From: Peter Zijlstra <peterz@infradead.org>
To: Paul Turner <pjt@google.com>
Cc: Ben Segall <bsegall@google.com>, Ingo Molnar <mingo@redhat.com>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 4/5] sched: Guarantee new group-entities always have weight
Date: Thu, 17 Oct 2013 00:22:25 +0200 [thread overview]
Message-ID: <20131016222225.GO10651@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <CAPM31RKnCkjL1Rf8_0-9ok_ZMi0G2iQtxi0KWnW4rSbsEtbQbw@mail.gmail.com>
On Wed, Oct 16, 2013 at 03:16:44PM -0700, Paul Turner wrote:
> > Hurm.. this gives new groups a massive weight; nr_cpus * NICE_0. ISTR
> > there being some issues with this; or was that on the wakeup path where
> > a task woke on a cpu who's group entity had '0' load because it used to
> > run on another cpu -- I can't remember.
>
> This could also arbitrarily be MIN_WEIGHT.
>
> I don't think it actually matters, in practice the set of conditions
> for this weight to ever see use are very specific (e.g. the race
> described above). Otherwise it's always going to be re-initialized (on
> first actual enqueue) to the right value. (NICE_0_LOAD seemed to make
> sense since this is what you'd "expect" for a new, single thread,
> autogroup/group.)
Fair enough. Thanks!
next prev parent reply other threads:[~2013-10-16 22:22 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-10-16 18:16 [PATCH 0/5] Fix several races in CFS_BANDWIDTH Ben Segall
2013-10-16 18:16 ` [PATCH 1/5] sched: Fix race on toggling cfs_bandwidth_used Ben Segall
2013-10-29 14:07 ` [tip:sched/core] " tip-bot for Ben Segall
2013-10-16 18:16 ` [PATCH 2/5] sched: Fix cfs_bandwidth misuse of hrtimer_expires_remaining Ben Segall
2013-10-16 18:41 ` Peter Zijlstra
2013-10-16 19:12 ` bsegall
2013-10-29 14:07 ` [tip:sched/core] " tip-bot for Ben Segall
2013-10-16 18:16 ` [PATCH 3/5] sched: Fix hrtimer_cancel/rq->lock deadlock Ben Segall
2013-10-29 14:07 ` [tip:sched/core] sched: Fix hrtimer_cancel()/rq->lock deadlock tip-bot for Ben Segall
2013-10-16 18:16 ` [PATCH 4/5] sched: Guarantee new group-entities always have weight Ben Segall
2013-10-16 22:01 ` Peter Zijlstra
2013-10-16 22:16 ` Paul Turner
2013-10-16 22:22 ` Peter Zijlstra [this message]
2013-10-16 22:40 ` bsegall
2013-10-29 14:07 ` [tip:sched/core] " tip-bot for Paul Turner
2013-10-16 18:16 ` [PATCH 5/5] sched: Avoid throttle_cfs_rq racing with period_timer stopping Ben Segall
2013-10-16 22:06 ` Peter Zijlstra
2013-10-29 14:07 ` [tip:sched/core] sched: Avoid throttle_cfs_rq() " tip-bot for Ben Segall
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=20131016222225.GO10651@twins.programming.kicks-ass.net \
--to=peterz@infradead.org \
--cc=bsegall@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=pjt@google.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