From: "Theodore Ts'o" <tytso@mit.edu>
To: Fengguang Wu <fengguang.wu@gmail.com>
Cc: linux RAID <linux-raid@vger.kernel.org>,
NeilBrown <neilb@suse.de>, Li Shaohua <shli@fusionio.com>,
Marti Raudsepp <marti@juffo.org>,
Kernel hackers <linux-kernel@vger.kernel.org>,
ext4 hackers <linux-ext4@vger.kernel.org>,
maze@google.com, "Shi, Alex" <alex.shi@intel.com>,
linux-fsdevel@vger.kernel.org
Subject: Re: ext4 write performance regression in 3.6-rc1 on RAID0/5
Date: Fri, 17 Aug 2012 11:37:35 -0400 [thread overview]
Message-ID: <20120817153735.GB31297@thunk.org> (raw)
In-Reply-To: <20120817151318.GA2341@localhost>
On Fri, Aug 17, 2012 at 11:13:18PM +0800, Fengguang Wu wrote:
>
> Obviously the major regressions happen to the 100dd over raid cases.
> Some 10dd cases are also impacted.
>
> The attached graphs show that everything becomes more fluctuated in
> 3.6.0-rc1 for the lkp-nex04/RAID0-12HDD-thresh=8G/ext4-100dd-1 case.
Hmm... I'm not seeing any differences in the block allocation code, or
in ext4's buffered writeback code paths, which would be the most
likely cause of such problems. Maybe a quick eyeball of the blktrace
to see if we're doing something pathalogically stupid?
You could also try running a filefrag -v on a few of the dd files to
see if there's any significant difference, although as I said, there
doesn't look like there was any significant changes in the block
allocation code between v3.5 and v3.6-rc1 --- although I suppose
changes in timeing could have have caused the block allocation
decisions to be different, so it's worth checking that out.
Thanks, regards,
- Ted
next prev parent reply other threads:[~2012-08-17 15:38 UTC|newest]
Thread overview: 31+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-08-15 18:33 NULL pointer dereference in ext4_ext_remove_space on 3.5.1 Marti Raudsepp
2012-08-16 2:46 ` Theodore Ts'o
2012-08-16 11:10 ` Fengguang Wu
2012-08-16 15:25 ` Theodore Ts'o
2012-08-16 20:21 ` Maciej Żenczykowski
2012-08-16 21:19 ` Theodore Ts'o
2012-08-16 21:40 ` Maciej Żenczykowski
2012-08-16 22:26 ` Theodore Ts'o
2012-08-16 22:44 ` Maciej Żenczykowski
2012-08-17 6:01 ` Fengguang Wu
2012-08-17 13:15 ` Theodore Ts'o
2012-08-17 13:22 ` Fengguang Wu
2012-08-17 17:48 ` Christoph Hellwig
2012-08-17 6:09 ` ext4 write performance regression in 3.6-rc1 Fengguang Wu
2012-08-17 13:40 ` Theodore Ts'o
2012-08-17 14:13 ` Fengguang Wu
2012-08-17 14:25 ` ext4 write performance regression in 3.6-rc1 on RAID0/5 Fengguang Wu
[not found] ` <20120817151318.GA2341@localhost>
2012-08-17 15:37 ` Theodore Ts'o [this message]
2012-08-17 20:44 ` NeilBrown
2012-08-21 9:42 ` Fengguang Wu
2012-08-21 12:07 ` Fengguang Wu
[not found] ` <20120822035702.GF2570@yliu-dev.sh.intel.com>
2012-08-22 4:07 ` Shaohua Li
2012-08-22 6:00 ` NeilBrown
2012-08-22 6:31 ` Yuanhan Liu
2012-08-22 7:14 ` Andreas Dilger
2012-08-22 20:47 ` Dan Williams
2012-08-22 21:59 ` NeilBrown
2012-09-17 12:21 ` NULL pointer dereference in ext4_ext_remove_space on 3.5.1 Dmitry Monakhov
2012-09-17 13:52 ` Theodore Ts'o
2012-09-17 14:48 ` Dmitry Monakhov
2012-08-16 9:00 ` Fengguang Wu
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=20120817153735.GB31297@thunk.org \
--to=tytso@mit.edu \
--cc=alex.shi@intel.com \
--cc=fengguang.wu@gmail.com \
--cc=linux-ext4@vger.kernel.org \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-raid@vger.kernel.org \
--cc=marti@juffo.org \
--cc=maze@google.com \
--cc=neilb@suse.de \
--cc=shli@fusionio.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).