Re: [Xen-devel] Xen 4.1 + Linux compiled with PVH == BOOM

[Ian Campbell <Ian.Campbell@xxxxxxxxxx>]

On Tue, 2014-01-07 at 11:50 +0000, Jan Beulich wrote:
> >>> On 07.01.14 at 12:31, Ian Campbell <Ian.Campbell@xxxxxxxxxx> wrote:
> > On Tue, 2014-01-07 at 08:23 +0000, Jan Beulich wrote:
> >> >> that feature statically in the first place - that should be done only 
> >> >> if 
> > the 
> >> > kernel
> >> >> could _only_ boot in PVH mode.
> >> > 
> >> > The feature is not marked as "required" but rather - it can utilize said
> >> > extension (so supported). I am advocating that the calleer checks that
> >> > all of the required pieces are correct - and it can ignore the ones it
> >> > has no idea off (which it does for some of the Xen ELF notes - ignores
> >> > them if it has no idea of what they are).
> >> 
> >> What would be to point of telling the hypervisor that the kernel
> >> can utilize a certain extension? The kernel could just utilize it, and
> >> the hypervisor would know by that simple fact.
> > 
> > But for PVH doesn't the hypervisor need to now at dom0 build time
> > whether to build a PV or PVH domain?
> 
> Which needs to be communicated via hypervisor command line option
> anyway.

I would expect that the plan is to eventually enable PVH by default if
the kernel can cope with it.

>  Specifying the option without have a suitable kernel is (of
> course) a user error (generally expected to result in a kernel crash).

In the case where the user has specified the option sure.

Note that the original issue was a PVH capable kernel under a non-PVH
capable Xen, although we've strayed a bit from that topic.

Ian.

> 
> Jan
> 
> > So it needs to know upfront if the
> > kernel could do PVH or not, and then pick, but once it has picked the
> > kernel had best follow that choice.
> > 
> > So in the PVH case it's not just a simple case of the kernel deciding to
> > utilize an optional feature, the optional feature has already been
> > enabled.
> > 
> > Ian
> 
> 
> 



_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel