[Intel-gfx] [PATCH 1/1] drm/i915: Move i915_gem_restore_fences to i915_gem_resume

[Chris Wilson]

Quoting Joonas Lahtinen (2017-09-29 08:08:31)
> On Fri, 2017-09-29 at 10:28 +0530, Sagar Arun Kamble wrote:
> > i915_gem_restore_fences is GEM resumption task hence it is moved to
> > i915_gem_resume from i915_restore_state.
> > 
> > Signed-off-by: Sagar Arun Kamble <sagar.a.kamble at intel.com>
> > Cc: Michal Wajdeczko <michal.wajdeczko at intel.com>
> > Cc: Michał Winiarski <michal.winiarski at intel.com>
> > Cc: Chris Wilson <chris at chris-wilson.co.uk>
> > Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
> 
> Yep, this is a more appropriate place. Locality to the display resume
> was probably reason for the odd placement previously (it was added to
> fix a bug of fences not being there for display resume).
> 
> Do we want to comment this requirement between i915_gem_resume and
> i915_restore_state? Chris?

I'd forgotten about restore_state. That's old school save-the-world,
restore-the-world for UMS code. Keep whittling it down and eventually it
will go away, everything we do now should be independent of
i915_suspend.c...
 
> Anyway,
> 
> Reviewed-by: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>

Added to my queue for pushing, thanks for the patch and review.

> Regards, Joonas
> 
> PS. My guess seems to be correct according to git blame (skipping one
> rename in between) info.

Yes, debugging fences across resume has always been a case of fixing
display corruption. :|
-Chris