On Fri, May 06, 2016 at 09:37:25AM +0100, Chris Wilson wrote: > On Fri, May 06, 2016 at 10:28:41AM +0200, Daniel Vetter wrote: > > Also, the unbind function is totally not sufficient, and it's calling > > the deprecated drm_put_dev. > > diff --git a/include/drm/drmP.h b/include/drm/drmP.h > index 360b2a74e1ef..7efc0b477e38 100644 > --- a/include/drm/drmP.h > +++ b/include/drm/drmP.h > @@ -1032,7 +1032,7 @@ extern void drm_vblank_post_modeset(struct drm_device *dev, unsigned int pipe); > extern struct drm_master *drm_master_get(struct drm_master *master); > extern void drm_master_put(struct drm_master **master); > > -extern void drm_put_dev(struct drm_device *dev); > +__deprecated extern void drm_put_dev(struct drm_device *dev); > extern void drm_unplug_dev(struct drm_device *dev); > extern unsigned int drm_debug; > extern bool drm_atomic; > > And you probably should markup all the others that are now deprecated and > being phased out. > > Maybe behind drm_deprecated so that the spam is hidden behind another > kconfig option rather than enabled-by-buildbots? I thought about this, but the problem is that we still have plenty drivers using them. And hiding behind a build option is a bit meh, or at least we'd need me and Dave to enable that option to make sure nothing slips through. Dave, do you think such an option would be useful? You'd need to integrate it into your build testing when pulling drivers to make sure we catch it all? Cheers, Daniel -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/dri-devel
