On Fri, May 30, 2014 at 7:40 AM, Dave Airlie <airlied@xxxxxxxxx> wrote: > Just wondering what the point of the WARN(!encoder->crtc) in that function is, > > I hit this with MST and I can't see what it should matter, where I hit > it is if I dock MST while X is running and VT switch, I get it, > > I first wondered when encoder would ever be false in non-MST world > anyways, but I've added a check to find a valid MST encoder at this > point, but none of them have a crtc assigned as of yet as they haven't > been configured, even though I expect X has configured them, fbdev > would turn them off on vt switch I assume. If drm_connector->encoder is set, then drm_encoder->crtc must also be set. Otherwise the hw state readout/sanitize step has left behind something which doesn't make sense. One of your older version (haven't checked the latest) used sw state in the get_hw_state functions, that might explain this if you know have a mix of them ;-) -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
