From: Christoph Hellwig <hch@lst.de>
To: Michael Tokarev <mjt@tls.msk.ru>
Cc: Avi Kivity <avi@redhat.com>, Christoph Hellwig <hch@lst.de>,
Matthew Tippett <tippettm@gmail.com>,
Dustin Kirkland <dustin.kirkland@gmail.com>,
Anthony Liguori <anthony@codemonkey.ws>, RW <kvm@tauceti.net>,
kvm@vger.kernel.org
Subject: Re: [PATCH] virtio-blk: fallback to draining the queue if barrier ops are not supported
Date: Wed, 14 Oct 2009 17:46:17 +0200 [thread overview]
Message-ID: <20091014154617.GA19549@lst.de> (raw)
In-Reply-To: <4AD5F085.8040909@msgid.tls.msk.ru>
On Wed, Oct 14, 2009 at 07:38:45PM +0400, Michael Tokarev wrote:
> Avi Kivity wrote:
> >Early implementations of virtio devices did not support barrier operations,
> >but did commit the data to disk. In such cases, drain the queue to emulate
> >barrier operations.
>
> Are there any implementation currently that actually supports barriers?
> As far as I remember there's no way to invoke barriers from a user-space
> application on linux, and this is how kvm/qemu is running on this OS.
Ignore all the barrier talk. The way Linux uses the various storage
transport the primitives are queue draining (done entirely in the guest
block layer) and cache flushes. Fdatasync is exactly the same primitive
as a WIN FLUSH CACHE in ATA or SYNCHRONIZE cache in SCSI module the lack
or ranges in fdatasync - but that is just a performance optimization and
not actually used by Linux guests for now.
next prev parent reply other threads:[~2009-10-14 15:47 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-09-25 7:00 sync guest calls made async on host - SQLite performance RW
2009-09-27 11:37 ` Avi Kivity
2009-09-27 12:07 ` Matthew Tippett
2009-09-29 19:02 ` Anthony Liguori
2009-09-29 19:32 ` Matthew Tippett
2009-09-29 20:46 ` Dustin Kirkland
2009-09-29 20:51 ` Anthony Liguori
2009-10-07 14:15 ` Matthew Tippett
2009-10-07 16:53 ` Matthew Tippett
2009-10-07 18:59 ` Dustin Kirkland
2009-10-07 19:31 ` Matthew Tippett
2009-10-07 20:12 ` Avi Kivity
2009-10-07 21:01 ` Matthew Tippett
2009-10-09 6:05 ` Dustin Kirkland
[not found] ` <4ACE0196.9010904@gmail.com>
[not found] ` <d9c105ea0910082312n7218e1abhc69a2be660838e89@mail.gmail.com>
[not found] ` <f28b49ce0910090425p385636c3he6dfb7595927d7e4@mail.gmail.com>
2009-10-09 15:18 ` Dustin Kirkland
2009-10-09 19:06 ` Matthew Tippett
2009-10-11 9:16 ` Avi Kivity
2009-10-13 22:37 ` Christoph Hellwig
2009-10-14 11:03 ` Avi Kivity
2009-10-14 12:03 ` [PATCH] virtio-blk: fallback to draining the queue if barrier ops are not supported Avi Kivity
2009-10-14 14:46 ` Javier Guerra
2009-10-14 16:53 ` Avi Kivity
2009-10-14 15:38 ` Michael Tokarev
2009-10-14 15:46 ` Christoph Hellwig [this message]
2009-10-14 13:41 ` sync guest calls made async on host - SQLite performance Christoph Hellwig
2009-10-14 16:56 ` Avi Kivity
2009-10-14 17:02 ` Christoph Hellwig
2009-10-14 22:54 ` Anthony Liguori
2009-10-14 23:37 ` Avi Kivity
2009-10-15 12:17 ` Christoph Hellwig
2009-10-15 12:36 ` Christoph Hellwig
2009-10-19 5:56 ` Avi Kivity
2009-10-13 23:08 ` Anthony Liguori
2009-10-14 2:09 ` Matthew Tippett
2009-10-14 4:12 ` Dustin Kirkland
2009-10-14 11:32 ` Matthew Tippett
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=20091014154617.GA19549@lst.de \
--to=hch@lst.de \
--cc=anthony@codemonkey.ws \
--cc=avi@redhat.com \
--cc=dustin.kirkland@gmail.com \
--cc=kvm@tauceti.net \
--cc=kvm@vger.kernel.org \
--cc=mjt@tls.msk.ru \
--cc=tippettm@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).