Re: [PATCH v4 01/11] pvh dom0: Introduce p2m_map_foreign

[Julien Grall <julien.grall@linaro.org>]

On 12/16/2013 10:59 AM, Tim Deegan wrote:
> At 10:45 +0000 on 16 Dec (1387187114), Ian Campbell wrote:
>> On Sat, 2013-12-14 at 00:00 +0000, Julien Grall wrote:
>>>
>>> On 12/13/2013 11:19 PM, Ian Campbell wrote:
>>>> On Fri, 2013-12-13 at 19:37 +0000, Julien Grall wrote:
>>>>> From: Mukesh Rathor <mukesh.rathor@oracle.com>
>>>>>
>>>>> In this patch, a new type p2m_map_foreign is introduced for pages
>>>>> that toolstack on PVH dom0 maps from foreign domains that its creating
>>>>> or supporting during it's run time.
>>>>>
>>>>> Signed-off-by: Mukesh Rathor <mukesh.rathor@oracle.com>
>>>>> Cc: Jan Beulich <jbeulich@suse.com>
>>>>> Cc: Keir Fraser <keir@xen.org>
>>>>> Cc: George Dunlap <george.dunlap@eu.citrix.com>
>>>>>
>>>>> ---
>>>>>       Changes in v4:
>>>>>           - Use the patch #4 from dom0 pvh v6 series. We need it for ARM
>>>>
>>>> Is this really the right patch? I'd have thought it would need to
>>>> include some generic/!x86 bits but:
>>>
>>> This patch introduce p2m_map_foreign type. I think it's safe to push it
>>> without the other patches.
>>
>> Is it really that simple? I suppose there is nothing to create a foreign
>> mapping on x86 yet so this is harmless?
>
> Presumably you're about to add some common code that creates foreign
> mappings -- otherwise there would be no need to take the x86 patch.
> In that case, it's not really safe to take only part of the x86
> implementation.  It'd be better to stub out any x86 interfaces with
> explicit errors, to be removed when x86 supports this properly.

This patch is no longer necessary. I have reworked my patch #8 to remove 
modification in common code.

-- 
Julien Grall