[Intel-gfx] [RFCv2 02/14] drm/i915/gvt: Introduce the basic architecture of GVT-g

[Tian, Kevin]

> From: Joonas Lahtinen [mailto:joonas.lahtinen at linux.intel.com]
> Sent: Tuesday, February 23, 2016 8:42 PM
> 
> Hi,
> 
> Code is looking a lot better.
> 
> A general question still; why you seem to be preferring multi-stage
> alloc and destroy?

One reason for multi-stage, IMO, is that GVT needs to do a snapshot
of initial MMIO state before i915 driver does actual initialization. That
snapshot will be presented to each VM which can then observe same
state as it would observe on a bare metal. Then the majority of
initialization needs to wait after i915 driver completes initialization. 
Zhi can correct me here.

Thanks
Kevin