On 7/6/06, Gaurav Dhiman <gauravd.chd@xxxxxxxxx> wrote:
On 7/6/06, Tharindu Rukshan Bamunuarachchi <tharindub@xxxxxxxxx> wrote: > > AFAIK, This is only applies to virtualization context. I think Xen uses > this, if i am not wrong. > > I thought Gaurav was asking about "traditional" ring zero. I was not asking .... rather telling about tradition ring 0 ... :-) well what is this ring 0 for vm and how it is different from tradition ring 0. As the mode of processor is defined by 2 LSB bits of CS register, how is this ring 0 for VM represented there and what previlieges are restriced in this ?? regards, Gaurav
Gaurav, I have only a general understanding, but when Xen first came out a couple of years ago they apparently blew away the VM industry with their concept of Paravirtualized drivers (PVDs). The world wanted to jump on and even PVDs were written for some of the Windows OSes. M$ decided this was just plain wrong. Only they should be able to do that, so they made the Xen people recall that offering. Apparently Intel thought Xen was a very good thing and determined that if the Guest OS ran in a restricted ring-0, then the guest could run totally standard code, but when it tried to access a real I/O device, mm controller, etc. then an interrupt of sorts would be generated and the Host OS running in full unrestricted ring-0 would be invoked to perform the work the Guest wanted done. For more technical hardware details see http://www.intel.com/technology/computing/vptech/ Particularily see the links on the right. FYI: AMD has a competitive offering, but I think they are playing catch-up. I don't remember the name of their offering, nor do I have any idea how it works. Greg -- Greg Freemyer The Norcross Group Forensics for the 21st Century -- Kernelnewbies: Help each other learn about the Linux kernel. Archive: http://mail.nl.linux.org/kernelnewbies/ FAQ: http://kernelnewbies.org/faq/
