All of lore.kernel.org
 help / color / mirror / Atom feed
From: Juan Quintela <quintela@redhat.com>
To: Kevin Wolf <kwolf@redhat.com>
Cc: stefanha@gmail.com, hch@lst.de, mjt@tls.msk.ru,
	qemu-devel@nongnu.org, avi@redhat.com
Subject: [Qemu-devel] Re: [RFC][STABLE 0.13] Revert "qcow2: Use bdrv_(p)write_sync for metadata writes"
Date: Tue, 24 Aug 2010 14:48:42 +0200	[thread overview]
Message-ID: <m3k4ngmb11.fsf@trasno.org> (raw)
In-Reply-To: <1282646430-5777-1-git-send-email-kwolf@redhat.com> (Kevin Wolf's message of "Tue, 24 Aug 2010 12:40:30 +0200")

Kevin Wolf <kwolf@redhat.com> wrote:
> This reverts commit 8b3b720620a1137a1b794fc3ed64734236f94e06.
>
> This fix has caused severe slowdowns on recent kernels that actually do flush
> when they are told so. Reverting this patch hurts correctness and means that we
> could get corrupted images in case of a host crash. This means that qcow2 might
> not be an option for some people without this fix. On the other hand, I get
> reports that the slowdown is so massive that not reverting it would mean that
> people can't use it either because it just takes ages to complete stuff. It
> probably can be fixed, but not in time for 0.13.0.
>
> Usually, if there's a possible tradeoff between correctness and performance, I
> tend to choose correctness, but I'm not so sure in this case. I'm not sure with
> reverting either, which is why I post this as an RFC only.
>
> I hope to get some more comments on how to proceed here for 0.13.

I think that we have to revert also.

It is a thoug decission so.  This patch makes things more "correct",
but it makes some load really slow, for instance, and install went from
15m to 50m on my hardware :(

My point here is that with the patch, qcow2 becomes almost unusable, so
... what to do?

Later, Juan.

  parent reply	other threads:[~2010-08-24 12:51 UTC|newest]

Thread overview: 44+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-24 10:40 [Qemu-devel] [RFC][STABLE 0.13] Revert "qcow2: Use bdrv_(p)write_sync for metadata writes" Kevin Wolf
2010-08-24 11:02 ` [Qemu-devel] " Stefan Hajnoczi
2010-08-24 11:06   ` Michael Tokarev
2010-08-24 11:40   ` Kevin Wolf
2010-08-24 11:56     ` Alexander Graf
2010-08-24 12:10       ` Kevin Wolf
2010-08-24 12:12         ` Alexander Graf
2010-08-24 12:18           ` Avi Kivity
2010-08-24 12:21             ` Alexander Graf
2010-08-24 12:27               ` Avi Kivity
2010-08-24 12:35                 ` Kevin Wolf
2010-08-24 12:39                   ` Avi Kivity
2010-08-24 12:53                     ` Kevin Wolf
2010-08-24 12:21     ` Stefan Hajnoczi
2010-08-24 12:23       ` Michael Tokarev
2010-08-24 12:48 ` Juan Quintela [this message]
2010-08-24 13:01 ` Anthony Liguori
2010-08-24 13:16   ` Kevin Wolf
2010-08-24 13:29     ` Anthony Liguori
2010-08-24 13:31       ` Avi Kivity
2010-08-24 13:35         ` Anthony Liguori
2010-08-24 13:39           ` Avi Kivity
2010-08-24 13:40             ` Anthony Liguori
2010-08-24 13:44               ` Avi Kivity
2010-08-24 13:56                 ` Anthony Liguori
2010-08-25  7:14                   ` Avi Kivity
2010-08-25 12:46                     ` Anthony Liguori
2010-08-25 13:07                       ` Avi Kivity
2010-08-25 13:37                         ` Anthony Liguori
2010-08-25 13:23                       ` Avi Kivity
2010-08-25 13:42                         ` Anthony Liguori
2010-08-25 14:00                           ` Avi Kivity
2010-08-25 14:14                             ` Anthony Liguori
2010-08-25 14:36                               ` Avi Kivity
2010-08-25 15:06                                 ` Anthony Liguori
2010-08-25 15:15                                   ` Avi Kivity
2010-08-25 15:21                                     ` Anthony Liguori
2010-08-25 13:46                         ` Anthony Liguori
2010-08-25 14:03                           ` Avi Kivity
2010-08-25 14:19                           ` Christoph Hellwig
2010-08-25 14:37                             ` Avi Kivity
2010-08-25 14:18                         ` Christoph Hellwig
2010-08-25 14:26                           ` Anthony Liguori
2010-08-25 14:49                             ` Daniel P. Berrange

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=m3k4ngmb11.fsf@trasno.org \
    --to=quintela@redhat.com \
    --cc=avi@redhat.com \
    --cc=hch@lst.de \
    --cc=kwolf@redhat.com \
    --cc=mjt@tls.msk.ru \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.