From: Christoph Lameter <cl@linux.com>
To: Michal Hocko <mhocko@kernel.org>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>,
Peter Zijlstra <peterz@infradead.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: fast path cycle muncher (vmstat: make vmstat_updater deferrable again and shut down on idle)
Date: Tue, 26 Jan 2016 10:25:21 -0600 (CST) [thread overview]
Message-ID: <alpine.DEB.2.20.1601261022140.27957@east.gentwo.org> (raw)
In-Reply-To: <20160125201319.GA19020@dhcp22.suse.cz>
On Mon, 25 Jan 2016, Michal Hocko wrote:
> > Why would the deferring cause this overhead?
>
> I guess the profile speaks for itself, doesn't it?
But the system is going idle? Why would this impact performance?
> > Also there is no cross core activity from quiet_vmstat(). It simply
> > disables the local vmstat updates.
>
> It doesn't go cross core but it still does nr_zones * counters atomic
> ops.
If there are updates then yes.
> > It reduces the OS events that the application experiences by folding it
> > into the tick events. If its not deferrable then a timer event will be
> > generated in addition to the tick. We do not want that.
>
> Yes this is what I have read in the changelog. But "how much" part is
> really missing. Is this even quantifiable?
Oh yes. If you want to have low latency responses yes. And event like that
causes a couple of microseconds delay which will cause a moneytary impact
if you have to react to stock trading events.
> Maybe this overhead can be reduced by outsourcing the functionality to
> vmstat_shepherd which can check idle CPUs, cancel the timer for them
> update the differentials and put them to cpu_stat_off?
Remote updating of differentials is problematic due to the counters being
per cpu values that are expected to only be updated from the cpu that
"owns" them.
> > The regression that I know of (independent of "RT") is due as far as I
> > know due to the switch of the parameters of some vmstat functions to 64
> > bit instead of 32 bit.
>
> I am not sure I am following.
An additional patch was merged for 4.5 that increases the arguments to the
counter operations to 64 bit which is known for regressions.
next prev parent reply other threads:[~2016-01-26 16:25 UTC|newest]
Thread overview: 77+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-19 0:33 mm, vmstat: kernel BUG at mm/vmstat.c:1408! Sasha Levin
2015-12-21 13:08 ` Christoph Lameter
2015-12-21 20:28 ` Sasha Levin
2015-12-21 21:07 ` Sasha Levin
2015-12-21 21:14 ` Christoph Lameter
2015-12-22 17:21 ` Christoph Lameter
2015-12-24 20:14 ` Sasha Levin
2015-12-29 17:01 ` Christoph Lameter
2015-12-29 17:18 ` Christoph Lameter
2016-01-04 18:05 ` Christoph Lameter
2016-01-04 18:46 ` Sasha Levin
2016-01-12 11:31 ` Shiraz Hashim
2016-01-12 12:23 ` Christoph Lameter
2016-01-12 12:27 ` Shiraz Hashim
2016-01-13 11:36 ` Shiraz Hashim
2016-01-13 12:32 ` Shiraz Hashim
2016-01-14 21:06 ` Sasha Levin
2016-01-20 14:37 ` Michal Hocko
2016-01-20 14:56 ` Sasha Levin
2016-01-20 15:10 ` Michal Hocko
2016-01-20 15:20 ` Christoph Lameter
2016-01-20 15:49 ` Sasha Levin
2016-01-20 15:55 ` Christoph Lameter
2016-01-20 21:28 ` Michal Hocko
2016-01-20 21:57 ` Christoph Lameter
2016-01-21 8:24 ` Michal Hocko
2016-01-21 15:45 ` Christoph Lameter
2016-01-21 16:51 ` Michal Hocko
2016-01-21 17:38 ` Christoph Lameter
2016-01-22 11:00 ` Shiraz Hashim
2016-01-22 14:04 ` Michal Hocko
2016-01-22 16:07 ` Christoph Lameter
2016-01-22 16:12 ` Michal Hocko
2016-01-22 16:46 ` Christoph Lameter
2016-01-22 17:12 ` Michal Hocko
2016-01-23 16:21 ` fast path cycle muncher (vmstat: make vmstat_updater deferrable again and shut down on idle) Mike Galbraith
2016-01-24 0:33 ` Christoph Lameter
2016-01-24 2:46 ` Mike Galbraith
2016-01-24 3:46 ` Christoph Lameter
2016-01-24 5:36 ` Mike Galbraith
2016-01-25 17:42 ` Michal Hocko
2016-01-25 18:02 ` Christoph Lameter
2016-01-25 20:13 ` Michal Hocko
2016-01-26 16:25 ` Christoph Lameter [this message]
2016-01-26 18:31 ` Mike Galbraith
2016-01-26 18:34 ` Christoph Lameter
2016-01-26 18:45 ` Mike Galbraith
2016-01-26 19:20 ` Christoph Lameter
2016-01-27 3:12 ` Mike Galbraith
2016-01-27 4:15 ` Mike Galbraith
2016-01-27 16:28 ` Christoph Lameter
2016-01-28 15:36 ` Frederic Weisbecker
2016-01-28 16:42 ` Christoph Lameter
2016-01-26 2:14 ` Mike Galbraith
2016-01-26 2:25 ` Mike Galbraith
2016-01-26 16:26 ` Christoph Lameter
2016-01-26 17:39 ` Mike Galbraith
2016-01-26 18:19 ` Christoph Lameter
2016-01-26 16:26 ` Christoph Lameter
2016-01-26 17:08 ` Mike Galbraith
2016-01-26 18:22 ` Christoph Lameter
2016-01-26 19:09 ` Mike Galbraith
2016-01-26 19:22 ` Christoph Lameter
2016-01-27 16:48 ` [PATCH] mm, vmstat: make quiet_vmstat lighter (was: Re: fast path cycle muncher (vmstat: make vmstat_updater deferrable) " Michal Hocko
2016-01-27 17:03 ` Mike Galbraith
2016-01-27 18:26 ` Christoph Lameter
2016-01-28 15:21 ` Michal Hocko
2016-01-28 16:40 ` Christoph Lameter
2016-01-28 16:53 ` Michal Hocko
2016-01-28 17:05 ` Christoph Lameter
2016-01-28 15:31 ` Michal Hocko
2016-01-28 15:37 ` [PATCH] vmstat: make vmstat_update deferrable (was: Re: [PATCH] mm, vmstat: make quiet_vmstat lighter) " Michal Hocko
2016-01-28 16:48 ` Christoph Lameter
2016-01-28 16:42 ` [PATCH] mm, vmstat: make quiet_vmstat lighter (was: Re: fast " Christoph Lameter
2016-01-24 16:57 ` mm, vmstat: kernel BUG at mm/vmstat.c:1408! Linus Torvalds
2016-01-20 15:14 ` Christoph Lameter
2016-01-20 15:20 ` Michal Hocko
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=alpine.DEB.2.20.1601261022140.27957@east.gentwo.org \
--to=cl@linux.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mhocko@kernel.org \
--cc=peterz@infradead.org \
--cc=umgwanakikbuti@gmail.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).