Re: [Xen-devel] [PATCH] x86/HVM: Merge HVM and PVH hypercall tables

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 10.12.15 at 17:53, <boris.ostrovsky@xxxxxxxxxx> wrote:
> On 12/10/2015 07:30 AM, Jan Beulich wrote:
>>>>> On 08.12.15 at 15:20, <boris.ostrovsky@xxxxxxxxxx> wrote:
>>> The tables are almost identical and therefore there is little reason to
>>> keep both sets.
>>>
>>> PVH needs 3 extra hypercalls:
>>> * mmuext_op. PVH uses MMUEXT_TLB_FLUSH_MULTI and MMUEXT_INVLPG_MULTI to
>>>    optimize TLB flushing. Since HVMlite guests may decide to use them as
>>>    well we can allow these two commands for all guests in an HVM container.
>> I must be missing something here: Especially for the INVLPG variant
>> I can't see what use it could be for a PVH guest, as it necessarily
>> would act on a different address space (the other one may have at
>> least some effect due to hvm_flush_guest_tlbs()).
> 
> This is done out of xen_flush_tlb_others(), which is what PVH guests use.
> 
> And yes --- there indeed seems to be little reason to do that. But it is 
> there now so I am not sure we can make this not work anymore for PVH guests.

PVH is experimental, so we certainly can. I'm pretty determined that
we shouldn't expose functionality to PVH (or HVMlite) that can't be
actually useful to it. I think it was a mistake to allow PVH general
access to all MMUEXT ops (I thought this had been discussed, but I
can't seem to find that discussion going back to v12 of the PVH series,
and I can't even spot the particular patch earlier than in v12).

Jan


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