*
* Scaled math optimizations by Thomas Gleixner
* Copyright (C) 2007, Thomas Gleixner <tglx@linutronix.de>
+ *
+ * Adaptive scheduling granularity, math enhancements by Peter Zijlstra
+ * Copyright (C) 2007 Red Hat, Inc., Peter Zijlstra <pzijlstr@redhat.com>
*/
/*
- * Preemption granularity:
- * (default: 10 msec, units: nanoseconds)
+ * Targeted preemption latency for CPU-bound tasks:
+ * (default: 20ms, units: nanoseconds)
*
- * NOTE: this granularity value is not the same as the concept of
- * 'timeslice length' - timeslices in CFS will typically be somewhat
- * larger than this value. (to see the precise effective timeslice
- * length of your workload, run vmstat and monitor the context-switches
- * field)
+ * NOTE: this latency value is not the same as the concept of
+ * 'timeslice length' - timeslices in CFS are of variable length.
+ * (to see the precise effective timeslice length of your workload,
+ * run vmstat and monitor the context-switches field)
*
* On SMP systems the value of this is multiplied by the log2 of the
* number of CPUs. (i.e. factor 2x on 2-way systems, 3x on 4-way
* systems, 4x on 8-way systems, 5x on 16-way systems, etc.)
+ * Targeted preemption latency for CPU-bound tasks:
+ */
+unsigned int sysctl_sched_latency __read_mostly = 20000000ULL;
+
+/*
+ * Minimal preemption granularity for CPU-bound tasks:
+ * (default: 2 msec, units: nanoseconds)
*/
-unsigned int sysctl_sched_granularity __read_mostly = 10000000UL;
+unsigned int sysctl_sched_min_granularity __read_mostly = 2000000ULL;
/*
* SCHED_BATCH wake-up granularity.
update_load_add(&cfs_rq->load, se->load.weight);
cfs_rq->nr_running++;
se->on_rq = 1;
+
+ schedstat_add(cfs_rq, wait_runtime, se->wait_runtime);
}
static inline void
update_load_sub(&cfs_rq->load, se->load.weight);
cfs_rq->nr_running--;
se->on_rq = 0;
+
+ schedstat_add(cfs_rq, wait_runtime, -se->wait_runtime);
}
static inline struct rb_node *first_fair(struct cfs_rq *cfs_rq)
* Scheduling class statistics methods:
*/
+/*
+ * Calculate the preemption granularity needed to schedule every
+ * runnable task once per sysctl_sched_latency amount of time.
+ * (down to a sensible low limit on granularity)
+ *
+ * For example, if there are 2 tasks running and latency is 10 msecs,
+ * we switch tasks every 5 msecs. If we have 3 tasks running, we have
+ * to switch tasks every 3.33 msecs to get a 10 msecs observed latency
+ * for each task. We do finer and finer scheduling up to until we
+ * reach the minimum granularity value.
+ *
+ * To achieve this we use the following dynamic-granularity rule:
+ *
+ * gran = lat/nr - lat/nr/nr
+ *
+ * This comes out of the following equations:
+ *
+ * kA1 + gran = kB1
+ * kB2 + gran = kA2
+ * kA2 = kA1
+ * kB2 = kB1 - d + d/nr
+ * lat = d * nr
+ *
+ * Where 'k' is key, 'A' is task A (waiting), 'B' is task B (running),
+ * '1' is start of time, '2' is end of time, 'd' is delay between
+ * 1 and 2 (during which task B was running), 'nr' is number of tasks
+ * running, 'lat' is the the period of each task. ('lat' is the
+ * sched_latency that we aim for.)
+ */
+static long
+sched_granularity(struct cfs_rq *cfs_rq)
+{
+ unsigned int gran = sysctl_sched_latency;
+ unsigned int nr = cfs_rq->nr_running;
+
+ if (nr > 1) {
+ gran = gran/nr - gran/nr/nr;
+ gran = max(gran, sysctl_sched_min_granularity);
+ }
+
+ return gran;
+}
+
/*
* We rescale the rescheduling granularity of tasks according to their
* nice level, but only linearly, not exponentially:
/*
* It will always fit into 'long':
*/
- return (long) (tmp >> WMULT_SHIFT);
+ return (long) (tmp >> (WMULT_SHIFT-NICE_0_SHIFT));
}
static inline void
delta_fair = calc_delta_fair(delta_exec, lw);
delta_mine = calc_delta_mine(delta_exec, curr->load.weight, lw);
- if (cfs_rq->sleeper_bonus > sysctl_sched_granularity) {
- delta = min(cfs_rq->sleeper_bonus, (u64)delta_exec);
- delta = calc_delta_mine(delta, curr->load.weight, lw);
- delta = min((u64)delta, cfs_rq->sleeper_bonus);
+ if (cfs_rq->sleeper_bonus > sysctl_sched_min_granularity) {
+ delta = min((u64)delta_mine, cfs_rq->sleeper_bonus);
+ delta = min(delta, (unsigned long)(
+ (long)sysctl_sched_runtime_limit - curr->wait_runtime));
cfs_rq->sleeper_bonus -= delta;
delta_mine -= delta;
}
{
unsigned long delta_fair;
+ if (unlikely(!se->wait_start_fair))
+ return;
+
delta_fair = (unsigned long)min((u64)(2*sysctl_sched_runtime_limit),
(u64)(cfs_rq->fair_clock - se->wait_start_fair));
unsigned long load = cfs_rq->load.weight, delta_fair;
long prev_runtime;
+ /*
+ * Do not boost sleepers if there's too much bonus 'in flight'
+ * already:
+ */
+ if (unlikely(cfs_rq->sleeper_bonus > sysctl_sched_runtime_limit))
+ return;
+
if (sysctl_sched_features & SCHED_FEAT_SLEEPER_LOAD_AVG)
load = rq_of(cfs_rq)->cpu_load[2];
* Track the amount of bonus we've given to sleepers:
*/
cfs_rq->sleeper_bonus += delta_fair;
- if (unlikely(cfs_rq->sleeper_bonus > sysctl_sched_runtime_limit))
- cfs_rq->sleeper_bonus = sysctl_sched_runtime_limit;
-
- schedstat_add(cfs_rq, wait_runtime, se->wait_runtime);
}
static void enqueue_sleeper(struct cfs_rq *cfs_rq, struct sched_entity *se)
if (tsk->state & TASK_UNINTERRUPTIBLE)
se->block_start = rq_of(cfs_rq)->clock;
}
- cfs_rq->wait_runtime -= se->wait_runtime;
#endif
}
__dequeue_entity(cfs_rq, se);
/*
* Preempt the current task with a newly woken task if needed:
*/
-static void
+static int
__check_preempt_curr_fair(struct cfs_rq *cfs_rq, struct sched_entity *se,
struct sched_entity *curr, unsigned long granularity)
{
* preempt the current task unless the best task has
* a larger than sched_granularity fairness advantage:
*/
- if (__delta > niced_granularity(curr, granularity))
+ if (__delta > niced_granularity(curr, granularity)) {
resched_task(rq_of(cfs_rq)->curr);
+ return 1;
+ }
+ return 0;
}
static inline void
static void entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
{
+ unsigned long gran, ideal_runtime, delta_exec;
struct sched_entity *next;
/*
if (next == curr)
return;
- __check_preempt_curr_fair(cfs_rq, next, curr, sysctl_sched_granularity);
+ gran = sched_granularity(cfs_rq);
+ ideal_runtime = niced_granularity(curr,
+ max(sysctl_sched_latency / cfs_rq->nr_running,
+ (unsigned long)sysctl_sched_min_granularity));
+ /*
+ * If we executed more than what the latency constraint suggests,
+ * reduce the rescheduling granularity. This way the total latency
+ * of how much a task is not scheduled converges to
+ * sysctl_sched_latency:
+ */
+ delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime;
+ if (delta_exec > ideal_runtime)
+ gran = 0;
+
+ if (__check_preempt_curr_fair(cfs_rq, next, curr, gran))
+ curr->prev_sum_exec_runtime = curr->sum_exec_runtime;
}
/**************************************************
static void task_new_fair(struct rq *rq, struct task_struct *p)
{
struct cfs_rq *cfs_rq = task_cfs_rq(p);
- struct sched_entity *se = &p->se;
+ struct sched_entity *se = &p->se, *curr = cfs_rq_curr(cfs_rq);
sched_info_queued(p);
+ update_curr(cfs_rq);
update_stats_enqueue(cfs_rq, se);
/*
* Child runs first: we let it run before the parent
* until it reschedules once. We set up the key so that
* it will preempt the parent:
*/
- p->se.fair_key = current->se.fair_key -
- niced_granularity(&rq->curr->se, sysctl_sched_granularity) - 1;
+ se->fair_key = curr->fair_key -
+ niced_granularity(curr, sched_granularity(cfs_rq)) - 1;
/*
* The first wait is dominated by the child-runs-first logic,
* so do not credit it with that waiting time yet:
*/
if (sysctl_sched_features & SCHED_FEAT_SKIP_INITIAL)
- p->se.wait_start_fair = 0;
+ se->wait_start_fair = 0;
/*
* The statistical average of wait_runtime is about
* -granularity/2, so initialize the task with that:
*/
if (sysctl_sched_features & SCHED_FEAT_START_DEBIT)
- p->se.wait_runtime = -(sysctl_sched_granularity / 2);
+ se->wait_runtime = -(sched_granularity(cfs_rq) / 2);
__enqueue_entity(cfs_rq, se);
}
git.openpandora.org / pandora-kernel.git / blobdiff