[Intel-gfx] [RFC 3/5] drm/i915: Infrastructure for supporting different GGTT views per object

[Tvrtko Ursulin]

On 10/30/2014 04:41 PM, Chris Wilson wrote:
> On Thu, Oct 30, 2014 at 04:39:36PM +0000, Tvrtko Ursulin wrote:
>> From: Tvrtko Ursulin <tvrtko.ursulin at intel.com>
>>
>> Things like reliable GGTT mappings and mirrored 3d display will need to be
>> to map the same object twice into the GGTT.
>
> What's a reliable GGTT mapping and how does it differ from today's
> notoriously unreliable mappings?

Ask Daniel. :P But you are probably aiming at the short commit message, 
which I justify with the RFC tag. :)

As far as I understand it idea is to allow mapping a subset of an object 
to work around the 256Mb limit of the mappable region. How it was 
described sounded a bit like VM inside a VM but I am more interested in 
the infrastructure than this particular justification.

Regards,

Tvrtko