[PATCH v4 0/3] drm/i915/gvt: optimizaton on shadow context population

[Zhao, Yan Y]

> -----Original Message-----
> From: intel-gvt-dev <intel-gvt-dev-bounces at lists.freedesktop.org> On Behalf
> Of Zhenyu Wang
> Sent: Thursday, April 16, 2020 12:57 PM
> To: Zhao, Yan Y <yan.y.zhao at intel.com>
> Cc: intel-gvt-dev at lists.freedesktop.org
> Subject: Re: [PATCH v4 0/3] drm/i915/gvt: optimizaton on shadow context
> population
> 
> On 2020.04.15 03:53:55 -0400, Yan Zhao wrote:
> > This patchset skips population of shadow contexts if their
> > corresponding guest contexts are known not changed.
> >
> > patch 1 implements the skipping logic, though intentionally disabled
> > the skip.
> > patch 2 checks ggtt entry to verify guest contexts are not changed.
> > patch 3 turns on the skipping logic
> >
> > Yan Zhao (3):
> >   drm/i915/gvt: skip populate shadow context if guest context not
> >     changed
> >   drm/i915/gvt: check ggtt entry modification status for guest ctxs
> >   drm/i915/gvt: turn on shadow context skipping
> >
> >  drivers/gpu/drm/i915/gvt/gtt.c       | 17 ++++++++++++++
> >  drivers/gpu/drm/i915/gvt/gvt.h       |  5 ++++
> >  drivers/gpu/drm/i915/gvt/scheduler.c | 35
> > ++++++++++++++++++++++++----
> >  3 files changed, 52 insertions(+), 5 deletions(-)
> >
> 
> The patch order looks strange to me... either order it as 1) new track struct; 2)
> ggtt update track; 3) actual optimization or just in one patch.
> 

Ok. I prefer one single patch 😊


> --
> Open Source Technology Center, Intel ltd.
> 
> $gpg --keyserver wwwkeys.pgp.net --recv-keys 4D781827