From: Gleb Natapov <gleb@redhat.com>
To: Avi Kivity <avi@redhat.com>
Cc: "Richard W.M. Jones" <rjones@redhat.com>,
Alexander Graf <agraf@suse.de>,
qemu-devel <qemu-devel@nongnu.org>
Subject: Re: [Qemu-devel] Slow kernel/initrd loading via fw_cfg; Was Re: Hack integrating SeaBios / LinuxBoot option rom with QEMU trace backends
Date: Tue, 11 Oct 2011 11:56:36 +0200 [thread overview]
Message-ID: <20111011095636.GV30105@redhat.com> (raw)
In-Reply-To: <4E94111C.403@redhat.com>
On Tue, Oct 11, 2011 at 11:49:16AM +0200, Avi Kivity wrote:
> >Whatever we do, the interface will never be as fast as DMA. We will always have to do sanity / permission checks for every IO operation, can batch up only so many IO requests and in QEMU again have to call our callbacks in a loop.
>
> We can batch per page, which makes the overhead negligible.
>
Current code batch userspace exit per 1024 bytes IIRC and changing it to
page didn't show significant improvement (also IIRC). But after io data
is copied into the kernel emulator process it byte by byte. Possible
optimization, which I didn't tried, is to check that destination memory is
not mmio and write back the whole buffer if it is the case.
--
Gleb.
next prev parent reply other threads:[~2011-10-11 9:56 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-10-10 17:08 [Qemu-devel] Hack integrating SeaBios / LinuxBoot option rom with QEMU trace backends Daniel P. Berrange
2011-10-10 18:53 ` Anthony Liguori
2011-10-10 19:01 ` Alexander Graf
2011-10-11 8:23 ` [Qemu-devel] Slow kernel/initrd loading via fw_cfg; Was " Daniel P. Berrange
2011-10-11 8:43 ` Richard W.M. Jones
2011-10-11 9:15 ` Avi Kivity
2011-10-11 9:19 ` Alexander Graf
2011-10-11 9:26 ` Avi Kivity
2011-10-11 9:38 ` Alexander Graf
2011-10-11 9:49 ` Avi Kivity
2011-10-11 9:56 ` Gleb Natapov [this message]
2011-10-11 9:59 ` Avi Kivity
2011-10-11 10:28 ` Gleb Natapov
2011-10-11 13:12 ` Anthony Liguori
2011-10-11 13:14 ` Alexander Graf
2011-10-11 13:17 ` Gleb Natapov
2011-10-11 13:19 ` Anthony Liguori
2011-10-11 13:23 ` Avi Kivity
2011-10-11 13:24 ` Gleb Natapov
2011-10-11 13:29 ` Anthony Liguori
2011-10-11 13:45 ` Avi Kivity
2011-10-11 13:58 ` Anthony Liguori
2011-10-11 14:01 ` Daniel P. Berrange
2011-10-11 14:33 ` Anthony Liguori
2011-10-11 14:34 ` Alexander Graf
2011-10-11 14:36 ` Daniel P. Berrange
2011-10-11 9:50 ` Gleb Natapov
2011-10-11 9:55 ` Avi Kivity
2011-10-11 13:17 ` Anthony Liguori
2011-10-11 13:22 ` Gleb Natapov
2011-10-15 14:19 ` Kevin O'Connor
2011-10-11 9:27 ` Daniel P. Berrange
2011-10-11 9:39 ` Avi Kivity
2011-10-11 9:49 ` Daniel P. Berrange
2011-10-11 9:50 ` Avi Kivity
2011-10-11 10:09 ` Daniel P. Berrange
2011-10-15 10:00 ` Blue Swirl
2011-10-15 16:16 ` Richard W.M. Jones
2011-10-16 17:20 ` Lluís Vilanova
2011-10-11 9:08 ` [Qemu-devel] " Avi Kivity
2011-10-11 9:18 ` Daniel P. Berrange
2011-10-11 9:35 ` Avi Kivity
2011-10-10 23:57 ` Kevin O'Connor
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=20111011095636.GV30105@redhat.com \
--to=gleb@redhat.com \
--cc=agraf@suse.de \
--cc=avi@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=rjones@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).