From: "Paul E. McKenney" <paulmck@linux.ibm.com>
To: Joel Fernandes <joel@joelfernandes.org>
Cc: Byungchul Park <byungchul.park@lge.com>,
linux-kernel@vger.kernel.org, Rao Shoaib <rao.shoaib@oracle.com>,
max.byungchul.park@gmail.com, kernel-team@android.com,
kernel-team@lge.com, Davidlohr Bueso <dave@stgolabs.net>,
Josh Triplett <josh@joshtriplett.org>,
Lai Jiangshan <jiangshanlai@gmail.com>,
Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
rcu@vger.kernel.org, Steven Rostedt <rostedt@goodmis.org>
Subject: Re: [PATCH RFC v1 1/2] rcu/tree: Add basic support for kfree_rcu batching
Date: Thu, 8 Aug 2019 13:51:29 -0700 [thread overview]
Message-ID: <20190808205129.GU28441@linux.ibm.com> (raw)
In-Reply-To: <20190808201333.GE261256@google.com>
On Thu, Aug 08, 2019 at 04:13:33PM -0400, Joel Fernandes wrote:
> On Thu, Aug 08, 2019 at 11:11:12AM -0700, Paul E. McKenney wrote:
> > On Thu, Aug 08, 2019 at 07:26:10PM +0900, Byungchul Park wrote:
> > > On Wed, Aug 07, 2019 at 05:45:04AM -0400, Joel Fernandes wrote:
> > > > On Tue, Aug 06, 2019 at 04:56:31PM -0700, Paul E. McKenney wrote:
> > >
> > > [snip]
> > >
> > > > > On Tue, Aug 06, 2019 at 05:20:40PM -0400, Joel Fernandes (Google) wrote:
> > > > > Of course, I am hoping that a later patch uses an array of pointers built
> > > > > at kfree_rcu() time, similar to Rao's patch (with or without kfree_bulk)
> > > > > in order to reduce per-object cache-miss overhead. This would make it
> > > > > easier for callback invocation to keep up with multi-CPU kfree_rcu()
> > > > > floods.
> > > >
> > > > I think Byungchul tried an experiment with array of pointers and wasn't
> > > > immediately able to see a benefit. Perhaps his patch needs a bit more polish
> > > > or another test-case needed to show benefit due to cache-misses, and the perf
> > > > tool could be used to show if cache misses were reduced. For this initial
> > > > pass, we decided to keep it without the array optimization.
> > >
> > > I'm still seeing no improvement with kfree_bulk().
> > >
> > > I've been thinking I could see improvement with kfree_bulk() because:
> > >
> > > 1. As you guys said, the number of cache misses will be reduced.
> > > 2. We can save (N - 1) irq-disable instructions while N kfrees.
> > > 3. As Joel said, saving/restoring CPU status that kfree() does inside
> > > is not required.
> > >
> > > But even with the following patch applied, the result was same as just
> > > batching test. We might need to get kmalloc objects from random
> > > addresses to maximize the result when using kfree_bulk() and this is
> > > even closer to real practical world too.
> > >
> > > And the second and third reasons doesn't seem to work as much as I
> > > expected.
> > >
> > > Do you have any idea? Or what do you think about it?
> >
> > I would not expect kfree_batch() to help all that much unless the
> > pre-grace-period kfree_rcu() code segregated the objects on a per-slab
> > basis.
>
> You mean kfree_bulk() instead of kfree_batch() right? I agree with you, would
> be nice to do per-slab optimization in the future.
Indeed I do mean kfree_bulk()! One of those mornings, I guess...
But again, without the per-slab locality, I doubt that we will see much
improvement from kfree_bulk() over kfree().
> Also, I am thinking that whenever we do per-slab optimization, then the
> kmem_cache_free_bulk() can be optimized further. If all pointers are on the
> same slab, then we can just do virt_to_cache on the first pointer and avoid
> repeated virt_to_cache() calls. That might also give a benefit -- but I could
> be missing something.
A sort might be required to make that work nicely, which would add some
overhead. Probably not that much, though, the increased locality would
have a fighting chance of overcoming the sort's overhead.
> Right now kmem_cache_free_bulk() just looks like a kmem_cache_free() in a
> loop except the small benefit of not disabling/enabling IRQs across each
> __cache_free, and the reduced cache miss benefit of using the array.
C'mon! Show some respect for the awesome power of temporal locality!!! ;-)
Thanx, Paul
next prev parent reply other threads:[~2019-08-08 20:51 UTC|newest]
Thread overview: 54+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-06 21:20 [PATCH RFC v1 1/2] rcu/tree: Add basic support for kfree_rcu batching Joel Fernandes (Google)
2019-08-06 21:20 ` [PATCH RFC v1 2/2] rcuperf: Add kfree_rcu performance Tests Joel Fernandes (Google)
2019-08-07 0:29 ` Paul E. McKenney
2019-08-07 10:22 ` Joel Fernandes
2019-08-07 17:56 ` Paul E. McKenney
2019-08-09 16:01 ` Joel Fernandes
2019-08-11 2:01 ` Joel Fernandes
2019-08-11 23:42 ` Paul E. McKenney
2019-08-06 23:56 ` [PATCH RFC v1 1/2] rcu/tree: Add basic support for kfree_rcu batching Paul E. McKenney
2019-08-07 9:45 ` Joel Fernandes
2019-08-07 17:52 ` Paul E. McKenney
2019-08-08 9:52 ` Byungchul Park
2019-08-08 12:56 ` Joel Fernandes
2019-08-08 14:23 ` Byungchul Park
2019-08-08 18:09 ` Paul E. McKenney
2019-08-11 8:36 ` Byungchul Park
2019-08-11 8:49 ` Byungchul Park
2019-08-11 23:49 ` Paul E. McKenney
2019-08-12 10:10 ` Byungchul Park
2019-08-12 13:12 ` Joel Fernandes
2019-08-13 5:29 ` Byungchul Park
2019-08-13 15:41 ` Paul E. McKenney
2019-08-14 0:11 ` Byungchul Park
2019-08-14 2:53 ` Paul E. McKenney
2019-08-14 3:43 ` Byungchul Park
2019-08-14 16:59 ` Paul E. McKenney
2019-08-11 10:37 ` Byungchul Park
2019-08-08 23:30 ` Joel Fernandes
2019-08-09 15:16 ` Paul E. McKenney
2019-08-09 15:39 ` Joel Fernandes
2019-08-09 16:33 ` Paul E. McKenney
2019-08-09 20:22 ` Joel Fernandes
2019-08-09 20:26 ` Joel Fernandes
2019-08-09 21:25 ` Joel Fernandes
2019-08-10 3:38 ` Paul E. McKenney
2019-08-09 20:29 ` Joel Fernandes
2019-08-09 20:42 ` Paul E. McKenney
2019-08-09 21:36 ` Joel Fernandes
2019-08-10 3:40 ` Paul E. McKenney
2019-08-10 3:52 ` Joel Fernandes
2019-08-10 2:42 ` Joel Fernandes
2019-08-10 3:38 ` Paul E. McKenney
2019-08-10 4:20 ` Joel Fernandes
2019-08-10 18:24 ` Paul E. McKenney
2019-08-11 2:26 ` Joel Fernandes
2019-08-11 23:35 ` Paul E. McKenney
2019-08-12 13:13 ` Joel Fernandes
2019-08-12 14:44 ` Paul E. McKenney
2019-08-08 10:26 ` Byungchul Park
2019-08-08 18:11 ` Paul E. McKenney
2019-08-08 20:13 ` Joel Fernandes
2019-08-08 20:51 ` Paul E. McKenney [this message]
2019-08-08 22:34 ` Joel Fernandes
2019-08-08 22:37 ` Paul E. McKenney
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=20190808205129.GU28441@linux.ibm.com \
--to=paulmck@linux.ibm.com \
--cc=byungchul.park@lge.com \
--cc=dave@stgolabs.net \
--cc=jiangshanlai@gmail.com \
--cc=joel@joelfernandes.org \
--cc=josh@joshtriplett.org \
--cc=kernel-team@android.com \
--cc=kernel-team@lge.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mathieu.desnoyers@efficios.com \
--cc=max.byungchul.park@gmail.com \
--cc=rao.shoaib@oracle.com \
--cc=rcu@vger.kernel.org \
--cc=rostedt@goodmis.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).