qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Qemu-devel] a user here - pci-assign VGA


From: lejeczek
Subject: [Qemu-devel] a user here - pci-assign VGA
Date: Tue, 06 Nov 2012 13:17:17 +0000
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:15.0) Gecko/20120911 Thunderbird/15.0.1

hi there, I wanted to share some fairly good news
a while ago when as was battling the problem - pci-assign a vga - I got it to work
I realized that it might have worked from the beginning, anyhow,
for me it works now with qemu-kvm 1.2 (which is fedora alpha rpm-rebuilt on SL 6.3) with various AMD vgas
but! (there is always a but, isn't there :) )
guest loses vnc?? meaning that VNC gets only blank(black screen (this was what deceived me earlier)) whereas guest seems to fine itself, I can rdp to it (it's win7), I can get it on a directly attached monitor via keyboard + mouse attached in the same fashion only win device manager shows "Standard VGA Graphic Adapter" exclamation marked. (I tied both -vga cirrus and -vga std)

would anybody know how to fix/troubleshoot it?
many thanks

On 03/10/12 15:46, Alex Williamson wrote:
On Wed, 2012-10-03 at 15:30 +0100, lejeczek wrote:
I cannot tell anything about upsteam, I'm going to try it
out now, but in general I tend to stay as close to
distro/yum version as possible,
so for 6.3 rhel it is tedious way of grabbing and compiling
everything that is needed
where it seems RHEL stands it the reservation that VGA
pass-though is not supported full stop
all I was saying was that their seabios might be worth
looking into/updating as slightly higher rev of it solved my
problem as described in
https://bugzilla.redhat.com/show_bug.cgi?id=860418
I do apologize if I was abrupt in closing that bug, but it needs to work
upstream first.  We've had a couple reports of some cards working, but
nothing worth declaring success, backporting and trying to call it
supported.  If a seabios update improves things for you, please report
it and maybe there are changes we can backport.  Personally I'd prefer
to see it working reliably upstream first and incorporate a full
solution rather than hearsay that card A worked on system B at least
once.  Thanks,

Alex







reply via email to

[Prev in Thread] Current Thread [Next in Thread]