From: Paolo Bonzini <pbonzini@redhat.com>
To: Ming Lei <tom.leiming@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>, Fam Zheng <famz@redhat.com>,
qemu-devel@nongnu.org, Stefan Hajnoczi <stefanha@redhat.com>,
"Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [Qemu-devel] [regression] dataplane: throughout -40% by commit 580b6b2aa2
Date: Thu, 26 Jun 2014 17:57:14 +0200 [thread overview]
Message-ID: <53AC42DA.2000703@redhat.com> (raw)
In-Reply-To: <CACVXFVOWcnJexn=64MRgqMfhtwP0vLKh45=KNQ+RBYeScp3qeA@mail.gmail.com>
This is indeed a difference between the ioq-based and block-based
backends. ioq could submit more than one request with the same
io_submit system call.
We can implement (advisory) calls like bdrv_plug/bdrv_unplug in order to
restore the previous levels of performance.
Note that some fallout of the conversion was expected. Dataplane told
us experimentally what level of performance could be reached, but was a
dead end in terms of functionality. Now Stefan added a whole lot of
functionality to dataplane (accounting, throttling, file formats and
protocols, thread-pool based I/O, etc.) and we need to bring back any
performance we lost in the process.
Out of curiosity, how does aio=threads (after the patch) fare in
comparison to aio=native (before the patch)?
Paolo
next prev parent reply other threads:[~2014-06-26 16:15 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-26 15:14 [Qemu-devel] [regression] dataplane: throughout -40% by commit 580b6b2aa2 Ming Lei
2014-06-26 15:29 ` Paolo Bonzini
2014-06-26 15:37 ` Ming Lei
2014-06-26 15:43 ` Paolo Bonzini
2014-06-26 15:47 ` Ming Lei
2014-06-26 15:57 ` Paolo Bonzini [this message]
2014-06-27 1:15 ` Ming Lei
2014-06-27 4:59 ` Paolo Bonzini
2014-06-27 6:23 ` Kevin Wolf
2014-06-27 7:35 ` Paolo Bonzini
2014-06-27 12:35 ` Ming Lei
2014-06-27 7:57 ` Ming Lei
2014-06-27 12:01 ` Stefan Hajnoczi
2014-06-27 12:21 ` Kevin Wolf
2014-06-27 14:50 ` Stefan Hajnoczi
2014-06-27 18:01 ` Ming Lei
2014-06-27 21:51 ` Paolo Bonzini
2014-06-28 9:58 ` Ming Lei
2014-06-30 8:08 ` Stefan Hajnoczi
2014-06-30 8:27 ` Ming Lei
2014-07-01 13:53 ` Ming Lei
2014-07-01 14:31 ` Stefan Hajnoczi
2014-07-01 14:49 ` Ming Lei
2014-07-01 16:49 ` Paolo Bonzini
2014-07-02 0:48 ` Ming Lei
2014-07-02 8:54 ` Stefan Hajnoczi
2014-07-02 9:13 ` Paolo Bonzini
2014-07-02 9:39 ` Kevin Wolf
2014-07-02 9:48 ` Paolo Bonzini
2014-07-02 10:01 ` Kevin Wolf
2014-07-02 10:23 ` Paolo Bonzini
2014-07-02 15:45 ` Ming Lei
2014-07-02 16:13 ` Ming Lei
2014-07-02 16:23 ` Paolo Bonzini
2014-07-02 16:27 ` Ming Lei
2014-07-02 16:38 ` Paolo Bonzini
2014-07-02 16:41 ` Ming Lei
2014-07-02 16:21 ` Paolo Bonzini
2014-07-03 4:54 ` Ming Lei
2014-07-03 10:29 ` Paolo Bonzini
2014-07-03 11:50 ` Ming Lei
2014-07-03 11:56 ` Paolo Bonzini
2014-07-03 12:09 ` Ming Lei
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=53AC42DA.2000703@redhat.com \
--to=pbonzini@redhat.com \
--cc=famz@redhat.com \
--cc=kwolf@redhat.com \
--cc=mst@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@redhat.com \
--cc=tom.leiming@gmail.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).