All of lore.kernel.org
 help / color / mirror / Atom feed
From: Luis Chamberlain <mcgrof@kernel.org>
To: Davidlohr Bueso <dave@stgolabs.net>
Cc: akpm@linux-foundation.org, linux-kernel@vger.kernel.org,
	broonie@kernel.org, alex.williamson@redhat.com,
	Davidlohr Bueso <dbueso@suse.de>
Subject: Re: [PATCH 1/5] lib/rbtree: introduce linked-list rbtree interface
Date: Mon, 10 Feb 2020 21:44:55 +0000	[thread overview]
Message-ID: <20200210214455.GO11244@42.do-not-panic.com> (raw)
In-Reply-To: <20200210212832.2i2i3kbybhk2rcee@linux-p48b>

On Mon, Feb 10, 2020 at 01:28:32PM -0800, Davidlohr Bueso wrote:
> On Mon, 10 Feb 2020, Luis Chamberlain wrote:
> > > access to the first node as well as
> > > both its in-order successor and predecessor. This is done at the cost of higher
> > > memory footprint: mainly additional prev and next pointers for each node. Such
> > > benefits can be seen in this table showing the amount of cycles it takes to
> > > do a full tree traversal:
> > > 
> > >    +--------+--------------+-----------+
> > >    | #nodes | plain rbtree | ll-rbtree |
> > >    +--------+--------------+-----------+
> > >    |     10 |          138 |        24 |
> > >    |    100 |        7,200 |       425 |
> > >    |   1000 |       17,000 |     8,000 |
> > >    |  10000 |      501,090 |   222,500 |
> > >    +--------+--------------+-----------+
> > 
> > Sold, however I wonder if we can have *one new API* where based on just one
> > Kconfig you either get the two pointers or not, the performance gain
> > then would only be observed if this new kconfig entry is enabled. The
> > benefit of this is that we don't shove the performance benefit down
> > all user's throughts but rather this can be decided by distributions
> > and system integrators.
> 
> I don't think we want an all or nothing approach

I'm not suggesting all or nothing for all rb tree users, I'm suggesting
an all or nothing approach to users of a *new* API. That is, users would
still need to be converted over, and *iff* the new kconfig entry is
enabled would the two pointers be added / used, otherwise we'd fallback
to the default current boring rbtree.

> as different users in the
> kernel have different needs and some users are simply unable to deal with
> enlarging data structures, while others have no problem.

The approach would allow distros which are not yet sure if two pointers
are worth it yet to continue to chug on as they used to, but also allow
them to explore this prospect in a flexible way.

  Luis

  reply	other threads:[~2020-02-10 21:45 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-07 18:03 [PATCH -next 0/5] rbtree: optimize frequent tree walks Davidlohr Bueso
2020-02-07 18:03 ` [PATCH 1/5] lib/rbtree: introduce linked-list rbtree interface Davidlohr Bueso
2020-02-10 20:07   ` Luis Chamberlain
2020-02-10 21:28     ` Davidlohr Bueso
2020-02-10 21:44       ` Luis Chamberlain [this message]
2020-02-07 18:03 ` [PATCH 2/5] proc/sysctl: optimize proc_sys_readdir Davidlohr Bueso
2020-02-10 20:08   ` Luis Chamberlain
2020-02-07 18:03 ` [PATCH 3/5] regmap: optimize sync() and drop() regcache callbacks Davidlohr Bueso
2020-02-10 13:11   ` Mark Brown
2020-02-07 18:03 ` [PATCH 4/5] vfio/type1: optimize dma_list tree iterations Davidlohr Bueso
2020-02-07 18:03 ` [PATCH 5/5] uprobes: optimize build_probe_list() Davidlohr Bueso
2020-02-10  1:46 ` [PATCH -next 0/5] rbtree: optimize frequent tree walks Andrew Morton
2020-02-10 15:56   ` Davidlohr Bueso
2020-02-10 22:35     ` Andrew Morton
2020-02-13 15:50       ` Davidlohr Bueso
2020-02-11 11:20     ` Mark Brown
2020-02-13 15:55       ` Davidlohr Bueso
2020-02-13 16:56         ` Mark Brown

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=20200210214455.GO11244@42.do-not-panic.com \
    --to=mcgrof@kernel.org \
    --cc=akpm@linux-foundation.org \
    --cc=alex.williamson@redhat.com \
    --cc=broonie@kernel.org \
    --cc=dave@stgolabs.net \
    --cc=dbueso@suse.de \
    --cc=linux-kernel@vger.kernel.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.