[Intel-gfx] [PATCH] drm/i915: replace snb_update_wm with haswell_update_wm on HSW

[Daniel Vetter]

On Fri, May 24, 2013 at 12:05:15PM -0300, Paulo Zanoni wrote:
> 2013/5/22 Ville Syrjälä <ville.syrjala at linux.intel.com>:
> > On Tue, May 21, 2013 at 06:24:38PM -0300, Paulo Zanoni wrote:
> >> Hi
> >>
> >> 2013/5/20 Ville Syrjälä <ville.syrjala at linux.intel.com>:
> >> > On Thu, May 09, 2013 at 05:13:41PM -0300, Paulo Zanoni wrote:
> >> >> From: Paulo Zanoni <paulo.r.zanoni at intel.com>
> >> >> +     /* We only use IF-ID interlacing. If we ever use PF-ID we'll need to
> >> >> +      * adjust the pixel_rate here. */
> >> >> +
> >> >> +     pfit_size = intel_crtc->config.pch_pfit.size;
> >> >> +     if (pfit_size) {
> >> >> +             int x, y, crtc_x, crtc_y, hscale, vscale, totscale;
> >> >> +
> >> >> +             x = (pfit_size >> 16) & 0xFFFF;
> >> >> +             y = pfit_size & 0xFFFF;
> >> >> +             crtc_x = intel_crtc->config.adjusted_mode.hdisplay;
> >> >> +             crtc_y = intel_crtc->config.adjusted_mode.vdisplay;
> >> >> +
> >> >> +             hscale = crtc_x * 1000 / x;
> >> >> +             vscale = crtc_y * 1000 / y;
> >> >
> >> > That should be 'requested_mode / pfit_size'
> >>
> >> Why? What's wrong with the current code?
> >
> > pfit_size is the panel fitter output size, requested_mode is the input
> > size (pipesrc).
> 
> If we use adjusted_mode we'll get the panel's native mode, if we use
> requested_mode we'll get the mode seen by xrandr. My interpretation by
> checking Haswell Watermarks Calculator and the BSpec description is
> that I should use adjusted_mode. And I'm already using pfit_size, so
> no need to discuss it. Besides this, I believe I have implemented all
> the other suggestions.

pfit_size stores the request mode, which matches crtc->config.requested
mode. Instead of jumping through hoops I think I'd be better to directly
use those values ...
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch