Re: [PATCH 00/11] Alternate p2m: support multiple copies of host p2m

[Ed White <edmund.h.white@intel.com>]

On 01/16/2015 12:12 AM, Jan Beulich wrote:
>>>> On 15.01.15 at 19:23, <edmund.h.white@intel.com> wrote:
>> On 01/15/2015 08:15 AM, Tim Deegan wrote:
>>> - Feature compatibilty/completeness.  You pointed out yourself that
>>>   it doesn't work with nested HVM or migration.  I think I'd have to
>>>   add mem_event/access/paging and PCI passthrough to the list of
>>>   features that ought to still work.  I'm resigned to the idea that
>>>   many new features don't work with shadow pagetables. :)
>>>
>>
>> The intention is that mem_event/access should still work. I haven't
>> specifically looked at paging, but I don't see any fundamental reason
>> why it shouldn't. PCI passthrough I suspect won't. Does nested HVM
>> work with migration? Is it simply not acceptable to submit a feature
>> as experimental, with known compatibility issues? I had assumed that
>> it was, based on the nested HVM status as documented in the release
>> notes.
> 
> It is generally acceptable, sure. But the sad thing here is that
> particularly with such code coming from Intel (with the nested
> VMX code being a good example, and certain IOMMU/pass-
> through things being another) my experience is that once that
> initial code drop happened, interest from the original authors is
> lost (possibly not maliciously, but because of being assigned other
> tasks) and the code therefore has to remain experimental for
> extended periods of time (often until someone else gets
> frustrated enough about its half baked state to spend non-
> negligible amounts of time to deal with the loose ends).

I can't make any guarantees, I'm just one person and I do what
the people who pay me tell me to do.

What I can tell you is that our primary motivation for producing
this patch series is that we have an internal customer already
shipping Xen-based products who wants to use these capabilities.
That customer isn't going to go away, and isn't going to want to
support something half-baked.

Ed