From: Nick Piggin <npiggin@suse.de>
To: David Chinner <dgc@sgi.com>
Cc: Chris Mason <chris.mason@oracle.com>,
Nick Piggin <nickpiggin@yahoo.com.au>,
Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
Linux Memory Management List <linux-mm@kvack.org>,
linux-fsdevel@vger.kernel.org
Subject: Re: [RFC] fsblock
Date: Thu, 28 Jun 2007 04:44:43 +0200 [thread overview]
Message-ID: <20070628024443.GB6038@wotan.suse.de> (raw)
In-Reply-To: <20070627223548.GS989688@sgi.com>
On Thu, Jun 28, 2007 at 08:35:48AM +1000, David Chinner wrote:
> On Wed, Jun 27, 2007 at 07:50:56AM -0400, Chris Mason wrote:
> > Lets look at a typical example of how IO actually gets done today,
> > starting with sys_write():
> >
> > sys_write(file, buffer, 1MB)
> > for each page:
> > prepare_write()
> > allocate contiguous chunks of disk
> > attach buffers
> > copy_from_user()
> > commit_write()
> > dirty buffers
> >
> > pdflush:
> > writepages()
> > find pages with contiguous chunks of disk
> > build and submit large bios
> >
> > So, we replace prepare_write and commit_write with an extent based api,
> > but we keep the dirty each buffer part. writepages has to turn that
> > back into extents (bio sized), and the result is completely full of dark
> > dark corner cases.
That's true but I don't think an extent data structure means we can
become too far divorced from the pagecache or the native block size
-- what will end up happening is that often we'll need "stuff" to map
between all those as well, even if it is only at IO-time.
But the point is taken, and I do believe that at least for APIs, extent
based seems like the best way to go. And that should allow fsblock to
be replaced or augmented in future without _too_ much pain.
> Yup - I've been on the painful end of those dark corner cases several
> times in the last few months.
>
> It's also worth pointing out that mpage_readpages() already works on
> an extent basis - it overloads bufferheads to provide a "map_bh" that
> can point to a range of blocks in the same state. The code then iterates
> the map_bh range a page at a time building bios (i.e. not even using
> buffer heads) from that map......
One issue I have with the current nobh and mpage stuff is that it
requires multiple calls into get_block (first to prepare write, then
to writepage), it doesn't allow filesystems to attach resources
required for writeout at prepare_write time, and it doesn't play nicely
with buffers in general. (not to mention that nobh error handling is
buggy).
I haven't done any mpage-like code for fsblocks yet, but I think they
wouldn't be too much trouble, and wouldn't have any of the above
problems...
next prev parent reply other threads:[~2007-06-28 2:45 UTC|newest]
Thread overview: 53+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-06-24 1:45 [RFC] fsblock Nick Piggin
2007-06-24 1:46 ` [patch 1/3] add the fsblock layer Nick Piggin
2007-06-24 15:28 ` Andi Kleen
2007-06-24 20:18 ` Arjan van de Ven
2007-06-25 8:58 ` Andi Kleen
2007-06-25 7:19 ` Nick Piggin
2007-06-24 23:01 ` Neil Brown
2007-06-25 7:41 ` Nick Piggin
2007-06-25 12:29 ` Chris Mason
2007-06-26 2:34 ` Nick Piggin
2007-06-26 2:48 ` Neil Brown
2007-06-26 3:07 ` Nick Piggin
2007-06-26 12:26 ` Chris Mason
2007-06-30 10:40 ` Christoph Hellwig
2007-06-30 10:40 ` Christoph Hellwig
2007-06-25 13:19 ` Chris Mason
2007-06-26 2:42 ` Nick Piggin
2007-06-24 1:46 ` [patch 2/3] block_dev: convert to fsblock Nick Piggin
2007-06-24 1:47 ` [patch 3/3] minix: " Nick Piggin
2007-06-24 1:53 ` [RFC] fsblock Nick Piggin
2007-06-24 3:07 ` Jeff Garzik
2007-06-24 3:47 ` Nick Piggin
2007-06-24 13:51 ` Chris Mason
2007-06-25 6:58 ` Nick Piggin
2007-06-25 12:25 ` Chris Mason
2007-06-30 10:44 ` Christoph Hellwig
2007-06-30 10:42 ` Christoph Hellwig
2007-06-30 11:10 ` Jeff Garzik
2007-06-30 11:13 ` Christoph Hellwig
2007-06-24 4:19 ` William Lee Irwin III
2007-06-24 14:16 ` Andi Kleen
2007-06-25 7:16 ` Nick Piggin
2007-06-26 3:06 ` David Chinner
2007-06-26 3:55 ` Nick Piggin
2007-06-26 9:23 ` David Chinner
2007-06-26 11:14 ` Nick Piggin
2007-06-27 12:39 ` Kyle Moffett
2007-06-26 12:34 ` Chris Mason
2007-06-27 5:32 ` Nick Piggin
2007-06-27 6:05 ` David Chinner
2007-06-27 11:50 ` Chris Mason
2007-06-27 15:18 ` Anton Altaparmakov
2007-06-27 22:35 ` David Chinner
2007-06-28 2:44 ` Nick Piggin [this message]
2007-06-28 12:20 ` Chris Mason
2007-06-29 2:08 ` David Chinner
2007-06-29 2:33 ` Nick Piggin
2007-06-30 11:05 ` Christoph Hellwig
2007-07-09 17:14 ` Christoph Lameter
2007-07-10 0:54 ` Nick Piggin
2007-07-10 0:59 ` Christoph Lameter
2007-07-10 1:07 ` Nick Piggin
2007-07-10 1:37 ` Dave McCracken
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=20070628024443.GB6038@wotan.suse.de \
--to=npiggin@suse.de \
--cc=chris.mason@oracle.com \
--cc=dgc@sgi.com \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=nickpiggin@yahoo.com.au \
/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).