diff options
author | Peter Zijlstra <peterz@infradead.org> | 2015-05-14 12:23:11 +0200 |
---|---|---|
committer | Thomas Gleixner <tglx@linutronix.de> | 2015-05-18 17:17:42 +0200 |
commit | 4cfafd3082afc707653aeb82e9f8e7b596fbbfd6 (patch) | |
tree | 8c3abec0535288f2b27bbae7bcedda5a458a6aba /kernel/sched | |
parent | c78c881824cc3354c0518c0d21febe04ffc69638 (diff) | |
download | linux-4cfafd3082afc707653aeb82e9f8e7b596fbbfd6.tar.gz linux-4cfafd3082afc707653aeb82e9f8e7b596fbbfd6.tar.bz2 linux-4cfafd3082afc707653aeb82e9f8e7b596fbbfd6.zip |
sched,perf: Fix periodic timers
In the below two commits (see Fixes) we have periodic timers that can
stop themselves when they're no longer required, but need to be
(re)-started when their idle condition changes.
Further complications is that we want the timer handler to always do
the forward such that it will always correctly deal with the overruns,
and we do not want to race such that the handler has already decided
to stop, but the (external) restart sees the timer still active and we
end up with a 'lost' timer.
The problem with the current code is that the re-start can come before
the callback does the forward, at which point the forward from the
callback will WARN about forwarding an enqueued timer.
Now, conceptually its easy to detect if you're before or after the fwd
by comparing the expiration time against the current time. Of course,
that's expensive (and racy) because we don't have the current time.
Alternatively one could cache this state inside the timer, but then
everybody pays the overhead of maintaining this extra state, and that
is undesired.
The only other option that I could see is the external timer_active
variable, which I tried to kill before. I would love a nicer interface
for this seemingly simple 'problem' but alas.
Fixes: 272325c4821f ("perf: Fix mux_interval hrtimer wreckage")
Fixes: 77a4d1a1b9a1 ("sched: Cleanup bandwidth timers")
Cc: pjt@google.com
Cc: tglx@linutronix.de
Cc: klamm@yandex-team.ru
Cc: mingo@kernel.org
Cc: bsegall@google.com
Cc: hpa@zytor.com
Cc: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: http://lkml.kernel.org/r/20150514102311.GX21418@twins.programming.kicks-ass.net
Diffstat (limited to 'kernel/sched')
-rw-r--r-- | kernel/sched/core.c | 12 | ||||
-rw-r--r-- | kernel/sched/fair.c | 17 | ||||
-rw-r--r-- | kernel/sched/rt.c | 8 | ||||
-rw-r--r-- | kernel/sched/sched.h | 5 |
4 files changed, 22 insertions, 20 deletions
diff --git a/kernel/sched/core.c b/kernel/sched/core.c index d8a6196465d5..e84aeb280777 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -90,18 +90,6 @@ #define CREATE_TRACE_POINTS #include <trace/events/sched.h> -void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period) -{ - /* - * Do not forward the expiration time of active timers; - * we do not want to loose an overrun. - */ - if (!hrtimer_active(period_timer)) - hrtimer_forward_now(period_timer, period); - - hrtimer_start_expires(period_timer, HRTIMER_MODE_ABS_PINNED); -} - DEFINE_MUTEX(sched_domains_mutex); DEFINE_PER_CPU_SHARED_ALIGNED(struct rq, runqueues); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index e3b32ebfe421..69be2825262d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -3870,8 +3870,9 @@ static void start_cfs_slack_bandwidth(struct cfs_bandwidth *cfs_b) if (runtime_refresh_within(cfs_b, min_left)) return; - start_bandwidth_timer(&cfs_b->slack_timer, - ns_to_ktime(cfs_bandwidth_slack_period)); + hrtimer_start(&cfs_b->slack_timer, + ns_to_ktime(cfs_bandwidth_slack_period), + HRTIMER_MODE_REL); } /* we know any runtime found here is valid as update_curr() precedes return */ @@ -4012,6 +4013,8 @@ static enum hrtimer_restart sched_cfs_period_timer(struct hrtimer *timer) idle = do_sched_cfs_period_timer(cfs_b, overrun); } + if (idle) + cfs_b->period_active = 0; raw_spin_unlock(&cfs_b->lock); return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; @@ -4025,7 +4028,7 @@ void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) cfs_b->period = ns_to_ktime(default_cfs_period()); INIT_LIST_HEAD(&cfs_b->throttled_cfs_rq); - hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); + hrtimer_init(&cfs_b->period_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED); cfs_b->period_timer.function = sched_cfs_period_timer; hrtimer_init(&cfs_b->slack_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); cfs_b->slack_timer.function = sched_cfs_slack_timer; @@ -4039,7 +4042,13 @@ static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) void start_cfs_bandwidth(struct cfs_bandwidth *cfs_b) { - start_bandwidth_timer(&cfs_b->period_timer, cfs_b->period); + lockdep_assert_held(&cfs_b->lock); + + if (!cfs_b->period_active) { + cfs_b->period_active = 1; + hrtimer_forward_now(&cfs_b->period_timer, cfs_b->period); + hrtimer_start_expires(&cfs_b->period_timer, HRTIMER_MODE_ABS_PINNED); + } } static void destroy_cfs_bandwidth(struct cfs_bandwidth *cfs_b) diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c index b0febf25d8f1..e43da5391dcd 100644 --- a/kernel/sched/rt.c +++ b/kernel/sched/rt.c @@ -31,6 +31,8 @@ static enum hrtimer_restart sched_rt_period_timer(struct hrtimer *timer) idle = do_sched_rt_period_timer(rt_b, overrun); raw_spin_lock(&rt_b->rt_runtime_lock); } + if (idle) + rt_b->rt_period_active = 0; raw_spin_unlock(&rt_b->rt_runtime_lock); return idle ? HRTIMER_NORESTART : HRTIMER_RESTART; @@ -54,7 +56,11 @@ static void start_rt_bandwidth(struct rt_bandwidth *rt_b) return; raw_spin_lock(&rt_b->rt_runtime_lock); - start_bandwidth_timer(&rt_b->rt_period_timer, rt_b->rt_period); + if (!rt_b->rt_period_active) { + rt_b->rt_period_active = 1; + hrtimer_forward_now(&rt_b->rt_period_timer, rt_b->rt_period); + hrtimer_start_expires(&rt_b->rt_period_timer, HRTIMER_MODE_ABS_PINNED); + } raw_spin_unlock(&rt_b->rt_runtime_lock); } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 08606a1f8c4d..f9a58ef373b4 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -131,6 +131,7 @@ struct rt_bandwidth { ktime_t rt_period; u64 rt_runtime; struct hrtimer rt_period_timer; + unsigned int rt_period_active; }; void __dl_clear_params(struct task_struct *p); @@ -215,7 +216,7 @@ struct cfs_bandwidth { s64 hierarchical_quota; u64 runtime_expires; - int idle; + int idle, period_active; struct hrtimer period_timer, slack_timer; struct list_head throttled_cfs_rq; @@ -1406,8 +1407,6 @@ static inline void sched_rt_avg_update(struct rq *rq, u64 rt_delta) { } static inline void sched_avg_update(struct rq *rq) { } #endif -extern void start_bandwidth_timer(struct hrtimer *period_timer, ktime_t period); - /* * __task_rq_lock - lock the rq @p resides on. */ |