From: "Paul E. McKenney" <paulmck@kernel.org>
To: Michal Hocko <mhocko@suse.com>
Cc: Uladzislau Rezki <urezki@gmail.com>,
LKML <linux-kernel@vger.kernel.org>, RCU <rcu@vger.kernel.org>,
linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>,
Peter Zijlstra <peterz@infradead.org>,
Vlastimil Babka <vbabka@suse.cz>,
Thomas Gleixner <tglx@linutronix.de>,
"Theodore Y . Ts'o" <tytso@mit.edu>,
Joel Fernandes <joel@joelfernandes.org>,
Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
Oleksiy Avramchenko <oleksiy.avramchenko@sonymobile.com>,
Mel Gorman <mgorman@suse.de>
Subject: Re: [RFC-PATCH 2/4] mm: Add __rcu_alloc_page_lockless() func.
Date: Tue, 22 Sep 2020 08:49:51 -0700 [thread overview]
Message-ID: <20200922154951.GX29330@paulmck-ThinkPad-P72> (raw)
In-Reply-To: <20200922075002.GU12990@dhcp22.suse.cz>
On Tue, Sep 22, 2020 at 09:50:02AM +0200, Michal Hocko wrote:
> [Cc Mel - the thread starts http://lkml.kernel.org/r/20200918194817.48921-1-urezki@gmail.com]
>
> On Mon 21-09-20 21:48:19, Uladzislau Rezki wrote:
> > Hello, Michal.
> >
> > > >
> > > > Yes, I do well remember that you are unhappy with this approach.
> > > > Unfortunately, thus far, there is no solution that makes all developers
> > > > happy. You might be glad to hear that we are also looking into other
> > > > solutions, each of which makes some other developers unhappy. So we
> > > > are at least not picking on you alone. :-/
> > >
> > > No worries I do not feel like a whipping boy here. But do expect me to
> > > argue against the approach. I would also appreciate it if there was some
> > > more information on other attempts, why they have failed. E.g. why
> > > pre-allocation is not an option that works well enough in most
> > > reasonable workloads.
> > Pre-allocating has some drawbacks:
> >
> > a) It is impossible to predict how many pages will be required to
> > cover a demand that is controlled by different workloads on
> > various systems.
>
> Yes, this is not trivial but not a rocket science either. Remember that
> you are relying on a very dumb watermark based pcp pool from the
> allocator. Mimicing a similar implementation shouldn't be all that hard
> and you will get your own pool which doesn't affect other page allocator
> users as much as a bonus.
>
> > b) Memory overhead since we do not know how much pages should be
> > preloaded: 100, 200 or 300
>
> Does anybody who really needs this optimization actually cares about 300
> pages?
That would be 100-300 (maybe more) pages -per- -CPU-, so yes, some people
will care quite deeply about this.
Thanx, Paul
> > As for memory overhead, it is important to reduce it because of
> > embedded devices like phones, where a low memory condition is a
> > big issue. In that sense pre-allocating is something that we strongly
> > would like to avoid.
>
> How big "machines" are we talking about here? I would expect that really
> tiny machines would have hard times to really fill up thousands of pages
> with pointers to free...
>
> Would a similar scaling as the page allocator feasible. Really I mostly
> do care about shared nature of the pcp allocator list that one user can
> easily monopolize with this API.
>
> > > I would also appreciate some more thoughts why we
> > > need to optimize for heavy abusers of RCU (like close(open) extremes).
> > >
> > I think here is a small misunderstanding. Please note, that is not only
> > about performance and corner cases. There is a single argument support
> > of the kvfree_rcu(ptr), where maintaining an array in time is needed.
> > The fallback of the single argument case is extrimely slow.
>
> This should be part of the changelog.
> >
> > Single-argument details is here: https://lkml.org/lkml/2020/4/28/1626
>
> Error 501
>
> > > > > I strongly agree with Thomas http://lkml.kernel.org/r/87tux4kefm.fsf@nanos.tec.linutronix.de
> > > > > that this optimization is not aiming at reasonable workloads. Really, go
> > > > > with pre-allocated buffer and fallback to whatever slow path you have
> > > > > already. Exposing more internals of the allocator is not going to do any
> > > > > good for long term maintainability.
> > > >
> > > > I suggest that you carefully re-read the thread following that email.
> > >
> > > I clearly remember Thomas not being particularly happy that you optimize
> > > for a corner case. I do not remember there being a consensus that this
> > > is the right approach. There was some consensus that this is better than
> > > a gfp flag. Still quite bad though if you ask me.
> > >
> > > > Given a choice between making users unhappy and making developers
> > > > unhappy, I will side with the users each and every time.
> > >
> > > Well, let me rephrase. It is not only about me (as a developer) being
> > > unhappy but also all the side effects this would have for users when
> > > performance of their favorite workload declines for no apparent reason
> > > just because pcp caches are depleted by an unrelated process.
> > >
> > If depleted, we have a special worker that charge it. From the other hand,
> > the pcplist can be depleted by its nature, what _is_ not wrong. But just
> > in case we secure it since you had a concern about it.
>
> pcp free lists should ever get empty when we run out of memory and need
> to reclaim. Otherwise they are constantly refilled/rebalanced on demand.
> The fact that you are refilling them from outside just suggest that you
> are operating on a wrong layer. Really, create your own pool of pages
> and rebalance them based on the workload.
>
> > Could you please specify a real test case or workload you are talking about?
>
> I am not a performance expert but essentially any memory allocator heavy
> workload might notice. I am pretty sure Mel would tell you more.
>
> --
> Michal Hocko
> SUSE Labs
next prev parent reply other threads:[~2020-09-22 15:49 UTC|newest]
Thread overview: 76+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-18 19:48 [PATCH 0/4] kvfree_rcu() and _LOCK_NESTING/_PREEMPT_RT Uladzislau Rezki (Sony)
2020-09-18 19:48 ` [PATCH 1/4] rcu/tree: Add a work to allocate pages from regular context Uladzislau Rezki (Sony)
2020-09-18 19:48 ` [RFC-PATCH 2/4] mm: Add __rcu_alloc_page_lockless() func Uladzislau Rezki (Sony)
2020-09-21 7:47 ` Michal Hocko
2020-09-21 15:45 ` Paul E. McKenney
2020-09-21 16:03 ` Michal Hocko
2020-09-21 19:48 ` Uladzislau Rezki
2020-09-22 7:50 ` Michal Hocko
2020-09-22 13:12 ` Uladzislau Rezki
2020-09-22 15:35 ` Michal Hocko
2020-09-23 10:37 ` Mel Gorman
2020-09-23 15:41 ` Paul E. McKenney
2020-09-23 23:22 ` Mel Gorman
2020-09-24 8:16 ` Uladzislau Rezki
2020-09-24 11:16 ` Peter Zijlstra
2020-09-24 15:16 ` Uladzislau Rezki
2020-09-24 11:19 ` Peter Zijlstra
2020-09-24 15:21 ` Uladzislau Rezki
2020-09-25 8:15 ` Peter Zijlstra
2020-09-25 10:25 ` Uladzislau Rezki
2020-09-24 15:38 ` Paul E. McKenney
2020-09-25 8:26 ` Peter Zijlstra
2020-09-26 14:37 ` Paul E. McKenney
2020-09-25 8:05 ` Michal Hocko
2020-09-25 15:31 ` Uladzislau Rezki
2020-09-25 15:47 ` Michal Hocko
2020-09-29 16:25 ` Uladzislau Rezki
2020-09-30 9:27 ` Michal Hocko
2020-09-30 12:35 ` Uladzislau Rezki
2020-09-30 12:44 ` Michal Hocko
2020-09-30 13:39 ` Uladzislau Rezki
2020-09-30 16:46 ` Michal Hocko
2020-09-30 20:36 ` Uladzislau Rezki
2020-09-30 15:25 ` Joel Fernandes
2020-09-30 16:48 ` Michal Hocko
2020-09-30 17:03 ` Joel Fernandes
2020-09-30 17:22 ` Michal Hocko
2020-09-30 17:48 ` Joel Fernandes
2020-09-25 16:17 ` Mel Gorman
2020-09-25 17:57 ` Uladzislau Rezki
2020-09-22 15:49 ` Paul E. McKenney [this message]
2020-09-22 3:35 ` Paul E. McKenney
2020-09-22 8:03 ` Michal Hocko
2020-09-22 15:46 ` Paul E. McKenney
2020-09-23 11:27 ` Uladzislau Rezki
2020-09-29 10:15 ` Vlastimil Babka
2020-09-29 22:07 ` Uladzislau Rezki
2020-09-30 10:35 ` Michal Hocko
2020-10-01 19:32 ` Uladzislau Rezki
2020-09-30 14:39 ` Vlastimil Babka
2020-09-30 15:37 ` Joel Fernandes
2020-10-01 19:26 ` Uladzislau Rezki
2020-10-02 7:11 ` Michal Hocko
2020-10-02 8:50 ` Mel Gorman
2020-10-02 9:05 ` Michal Hocko
2020-10-05 15:08 ` Uladzislau Rezki
2020-10-05 15:41 ` Michal Hocko
2020-10-06 22:25 ` Uladzislau Rezki
2020-10-07 10:02 ` Michal Hocko
2020-10-07 11:02 ` Uladzislau Rezki
2020-10-02 9:07 ` Peter Zijlstra
2020-10-02 9:45 ` Mel Gorman
2020-10-02 9:58 ` Peter Zijlstra
2020-10-02 10:19 ` Mel Gorman
2020-10-02 14:41 ` Paul E. McKenney
2020-10-06 10:03 ` Mel Gorman
2020-10-06 15:41 ` Paul E. McKenney
2020-10-05 13:58 ` Uladzislau Rezki
2020-10-02 8:06 ` Mel Gorman
2020-10-05 14:12 ` Uladzislau Rezki
2020-09-18 19:48 ` [PATCH 3/4] rcu/tree: use " Uladzislau Rezki (Sony)
2020-09-18 19:48 ` [PATCH 4/4] rcu/tree: Use schedule_delayed_work() instead of WQ_HIGHPRI queue Uladzislau Rezki (Sony)
2020-09-20 15:06 ` Paul E. McKenney
2020-09-21 13:27 ` Uladzislau Rezki
2020-09-18 22:15 ` [PATCH 0/4] kvfree_rcu() and _LOCK_NESTING/_PREEMPT_RT Paul E. McKenney
2020-09-30 15:52 ` Joel Fernandes
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=20200922154951.GX29330@paulmck-ThinkPad-P72 \
--to=paulmck@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=bigeasy@linutronix.de \
--cc=joel@joelfernandes.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@suse.com \
--cc=oleksiy.avramchenko@sonymobile.com \
--cc=peterz@infradead.org \
--cc=rcu@vger.kernel.org \
--cc=tglx@linutronix.de \
--cc=tytso@mit.edu \
--cc=urezki@gmail.com \
--cc=vbabka@suse.cz \
/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).