On Wed, Oct 24, 2012 at 04:45:04PM +0100, Lespiau, Damien wrote: > On Wed, Oct 24, 2012 at 1:38 PM, Daniel Vetter <daniel at ffwll.ch> wrote: > > On Tue, Oct 23, 2012 at 06:29:58PM -0200, Paulo Zanoni wrote: > >> From: Paulo Zanoni <paulo.r.zanoni at intel.com> > >> > >> We need to check if any of the pipes is using TRANSCODER_EDP. > >> > >> Signed-off-by: Paulo Zanoni <paulo.r.zanoni at intel.com> > > > > One thing that irks me is that you add new state readout code here, but > > don't call the same code in the modeset_check_state code > > Isn't what the previous patch introduce? (look at the > intel_ddi_get_hw_state() hunk in the patch 07/18 of the series). > AFAICS we're are already checking if the pipe returned by the encoder > is what we think it is. In a way, yes. But in the best case we should have the same code in both places (now it's copied), and also should be able to check which parts of the hw we actually need (since the get_hw_state now hides the cpu_transcoder in between the ddi and the pipe that's not so easy to do). So I still think we should try to unify/improve this a bit, especially since for the other fastboo stuff we need more cleverness for pipe state anyway (otherwise handling plls&friends will be a pain). -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch
