From: "Arnaldo Carvalho de Melo" <acme@redhat.com>
To: David Miller <davem@davemloft.net>
Cc: dada1@cosmosbay.com, netdev@vger.kernel.org, ak@suse.de, acme@redhat.com
Subject: Re: [PATCH] INET : removes per bucket rwlock in tcp/dccp ehash table
Date: Thu, 1 Nov 2007 09:20:09 -0200 [thread overview]
Message-ID: <20071101112009.GD4928@ghostprotocols.net> (raw)
In-Reply-To: <20071101.040340.60088560.davem@davemloft.net>
Em Thu, Nov 01, 2007 at 04:03:40AM -0700, David Miller escreveu:
> From: Eric Dumazet <dada1@cosmosbay.com>
> Date: Thu, 01 Nov 2007 11:16:20 +0100
>
> > As done two years ago on IP route cache table (commit
> > 22c047ccbc68fa8f3fa57f0e8f906479a062c426) , we can avoid using one lock per
> > hash bucket for the huge TCP/DCCP hash tables.
> >
> > On a typical x86_64 platform, this saves about 2MB or 4MB of ram, for litle
> > performance differences. (we hit a different cache line for the rwlock, but
> > then the bucket cache line have a better sharing factor among cpus, since we
> > dirty it less often)
> >
> > Using a 'small' table of hashed rwlocks should be more than enough to provide
> > correct SMP concurrency between different buckets, without using too much
> > memory. Sizing of this table depends on NR_CPUS and various CONFIG settings.
> >
> > This patch provides some locking abstraction that may ease a future work using
> > a different model for TCP/DCCP table.
> >
> > Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>
>
> Nice work Eric.
>
> I've tossed this into my local tree and we'll let this cook
> for a few days. If no problems crop up I will submit it
> for 2.6.24 because the memory savings is non-trivial.
Agreed, thanks!
Acked-by: Arnaldo Carvalho de Melo <acme@redhat.com>
next prev parent reply other threads:[~2007-11-01 11:20 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-11-01 10:16 [PATCH] INET : removes per bucket rwlock in tcp/dccp ehash table Eric Dumazet
2007-11-01 11:03 ` David Miller
2007-11-01 11:20 ` Arnaldo Carvalho de Melo [this message]
2007-11-01 11:15 ` Ilpo Järvinen
2007-11-01 16:06 ` Jarek Poplawski
2007-11-01 18:00 ` Eric Dumazet
2007-11-01 16:14 ` Stephen Hemminger
2007-11-01 17:54 ` Eric Dumazet
2007-11-01 18:48 ` Rick Jones
2007-11-01 19:00 ` Eric Dumazet
2007-11-01 19:17 ` Eric Dumazet
2007-11-01 21:52 ` David Miller
2007-11-01 21:46 ` David Miller
2007-11-03 23:18 ` Andi Kleen
2007-11-03 23:23 ` David Miller
2007-11-04 0:54 ` Andi Kleen
2007-11-04 11:31 ` Eric Dumazet
2007-11-04 12:26 ` Andi Kleen
2007-11-04 13:05 ` Eric Dumazet
2007-11-04 21:56 ` David Miller
2007-11-04 23:01 ` Andi Kleen
2007-11-05 4:24 ` David Miller
2007-11-05 4:35 ` David Miller
2007-11-04 17:58 ` Jarek Poplawski
2007-11-04 18:15 ` Jarek Poplawski
2007-11-04 21:23 ` Eric Dumazet
2007-11-04 23:08 ` Jarek Poplawski
2007-11-07 10:41 ` David Miller
2007-11-07 12:13 ` Jarek Poplawski
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=20071101112009.GD4928@ghostprotocols.net \
--to=acme@redhat.com \
--cc=ak@suse.de \
--cc=dada1@cosmosbay.com \
--cc=davem@davemloft.net \
--cc=netdev@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 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).