On Tue, May 12, 2015 at 02:41:17PM +0200, Maarten Lankhorst wrote: > Op 12-05-15 om 08:59 schreef Daniel Vetter: > > On Mon, May 11, 2015 at 04:24:45PM +0200, Maarten Lankhorst wrote: > >> This should be safe. > > Usual request: A few more details about what you've changed to help guide > > the review would be great. E.g. which functions from the atomic helpers > > you're trying to copy here exactly. > That's in the diff. :-) Indeed ;-) > > It looks like this models set_routing_links. I think it would be rather > > useful to expose this to drivers as a helper function, maybe with a more > > useful name like drm_atomic_helper_update_legacy_state or similar. > I think you're right about this one. > > Another thing I've noticed is that atomic helpers lost the call to > > drm_calc_timestamping_constants. Would be good to add that to the same > > function. > Yeah, but the precise vblanking stuff is not really atomic friendly, > so I'm not sure it should be done outside the driver. Hm what's unfriendly about it? It's definitely a regression to not update the constants needed by it compared to crtc helpers. I'll go about and fix that now. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
