[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [Xen-users] setting up private networking between dom0 and domU?
> What is known about Nvidia plans - do they plan to develop a dom0-only, > or maybe even a domU drivers as well? dom0 3D drivers are certainly easier (these should also work in a domU that has access to PCI graphics card, once PCI-passthrough is working again - someone will have to figure out how to make X behave correctly, tho). domU drivers would ideally be more generic than a particular manufacturer's drivers, although there is a performance / generic-ness tradeoff here. > Do I understand properly that > with proper domU-drivers it will be possible... maybe even run > OpenGL/DirectX software under guest Windows? In theory, certainly. It just requires somebody to design and implement a 3D virtual graphics interface *and* write the appropriate Windows driver. > As far as I understand, ATI > drivers were made compatible with dom0, but no domU yet - so is Nvidia > behind ATI in this area yet? I'm not sure the patches to the ATI driver to make it behave itself under Xen have been upstreamed yet. Once Xen is in the mainline kernel and its popularity becomes apparent, hopefully both manufacturers will decide its worth testing under Xen. > Is sound card virtualization now working in any form (at least in > "3"-mode, by emulation)? I'd imagine that's reasonably easy to make work for VT guests. Paravirt guests need to use network sound daemons at the moment, (artsd, esoundd, or use FreeNX if you can make it work). A number of people have been looking at paravirtualised sound drivers for native Xen guests. > And, probably, the most important: how hard drive is emulated for > guest Windows? Should it use a dedicated drive, a dedicated partition, > or maybe it requires an "image" file (visible to domU-Windows as hard > drive, but stored as a file on a real FS for dom-Linux)? Right now the emulation is provided by userspace in dom0, so it could be any of those options. In future I understand it's likely to be provided by some kind of shadow IO domain that will run alongside the VT domain and emulate device accesses, using the normal paravirtualised interface to dom0 to actually transfer data. This option should help with performance and also make live migration, suspend/resume, etc closer to the process required for native domains. > Or maybe even > file-level access is virtualized, so that we can be able to have paths > like /var/Windows/Program Files/ on our Reisers and Exts and to > distribute Windows files over LVM? You'd want something like my XenFS for this. It'll be quite a long time before a XenFS client for Windows is available, it'll be Linux-only to start with. > And also: currently mostly Intel VT is being discussed, as the only > available on the market now. But will advanced AMD Pacifica's features > help with anything of above (video/audio/hard disk), or they will give > only performance benefits? Device emulation will be rather similar for both. I get the impression Pacifica has more features for safely exposing real devices for dedicated access by guests, but I'm not certain on this. Future iterations of virtualisation-aware hardware are likely to include more support for virtualisation. Smart NICS (like Infiniband) are already well disposed towards efficient virtualisation, since they provide virtual network interfaces in hardware. HTH, Mark -- Dave: Just a question. What use is a unicyle with no seat? And no pedals! Mark: To answer a question with a question: What use is a skateboard? Dave: Skateboards have wheels. Mark: My wheel has a wheel! _______________________________________________ Xen-users mailing list Xen-users@xxxxxxxxxxxxxxxxxxx http://lists.xensource.com/xen-users
|
Lists.xenproject.org is hosted with RackSpace, monitoring our |