[PATCH] drm: Use atomic state for FB in legacy ioctls

[Daniel Vetter]

On Tue, Dec 13, 2016 at 11:14:47PM +0200, Ville Syrjälä wrote:
> On Tue, Dec 13, 2016 at 07:02:53PM +0000, Daniel Stone wrote:
> > Hi,
> > 
> > > On 13 Dec 2016, at 6:48 pm, Ville Syrjälä <ville.syrjala at linux.intel.com> wrote:
> > > 
> > >> On Tue, Dec 13, 2016 at 06:19:12PM +0000, Daniel Stone wrote:
> > >> @@ -357,7 +357,10 @@ int drm_mode_getcrtc(struct drm_device *dev,
> > >> 
> > >>    drm_modeset_lock_crtc(crtc, crtc->primary);
> > >>    crtc_resp->gamma_size = crtc->gamma_size;
> > >> -    if (crtc->primary->fb)
> > >> +
> > >> +    if (crtc->primary->state && crtc->primary->state->fb)
> > >> +        crtc_resp->fb_id = crtc->primary->state->fb->base.id;
> > >> +    else if (!crtc->primary->state && crtc->primary->fb)
> > >>        crtc_resp->fb_id = crtc->primary->fb->base.id;
> > > 
> > > I think what we do elsewhere is totally ignore the legacy junk if the
> > > ->state pointer exists.
> > 
> > Indeed, hence the negative state check on the second branch; having nested if statements instead seemed unnecessarily unwieldy, but the effect is the same.
> 
> My bad. You hid it well though ;)
> 
> Reviewed-by: Ville Syrjälä <ville.syrjala at linux.intel.com>

Applied to drm-misc, thanks.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch