From: Jeff Garzik <jgarzik@mandrakesoft.com>
To: Daniel Phillips <phillips@bonn-fries.net>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
linux-fsdevel@vger.kernel.org
Subject: Re: Linux/Pro -- clusters
Date: Tue, 04 Dec 2001 12:20:07 -0500 [thread overview]
Message-ID: <3C0D05C7.6C566FF2@mandrakesoft.com> (raw)
In-Reply-To: <Pine.LNX.4.10.10112032057070.978-100000@vaio.greennet> <E16BGhe-0000Pq-00@starship.berlin> <3C0CE97F.74AFFDBC@mandrakesoft.com> <E16BJBR-0000RM-00@starship.berlin>
Daniel Phillips wrote:
> On December 4, 2001 04:19 pm, Jeff Garzik wrote:
> > ug. what's wrong with a single additional alloc for generic_ip? [if a
> > filesystem needs to do multiple allocs post-conversion, somebody's doing
> > something wrong]
>
> Single additional alloc -> twice as many allocs, two slabs, more cachelines
> dirty. This was hashed out on fsdevel, though apparently not to everyone's
> satisfaction.
>
> > Using generic_ip in its current form has the advantage of being able to
> > create a nicely-aligned kmem cache for your private inode data.
>
> I don't see why that's hard with the combined struct.
The advantage of having two structs means that both struct inode and the
private info can be aligned nicely. Yes it potentially wastes a tiny
bit more memory, but I challenge you to find an architecture where doing
this isn't a win. In a couple cases I looked at, additional slabs are
not even necessary, as kmalloc's standard ones do the job quite well.
'cat /proc/slabinfo' for a list of the sizes.
Note this only applies to inodes. There aren't enough superblocks in a
running system to worry about doing anything but simple kmalloc on the
superblock private info (before assigning to generic_sbp).
Jeff
--
Jeff Garzik | Only so many songs can be sung
Building 1024 | with two lips, two lungs, and one tongue.
MandrakeSoft | - nomeansno
next prev parent reply other threads:[~2001-12-04 17:22 UTC|newest]
Thread overview: 70+ messages / expand[flat|nested] mbox.gz Atom feed top
2001-12-03 18:12 Linux/Pro -- clusters Donald Becker
2001-12-04 1:55 ` Davide Libenzi
2001-12-04 2:09 ` Donald Becker
2001-12-04 2:23 ` Davide Libenzi
2001-12-04 2:34 ` Alexander Viro
2001-12-04 9:10 ` Alan Cox
2001-12-04 9:30 ` Thomas Langås
2001-12-04 9:45 ` Alan Cox
2001-12-04 11:34 ` Thomas Langås
2001-12-05 21:57 ` Linus Torvalds
2001-12-05 23:05 ` Andre Hedrick
2001-12-06 4:31 ` Daniel Phillips
2001-12-05 23:49 ` Alan Cox
2001-12-05 23:48 ` Andre Hedrick
2001-12-06 16:58 ` Linus Torvalds
2001-12-06 18:02 ` Alan Cox
2001-12-06 18:07 ` Linus Torvalds
2001-12-06 18:12 ` Kai Henningsen
2001-12-06 20:46 ` Linus Torvalds
2001-12-06 22:40 ` Alan Cox
2001-12-06 18:33 ` Alan Cox
2001-12-06 18:55 ` Linus Torvalds
2001-12-06 19:19 ` Alan Cox
2001-12-06 20:37 ` Linus Torvalds
2001-12-06 22:35 ` Alan Cox
2001-12-06 22:34 ` Linus Torvalds
2001-12-06 22:58 ` Alexander Viro
2001-12-07 10:14 ` Martin Dalecki
2001-12-07 10:37 ` Alan Cox
2001-12-07 10:56 ` Martin Dalecki
2001-12-07 12:08 ` Alan Cox
2001-12-07 20:51 ` On re-working the major/minor system Erik Andersen
2001-12-07 21:21 ` H. Peter Anvin
2001-12-07 21:55 ` Erik Andersen
2001-12-07 22:04 ` H. Peter Anvin
2001-12-07 23:07 ` Erik Andersen
2001-12-07 23:12 ` H. Peter Anvin
2001-12-08 11:42 ` Alan Cox
2001-12-08 20:37 ` H. Peter Anvin
2001-12-09 12:06 ` Kai Henningsen
2001-12-09 21:57 ` H. Peter Anvin
2001-12-11 20:45 ` Kai Henningsen
2001-12-06 18:38 ` Linux/Pro -- clusters Doug Ledford
2001-12-04 14:37 ` Daniel Phillips
2001-12-04 15:19 ` Jeff Garzik
2001-12-04 17:16 ` Daniel Phillips
2001-12-04 17:20 ` Jeff Garzik [this message]
2001-12-04 18:04 ` Alan Cox
2001-12-04 18:16 ` Daniel Phillips
2001-12-04 20:20 ` Andrew Morton
2001-12-05 13:11 ` Deep look into VFS Martin Dalecki
2001-12-05 15:19 ` Alexander Viro
2001-12-05 15:30 ` Martin Dalecki
-- strict thread matches above, loose matches on Subject: below --
2001-12-08 1:50 Linux/Pro -- clusters Andries.Brouwer
2001-12-08 3:42 ` H. Peter Anvin
2001-12-08 17:26 Andries.Brouwer
2001-12-09 4:22 ` Linus Torvalds
2001-12-09 5:49 ` Alexander Viro
2001-12-09 8:59 Andries.Brouwer
2001-12-10 16:49 ` Alexander Viro
2001-12-10 17:09 ` Alan Cox
2001-12-11 8:39 ` Albert D. Cahalan
2001-12-10 19:36 Andries.Brouwer
2001-12-10 22:55 ` Alexander Viro
2001-12-10 19:51 Andries.Brouwer
2001-12-10 20:34 ` Alan Cox
2001-12-10 21:31 Andries.Brouwer
2001-12-10 21:44 ` Alan Cox
2001-12-10 22:48 Andries.Brouwer
2001-12-10 23:33 Andries.Brouwer
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=3C0D05C7.6C566FF2@mandrakesoft.com \
--to=jgarzik@mandrakesoft.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=phillips@bonn-fries.net \
/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