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

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

On 01/20/2015 12:47 AM, Jan Beulich wrote:
>>>> On 19.01.15 at 22:54, <edmund.h.white@intel.com> wrote:
>> There's also the issue that access permissions
>> are soft state and can be reverted to default in certain cases.
> 
> Some instances of which have got removed during the 4.5 cycle,
> and at least some of the remaining ones are deemed at least
> questionable too. I.e. if this reverting to default permissions
> stands in the way, propose (separate!) patches to eliminate it.

The less frequently changed state disappears the better, but it
seems there isn't any way to prevent all such events, which is
a new problem for the software these patches are meant to
enable.

So, there has to be a mechanism to notify the agent that
a piece of its state is no longer valid. I think the
challenge is figuring out how fine-grained the notification
should be.

One idea I had since yesterday is to track the minimum and
maximum mfn's ever subject to any altp2m-type modification
in a p2m. That shouldn't be very expensive, and would likely
avoid a lot of unnecessary p2m invalidation.

Ed