On Mon, Nov 18, 2013 at 10:14:26PM +0200, Ville Syrjälä wrote: > On Mon, Nov 18, 2013 at 09:00:58AM +0100, Daniel Vetter wrote: > > We need this to properly fill in adjusted_mode.crtc_clock, otherwise > > the state checker gets unhappy. This seems to have been forgotten in > > the big clock rework in > > > > commit 18442d08786472c63a0a80c27f92b033dffc26de > > Author: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > Date: Fri Sep 13 16:00:08 2013 +0300 > > > > drm/i915: Fix port_clock and adjusted_mode.clock readout all over > > > > Cc: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> > > Cc: Jani Nikula <jani.nikula@xxxxxxxxx> > > Signed-off-by: Daniel Vetter <daniel.vetter@xxxxxxxx> > > For the series: > Reviewed-by: Ville Syrjälä <ville.syrjala@xxxxxxxxxxxxxxx> Thanks for the review, patches merged to -fixes > Although using adjusted_mode.crtc_clock in intel_tv compute_config and > get_config is a bit wrong I think. That's not really the pixel clock > we're shoveling into it, so we're going to be computing the watermarks > incorrectly. > > To do it really right, I think we should stick the tv_mode clock to > port_clock, and then compute the pixel rate based on the input mode > and the refresh rate. Or maybe we just need a TV out specific version > of ilk_pipe_pixel_rate() (just like we'd need one for GMCH panel fitter). > The TV out scaler is essentially just another panel fitter anyway. Yeah, it's just duct-tape over duct-tape at this point. But meh, it's tv-out. I have a similar series bubbling for sdvo-tv, but that's still stalled since my g33 decided to be a bit more hang-happy than I'd prefer. -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
