LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Joonsoo Kim <js1304@gmail.com>
To: Christoph Lameter <cl@linux.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	akpm@linuxfoundation.org, LKML <linux-kernel@vger.kernel.org>,
	Linux Memory Management List <linux-mm@kvack.org>,
	Pekka Enberg <penberg@kernel.org>,
	iamjoonsoo@lge.com, Jesper Dangaard Brouer <brouer@redhat.com>
Subject: Re: [RFC 1/3] Slab infrastructure for array operations
Date: Wed, 28 Jan 2015 10:33:14 +0900	[thread overview]
Message-ID: <CAAmzW4MzNfcRucHeTxJtXLks5T-Def=O1sRpQY6fo5ybTzKsBA@mail.gmail.com> (raw)
In-Reply-To: <alpine.DEB.2.11.1501271054310.25124@gentwo.org>

2015-01-28 1:57 GMT+09:00 Christoph Lameter <cl@linux.com>:
> On Tue, 27 Jan 2015, Joonsoo Kim wrote:
>
>> IMHO, exposing these options is not a good idea. It's really
>> implementation specific. And, this flag won't show consistent performance
>> according to specific slab implementation. For example, to get best
>> performance, if SLAB is used, GFP_SLAB_ARRAY_LOCAL would be the best option,
>> but, for the same purpose, if SLUB is used, GFP_SLAB_ARRAY_NEW would
>> be the best option. And, performance could also depend on number of objects
>> and size.
>
> Why would slab show a better performance? SLUB also can have partial
> allocated pages per cpu and could also get data quite fast if only a
> minimal number of objects are desired. SLAB is slightly better because the
> number of cachelines touches stays small due to the arrangement of the freelist
> on the slab page and the queueing approach that does not involve linked
> lists.
>
>
> GFP_SLAB_ARRAY new is best for large quantities in either allocator since
> SLAB also has to construct local metadata structures.

In case of SLAB, there is just a little more work to construct local metadata so
GFP_SLAB_ARRAY_NEW would not show better performance
than GFP_SLAB_ARRAY_LOCAL, because it would cause more overhead due to
more page allocations. Because of this characteristic, I said that
which option is
the best is implementation specific and therefore we should not expose it.

Even if we narrow down the problem to the SLUB, choosing correct option is
difficult enough. User should know how many objects are cached in this
kmem_cache
in order to choose best option since relative quantity would make
performance difference.

And, how many objects are cached in this kmem_cache could be changed
whenever implementation changed.

Thanks.

  reply	other threads:[~2015-01-28  1:33 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-23 21:37 [RFC 0/3] Slab allocator " Christoph Lameter
2015-01-23 21:37 ` [RFC 1/3] Slab infrastructure for " Christoph Lameter
2015-01-27  8:21   ` Joonsoo Kim
2015-01-27 16:57     ` Christoph Lameter
2015-01-28  1:33       ` Joonsoo Kim [this message]
2015-01-28 15:30         ` Christoph Lameter
2015-01-29  7:44           ` Joonsoo Kim
2015-02-03 22:55             ` Jesper Dangaard Brouer
2015-01-23 21:37 ` [RFC 2/3] slub: Support " Christoph Lameter
2015-01-23 21:37 ` [RFC 3/3] Array alloc test code Christoph Lameter
2015-01-23 22:57 ` [RFC 0/3] Slab allocator array operations Andrew Morton
2015-01-24  0:28   ` Christoph Lameter
2015-02-03 23:19     ` Jesper Dangaard Brouer
2015-02-06 18:39       ` Christoph Lameter

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='CAAmzW4MzNfcRucHeTxJtXLks5T-Def=O1sRpQY6fo5ybTzKsBA@mail.gmail.com' \
    --to=js1304@gmail.com \
    --cc=akpm@linuxfoundation.org \
    --cc=brouer@redhat.com \
    --cc=cl@linux.com \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=iamjoonsoo@lge.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=penberg@kernel.org \
    --subject='Re: [RFC 1/3] Slab infrastructure for array operations' \
    /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

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).