From: Anton Blanchard <anton@samba.org>
To: Thomas Gleixner <tglx@linutronix.de>
Cc: LKML <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Ingo Molnar <mingo@kernel.org>,
rt@linutronix.de, Michael Ellerman <mpe@ellerman.id.au>,
Vaidyanathan Srinivasan <svaidy@linux.vnet.ibm.com>,
shreyas@linux.vnet.ibm.com
Subject: Re: [patch 10/15] sched/migration: Move calc_load_migrate() into CPU_DYING
Date: Tue, 12 Jul 2016 14:37:45 +1000 [thread overview]
Message-ID: <20160712143745.430733e7@kryten> (raw)
In-Reply-To: <20160310120025.328739226@linutronix.de>
Hi Thomas,
> It really does not matter when we fold the load for the outgoing cpu.
> It's almost dead anyway, so there is no harm if we fail to fold the
> few microseconds which are required for going fully away.
We are seeing the load average shoot up when hot unplugging CPUs (+1
for every CPU we offline) on ppc64. This reproduces on bare metal as
well as inside a KVM guest. A bisect points at this commit.
As an example, a completely idle box with 128 CPUS and 112 hot
unplugged:
# uptime
04:35:30 up 1:23, 2 users, load average: 112.43, 122.94, 125.54
Anton
next prev parent reply other threads:[~2016-07-12 5:02 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-03-10 12:04 [patch 00/15] cpu/hotplug: Convert scheduler to hotplug state machine Thomas Gleixner
2016-03-10 12:04 ` [patch 01/15] cpu/hotplug: Document states better Thomas Gleixner
2016-03-10 12:04 ` [patch 03/15] sched: Make set_cpu_rq_start_time() a built in hotplug state Thomas Gleixner
2016-03-10 12:04 ` [patch 05/15] sched: Consolidate the notifier maze Thomas Gleixner
2016-03-10 12:04 ` [patch 04/15] sched: Allow hotplug notifiers to be setup early Thomas Gleixner
2016-03-10 12:04 ` [patch 06/15] sched: Move sched_domains_numa_masks_clear() to DOWN_PREPARE Thomas Gleixner
2016-03-10 12:04 ` [patch 08/15] sched, hotplug: Move sync_rcu to be with set_cpu_active(false) Thomas Gleixner
2016-05-05 11:24 ` [tip:smp/hotplug] sched/hotplug: " tip-bot for Peter Zijlstra
2016-05-06 13:06 ` tip-bot for Peter Zijlstra
2016-03-10 12:04 ` [patch 07/15] sched/hotplug: Convert cpu_[in]active notifiers to state machine Thomas Gleixner
2016-03-10 12:04 ` [patch 10/15] sched/migration: Move calc_load_migrate() into CPU_DYING Thomas Gleixner
2016-05-05 11:24 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:06 ` tip-bot for Thomas Gleixner
2016-07-12 4:37 ` Anton Blanchard [this message]
2016-07-12 16:33 ` [patch 10/15] " Thomas Gleixner
2016-07-12 18:49 ` Vaidyanathan Srinivasan
2016-07-12 20:05 ` Shilpasri G Bhat
2016-07-13 7:49 ` Peter Zijlstra
2016-07-13 13:40 ` [tip:sched/urgent] sched/core: Correct off by one bug in load migration calculation tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 09/15] sched/migration: Move prepare transition to SCHED_STARTING state Thomas Gleixner
2016-05-05 11:24 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:06 ` tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 11/15] sched/migration: Move CPU_ONLINE into scheduler state Thomas Gleixner
2016-05-05 11:25 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:07 ` tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 12/15] sched/hotplug: Move migration CPU_DYING to sched_cpu_dying() Thomas Gleixner
2016-05-05 11:25 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:07 ` tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 14/15] sched/fair: Make ilb_notifier an explicit call Thomas Gleixner
2016-05-05 11:26 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:08 ` tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 13/15] sched/hotplug: Make activate() the last hotplug step Thomas Gleixner
2016-05-05 11:25 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:07 ` tip-bot for Thomas Gleixner
2016-03-10 12:04 ` [patch 15/15] sched: Make hrtick_notifier an explicit call Thomas Gleixner
2016-05-05 11:26 ` [tip:smp/hotplug] " tip-bot for Thomas Gleixner
2016-05-06 13:08 ` tip-bot for Thomas Gleixner
2016-04-04 7:54 ` [patch 00/15] cpu/hotplug: Convert scheduler to hotplug state machine 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=20160712143745.430733e7@kryten \
--to=anton@samba.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=mpe@ellerman.id.au \
--cc=peterz@infradead.org \
--cc=rt@linutronix.de \
--cc=shreyas@linux.vnet.ibm.com \
--cc=svaidy@linux.vnet.ibm.com \
--cc=tglx@linutronix.de \
/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