From: Alex Williamson <alex.williamson@redhat.com>
To: Gerd Hoffmann <kraxel@redhat.com>
Cc: Peter Maydell <peter.maydell@linaro.org>,
QEMU Developers <qemu-devel@nongnu.org>,
Tina Zhang <tina.zhang@intel.com>,
intel-gvt-dev@lists.freedesktop.org,
Kirti Wankhede <kwankhede@nvidia.com>,
Markus Armbruster <armbru@redhat.com>,
Eric Blake <eblake@redhat.com>,
Paolo Bonzini <pbonzini@redhat.com>,
Laine Stump <laine@redhat.com>,
Erik Skultety <eskultet@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v4 4/7] qdev: add hotpluggable to DeviceState
Date: Fri, 23 Feb 2018 08:58:41 -0700 [thread overview]
Message-ID: <20180223085841.0f8362ed@w520.home> (raw)
In-Reply-To: <20180223085322.vhc4awb5aprohe2s@sirius.home.kraxel.org>
On Fri, 23 Feb 2018 09:53:22 +0100
Gerd Hoffmann <kraxel@redhat.com> wrote:
> Hi,
>
> > > What type of device is only sometimes hotpluggable ?
> > > The commit message says "display devices" and "consoles",
> > > but I would expect those to both be types of device which
> > > have a class which is never hotpluggable, so you can mark
> > > them non-hotpluggable with the existing class flag rather
> > > than needing a per-instance flag.
> >
> > With this series, a vfio-pci device optionally supports a display. The
> > vfio-pci device is hotpluggable, but QEMU display support is not. So
> > the solution here is to make the vfio-pci device non-hotpluggable only
> > when it supports and enables a display.
> >
> > Gerd, is there another solution that the display object is instantiated
> > separately from the vfio-pci object and the display support in the
> > vfio-pci device references the display object via an id.
>
> Well, not really. At least not without putting much of qemu display
> support upside down. The qemu display (aka QemuConsole) is created and
> managed by the display devices, they can't be created independant from
> a device ...
>
> The connection between QemuConsole and User Interface (i.e. gtk, spice,
> ...) is a bit more flexible. But also not really designed for hotplug
> as QemuConsole is not hotpluggable in the first place ...
>
> We could drop the display property and use two devices instead.
>
> new vfio-pci would behave like display=off with this series.
> added vfio-pci-display has display=on behavior.
> display=auto is not possible.
I expect libvirt and above would balk at creating a separate QEMU
device for this purpose, easy for QEMU, hard for anything that manages
QEMU. Thanks,
Alex
next prev parent reply other threads:[~2018-02-23 15:58 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-02-19 11:14 [Qemu-devel] [PATCH v4 0/7] vfio: add display support Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 1/7] linux-headers: update to 4.16-rc1 Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 2/7] standard-headers: add drm/drm_fourcc.h Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 3/7] ui/pixman: add qemu_drm_format_to_pixman() Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 4/7] qdev: add hotpluggable to DeviceState Gerd Hoffmann
2018-02-22 15:37 ` Gerd Hoffmann
2018-02-22 15:46 ` Peter Maydell
2018-02-22 16:59 ` Alex Williamson
2018-02-22 17:21 ` Peter Maydell
2018-02-23 3:14 ` Zhenyu Wang
2018-02-23 8:53 ` Gerd Hoffmann
2018-02-23 15:58 ` Alex Williamson [this message]
2018-02-26 8:23 ` Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 5/7] vfio/common: cleanup in vfio_region_finalize Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 6/7] vfio/display: core & wireup Gerd Hoffmann
2018-02-19 11:14 ` [Qemu-devel] [PATCH v4 7/7] vfio/display: adding region support Gerd Hoffmann
2018-02-19 22:16 ` [Qemu-devel] [PATCH v4 0/7] vfio: add display support Alex Williamson
2018-02-20 10:28 ` Gerd Hoffmann
2018-02-20 17:04 ` Gerd Hoffmann
2018-02-22 6:40 ` Zhenyu Wang
2018-02-22 9:30 ` Gerd Hoffmann
2018-02-23 2:42 ` Zhenyu Wang
2018-02-23 7:06 ` Zhang, Tina
2018-02-22 16:28 ` Alex Williamson
2018-02-23 9:05 ` Gerd Hoffmann
2018-02-26 8:35 ` Gerd Hoffmann
2018-02-26 20:59 ` Alex Williamson
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=20180223085841.0f8362ed@w520.home \
--to=alex.williamson@redhat.com \
--cc=armbru@redhat.com \
--cc=eblake@redhat.com \
--cc=eskultet@redhat.com \
--cc=intel-gvt-dev@lists.freedesktop.org \
--cc=kraxel@redhat.com \
--cc=kwankhede@nvidia.com \
--cc=laine@redhat.com \
--cc=pbonzini@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=tina.zhang@intel.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).