Re: [Qemu-devel] FreeOSZoo will stop March 1, 2005

[Daniel Egger]

On 12.02.2005, at 19:11, Jernej Simončič wrote:

> Isn't the performance penalty only if you compare current Qemu with and
> without the kernel module, and the speed without the module hasn't 
> changed
> compared to Qemu from before KQemu was available?

Nope, qemu-fast uses a patched Linux kernel (yes, the guest can
*only* be Linux) to run Linux at very-close-to-native speed
with very little translation in a different address space.

If you only intend to run Linux VMs and don't have a problem
patching the kernel sources this is a very viable approach to
get something which is not possible anymore *without* kqemu
which is more flexible but does only run on 32bit kernels ATM.

Servus,
      Daniel

PGP.sig
Description: This is a digitally signed message part