From: Rakib Mullick <rakib.mullick@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: mingo@kernel.org, linux-kernel@vger.kernel.org,
paulmck <paulmck@linux.vnet.ibm.com>
Subject: Re: Add rq->nr_uninterruptible count to dest cpu's rq while CPU goes down.
Date: Thu, 16 Aug 2012 21:32:45 +0600 [thread overview]
Message-ID: <CADZ9YHjzNVz6nSqbnTgVYfwi2hz0KYuXC89GHw6UjL2GyGZ64A@mail.gmail.com> (raw)
In-Reply-To: <1345128138.29668.42.camel@twins>
On 8/16/12, Peter Zijlstra <peterz@infradead.org> wrote:
> On Thu, 2012-08-16 at 20:28 +0600, Rakib Mullick wrote:
>
>> nr_uninterruptible is coupled with tasks on the runqueue to calculate
>> nr_active numbers.
>
> It is not.. nr_uninterruptible is incremented on the cpu the task goes
> to sleep and decremented on the cpu doing the wakeup.
>
If nr_uninterruptible's life cycle is this simple then, while CPU
goes down, nr_uninterruptible count will be decremented when all the
tasks are moved to other CPUs and should be fine.
> This means that nr_uninterruptible is a complete mess and any per-cpu
> value isn't meaningful at all.
>
Well, if nr_uninterruptible is a mess, then this patch has no meaning.
And also I think migrate_nr_uninterruptible() is meaning less too.
> It is quite possible to always have the inc on cpu0 and the decrement on
> cpu1, yielding results like:
>
> {1000, -1000} for an effective nr_uninterruptible = 0. Taking either cpu
> down will then migrate whatever delta it has to another cpu, but there
> might only be a single task, yet the delta is +-1000.
>
>> In calc_load_fold_active(), this nr_active numbers are used to
>> calculate delta. This is how I understand this part and seeing some
>> impact.
>
> You understand wrong, please re-read the comment added in commit
> 5167e8d5.
>
Yes, reading.
Thanks,
Rakib.
next prev parent reply other threads:[~2012-08-16 15:32 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-16 13:45 Add rq->nr_uninterruptible count to dest cpu's rq while CPU goes down Rakib Mullick
2012-08-16 13:56 ` Peter Zijlstra
2012-08-16 14:28 ` Rakib Mullick
2012-08-16 14:42 ` Peter Zijlstra
2012-08-16 15:32 ` Rakib Mullick [this message]
2012-08-16 17:46 ` Peter Zijlstra
2012-08-17 13:39 ` Rakib Mullick
2012-08-20 9:26 ` Peter Zijlstra
2012-08-20 16:10 ` Rakib Mullick
2012-08-20 16:16 ` Peter Zijlstra
2012-08-20 16:26 ` Paul E. McKenney
2012-08-27 18:44 ` Paul E. McKenney
2012-08-28 6:57 ` Rakib Mullick
2012-08-28 13:42 ` Paul E. McKenney
2012-08-28 16:52 ` Rakib Mullick
2012-08-28 17:07 ` Paul E. McKenney
2012-08-29 1:05 ` Rakib Mullick
2012-09-04 18:43 ` [tip:sched/core] sched: Fix load avg vs cpu-hotplug tip-bot for Peter Zijlstra
2012-09-05 12:36 ` Peter Zijlstra
2012-09-05 13:29 ` Ingo Molnar
2012-09-05 17:01 ` Peter Zijlstra
2012-09-05 17:34 ` Ingo Molnar
2012-09-05 22:03 ` Peter Zijlstra
2012-09-05 23:39 ` Paul E. McKenney
2012-09-06 3:30 ` Rakib Mullick
2012-09-14 6:14 ` [tip:sched/core] sched: Fix load avg vs. cpu-hotplug tip-bot for 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=CADZ9YHjzNVz6nSqbnTgVYfwi2hz0KYuXC89GHw6UjL2GyGZ64A@mail.gmail.com \
--to=rakib.mullick@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@kernel.org \
--cc=paulmck@linux.vnet.ibm.com \
--cc=peterz@infradead.org \
/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).