All of lore.kernel.org
 help / color / mirror / Atom feed
From: luca abeni <luca.abeni@santannapisa.it>
To: linux-kernel@vger.kernel.org
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>, Juri Lelli <juri.lelli@arm.com>,
	Claudio Scordino <claudio@evidence.eu.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Tommaso Cucinotta <tommaso.cucinotta@sssup.it>,
	Daniel Bristot de Oliveira <bristot@redhat.com>,
	Joel Fernandes <joelaf@google.com>,
	Mathieu Poirier <mathieu.poirier@linaro.org>,
	Luca Abeni <luca.abeni@santannapisa.it>
Subject: [RFC v5 5/9] sched/deadline: do not reclaim the whole CPU bandwidth
Date: Fri, 24 Mar 2017 04:52:58 +0100	[thread overview]
Message-ID: <1490327582-4376-6-git-send-email-luca.abeni@santannapisa.it> (raw)
In-Reply-To: <1490327582-4376-1-git-send-email-luca.abeni@santannapisa.it>

From: Luca Abeni <luca.abeni@santannapisa.it>

Original GRUB tends to reclaim 100% of the CPU time... And this
allows a CPU hog to starve non-deadline tasks.
To address this issue, allow the scheduler to reclaim only a
specified fraction of CPU time.

Signed-off-by: Luca Abeni <luca.abeni@santannapisa.it>
Tested-by: Daniel Bristot de Oliveira <bristot@redhat.com>
---
 kernel/sched/core.c     | 6 ++++++
 kernel/sched/deadline.c | 7 ++++++-
 kernel/sched/sched.h    | 6 ++++++
 3 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 20c62e7..efa88eb 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -6716,6 +6716,12 @@ static void sched_dl_do_global(void)
 		raw_spin_unlock_irqrestore(&dl_b->lock, flags);
 
 		rcu_read_unlock_sched();
+		if (dl_b->bw == -1)
+			cpu_rq(cpu)->dl.deadline_bw_inv = 1 << 8;
+		else
+			cpu_rq(cpu)->dl.deadline_bw_inv =
+				to_ratio(global_rt_runtime(),
+					 global_rt_period()) >> 12;
 	}
 }
 
diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
index 6035311..e964051 100644
--- a/kernel/sched/deadline.c
+++ b/kernel/sched/deadline.c
@@ -212,6 +212,11 @@ void init_dl_rq(struct dl_rq *dl_rq)
 #else
 	init_dl_bw(&dl_rq->dl_bw);
 #endif
+	if (global_rt_runtime() == RUNTIME_INF)
+		dl_rq->deadline_bw_inv = 1 << 8;
+	else
+		dl_rq->deadline_bw_inv =
+			to_ratio(global_rt_runtime(), global_rt_period()) >> 12;
 }
 
 #ifdef CONFIG_SMP
@@ -871,7 +876,7 @@ extern bool sched_rt_bandwidth_account(struct rt_rq *rt_rq);
  */
 u64 grub_reclaim(u64 delta, struct rq *rq)
 {
-	return (delta * rq->dl.running_bw) >> 20;
+	return (delta * rq->dl.running_bw * rq->dl.deadline_bw_inv) >> 20 >> 8;
 }
 
 /*
diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
index 57bb79b..141549b 100644
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -565,6 +565,12 @@ struct dl_rq {
 	 * task blocks
 	 */
 	u64 running_bw;
+
+	/*
+	 * Inverse of the fraction of CPU utilization that can be reclaimed
+	 * by the GRUB algorithm.
+	 */
+	u64 deadline_bw_inv;
 };
 
 #ifdef CONFIG_SMP
-- 
2.7.4

  parent reply	other threads:[~2017-03-23 19:53 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-24  3:52 [RFC v5 0/9] CPU reclaiming for SCHED_DEADLINE luca abeni
2017-03-24  3:52 ` [RFC v5 1/9] sched/deadline: track the active utilization luca abeni
2017-03-26 17:04   ` Mathieu Poirier
2017-03-26 20:55     ` luca abeni
2017-03-24  3:52 ` [RFC v5 2/9] sched/deadline: improve the tracking of " luca abeni
2017-03-24 13:20   ` Peter Zijlstra
2017-03-24 21:47     ` luca abeni
2017-03-25  2:31       ` Steven Rostedt
2017-03-27  8:20         ` Luca Abeni
2017-03-27  8:54           ` Claudio Scordino
2017-03-27  7:17       ` Juri Lelli
2017-03-27  7:43         ` Luca Abeni
2017-03-27  8:45           ` Juri Lelli
2017-03-27  7:36       ` Luca Abeni
2017-07-24  8:06       ` Luca Abeni
2017-07-24  9:04         ` Peter Zijlstra
2017-07-25  6:41           ` Luca Abeni
2017-03-24 13:23   ` Peter Zijlstra
2017-07-24  7:54     ` Luca Abeni
2017-07-24  9:11       ` Peter Zijlstra
2017-07-25  6:46         ` Luca Abeni
2017-03-26 17:32   ` Mathieu Poirier
2017-03-26 21:01     ` luca abeni
2017-03-24  3:52 ` [RFC v5 3/9] sched/deadline: fix the update of the total -deadline utilization luca abeni
2017-03-24  3:52 ` [RFC v5 4/9] sched/deadline: implement GRUB accounting luca abeni
2017-03-24  3:52 ` luca abeni [this message]
2017-03-24 14:00   ` [RFC v5 5/9] sched/deadline: do not reclaim the whole CPU bandwidth Peter Zijlstra
2017-03-24 21:58     ` luca abeni
2017-03-25  2:38       ` Steven Rostedt
2017-03-27  8:35         ` Peter Zijlstra
2017-03-24  3:52 ` [RFC v5 6/9] sched/deadline: make GRUB a task's flag luca abeni
2017-03-24  3:53 ` [RFC v5 7/9] sched/deadline: track the "total rq utilization" too luca abeni
2017-03-24  3:53 ` [RFC v5 8/9] sched/deadline: base GRUB reclaiming on the inactive utilization luca abeni
2017-03-27 14:26   ` Peter Zijlstra
2017-03-27 14:56     ` Luca Abeni
2017-03-27 15:53       ` Peter Zijlstra
2017-03-27 17:02         ` luca abeni
2017-05-08  7:41     ` Luca Abeni
2017-05-08  8:06       ` Peter Zijlstra
2017-05-09  9:37         ` Luca Abeni
2017-03-24  3:53 ` [RFC v5 9/9] sched/deadline: also reclaim bandwidth not used by dl tasks luca abeni
2017-03-27 14:03   ` Peter Zijlstra
2017-03-27 14:48     ` Luca Abeni

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=1490327582-4376-6-git-send-email-luca.abeni@santannapisa.it \
    --to=luca.abeni@santannapisa.it \
    --cc=bristot@redhat.com \
    --cc=claudio@evidence.eu.com \
    --cc=joelaf@google.com \
    --cc=juri.lelli@arm.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathieu.poirier@linaro.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=rostedt@goodmis.org \
    --cc=tommaso.cucinotta@sssup.it \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.