[Intel-gfx] [PATCH 14/15] drm/i915: Track frontbuffer invalidation/flushing

[Chris Wilson]

On Mon, Jun 16, 2014 at 07:51:34PM +0200, Daniel Vetter wrote:
>  void i915_gem_release(struct drm_device *dev, struct drm_file *file)
> diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> index 93d7f7246588..2d06aad4a954 100644
> --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> @@ -978,7 +978,7 @@ i915_gem_execbuffer_move_to_active(struct list_head *vmas,
>  			/* check for potential scanout */
>  			if (i915_gem_obj_ggtt_bound(obj) &&
>  			    i915_gem_obj_to_ggtt(obj)->pin_count)

What is it about these checks you like so much? You haven't taken the
patches to convert this to if (obj->pin_display) and you haven't
converted them to obj->frontbuffer_bits in this patch either!

> -				intel_mark_fb_busy(obj, ring);
> +				intel_fb_invalidate(obj, ring);
>  
>  			/* update for the implicit flush after a batch */
>  			obj->base.write_domain &= ~I915_GEM_GPU_DOMAINS;

-- 
Chris Wilson, Intel Open Source Technology Centre