On Fri, Feb 24, 2006 at 06:06:22PM +0100, Bastian Blank wrote: Hi,
> It is a sort of kernel. Yeah, it's a sort of kernel, but it's not the linux kernel... And it seems the kernel team is about the linux kernel, not just any kernel, isn't it? > Just to say, how connected xen to linux is: > > For example: There are three kernel trees of xen: > - from xen-3.0-testing, 2.6.12 > - from linux-2.6-xen, 2.6.16-rc4 > - from linux-2.6-merge, 2.6.16-rc3 > All of them have different needs from xen. > > The kernels from xen-3.0-testing and linux-2.6-merge works with a 3.0 > and unstable hypervisor. > > The 3.0 utils only works on the kernel from xen-3.0-testing. The > unstable utils with the other. But with both hypervisors. > That's I think because xen is still young, and is starting just now its distribution integration, and probably will happen a lot less when it will be integrated with Linux (Linus' tree) and the development of xenolinux will proceed at a different pace than the hypervisor. Then probably it will just be that any xen version will have a minimum linux version needed, just as now a lot of other stuff does, and there will be nothing special in it, except the fact that it needs a kernel compiled for the appropriate subarch). > I won't reject the help of volunteers but I strongly think that the > kernel team needs to have its hands on them. > What do other people in the kernel team think? If the majority of them agree fine, otherwise are you sure it's not counterproductive to force xen in the kernel team hands if most of them don't want to touch it, and on the other hand to risk driving away other people who just cannot follow the whole linux business but could work on the xen hypervisor and tools, help coordinate with xen's upstream, debian glibc and d-i, etc! Especially if you and other people who would do both can still do it! :) Guido -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
