From: Michal Hocko <mhocko@kernel.org>
To: Joel Fernandes <joel@joelfernandes.org>
Cc: linux-kernel@vger.kernel.org, Tim Murray <timmurray@google.com>,
carmenjackson@google.com, mayankgupta@google.com,
dancol@google.com, rostedt@goodmis.org, minchan@kernel.org,
akpm@linux-foundation.org, kernel-team@android.com,
"Aneesh Kumar K.V" <aneesh.kumar@linux.ibm.com>,
Dan Williams <dan.j.williams@intel.com>,
Jerome Glisse <jglisse@redhat.com>,
linux-mm@kvack.org, Matthew Wilcox <willy@infradead.org>,
Ralph Campbell <rcampbell@nvidia.com>,
Vlastimil Babka <vbabka@suse.cz>
Subject: Re: [PATCH v2] mm: emit tracepoint when RSS changes by threshold
Date: Thu, 5 Sep 2019 16:20:10 +0200 [thread overview]
Message-ID: <20190905142010.GC3838@dhcp22.suse.cz> (raw)
In-Reply-To: <20190905141452.GA26466@google.com>
On Thu 05-09-19 10:14:52, Joel Fernandes wrote:
> On Thu, Sep 05, 2019 at 12:54:24PM +0200, Michal Hocko wrote:
> > On Wed 04-09-19 12:28:08, Joel Fernandes wrote:
> > > On Wed, Sep 4, 2019 at 11:38 AM Michal Hocko <mhocko@kernel.org> wrote:
> > > >
> > > > On Wed 04-09-19 11:32:58, Joel Fernandes wrote:
> > > > > On Wed, Sep 04, 2019 at 10:45:08AM +0200, Michal Hocko wrote:
> > > > > > On Tue 03-09-19 16:09:05, Joel Fernandes (Google) wrote:
> > > > > > > Useful to track how RSS is changing per TGID to detect spikes in RSS and
> > > > > > > memory hogs. Several Android teams have been using this patch in various
> > > > > > > kernel trees for half a year now. Many reported to me it is really
> > > > > > > useful so I'm posting it upstream.
> > > > > > >
> > > > > > > Initial patch developed by Tim Murray. Changes I made from original patch:
> > > > > > > o Prevent any additional space consumed by mm_struct.
> > > > > > > o Keep overhead low by checking if tracing is enabled.
> > > > > > > o Add some noise reduction and lower overhead by emitting only on
> > > > > > > threshold changes.
> > > > > >
> > > > > > Does this have any pre-requisite? I do not see trace_rss_stat_enabled in
> > > > > > the Linus tree (nor in linux-next).
> > > > >
> > > > > No, this is generated automatically by the tracepoint infrastructure when a
> > > > > tracepoint is added.
> > > >
> > > > OK, I was not aware of that.
> > > >
> > > > > > Besides that why do we need batching in the first place. Does this have a
> > > > > > measurable overhead? How does it differ from any other tracepoints that we
> > > > > > have in other hotpaths (e.g. page allocator doesn't do any checks).
> > > > >
> > > > > We do need batching not only for overhead reduction,
> > > >
> > > > What is the overhead?
> > >
> > > The overhead is occasionally higher without the threshold (that is if we
> > > trace every counter change). I would classify performance benefit to be
> > > almost the same and within the noise.
> >
> > OK, so the additional code is not really justified.
>
> It is really justified. Did you read the whole of the last email?
Of course I have. The information that numbers are in noise with some
outliers (without any details about the underlying reason) is simply
showing that you are optimizing something probably not worth it.
I would recommend adding a simple tracepoint. That should be pretty non
controversial. And if you want to add an optimization on top then
provide data to justify it.
--
Michal Hocko
SUSE Labs
next prev parent reply other threads:[~2019-09-05 14:20 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-03 20:09 Joel Fernandes (Google)
2019-09-04 4:44 ` Suren Baghdasaryan
2019-09-04 4:51 ` Daniel Colascione
2019-09-04 5:15 ` Joel Fernandes
2019-09-04 5:42 ` Daniel Colascione
2019-09-04 14:59 ` Joel Fernandes
2019-09-04 17:15 ` Daniel Colascione
2019-09-04 23:59 ` sspatil
2019-09-04 5:02 ` Joel Fernandes
2019-09-04 5:38 ` Suren Baghdasaryan
2019-09-04 8:45 ` Michal Hocko
2019-09-04 15:32 ` Joel Fernandes
2019-09-04 15:37 ` Michal Hocko
2019-09-04 16:28 ` Joel Fernandes
2019-09-05 10:54 ` Michal Hocko
2019-09-05 14:14 ` Joel Fernandes
2019-09-05 14:20 ` Michal Hocko [this message]
2019-09-05 14:23 ` Joel Fernandes
2019-09-05 14:43 ` Michal Hocko
2019-09-05 16:03 ` Suren Baghdasaryan
2019-09-05 17:35 ` Steven Rostedt
2019-09-05 17:39 ` Suren Baghdasaryan
2019-09-05 17:43 ` Tim Murray
2019-09-05 17:47 ` Joel Fernandes
2019-09-05 17:51 ` Joel Fernandes
2019-09-05 19:56 ` Tom Zanussi
2019-09-05 20:24 ` Daniel Colascione
2019-09-05 20:32 ` Tom Zanussi
2019-09-05 21:14 ` Tom Zanussi
2019-09-05 22:12 ` Daniel Colascione
2019-09-05 22:51 ` Daniel Colascione
2019-09-05 17:50 ` Daniel Colascione
2019-09-06 0:59 ` Joel Fernandes
2019-09-06 1:15 ` Daniel Colascione
2019-09-06 3:01 ` Joel Fernandes
2019-09-04 17:17 ` Daniel Colascione
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=20190905142010.GC3838@dhcp22.suse.cz \
--to=mhocko@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=aneesh.kumar@linux.ibm.com \
--cc=carmenjackson@google.com \
--cc=dan.j.williams@intel.com \
--cc=dancol@google.com \
--cc=jglisse@redhat.com \
--cc=joel@joelfernandes.org \
--cc=kernel-team@android.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mayankgupta@google.com \
--cc=minchan@kernel.org \
--cc=rcampbell@nvidia.com \
--cc=rostedt@goodmis.org \
--cc=timmurray@google.com \
--cc=vbabka@suse.cz \
--cc=willy@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