linux-fsdevel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Chris Mason <chris.mason@fusionio.com>
To: Chris Mason <chris.mason@fusionio.com>,
	Eric Sandeen <sandeen@redhat.com>,
	Dmitry Monakhov <dmonakhov@openvz.org>
Cc: ext4 development <linux-ext4@vger.kernel.org>,
	"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
	"axboe@kernel.dk" <axboe@kernel.dk>, Jan Kara <jack@suse.cz>
Subject: Re: EXT4 nodelalloc => back to stone age.
Date: Mon, 1 Apr 2013 11:57:05 -0400	[thread overview]
Message-ID: <20130401155705.16754.31446@localhost.localdomain> (raw)
In-Reply-To: <20130401154541.16754.22860@localhost.localdomain>

Quoting Chris Mason (2013-04-01 11:45:41)
> Quoting Eric Sandeen (2013-04-01 11:18:51)
> > On 4/1/13 6:06 AM, Dmitry Monakhov wrote:
> > 
> > > 1)Do we really have to use WRITE_SYNC in case of WB_SYNC_ALL ?
> 
> Yes?  The stuff we wait on should be WRITE_SYNC.
> 
> > 
> > ...
> > 
> > > 2) Why don't we have writepages for non delalloc case ?
> > 
> > ...
> > 
> > I'd add:
> > 
> > 3) Why do we have a "nodelalloc" mount option at all?
> > 
> > but then I thought:
> > 
> > Is it also this bad when using the ext4 driver to run an ext3 fs?
> 
> Quick comparison on a single iodrive:

On the theory that writepages is the problem try echo 1 >
/sys/block/xxx/queue/rotational.  With request merging on here in
nodelalloc mode:

dd if=/dev/zero of=foo bs=1M count=1024 conv=fsync,notrunc
1073741824 bytes (1.1 GB) copied, 2.53741 s, 423 MB/s

dd if=/dev/zero of=foo bs=1M count=1024 conv=fsync,notrunc
1073741824 bytes (1.1 GB) copied, 1.37795 s, 779 MB/s

-chris

  reply	other threads:[~2013-04-01 15:57 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-04-01 11:06 EXT4 nodelalloc => back to stone age Dmitry Monakhov
2013-04-01 15:18 ` Eric Sandeen
2013-04-01 15:39   ` Theodore Ts'o
2013-04-01 16:00     ` Eric Sandeen
2013-04-01 16:34       ` Zheng Liu
2013-04-01 15:45   ` Chris Mason
2013-04-01 15:57     ` Chris Mason [this message]
2013-04-02 13:46 ` Jan Kara

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=20130401155705.16754.31446@localhost.localdomain \
    --to=chris.mason@fusionio.com \
    --cc=axboe@kernel.dk \
    --cc=dmonakhov@openvz.org \
    --cc=jack@suse.cz \
    --cc=linux-ext4@vger.kernel.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=sandeen@redhat.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;
as well as URLs for NNTP newsgroup(s).