From: "Jörn Engel" <joern@logfs.org>
To: Josh Boyer <jwboyer@gmail.com>
Cc: "David Woodhouse" <dwmw2@infradead.org>,
"Jörn Engel" <joern@logfs.org>,
linux-mtd@lists.infradead.org,
"Matthieu CASTET" <matthieu.castet@parrot.com>
Subject: Re: Jffs2 and big file = very slow jffs2_garbage_collect_pass
Date: Fri, 18 Jan 2008 19:17:44 +0100 [thread overview]
Message-ID: <20080118181744.GA15039@lazybastard.org> (raw)
In-Reply-To: <20080118115531.2deef9fb@zod.rchland.ibm.com>
On Fri, 18 January 2008 11:55:31 -0600, Josh Boyer wrote:
>
> That seems consistent with JFFS2 doing the CRC checks and constructing
> the in-memory representation of your large file. I suspect the older
> list-based in-memory implementation would have taken even longer, but
> there could be something amiss with the rb-tree stuff perhaps.
There is something conceptually amiss with rb-trees. Each node
effectively occupies its own cacheline. With those 40k+ nodes, you
would need a rather sizeable cache with at least 20k cachelines to have
an impact. Noone does. So for all practical purposes, every single
lookup will go to main memory.
Maybe it is about time to suggest trying logfs?
Jörn
--
People will accept your ideas much more readily if you tell them
that Benjamin Franklin said it first.
-- unknown
next prev parent reply other threads:[~2008-01-18 18:26 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-17 16:12 Jffs2 and big file = very slow jffs2_garbage_collect_pass Matthieu CASTET
2008-01-17 16:26 ` Jörn Engel
2008-01-17 17:43 ` Josh Boyer
2008-01-18 9:39 ` Matthieu CASTET
2008-01-18 12:48 ` Josh Boyer
2008-01-18 16:17 ` Matthieu CASTET
2008-01-18 17:55 ` Josh Boyer
2008-01-18 18:17 ` Jörn Engel [this message]
2008-01-21 15:57 ` Matthieu CASTET
2008-01-21 21:25 ` Jörn Engel
2008-01-21 22:16 ` Josh Boyer
2008-01-21 22:29 ` Jörn Engel
2008-01-22 8:57 ` Matthieu CASTET
2008-01-22 12:03 ` Jörn Engel
2008-01-22 13:24 ` Ricard Wanderlof
2008-01-22 15:05 ` Jörn Engel
2008-01-23 9:23 ` Ricard Wanderlof
2008-01-23 10:19 ` Jörn Engel
2008-01-23 10:41 ` Ricard Wanderlof
2008-01-23 10:57 ` Jörn Engel
2008-01-23 11:57 ` Ricard Wanderlof
2008-01-23 13:01 ` Jörn Engel
2008-01-23 13:16 ` Ricard Wanderlof
2008-01-23 14:06 ` Jörn Engel
2008-01-23 14:25 ` Ricard Wanderlof
2008-01-21 22:36 ` Glenn Henshaw
2008-01-18 17:20 ` Glenn Henshaw
2008-01-18 18:39 ` Jamie Lokier
2008-01-18 21:00 ` Jörn Engel
2008-01-19 0:23 ` Jamie Lokier
2008-01-19 2:38 ` Jörn Engel
2008-01-17 23:22 ` David Woodhouse
2008-01-18 9:45 ` Matthieu CASTET
2008-01-18 18:20 ` Jamie Lokier
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=20080118181744.GA15039@lazybastard.org \
--to=joern@logfs.org \
--cc=dwmw2@infradead.org \
--cc=jwboyer@gmail.com \
--cc=linux-mtd@lists.infradead.org \
--cc=matthieu.castet@parrot.com \
/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