From: Jeff Moyer <jmoyer@redhat.com>
To: Jan Kara <jack@suse.cz>
Cc: Corrado Zoccolo <czoccolo@gmail.com>,
"Alex\,Shi" <alex.shi@intel.com>, "Li\,
Shaohua" <shaohua.li@intel.com>, Vivek Goyal <vgoyal@redhat.com>,
"tytso\@mit.edu" <tytso@mit.edu>,
"jaxboe\@fusionio.com" <jaxboe@fusionio.com>,
"linux-kernel\@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"Chen\, Tim C" <tim.c.chen@intel.com>
Subject: Re: [performance bug] kernel building regression on 64 LCPUs machine
Date: Fri, 04 Mar 2011 10:50:32 -0500 [thread overview]
Message-ID: <x49d3m6j35j.fsf@segfault.boston.devel.redhat.com> (raw)
In-Reply-To: <20110304153248.GC2649@quack.suse.cz> (Jan Kara's message of "Fri, 4 Mar 2011 16:32:48 +0100")
Jan Kara <jack@suse.cz> writes:
> I'm not so happy with ext4 results. The difference between ext3 and ext4
> might be that amount of data written by kjournald in ext3 is considerably
> larger if it ends up pushing out data (because of data=ordered mode) as
> well. With ext4, all data are written by filemap_fdatawrite() from fsync
> because of delayed allocation. And thus maybe for ext4 WRITE_SYNC_PLUG
> is hurting us with your fast storage and small amount of written data? With
> WRITE_SYNC, data would be already on it's way to storage before we get to
> wait for them...
> Or it could be that we really send more data in WRITE mode rather than in
> WRITE_SYNC mode with the patch on ext4 (that should be verifiable with
> blktrace). But I wonder how that could happen...
It looks like this is the case, the I/O isn't coming down as
synchronous. I'm seeing a lot of writes, very few write sync's, which
means that the write stream will be preempted by the incoming reads.
Time to audit that fsync path and make sure it's marked properly, I
guess.
Cheers,
Jeff
next prev parent reply other threads:[~2011-03-04 15:50 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-19 1:55 [performance bug] kernel building regression on 64 LCPUs machine Alex,Shi
2011-01-19 2:03 ` Shaohua Li
2011-01-19 12:56 ` Jan Kara
2011-01-20 7:52 ` Alex,Shi
2011-01-20 15:16 ` Vivek Goyal
2011-01-21 7:17 ` Shaohua Li
2011-01-26 8:15 ` Shaohua Li
2011-02-12 9:21 ` Alex,Shi
2011-02-12 18:25 ` Corrado Zoccolo
2011-02-14 2:25 ` Alex,Shi
2011-02-15 1:10 ` Shaohua Li
2011-02-21 16:49 ` Jan Kara
2011-02-23 8:24 ` Alex,Shi
2011-02-24 12:13 ` Jan Kara
2011-02-25 0:44 ` Alex Shi
2011-02-26 14:45 ` Corrado Zoccolo
2011-03-01 19:56 ` Jeff Moyer
2011-03-02 9:42 ` Jan Kara
2011-03-02 16:13 ` Jeff Moyer
2011-03-02 21:17 ` Jan Kara
2011-03-02 21:20 ` Jeff Moyer
2011-03-03 1:14 ` Jeff Moyer
2011-03-04 15:32 ` Jan Kara
2011-03-04 15:40 ` Jeff Moyer
2011-03-04 15:50 ` Jeff Moyer [this message]
2011-03-04 18:27 ` Jeff Moyer
2011-03-22 7:38 ` Alex,Shi
2011-03-22 16:14 ` Jan Kara
2011-03-22 17:46 ` Jeff Moyer
2011-03-24 6:45 ` Alex,Shi
2011-03-28 19:48 ` Jan Kara
2011-01-19 14:32 ` Ted Ts'o
2011-01-20 2:12 ` Shaohua Li
2011-01-21 7:23 ` Corrado Zoccolo
2011-01-21 7:47 ` Alex,Shi
2011-01-21 7:52 ` Alex,Shi
2011-01-21 8:13 ` Corrado Zoccolo
2011-01-21 8:20 ` Shaohua Li
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=x49d3m6j35j.fsf@segfault.boston.devel.redhat.com \
--to=jmoyer@redhat.com \
--cc=alex.shi@intel.com \
--cc=czoccolo@gmail.com \
--cc=jack@suse.cz \
--cc=jaxboe@fusionio.com \
--cc=linux-kernel@vger.kernel.org \
--cc=shaohua.li@intel.com \
--cc=tim.c.chen@intel.com \
--cc=tytso@mit.edu \
--cc=vgoyal@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).