Re: [PATCH][v3] PV featured HVM(hybrid) support in Xen

[Sheng Yang <sheng@linux.intel.com>]

On Monday 08 February 2010 16:12:54 Keir Fraser wrote:
> On 08/02/2010 08:05, "Yang, Sheng" <sheng.yang@intel.com> wrote:
> > Hi Keir
> >
> > Here is the latest (hybrid) patchset.
> >
> > Change from v2:
> >
> > 1. Change the name "hybrid" to "PV featured HVM", as well as flag and
> > macro names.
> > 2. Merge the hypercall page with HVM.
> > 3. Clean up VIRQ delivery mechanism, fixing the lock issue.
> > 4. Remove the reserved the region in E820(but another issue remains, I
> > can't get location of grant table elegantly, as I described in Linux part
> > patches)
> >
> > Please review, thanks!
> 
> Why can't you leave LAPIC available for things like INIT-SIPI-SIPI, and
> simply not use it for things you need higher perf for?

I am not quite understand why we need to stick to INIT-SIPI-SIPI. For the 
physical cpus, we can't tell their state directly, so we need a protocol, like 
INIT-SIPI-SIPI; but for virtual cpus, we have hypercall as our protocol, so we 
can easily tell them when and where to start, which is much more directly like 
PV domU did.

-- 
regards
Yang, Sheng