[Intel-gfx] [PATCH 00/15] Atomic watermark updates (v5)

[Zanoni, Paulo R]

Em Qua, 2015-09-30 às 17:20 +0200, Daniel Vetter escreveu:
> On Thu, Sep 24, 2015 at 03:53:05PM -0700, Matt Roper wrote:
> > Previous version of the series was here:
> >   http://lists.freedesktop.org/archives/intel-gfx/2015-September/07
> > 5883.html
> > 
> > Pretty minimal changes since the last series:
> >  * General rebasing on di-nightly
> >  * Some minor SKL-specific bugfixes on patch #6 based on Maarten's
> > review of
> >    v4 of this series.
> >  * Added a new patch #14 to try to sanitize watermarks after
> > hardware state
> >    readout.  Once we've read general state out of the hardware, we
> > should
> >    recalculate what we think the watermarks for that state should
> > be and not
> >    just trust whatever the system firmware happens to have
> > programmed them to.
> >  * Added Maarten's r-b to all patches except the new #14; he gave
> > it on his
> >    feedback to patch #6 of the last revision of the series and I
> > didn't notice
> >    that it applied to the whole series until I re-read his
> > feedback.  :-)
> 
> Pulled in all into dinq, thanks!

And it's causing problems on my SKL machine. It gives me a new
DRM_ERROR about DDB state mismatch, and I'm getting FIFO underruns
sometimes (and due to the underruns, FBC works 0% of the times and
gives us nice black screens).

I tried bisecting the underruns but since they don't happen 100% of the
times I got lost very quickly. It *seems* that "drm/i915: Drop
intel_update_sprite_watermarks" is still a good commit, and the
intermittent problems (and DRM_ERROR) start after that, but I'm not
even 100% sure.

In case you want to try to reproduce the problem, my SKL machine has a
3200x1800 eDP panel.

I'm glad we don't have PRTS to catch these things :)

Thanks,
Paulo

> -Daniel
> 
> > 
> > Matt Roper (14):
> >   drm/i915: Drop redundant watermark programming
> >   drm/i915: Eliminate usage of plane_wm_parameters from ILK-style
> > WM
> >     code (v2)
> >   drm/i915: Eliminate usage of pipe_wm_parameters from ILK-style WM
> > (v2)
> >   drm/i915: Determine I915_MAX_PLANES from plane enum
> >   drm/i915/skl: Simplify wm structures slightly (v2)
> >   drm/i915/skl: Eliminate usage of pipe_wm_parameters from SKL
> > -style WM
> >     (v3)
> >   drm/i915/ivb: Move WaCxSRDisabledForSpriteScaling w/a to atomic
> > check
> >   drm/i915: Drop intel_update_sprite_watermarks
> >   drm/i915: Calculate pipe watermarks into CRTC state (v3)
> >   drm/i915: Calculate ILK-style watermarks during atomic check (v3)
> >   drm/i915: Don't set plane visible during HW readout if CRTC is
> > off
> >   drm/i915: Calculate watermark configuration during atomic check
> > (v2)
> >   drm/i915: Sanitize watermarks after hardware state readout
> >   drm/i915: Add two-stage ILK-style watermark programming (v5)
> > 
> > Ville Syrjälä (1):
> >   drm/i915: Refactor ilk_update_wm (v3)
> > 
> >  drivers/gpu/drm/i915/i915_debugfs.c  |   2 +-
> >  drivers/gpu/drm/i915/i915_drv.h      |  41 +-
> >  drivers/gpu/drm/i915/intel_atomic.c  |   2 +
> >  drivers/gpu/drm/i915/intel_display.c | 255 +++++++++--
> >  drivers/gpu/drm/i915/intel_drv.h     |  85 +++-
> >  drivers/gpu/drm/i915/intel_pm.c      | 855 ++++++++++++++++-------
> > ------------
> >  drivers/gpu/drm/i915/intel_sprite.c  |  15 -
> >  7 files changed, 707 insertions(+), 548 deletions(-)
> > 
> > -- 
> > 2.1.4
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
>