From: Stefan Hajnoczi <stefanha@gmail.com>
To: Christoph Hellwig <hch@infradead.org>
Cc: Antoine Martin <antoine@nagafix.co.uk>,
Avi Kivity <avi@redhat.com>, Michael Tokarev <mjt@tls.msk.ru>,
kvm@vger.kernel.org
Subject: Re: raw disks no longer work in latest kvm (kvm-88 was fine)
Date: Sat, 29 May 2010 12:06:27 +0100 [thread overview]
Message-ID: <AANLkTin_Nom73bYdqrywjFKFcesikjpQRuVHNA0iVxSi@mail.gmail.com> (raw)
In-Reply-To: <20100529103448.GB16311@infradead.org>
On Sat, May 29, 2010 at 11:34 AM, Christoph Hellwig <hch@infradead.org> wrote:
> In what benchmark do you see worse results for aio=native compared to
> aio=threads?
Sequential reads using 4 concurrent dd if=/dev/vdb iflag=direct
of=/dev/null bs=8k processes. 2 vcpu guest with 4 GB RAM, virtio
block devices, cache=none. Host storage is a striped LVM volume.
Host kernel kvm.git and qemu-kvm.git userspace.
aio=native and aio=threads each run 3 times.
Result: aio=native has 15% lower throughput than aio=threads.
I haven't looked into this so I don't know what is causes these results.
Stefan
next prev parent reply other threads:[~2010-05-29 11:06 UTC|newest]
Thread overview: 51+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-03-06 20:48 raw disks no longer work in latest kvm (kvm-88 was fine) Antoine Martin
2010-03-06 21:28 ` Michael Tokarev
2010-03-07 4:14 ` Antoine Martin
2010-03-07 9:32 ` Michael Tokarev
2010-03-07 10:00 ` Christoph Hellwig
2010-03-07 13:12 ` Antoine Martin
2010-03-07 13:52 ` Antoine Martin
2010-03-07 15:15 ` Michael Tokarev
2010-03-07 17:11 ` Antoine Martin
2010-03-07 17:18 ` Avi Kivity
2010-03-07 17:21 ` Christoph Hellwig
2010-03-07 17:30 ` Avi Kivity
2010-03-07 17:34 ` Christoph Hellwig
2010-03-07 18:43 ` raw disks no longer work in latest kvm (kvm-88 was fine) [SOLVED] Antoine Martin
2010-03-07 18:55 ` Avi Kivity
2010-03-07 18:01 ` raw disks no longer work in latest kvm (kvm-88 was fine) Antoine Martin
2010-03-07 18:47 ` Avi Kivity
2010-03-07 19:07 ` Antoine Martin
2010-03-07 19:10 ` Avi Kivity
2010-03-07 19:13 ` Antoine Martin
2010-03-07 19:17 ` Avi Kivity
2010-03-07 19:25 ` Antoine Martin
2010-03-07 19:35 ` Avi Kivity
2010-03-13 9:51 ` Antoine Martin
2010-03-14 7:34 ` Avi Kivity
2010-04-08 22:00 ` Antoine Martin
2010-05-22 10:44 ` Antoine Martin
2010-05-22 11:17 ` Michael Tokarev
2010-05-22 11:35 ` Antoine Martin
2010-05-23 8:53 ` Antoine Martin
2010-05-23 11:57 ` Avi Kivity
2010-05-23 14:07 ` Antoine Martin
2010-05-23 14:18 ` Avi Kivity
2010-05-23 14:43 ` Antoine Martin
2010-05-23 14:53 ` Antoine Martin
2010-05-23 14:56 ` Avi Kivity
2010-05-23 15:06 ` Antoine Martin
2010-05-23 15:12 ` Avi Kivity
2010-05-23 16:18 ` Antoine Martin
2010-05-23 17:47 ` Stefan Hajnoczi
2010-05-29 9:42 ` Antoine Martin
2010-05-29 9:55 ` Stefan Hajnoczi
2010-05-29 10:34 ` Christoph Hellwig
2010-05-29 11:06 ` Stefan Hajnoczi [this message]
2010-05-29 10:34 ` Christoph Hellwig
2010-03-07 19:09 ` Asdo
2010-03-07 19:11 ` Antoine Martin
2010-03-07 19:12 ` Avi Kivity
2010-03-07 16:21 ` Avi Kivity
2010-03-08 16:45 ` Anthony Liguori
2010-03-07 9:36 ` Gleb Natapov
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=AANLkTin_Nom73bYdqrywjFKFcesikjpQRuVHNA0iVxSi@mail.gmail.com \
--to=stefanha@gmail.com \
--cc=antoine@nagafix.co.uk \
--cc=avi@redhat.com \
--cc=hch@infradead.org \
--cc=kvm@vger.kernel.org \
--cc=mjt@tls.msk.ru \
/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).