Are there any plans to get Fedora's 32-bit qemu-kvm package working with Spice support? Is this a limitation of the buildsystem, or something else?
- Ken
On Thu, 2011-05-19 at 14:31 -0600, Ken Dreyer wrote:
Are there any plans to get Fedora's 32-bit qemu-kvm package working with Spice support? Is this a limitation of the buildsystem, or something else?
I am unaware of any plans to make this work, though that does not mean that none exist. It is a matter of code, and the spice-server is a 64bit only build as well. I have never heard of any planned 32bit host spice support.
Justin
On Thu, May 19, 2011 at 02:31:23PM -0600, Ken Dreyer wrote:
Are there any plans to get Fedora's 32-bit qemu-kvm package working with Spice support? Is this a limitation of the buildsystem, or something else?
No plans, but no limitation either AFAICT.
- Ken
virt mailing list virt@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/virt
On Fri, May 20, 2011 at 04:35:58PM +0300, Alon Levy wrote:
On Thu, May 19, 2011 at 02:31:23PM -0600, Ken Dreyer wrote:
Are there any plans to get Fedora's 32-bit qemu-kvm package working with Spice support? Is this a limitation of the buildsystem, or something else?
No plans, but no limitation either AFAICT.
IIUC, in the past, the SPICE protocol itself was not endian-safe, nor 32/64-bit invariant. I assume those problems have all been fixed in latest SPICE protocol specs, and it just remains that the actual impl code which needs fixing up ?
Daniel
On Fri, May 20, 2011 at 02:39:37PM +0100, Daniel P. Berrange wrote:
On Fri, May 20, 2011 at 04:35:58PM +0300, Alon Levy wrote:
On Thu, May 19, 2011 at 02:31:23PM -0600, Ken Dreyer wrote:
Are there any plans to get Fedora's 32-bit qemu-kvm package working with Spice support? Is this a limitation of the buildsystem, or something else?
No plans, but no limitation either AFAICT.
IIUC, in the past, the SPICE protocol itself was not endian-safe, nor 32/64-bit invariant. I assume those problems have all been fixed in latest SPICE protocol specs, and it just remains that the actual impl code which needs fixing up ?
All I meant was that there is nothing inherently 64 bit about spice. The marshalling code does make for an easy place to deal with endianess of the protocol vs the machine. A quick grep shows there is SPICE_BYTESWAP in macros.h and it is used by generated_demarshaller.c (and I'll assume the others). 64/32 is something that needs other checking. I'd assume if someone got to building it on 32 bit they will find pretty fast if there is anything to fix.
Daniel
|: http://berrange.com -o- http://www.flickr.com/photos/dberrange/ :| |: http://libvirt.org -o- http://virt-manager.org :| |: http://autobuild.org -o- http://search.cpan.org/~danberr/ :| |: http://entangle-photo.org -o- http://live.gnome.org/gtk-vnc :|
-
Alon Levy -
Daniel P. Berrange -
Justin Forbes -
Ken Dreyer