From: Peter Zijlstra <peterz@infradead.org>
To: mingo@kernel.org, juri.lelli@redhat.com
Cc: linux-kernel@vger.kernel.org, dietmar.eggemann@arm.com,
luca.abeni@santannapisa.it, bristot@redhat.com,
balsini@android.com, dvyukov@google.com, tglx@linutronix.de,
vpillai@digitalocean.com, rostedt@goodmis.org,
peterz@infradead.org
Subject: [RFC][PATCH 08/13] sched: Rework pick_next_task() slow-path
Date: Fri, 26 Jul 2019 16:54:17 +0200 [thread overview]
Message-ID: <20190726161357.812535015@infradead.org> (raw)
In-Reply-To: 20190726145409.947503076@infradead.org
Avoid the RETRY_TASK case in the pick_next_task() slow path.
By doing the put_prev_task() early, we get the rt/deadline pull done,
and by testing rq->nr_running we know if we need newidle_balance().
This then gives a stable state to pick a task from.
Since the fast-path is fair only; it means the other classes will
always have pick_next_task(.prev=NULL, .rf=NULL) and we can simplify.
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
kernel/sched/core.c | 19 ++++++++++++-------
kernel/sched/deadline.c | 30 ++----------------------------
kernel/sched/fair.c | 9 ++++++---
kernel/sched/idle.c | 4 +++-
kernel/sched/rt.c | 29 +----------------------------
kernel/sched/sched.h | 13 ++++++++-----
kernel/sched/stop_task.c | 3 ++-
7 files changed, 34 insertions(+), 73 deletions(-)
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -3739,7 +3739,7 @@ pick_next_task(struct rq *rq, struct tas
p = fair_sched_class.pick_next_task(rq, prev, rf);
if (unlikely(p == RETRY_TASK))
- goto again;
+ goto restart;
/* Assumes fair_sched_class->next == idle_sched_class */
if (unlikely(!p))
@@ -3748,14 +3748,19 @@ pick_next_task(struct rq *rq, struct tas
return p;
}
-again:
+restart:
+ /*
+ * Ensure that we put DL/RT tasks before the pick loop, such that they
+ * can PULL higher prio tasks when we lower the RQ 'priority'.
+ */
+ prev->sched_class->put_prev_task(rq, prev, rf);
+ if (!rq->nr_running)
+ newidle_balance(rq, rf);
+
for_each_class(class) {
- p = class->pick_next_task(rq, prev, rf);
- if (p) {
- if (unlikely(p == RETRY_TASK))
- goto again;
+ p = class->pick_next_task(rq, NULL, NULL);
+ if (p)
return p;
- }
}
/* The idle class should always have a runnable task: */
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -1729,39 +1729,13 @@ pick_next_task_dl(struct rq *rq, struct
struct task_struct *p;
struct dl_rq *dl_rq;
- dl_rq = &rq->dl;
-
- if (need_pull_dl_task(rq, prev)) {
- /*
- * This is OK, because current is on_cpu, which avoids it being
- * picked for load-balance and preemption/IRQs are still
- * disabled avoiding further scheduler activity on it and we're
- * being very careful to re-start the picking loop.
- */
- rq_unpin_lock(rq, rf);
- pull_dl_task(rq);
- rq_repin_lock(rq, rf);
- /*
- * pull_dl_task() can drop (and re-acquire) rq->lock; this
- * means a stop task can slip in, in which case we need to
- * re-start task selection.
- */
- if (rq->stop && task_on_rq_queued(rq->stop))
- return RETRY_TASK;
- }
+ WARN_ON_ONCE(prev || rf);
- /*
- * When prev is DL, we may throttle it in put_prev_task().
- * So, we update time before we check for dl_nr_running.
- */
- if (prev->sched_class == &dl_sched_class)
- update_curr_dl(rq);
+ dl_rq = &rq->dl;
if (unlikely(!dl_rq->dl_nr_running))
return NULL;
- put_prev_task(rq, prev);
-
dl_se = pick_next_dl_entity(rq, dl_rq);
BUG_ON(!dl_se);
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -6740,7 +6740,7 @@ pick_next_task_fair(struct rq *rq, struc
goto idle;
#ifdef CONFIG_FAIR_GROUP_SCHED
- if (prev->sched_class != &fair_sched_class)
+ if (!prev || prev->sched_class != &fair_sched_class)
goto simple;
/*
@@ -6817,8 +6817,8 @@ pick_next_task_fair(struct rq *rq, struc
goto done;
simple:
#endif
-
- put_prev_task(rq, prev);
+ if (prev)
+ put_prev_task(rq, prev);
do {
se = pick_next_entity(cfs_rq, NULL);
@@ -6846,6 +6846,9 @@ done: __maybe_unused;
return p;
idle:
+ if (!rf)
+ return NULL;
+
new_tasks = newidle_balance(rq, rf);
/*
--- a/kernel/sched/idle.c
+++ b/kernel/sched/idle.c
@@ -389,7 +389,9 @@ pick_next_task_idle(struct rq *rq, struc
{
struct task_struct *next = rq->idle;
- put_prev_task(rq, prev);
+ if (prev)
+ put_prev_task(rq, prev);
+
set_next_task_idle(rq, next);
return next;
--- a/kernel/sched/rt.c
+++ b/kernel/sched/rt.c
@@ -1554,38 +1554,11 @@ pick_next_task_rt(struct rq *rq, struct
struct task_struct *p;
struct rt_rq *rt_rq = &rq->rt;
- if (need_pull_rt_task(rq, prev)) {
- /*
- * This is OK, because current is on_cpu, which avoids it being
- * picked for load-balance and preemption/IRQs are still
- * disabled avoiding further scheduler activity on it and we're
- * being very careful to re-start the picking loop.
- */
- rq_unpin_lock(rq, rf);
- pull_rt_task(rq);
- rq_repin_lock(rq, rf);
- /*
- * pull_rt_task() can drop (and re-acquire) rq->lock; this
- * means a dl or stop task can slip in, in which case we need
- * to re-start task selection.
- */
- if (unlikely((rq->stop && task_on_rq_queued(rq->stop)) ||
- rq->dl.dl_nr_running))
- return RETRY_TASK;
- }
-
- /*
- * We may dequeue prev's rt_rq in put_prev_task().
- * So, we update time before rt_queued check.
- */
- if (prev->sched_class == &rt_sched_class)
- update_curr_rt(rq);
+ WARN_ON_ONCE(prev || rf);
if (!rt_rq->rt_queued)
return NULL;
- put_prev_task(rq, prev);
-
p = _pick_next_task_rt(rq);
set_next_task_rt(rq, p);
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1704,12 +1704,15 @@ struct sched_class {
void (*check_preempt_curr)(struct rq *rq, struct task_struct *p, int flags);
/*
- * It is the responsibility of the pick_next_task() method that will
- * return the next task to call put_prev_task() on the @prev task or
- * something equivalent.
+ * Both @prev and @rf are optional and may be NULL, in which case the
+ * caller must already have invoked put_prev_task(rq, prev, rf).
*
- * May return RETRY_TASK when it finds a higher prio class has runnable
- * tasks.
+ * Otherwise it is the responsibility of the pick_next_task() to call
+ * put_prev_task() on the @prev task or something equivalent, IFF it
+ * returns a next task.
+ *
+ * In that case (@rf != NULL) it may return RETRY_TASK when it finds a
+ * higher prio class has runnable tasks.
*/
struct task_struct * (*pick_next_task)(struct rq *rq,
struct task_struct *prev,
--- a/kernel/sched/stop_task.c
+++ b/kernel/sched/stop_task.c
@@ -33,10 +33,11 @@ pick_next_task_stop(struct rq *rq, struc
{
struct task_struct *stop = rq->stop;
+ WARN_ON_ONCE(prev || rf);
+
if (!stop || !task_on_rq_queued(stop))
return NULL;
- put_prev_task(rq, prev);
set_next_task_stop(rq, stop);
return stop;
next prev parent reply other threads:[~2019-07-26 16:20 UTC|newest]
Thread overview: 63+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-07-26 14:54 [RFC][PATCH 00/13] SCHED_DEADLINE server infrastructure Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 01/13] sched/deadline: Impose global limits on sched_attr::sched_period Peter Zijlstra
2019-07-29 8:57 ` Juri Lelli
2019-07-29 11:45 ` Daniel Bristot de Oliveira
2019-08-02 17:21 ` Alessio Balsini
2019-08-05 11:53 ` Peter Zijlstra
2019-08-22 12:29 ` Alessio Balsini
2019-08-22 16:51 ` Peter Zijlstra
2019-08-31 14:41 ` Alessio Balsini
2019-09-02 9:16 ` Peter Zijlstra
2019-09-02 12:31 ` Peter Zijlstra
2019-09-04 10:16 ` Steven Rostedt
2019-09-04 11:30 ` Peter Zijlstra
2019-09-04 13:24 ` Joel Fernandes
2019-09-04 14:11 ` Will Deacon
2019-09-04 14:35 ` Joel Fernandes
2019-09-04 15:52 ` Peter Zijlstra
2019-10-23 17:17 ` [PATCH 4.4 4.9 4.14] loop: Add LOOP_SET_DIRECT_IO to compat ioctl Alessio Balsini
2019-10-23 17:22 ` Alessio Balsini
2019-10-25 0:17 ` Sasha Levin
2020-05-20 18:38 ` [RFC][PATCH 01/13] sched/deadline: Impose global limits on sched_attr::sched_period Juri Lelli
2020-05-21 13:45 ` Daniel Bristot de Oliveira
2020-06-16 12:21 ` [tip: sched/core] " tip-bot2 for Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 02/13] stop_machine: Fix stop_cpus_in_progress ordering Peter Zijlstra
2019-07-30 13:16 ` Phil Auld
2019-07-30 13:22 ` Steven Rostedt
2019-07-26 14:54 ` [RFC][PATCH 03/13] sched: Fix kerneldoc comment for ia64_set_curr_task Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 04/13] sched/{rt,deadline}: Fix set_next_task vs pick_next_task Peter Zijlstra
2019-07-29 9:25 ` Juri Lelli
2019-07-29 11:15 ` Peter Zijlstra
2019-07-29 11:27 ` Juri Lelli
2019-07-29 13:04 ` Peter Zijlstra
2019-07-29 13:17 ` Juri Lelli
2019-07-29 14:40 ` Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 05/13] sched: Add task_struct pointer to sched_class::set_curr_task Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 06/13] sched/fair: Export newidle_balance() Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 07/13] sched: Allow put_prev_task() to drop rq->lock Peter Zijlstra
2019-07-26 14:54 ` Peter Zijlstra [this message]
2019-07-26 14:54 ` [RFC][PATCH 09/13] sched: Unify runtime accounting across classes Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 10/13] sched/deadline: Collect sched_dl_entity initialization Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 11/13] sched/deadline: Move bandwidth accounting into {en,de}queue_dl_entity Peter Zijlstra
2019-07-26 14:54 ` [RFC][PATCH 12/13] sched/deadline: Introduce deadline servers Peter Zijlstra
2019-08-07 16:31 ` Dietmar Eggemann
2019-08-08 6:52 ` Juri Lelli
2019-08-08 7:52 ` Dietmar Eggemann
2019-08-08 7:56 ` Peter Zijlstra
2019-08-08 8:11 ` Dietmar Eggemann
2019-08-08 8:46 ` Juri Lelli
2019-08-08 8:57 ` Dietmar Eggemann
2019-08-08 9:27 ` Juri Lelli
2019-08-08 9:45 ` Juri Lelli
2019-08-30 11:24 ` Peter Zijlstra
2019-09-06 9:36 ` Juri Lelli
2019-08-08 10:31 ` Peter Zijlstra
2019-08-09 7:13 ` Juri Lelli
2019-08-08 6:59 ` Juri Lelli
2019-08-09 9:17 ` Dietmar Eggemann
2019-08-09 12:16 ` Juri Lelli
2019-07-26 14:54 ` [RFC][PATCH 13/13] sched/fair: Add trivial fair server Peter Zijlstra
2019-07-26 20:01 ` [RFC][PATCH 00/13] SCHED_DEADLINE server infrastructure luca abeni
2019-09-03 14:27 ` Alessio Balsini
2019-09-04 10:50 ` Juri Lelli
2019-09-04 11:32 ` Peter Zijlstra
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190726161357.812535015@infradead.org \
--to=peterz@infradead.org \
--cc=balsini@android.com \
--cc=bristot@redhat.com \
--cc=dietmar.eggemann@arm.com \
--cc=dvyukov@google.com \
--cc=juri.lelli@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luca.abeni@santannapisa.it \
--cc=mingo@kernel.org \
--cc=rostedt@goodmis.org \
--cc=tglx@linutronix.de \
--cc=vpillai@digitalocean.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).