Re: [Draft C] Boot ABI for HVM guests without a device-model

["Roger Pau Monné" <roger.pau@citrix.com>]

El 07/09/15 a les 12.05, Jan Beulich ha escrit:
>>>> On 07.09.15 at 11:34, <roger.pau@citrix.com> wrote:
>> So AFAICS we have 3 options:
>>
>> 1. Overload VCPUOP_initialise like it's done in the current series (v6).
>> For PV guests the hypercall parameter is of type vcpu_guest_context,
>> while for HVM guests the parameter is of type vcpu_hvm_context.
>>
>> 2. Create a new hypercall (VCPUOP_hvm_initialise) only available to HVM
>> guests, that only allows vcpu_hvm_context as a parameter.
>>
>> 3. Deprecate current VCPUOP_initialise, introduce a new
>> VCPUOP_initialise, that takes the following parameter:
>>
>> union vcpu_context {
>> 	struct vcpu_guest_context pv_ctx;
>> 	struct vcpu_hvm_context hvm_ctx;
>> };
>>
>> TBH, I don't have an opinion between 2 and 3, but I would like to get a
>> consensus before I start implementing any of those.
> 
> Let me first take a look at how your implementation of 1 looks like.

Ping?