Re: [PATCH v2 2/2] drm/i915: Move GEM BO inside drm_framebuffer

["Ville Syrjälä" <ville.syrjala@linux.intel.com>]

On Fri, May 18, 2018 at 02:48:44PM +0100, Daniel Stone wrote:
> Since drm_framebuffer can now store GEM objects directly, place them
> there rather than in our own subclass.
> 
> v2: Only hold a single reference per framebuffer, not per plane. (Ville)
> 
> Signed-off-by: Daniel Stone <daniels@collabora.com>
> Cc: Ville Syrjälä <ville.syrjala@intel.com>
> Cc: Jani Nikula <jani.nikula@linux.intel.com>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Cc: intel-gfx@lists.freedesktop.org
> ---
>  drivers/gpu/drm/i915/intel_display.c | 4 ++--
>  drivers/gpu/drm/i915/intel_drv.h     | 3 +--
>  2 files changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 1b2cf631305e..12226a2c8d39 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -14370,9 +14370,9 @@ static int intel_framebuffer_init(struct intel_framebuffer *intel_fb,
>  				      i, fb->pitches[i], stride_alignment);
>  			goto err;
>  		}
> -	}
>  
> -	intel_fb->obj = obj;
> +		fb->obj[i] = &obj->base;
> +	}
>  
>  	ret = intel_fill_fb_info(dev_priv, fb);
>  	if (ret)
> diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
> index 12002fc77235..03e1d1d7fb58 100644
> --- a/drivers/gpu/drm/i915/intel_drv.h
> +++ b/drivers/gpu/drm/i915/intel_drv.h
> @@ -194,7 +194,6 @@ enum intel_output_type {
>  
>  struct intel_framebuffer {
>  	struct drm_framebuffer base;
> -	struct drm_i915_gem_object *obj;
>  	struct intel_rotation_info rot_info;
>  
>  	/* for each plane in the normal GTT view */
> @@ -1005,7 +1004,7 @@ struct cxsr_latency {
>  #define to_intel_framebuffer(x) container_of(x, struct intel_framebuffer, base)
>  #define to_intel_plane(x) container_of(x, struct intel_plane, base)
>  #define to_intel_plane_state(x) container_of(x, struct intel_plane_state, base)
> -#define intel_fb_obj(x) (x ? to_intel_framebuffer(x)->obj : NULL)
> +#define intel_fb_obj(x) (((x) && (x)->obj[0]) ? to_intel_bo((x)->obj[0]) : NULL)

We don't need the obj[0] null check. For most things we just assume
that the base object is at offset 0. And in case of drm_i915_gem_object
it looks like we even have a BUILD_BUG_ON() to make sure. So you may
want to drop that part.

Series is
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

>  
>  struct intel_hdmi {
>  	i915_reg_t hdmi_reg;
> -- 
> 2.17.0
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Ville Syrjälä
Intel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx