qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Alon Levy <alevy@redhat.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: qemu-devel@nongnu.org, elmarco@redhat.com,
	Luiz Capitulino <lcapitulino@redhat.com>
Subject: Re: [Qemu-devel] [RFC 5/7] qxl-render: call ppm_save on callback
Date: Mon, 20 Feb 2012 14:36:28 +0200	[thread overview]
Message-ID: <20120220123628.GF23926@garlic.redhat.com> (raw)
In-Reply-To: <4F422F5C.9060202@redhat.com>

On Mon, Feb 20, 2012 at 12:32:44PM +0100, Gerd Hoffmann wrote:
> On 02/19/12 22:28, Alon Levy wrote:
> > This changes the behavior of the monitor command. After the previous
> > patch, there is no longer an option of deadlock with virt-manager, but
> > ppm_save is called too early, before the update has completed. With this
> > patch it is called at the correct moment, but that means there is a race
> > between the monitor command completing and the screendump file being saved.
> > 
> > The only solution is to use an asynchronous monitor command. For a
> > previous round of this see:
> >  http://lists.gnu.org/archive/html/qemu-devel/2011-10/msg02810.html
> > 
> > Since that's contentious, I'm suggesting we do something that is almost
> > correct and doesn't hang, instead of correct and hangs. The screendump
> > user can inotify on the directory and the file if need be. For casual
> > monitor usage there is no difference.
> 
> Hmm, that is pretty lame.  There are users like autotest which expect
> the screen dump being there when the monitor command is finished, that
> change will break them.
> 
> Unfortunaly there is no easy way out.  I think the options are:
> 
>  (1) Keep existing behavior.  That means the screenshot might show old
>      screen content.  Not very nice too.  Would work sort-of ok for
>      autotest though as autotest does screenshots every second and thus
>      the screen content wouldn't be older than a second.
> 
>  (2) Async monitor command.  Keeps interface and works nicely.  A bunch
>      of QAPI bits tickled into master meanwhile, so we could look at
>      this again.  Luiz?  What is the status here?
> 
>  (3) Something like this patch + additionally introduce a
>      "your-screenshot-is-finished-now" qmp event.  Will break existing
>      users too.  But at least they can be adapted without requiring
>      some external, nonportable service like inotify ...
> 

I was going to look for QAPI bits after this series (i.e. (2)). Doing
(3) is also possible.

> cheers,
>   Gerd
> 

  reply	other threads:[~2012-02-20 12:36 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-02-19 21:27 [Qemu-devel] [RFC 0/7] qxl: fix hangs caused by qxl_render_update Alon Levy
2012-02-19 21:28 ` [Qemu-devel] [RFC 1/7] sdl: remove NULL check, g_malloc0 can't fail Alon Levy
2012-02-19 21:28 ` [Qemu-devel] [RFC 2/7] qxl: drop qxl_spice_update_area_async definition Alon Levy
2012-02-19 21:28 ` [Qemu-devel] [RFC 3/7] qxl: introduce QXLCookie Alon Levy
2012-02-20 10:56   ` Gerd Hoffmann
2012-02-20 12:31     ` Alon Levy
2012-02-20 12:39       ` Gerd Hoffmann
2012-02-19 21:28 ` [Qemu-devel] [RFC 4/7] qxl: make qxl_render_update async Alon Levy
2012-02-20 11:10   ` Gerd Hoffmann
2012-02-20 12:32     ` Alon Levy
2012-02-20 12:45       ` Gerd Hoffmann
2012-02-19 21:28 ` [Qemu-devel] [RFC 5/7] qxl-render: call ppm_save on callback Alon Levy
2012-02-20 11:32   ` Gerd Hoffmann
2012-02-20 12:36     ` Alon Levy [this message]
2012-02-20 12:49       ` Gerd Hoffmann
2012-02-20 21:29     ` Eric Blake
2012-02-21  8:19       ` Alon Levy
2012-02-21 16:15         ` Eric Blake
2012-02-21 17:40           ` Alon Levy
2012-02-22 13:17             ` Luiz Capitulino
2012-02-22 13:22               ` Alon Levy
2012-02-22 13:49                 ` Luiz Capitulino
2012-02-22 14:22                   ` Gerd Hoffmann
2012-02-22 14:29                     ` Alon Levy
2012-02-22 15:55                       ` Luiz Capitulino
2012-02-22 16:35                         ` Alon Levy
2012-02-22 19:27                           ` Luiz Capitulino
2012-02-22 14:28                   ` Alon Levy
2012-02-22 14:47                     ` Gerd Hoffmann
2012-02-22 15:26                       ` Alon Levy
2012-02-19 21:28 ` [Qemu-devel] [RFC 6/7] qxl: use spice_qxl_update_area_dirty_async Alon Levy
2012-02-19 21:28 ` [Qemu-devel] [RFC 7/7] qxl: add allocator Alon Levy
2012-02-20 11:41   ` Gerd Hoffmann
2012-02-20 12:38     ` Alon Levy
2012-02-20 13:18       ` Gerd Hoffmann
2012-02-20 17:36         ` Alon Levy
2012-02-21  7:57           ` Gerd Hoffmann
2012-02-21  8:26             ` Alon Levy
2012-02-21  9:20               ` Gerd Hoffmann
2012-02-21  9:59                 ` Alon Levy

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=20120220123628.GF23926@garlic.redhat.com \
    --to=alevy@redhat.com \
    --cc=elmarco@redhat.com \
    --cc=kraxel@redhat.com \
    --cc=lcapitulino@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /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).