +471,7 @@ struct rt_rq { #ifdef CONFIG_SMP unsigned long rt_nr_migratory; int overloaded; + struct plist_head pushable_tasks; #endif int rt_throttled; u64 rt_time; @@ -2481,6 +2482,8 @@ void sched_fork(struct task_struct *p, int clone_flags) /* Want to start with kernel preemption disabled. */ task_thread_info(p)->preempt_count = 1; #endif + plist_node_init(&p->pushable_tasks, MAX_PRIO); + put_cpu(); } @@ -8237,6 +8240,7 @@ static void init_rt_rq(struct rt_rq *rt_rq, struct rq *rq) #ifdef CONFIG_SMP rt_rq->rt_nr_migratory = 0; rt_rq->overloaded = 0; + plist_head_init(&rq->rt.pushable_tasks, &rq->lock); #endif rt_rq->rt_time = 0; diff --git a/kernel/sched_rt.c b/kernel/sched_rt.c index b0b6ea4ed67..fe9da6084c8 100644 --- a/kernel/sched_rt.c +++ b/kernel/sched_rt.c @@ -49,6 +49,24 @@ static void update_rt_migration(struct rq *rq) rq->rt.overloaded = 0; } } + +static void enqueue_pushable_task(struct rq *rq, struct task_struct *p) +{ + plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); + plist_node_init(&p->pushable_tasks, p->prio); + plist_add(&p->pushable_tasks, &rq->rt.pushable_tasks); +} + +static void dequeue_pushable_task(struct rq *rq, struct task_struct *p) +{ + plist_del(&p->pushable_tasks, &rq->rt.pushable_tasks); +} + +#else + +#define enqueue_pushable_task(rq, p) do { } while (0) +#define dequeue_pushable_task(rq, p) do { } while (0) + #endif /* CONFIG_SMP */ static inline struct task_struct *rt_task_of(struct sched_rt_entity *rt_se) @@ -751,6 +769,9 @@ static void enqueue_task_rt(struct rq *rq, struct task_struct *p, int wakeup) enqueue_rt_entity(rt_se); + if (!task_current(rq, p) && p->rt.nr_cpus_allowed > 1) + enqueue_pushable_task(rq, p); + inc_cpu_load(rq, p->se.load.weight); } @@ -761,6 +782,8 @@ static void dequeue_task_rt(struct rq *rq, struct task_struct *p, int sleep) update_curr_rt(rq); dequeue_rt_entity(rt_se); + dequeue_pushable_task(rq, p); + dec_cpu_load(rq, p->se.load.weight); } @@ -911,7 +934,7 @@ static struct sched_rt_entity *pick_next_rt_entity(struct rq *rq, return next; } -static struct task_struct *pick_next_task_rt(struct rq *rq) +static struct task_struct *_pick_next_task_rt(struct rq *rq) { struct sched_rt_entity *rt_se; struct task_struct *p; @@ -933,6 +956,18 @@ static struct task_struct *pick_next_task_rt(struct rq *rq) p = rt_task_of(rt_se); p->se.exec_start = rq->clock; + + return p; +} + +static struct task_struct *pick_next_task_rt(struct rq *rq) +{ + struct task_struct *p = _pick_next_task_rt(rq); + + /* The running task is never eligible for pushing */ + if (p) + dequeue_pushable_task(rq, p); + return p; } @@ -940,6 +975,13 @@ static void put_prev_task_rt(struct rq *rq, struct task_struct *p) { update_curr_rt(rq); p->se.exec_start = 0; + + /* + * The previous task needs to be made eligible for pushing + * if it is still active + */ + if (p->se.on_rq && p->rt.nr_cpus_allowed > 1) + enqueue_pushable_task(rq, p); } #ifdef CONFIG_SMP @@ -1116,6 +1158,31 @@ static struct rq *find_lock_lowest_rq(struct task_struct *task, struct rq *rq) return lowest_rq; } +static inline int has_pushable_tasks(struct rq *rq) +{ + return !plist_head_empty(&rq->rt.pushable_tasks); +} + +static struct task_struct *pick_next_pushable_task(struct rq *rq) +{ + struct task_struct *p; + + if (!has_pushable_tasks(rq)) + return NULL; + + p = plist_first_entry(&rq->rt.pushable_tasks, + struct task_struct, pushable_tasks); + + BUG_ON(rq->cpu != task_cpu(p)); + BUG_ON(task_current(rq, p)); + BUG_ON(p->rt.nr_cpus_allowed <= 1); + + BUG_ON(!p->se.on_rq); + BUG_ON(!rt_task(p)); + + return p; +} + /* * If the current CPU has more than one RT task, see if the non * running task can migrate over to a CPU that is running a task @@ -1125,13 +1192,12 @@ static int push_rt_task(struct rq *rq) { struct task_struct *next_task; struct rq *lowest_rq; - int ret = 0; int paranoid = RT_MAX_TRIES; if (!rq->rt.overloaded) return 0; - next_task = pick_next_highest_task_rt(rq, -1); + next_task = pick_next_pushable_task(rq); if (!next_task) return 0; @@ -1163,12 +1229,19 @@ static int push_rt_task(struct rq *rq) * so it is possible that next_task has changed. * If it has, then try again. */ - task = pick_next_highest_task_rt(rq, -1); + task = pick_next_pushable_task(rq); if (unlikely(task != next_task) && task && paranoid--) { put_task_struct(next_task); next_task = task; goto retry; } + + /* + * Once we have failed to push this task, we will not + * try again, since the other cpus will pull from us + * when they are ready + */ + dequeue_pushable_task(rq, next_task); goto out; } @@ -1180,23 +1253,12 @@ static int push_rt_task(struct rq *rq) double_unlock_balance(rq, lowest_rq); - ret = 1; out: put_task_struct(next_task); - return ret; + return 1; } -/* - * TODO: Currently we just use the second highest prio task on - * the queue, and stop when it can't migrate (or there's - * no more RT tasks). There may be a case where a lower - * priority RT task has a different affinity than the - * higher RT task. In this case the lower RT task could - * possibly be able to migrate where as the higher priority - * RT task could not. We currently ignore this issue. - * Enhancements are welcome! - */ static void push_rt_tasks(struct rq *rq) { /* push_rt_task will return true if it moved an RT */ @@ -1295,7 +1357,7 @@ static void pre_schedule_rt(struct rq *rq, struct task_struct *prev) */ static int needs_post_schedule_rt(struct rq *rq) { - return rq->rt.overloaded ? 1 : 0; + return has_pushable_tasks(rq); } static void post_schedule_rt(struct rq *rq) @@ -1317,7 +1379,7 @@ static void task_wake_up_rt(struct rq *rq, struct task_struct *p) { if (!task_running(rq, p) && !test_tsk_need_resched(rq->curr) && - rq->rt.overloaded && + has_pushable_tasks(rq) && p->rt.nr_cpus_allowed > 1) push_rt_tasks(rq); } @@ -1354,6 +1416,24 @@ static void set_cpus_allowed_rt(struct task_struct *p, if (p->se.on_rq && (weight != p->rt.nr_cpus_allowed)) { struct rq *rq = task_rq(p); + if (!task_current(rq, p)) { + /* + * Make sure we dequeue this task from the pushable list + * before going further. It will either remain off of + * the list because we are no longer pushable, or it + * will be requeued. + */ + if (p->rt.nr_cpus_allowed > 1) + dequeue_pushable_task(rq, p); + + /* + * Requeue if our weight is changing and still > 1 + */ + if (weight > 1) + enqueue_pushable_task(rq, p); + + } + if ((p->rt.nr_cpus_allowed <= 1) && (weight > 1)) { rq->rt.rt_nr_migratory++; } else if ((p->rt.nr_cpus_allowed > 1) && (weight <= 1)) { @@ -1538,6 +1618,9 @@ static void set_curr_task_rt(struct rq *rq) struct task_struct *p = rq->curr; p->se.exec_start = rq->clock; + + /* The running task is never eligible for pushing */ + dequeue_pushable_task(rq, p); } static const struct sched_class rt_sched_class = { -- cgit v1.2.3-70-g09d2 From 831451ac4e44d3a20b581ce726ef1d1144373f7d Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 14 Jan 2009 12:39:18 +0100 Subject: sched: introduce avg_wakeup Introduce a new avg_wakeup statistic. avg_wakeup is a measure of how frequently a task wakes up other tasks, it represents the average time between wakeups, with a limit of avg_runtime for when it doesn't wake up anybody. Signed-off-by: Peter Zijlstra Signed-off-by: Mike Galbraith Signed-off-by: Ingo Molnar --- include/linux/sched.h | 3 +++ kernel/sched.c | 36 ++++++++++++++++++++++++++++++------ kernel/sched_debug.c | 1 + 3 files changed, 34 insertions(+), 6 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 4cae9b81a1f..daf4e07bc97 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1046,6 +1046,9 @@ struct sched_entity { u64 exec_max; u64 slice_max; + u64 start_runtime; + u64 avg_wakeup; + u64 nr_migrations; u64 nr_migrations_cold; u64 nr_failed_migrations_affine; diff --git a/kernel/sched.c b/kernel/sched.c index 8be2c13b50d..86f5a063f0b 100644 --- a/kernel/sched.c +++ b/kernel/sched.c @@ -1705,6 +1705,9 @@ static void update_avg(u64 *avg, u64 sample) static void enqueue_task(struct rq *rq, struct task_struct *p, int wakeup) { + if (wakeup) + p->se.start_runtime = p->se.sum_exec_runtime; + sched_info_queued(p); p->sched_class->enqueue_task(rq, p, wakeup); p->se.on_rq = 1; @@ -1712,10 +1715,15 @@ static void enqueue_task(struct rq *rq, struct task_struct *p, int wakeup) static void dequeue_task(struct rq *rq, struct task_struct *p, int sleep) { - if (sleep && p->se.last_wakeup) { - update_avg(&p->se.avg_overlap, - p->se.sum_exec_runtime - p->se.last_wakeup); - p->se.last_wakeup = 0; + if (sleep) { + if (p->se.last_wakeup) { + update_avg(&p->se.avg_overlap, + p->se.sum_exec_runtime - p->se.last_wakeup); + p->se.last_wakeup = 0; + } else { + update_avg(&p->se.avg_wakeup, + sysctl_sched_wakeup_granularity); + } } sched_info_dequeued(p); @@ -2345,6 +2353,22 @@ out_activate: activate_task(rq, p, 1); success = 1; + /* + * Only attribute actual wakeups done by this task. + */ + if (!in_interrupt()) { + struct sched_entity *se = ¤t->se; + u64 sample = se->sum_exec_runtime; + + if (se->last_wakeup) + sample -= se->last_wakeup; + else + sample -= se->start_runtime; + update_avg(&se->avg_wakeup, sample); + + se->last_wakeup = se->sum_exec_runtime; + } + out_running: trace_sched_wakeup(rq, p, success); check_preempt_curr(rq, p, sync); @@ -2355,8 +2379,6 @@ out_running: p->sched_class->task_wake_up(rq, p); #endif out: - current->se.last_wakeup = current->se.sum_exec_runtime; - task_rq_unlock(rq, &flags); return success; @@ -2386,6 +2408,8 @@ static void __sched_fork(struct task_struct *p) p->se.prev_sum_exec_runtime = 0; p->se.last_wakeup = 0; p->se.avg_overlap = 0; + p->se.start_runtime = 0; + p->se.avg_wakeup = sysctl_sched_wakeup_granularity; #ifdef CONFIG_SCHEDSTATS p->se.wait_start = 0; diff --git a/kernel/sched_debug.c b/kernel/sched_debug.c index 16eeba4e416..2b1260f0e80 100644 --- a/kernel/sched_debug.c +++ b/kernel/sched_debug.c @@ -397,6 +397,7 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m) PN(se.vruntime); PN(se.sum_exec_runtime); PN(se.avg_overlap); + PN(se.avg_wakeup); nr_switches = p->nvcsw + p->nivcsw; -- cgit v1.2.3-70-g09d2 From 34cb61359b503d7aff6447acb037a5efd6ce93b2 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Fri, 16 Jan 2009 13:36:06 +0100 Subject: sched: fix !CONFIG_SCHEDSTATS build failure Stephen Rothwell reported this linux-next build failure with !CONFIG_SCHEDSTATS: | In file included from kernel/sched.c:1703: | kernel/sched_fair.c: In function 'adaptive_gran': | kernel/sched_fair.c:1324: error: 'struct sched_entity' has no member named 'avg_wakeup' The start_runtime and avg_wakeup metrics are now not just for statistics, but also for scheduling - so they always need to be available. (Also move out the nr_migrations fields - for future perfcounters usage.) Reported-by: Stephen Rothwell Signed-off-by: Ingo Molnar --- include/linux/sched.h | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index daf4e07bc97..5d56b54350a 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1031,6 +1031,10 @@ struct sched_entity { u64 last_wakeup; u64 avg_overlap; + u64 start_runtime; + u64 avg_wakeup; + u64 nr_migrations; + #ifdef CONFIG_SCHEDSTATS u64 wait_start; u64 wait_max; @@ -1046,10 +1050,6 @@ struct sched_entity { u64 exec_max; u64 slice_max; - u64 start_runtime; - u64 avg_wakeup; - - u64 nr_migrations; u64 nr_migrations_cold; u64 nr_failed_migrations_affine; u64 nr_failed_migrations_running; -- cgit v1.2.3-70-g09d2 From ad0b0fd554dfc126b5750d14908dccc3bbf602be Mon Sep 17 00:00:00 2001 From: Arjan van de Ven Date: Tue, 10 Feb 2009 11:42:26 -0800 Subject: sched, latencytop: incorporate review feedback from Andrew Morton Andrew had some suggestions for the latencytop file; this patch takes care of most of these: * Add documentation * Turn account_scheduler_latency into an inline function * Don't report negative values to userspace * Make the file operations struct const * Fix a few checkpatch.pl warnings Signed-off-by: Arjan van de Ven Signed-off-by: Andrew Morton Signed-off-by: Ingo Molnar --- include/linux/latencytop.h | 10 +++++- kernel/latencytop.c | 83 +++++++++++++++++++++++++++++++++++++++------- 2 files changed, 80 insertions(+), 13 deletions(-) (limited to 'include') diff --git a/include/linux/latencytop.h b/include/linux/latencytop.h index 901c2d6377a..b0e99898527 100644 --- a/include/linux/latencytop.h +++ b/include/linux/latencytop.h @@ -9,6 +9,7 @@ #ifndef _INCLUDE_GUARD_LATENCYTOP_H_ #define _INCLUDE_GUARD_LATENCYTOP_H_ +#include #ifdef CONFIG_LATENCYTOP #define LT_SAVECOUNT 32 @@ -24,7 +25,14 @@ struct latency_record { struct task_struct; -void account_scheduler_latency(struct task_struct *task, int usecs, int inter); +extern int latencytop_enabled; +void __account_scheduler_latency(struct task_struct *task, int usecs, int inter); +static inline void +account_scheduler_latency(struct task_struct *task, int usecs, int inter) +{ + if (unlikely(latencytop_enabled)) + __account_scheduler_latency(task, usecs, inter); +} void clear_all_latency_tracing(struct task_struct *p); diff --git a/kernel/latencytop.c b/kernel/latencytop.c index 449db466bdb..ca07c5c0c91 100644 --- a/kernel/latencytop.c +++ b/kernel/latencytop.c @@ -9,6 +9,44 @@ * as published by the Free Software Foundation; version 2 * of the License. */ + +/* + * CONFIG_LATENCYTOP enables a kernel latency tracking infrastructure that is + * used by the "latencytop" userspace tool. The latency that is tracked is not + * the 'traditional' interrupt latency (which is primarily caused by something + * else consuming CPU), but instead, it is the latency an application encounters + * because the kernel sleeps on its behalf for various reasons. + * + * This code tracks 2 levels of statistics: + * 1) System level latency + * 2) Per process latency + * + * The latency is stored in fixed sized data structures in an accumulated form; + * if the "same" latency cause is hit twice, this will be tracked as one entry + * in the data structure. Both the count, total accumulated latency and maximum + * latency are tracked in this data structure. When the fixed size structure is + * full, no new causes are tracked until the buffer is flushed by writing to + * the /proc file; the userspace tool does this on a regular basis. + * + * A latency cause is identified by a stringified backtrace at the point that + * the scheduler gets invoked. The userland tool will use this string to + * identify the cause of the latency in human readable form. + * + * The information is exported via /proc/latency_stats and /proc//latency. + * These files look like this: + * + * Latency Top version : v0.1 + * 70 59433 4897 i915_irq_wait drm_ioctl vfs_ioctl do_vfs_ioctl sys_ioctl + * | | | | + * | | | +----> the stringified backtrace + * | | +---------> The maximum latency for this entry in microseconds + * | +--------------> The accumulated latency for this entry (microseconds) + * +-------------------> The number of times this entry is hit + * + * (note: the average latency is the accumulated latency divided by the number + * of times) + */ + #include #include #include @@ -72,7 +110,7 @@ account_global_scheduler_latency(struct task_struct *tsk, struct latency_record firstnonnull = i; continue; } - for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) { + for (q = 0; q < LT_BACKTRACEDEPTH; q++) { unsigned long record = lat->backtrace[q]; if (latency_record[i].backtrace[q] != record) { @@ -101,31 +139,52 @@ account_global_scheduler_latency(struct task_struct *tsk, struct latency_record memcpy(&latency_record[i], lat, sizeof(struct latency_record)); } -static inline void store_stacktrace(struct task_struct *tsk, struct latency_record *lat) +/* + * Iterator to store a backtrace into a latency record entry + */ +static inline void store_stacktrace(struct task_struct *tsk, + struct latency_record *lat) { struct stack_trace trace; memset(&trace, 0, sizeof(trace)); trace.max_entries = LT_BACKTRACEDEPTH; trace.entries = &lat->backtrace[0]; - trace.skip = 0; save_stack_trace_tsk(tsk, &trace); } +/** + * __account_scheduler_latency - record an occured latency + * @tsk - the task struct of the task hitting the latency + * @usecs - the duration of the latency in microseconds + * @inter - 1 if the sleep was interruptible, 0 if uninterruptible + * + * This function is the main entry point for recording latency entries + * as called by the scheduler. + * + * This function has a few special cases to deal with normal 'non-latency' + * sleeps: specifically, interruptible sleep longer than 5 msec is skipped + * since this usually is caused by waiting for events via select() and co. + * + * Negative latencies (caused by time going backwards) are also explicitly + * skipped. + */ void __sched -account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) +__account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) { unsigned long flags; int i, q; struct latency_record lat; - if (!latencytop_enabled) - return; - /* Long interruptible waits are generally user requested... */ if (inter && usecs > 5000) return; + /* Negative sleeps are time going backwards */ + /* Zero-time sleeps are non-interesting */ + if (usecs <= 0) + return; + memset(&lat, 0, sizeof(lat)); lat.count = 1; lat.time = usecs; @@ -143,12 +202,12 @@ account_scheduler_latency(struct task_struct *tsk, int usecs, int inter) if (tsk->latency_record_count >= LT_SAVECOUNT) goto out_unlock; - for (i = 0; i < LT_SAVECOUNT ; i++) { + for (i = 0; i < LT_SAVECOUNT; i++) { struct latency_record *mylat; int same = 1; mylat = &tsk->latency_record[i]; - for (q = 0 ; q < LT_BACKTRACEDEPTH ; q++) { + for (q = 0; q < LT_BACKTRACEDEPTH; q++) { unsigned long record = lat.backtrace[q]; if (mylat->backtrace[q] != record) { @@ -186,7 +245,7 @@ static int lstats_show(struct seq_file *m, void *v) for (i = 0; i < MAXLR; i++) { if (latency_record[i].backtrace[0]) { int q; - seq_printf(m, "%i %li %li ", + seq_printf(m, "%i %lu %lu ", latency_record[i].count, latency_record[i].time, latency_record[i].max); @@ -223,7 +282,7 @@ static int lstats_open(struct inode *inode, struct file *filp) return single_open(filp, lstats_show, NULL); } -static struct file_operations lstats_fops = { +static const struct file_operations lstats_fops = { .open = lstats_open, .read = seq_read, .write = lstats_write, @@ -236,4 +295,4 @@ static int __init init_lstats_procfs(void) proc_create("latency_stats", 0644, NULL, &lstats_fops); return 0; } -__initcall(init_lstats_procfs); +device_initcall(init_lstats_procfs); -- cgit v1.2.3-70-g09d2 From b342501cd31e5546d0c9ca8ceff5ded1832f9e5b Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 26 Feb 2009 20:20:29 +0100 Subject: sched: allow architectures to specify sched_clock_stable Allow CONFIG_HAVE_UNSTABLE_SCHED_CLOCK architectures to still specify that their sched_clock() implementation is reliable. This will be used by x86 to switch on a faster sched_clock_cpu() implementation on certain CPU types. Signed-off-by: Ingo Molnar --- include/linux/sched.h | 10 ++++++++++ kernel/sched_clock.c | 45 ++++++++++++++++++++------------------------- 2 files changed, 30 insertions(+), 25 deletions(-) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 8981e52c714..a063d19b7a7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1670,6 +1670,16 @@ static inline int set_cpus_allowed(struct task_struct *p, cpumask_t new_mask) return set_cpus_allowed_ptr(p, &new_mask); } +/* + * Architectures can set this to 1 if they have specified + * CONFIG_HAVE_UNSTABLE_SCHED_CLOCK in their arch Kconfig, + * but then during bootup it turns out that sched_clock() + * is reliable after all: + */ +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK +extern int sched_clock_stable; +#endif + extern unsigned long long sched_clock(void); extern void sched_clock_init(void); diff --git a/kernel/sched_clock.c b/kernel/sched_clock.c index a0b0852414c..a755d023805 100644 --- a/kernel/sched_clock.c +++ b/kernel/sched_clock.c @@ -24,11 +24,11 @@ * The clock: sched_clock_cpu() is monotonic per cpu, and should be somewhat * consistent between cpus (never more than 2 jiffies difference). */ -#include -#include #include -#include #include +#include +#include +#include /* * Scheduler clock - returns current time in nanosec units. @@ -43,6 +43,10 @@ unsigned long long __attribute__((weak)) sched_clock(void) static __read_mostly int sched_clock_running; #ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK +__read_mostly int sched_clock_stable; +#else +static const int sched_clock_stable = 1; +#endif struct sched_clock_data { /* @@ -87,7 +91,7 @@ void sched_clock_init(void) } /* - * min,max except they take wrapping into account + * min, max except they take wrapping into account */ static inline u64 wrap_min(u64 x, u64 y) @@ -116,10 +120,13 @@ static u64 __update_sched_clock(struct sched_clock_data *scd, u64 now) if (unlikely(delta < 0)) delta = 0; + if (unlikely(!sched_clock_running)) + return 0ull; + /* * scd->clock = clamp(scd->tick_gtod + delta, - * max(scd->tick_gtod, scd->clock), - * scd->tick_gtod + TICK_NSEC); + * max(scd->tick_gtod, scd->clock), + * scd->tick_gtod + TICK_NSEC); */ clock = scd->tick_gtod + delta; @@ -148,12 +155,13 @@ static void lock_double_clock(struct sched_clock_data *data1, u64 sched_clock_cpu(int cpu) { - struct sched_clock_data *scd = cpu_sdc(cpu); u64 now, clock, this_clock, remote_clock; + struct sched_clock_data *scd; - if (unlikely(!sched_clock_running)) - return 0ull; + if (sched_clock_stable) + return sched_clock(); + scd = cpu_sdc(cpu); WARN_ON_ONCE(!irqs_disabled()); now = sched_clock(); @@ -193,6 +201,8 @@ u64 sched_clock_cpu(int cpu) return clock; } +#ifdef CONFIG_HAVE_UNSTABLE_SCHED_CLOCK + void sched_clock_tick(void) { struct sched_clock_data *scd = this_scd(); @@ -235,22 +245,7 @@ void sched_clock_idle_wakeup_event(u64 delta_ns) } EXPORT_SYMBOL_GPL(sched_clock_idle_wakeup_event); -#else /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ - -void sched_clock_init(void) -{ - sched_clock_running = 1; -} - -u64 sched_clock_cpu(int cpu) -{ - if (unlikely(!sched_clock_running)) - return 0; - - return sched_clock(); -} - -#endif +#endif /* CONFIG_HAVE_UNSTABLE_SCHED_CLOCK */ unsigned long long cpu_clock(int cpu) { -- cgit v1.2.3-70-g09d2