From: Gerd Hoffmann <kraxel@redhat.com>
To: "Michal Prívozník" <mprivozn@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
qemu-devel@nongnu.org, "Michael S. Tsirkin" <mst@redhat.com>
Subject: Re: [PULL 12/25] virtio-gpu: move virgl realize + properties
Date: Fri, 21 May 2021 15:57:12 +0200 [thread overview]
Message-ID: <20210521135712.lqkdn4xhqhhqaxw4@sirius.home.kraxel.org> (raw)
In-Reply-To: <a86e2a07-a803-cd00-849e-96a30e8839d3@redhat.com>
Hi,
> Sorry for catching this a bit late, but libvirt is looking for "virgl"
> property when guest XML has 3D acceleration enabled:
Yes, libvirt must be adapted to this.
https://gitlab.com/libvirt/libvirt/-/issues/167
As far I know libvirt checks whenever the virgl property exists to
figure whenever virgl support is available (as you can compile qemu
without virgl support). So without changes libvirt will simply
think there is no 3d support and configurations without virgl enables
should continue to work fine.
Configurations with virgl enabled will break though, and unfortunaly
there is no easy way to avoid that.
> The commit message suggests that virtio-gpu-gl-device should be used
> instead. Fair enough, so IIUC the cmd line should be changed to:
>
> -device virtio-gpu-gl-device,id=video0,max_outputs=1,bus=pci.0,addr=0x2
virtio-gpu-gl-device is the mmio variant, for pci you need
virtio-gpu-gl-pci. But otherwise yes, this is what libvirt should use
in case it figures qemu supports the virtio-gpu-gl-pci device (again,
when compiling with virgl disabled the device will not be there).
take care,
Gerd
next prev parent reply other threads:[~2021-05-21 13:58 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-05-10 13:20 [PULL 00/25] Vga 20210510 patches Gerd Hoffmann
2021-05-10 13:20 ` [PULL 01/25] qemu-edid: use qemu_edid_size() Gerd Hoffmann
2021-05-10 13:20 ` [PULL 02/25] edid: edid_desc_next Gerd Hoffmann
2021-05-10 13:20 ` [PULL 03/25] edid: move xtra3 descriptor Gerd Hoffmann
2021-05-10 13:20 ` [PULL 04/25] edid: use dta extension block descriptors Gerd Hoffmann
2021-05-10 13:20 ` [PULL 05/25] edid: Make refresh rate configurable Gerd Hoffmann
2021-05-10 13:20 ` [PULL 06/25] edid: move timing generation into a separate function Gerd Hoffmann
2021-05-10 13:20 ` [PULL 07/25] edid: allow arbitrary-length checksums Gerd Hoffmann
2021-05-10 13:20 ` [PULL 08/25] edid: add support for DisplayID extension (5k resolution) Gerd Hoffmann
2021-05-10 13:20 ` [PULL 09/25] virtio-gpu: handle partial maps properly Gerd Hoffmann
2021-05-10 13:20 ` [PULL 10/25] virtio-gpu: rename virgl source file Gerd Hoffmann
2021-05-10 13:20 ` [PULL 11/25] virtio-gpu: add virtio-gpu-gl-device Gerd Hoffmann
2021-05-10 13:20 ` [PULL 12/25] virtio-gpu: move virgl realize + properties Gerd Hoffmann
2021-05-21 9:32 ` Michal Prívozník
2021-05-21 10:50 ` Marc-André Lureau
2021-05-21 13:45 ` Gerd Hoffmann
2021-05-21 13:57 ` Gerd Hoffmann [this message]
2021-05-10 13:20 ` [PULL 13/25] virtio-gpu: move virgl reset Gerd Hoffmann
2021-05-10 13:20 ` [PULL 14/25] virtio-gpu: use class function for ctrl queue handlers Gerd Hoffmann
2021-05-10 13:20 ` [PULL 15/25] virtio-gpu: move virgl handle_ctrl Gerd Hoffmann
2021-05-10 13:20 ` [PULL 16/25] virtio-gpu: move virgl gl_flushed Gerd Hoffmann
2021-05-10 13:20 ` [PULL 17/25] virtio-gpu: move virgl process_cmd Gerd Hoffmann
2021-05-10 13:20 ` [PULL 18/25] virtio-gpu: move update_cursor_data Gerd Hoffmann
2021-05-10 13:20 ` [PULL 19/25] virtio-gpu: drop VIRGL() macro Gerd Hoffmann
2021-05-10 13:20 ` [PULL 20/25] virtio-gpu: move virtio-gpu-gl-device to separate module Gerd Hoffmann
2021-05-10 13:20 ` [PULL 21/25] virtio-gpu: drop use_virgl_renderer Gerd Hoffmann
2021-05-10 13:20 ` [PULL 22/25] virtio-gpu: move fields to struct VirtIOGPUGL Gerd Hoffmann
2021-05-10 13:20 ` [PULL 23/25] virtio-gpu: add virtio-gpu-gl-pci Gerd Hoffmann
2021-05-10 13:20 ` [PULL 24/25] modules: add have_vga Gerd Hoffmann
2021-05-10 13:20 ` [PULL 25/25] virtio-gpu: add virtio-vga-gl Gerd Hoffmann
2021-05-10 13:43 ` [PULL 00/25] Vga 20210510 patches no-reply
2021-05-12 15:05 ` Peter Maydell
2021-06-30 18:16 ` Marc-André Lureau
2021-07-01 6:29 ` Marc-André Lureau
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=20210521135712.lqkdn4xhqhhqaxw4@sirius.home.kraxel.org \
--to=kraxel@redhat.com \
--cc=mprivozn@redhat.com \
--cc=mst@redhat.com \
--cc=pbonzini@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).