[Intel-gfx] [PATCH 00/11] drm/i915: adjusted_mode.clock vs. port_clock v3

[Daniel Vetter]

On Fri, Sep 06, 2013 at 11:28:57PM +0300, ville.syrjala at linux.intel.com wrote:
> Another day, another attempt.
> 
> This series replaces the earlier attempt:
> [PATCH 0/8] drm/i915: adjusted_mode.clock vs. port_clock v2 
> 
> The main new idea is to make it the encoders' responsibility to calculate
> adjusted_mode.clock. In order to do that I had to add DP M/N extraction and
> i9xx_crtc_clock_get() had to be improved so that it can dig out the PCH DPLL
> frequency. The rest is more or less the same old.
> 
> As a bonus we now get to cross check the FDI's idea of dotclock against the
> encoder's idea.
> 
> Also now CTG too can calculate the dotclock from the DP port_clock. I don't
> think that could have worked before as the DPLL was running at the DP link
> frequency, so we read that out into adjusted_mode.clock and then compared
> it with whatever the user requested. Not the same thing at all. But now it
> should just work (tm).
> 
> I gave it a quick whirl on my ILK eDP + DP, and IVB HDMI and DP. Neither
> machine started to yell at me, so I think it's in a semi-decent shape at
> least. Unfortunately I don't have a CTG w/ DP so I can't test that part,
> nor do I have any SDVO/DVO hardware.

Ok, I've read through this a bit and commented on the oddball cases where
I expect pipe config mismatches might happen. Otherwise I guess we can
just merge this and then fix any oddball fallout - it's early for 3.13
after all ;-)

Cheers, Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch