[Intel-gfx] [PATCH] drm/i915: Workaround CHV pipe C cursor fail

[Daniel Vetter]

On Mon, Dec 21, 2015 at 02:49:18PM +0100, Daniel Vetter wrote:
> On Fri, Dec 18, 2015 at 07:24:39PM +0200, ville.syrjala at linux.intel.com wrote:
> > From: Ville Syrjälä <ville.syrjala at linux.intel.com>
> > 
> > Turns out CHV pipe C was glued on somewhat poorly, and there's something
> > wrong with the cursor. If the cursor straddles the left screen edge,
> > and is then moved away from the edge or disabled, the pipe will often
> > underrun. If enough underruns are triggered quickly enough the pipe
> > will fall over and die (it just scans out a solid color and reports
> > a constant underrun). We need to turn the disp2d power well off and
> > on again to recover the pipe.
> > 
> > None of that is very nice for the user, so let's just refuse to place
> > the cursor in the compromised position. The ddx appears to fall back
> > to swcursor when the ioctl returns an error, so theoretically there's
> > no loss of functionality for the user (discounting swcursor bugs).
> > I suppose most cursors images actually have the hotspot not exactly
> > at 0,0 so under typical conditions the fallback will in fact kick in
> > as soon as the cursor touches the left edge of the screen.
> > 
> > Any atomic compositor should anyway be prepared to fall back to
> > GPU composition when things don't work out, so there should be no
> > problem with those.
> > 
> > Other things that I tried to solve this include flipping all
> > display related clock gating knobs I could find, increasing the
> > minimum gtt alignment all the way up to 512k. I also tried to see
> > if there are more specific screen coordinates that hit the bug, but
> > the findings were somewhat inconclusive. Sometimes the failures
> > happen almost across the whole left edge, sometimes more at the very
> > top and around the bottom half. I wasn't able to find any real pattern
> > to these variations, so it seems our only choice is to just refuse
> > to straddle the left screen edge at all.
> > 
> > Cc: stable at vger.kernel.org
> > Cc: Jason Plum <max at warheads.net>
> > Testcase: igt/kms_chv_cursor_fail
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=92826
> > Signed-off-by: Ville Syrjälä <ville.syrjala at linux.intel.com>
> 
> Well can't really review with Bspec and I don't expect we can get
> confirmation from hw engineers. But does what it says in the commit
> message.
> 
> Reviewed-by: Daniel Vetter <daniel.vetter at ffwll.ch>

And applied to dinq since that's our new flow even for fixes.
-Daniel

> 
> > ---
> >  drivers/gpu/drm/i915/intel_display.c | 17 +++++++++++++++++
> >  1 file changed, 17 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> > index abd2d2944022..8acc66b95139 100644
> > --- a/drivers/gpu/drm/i915/intel_display.c
> > +++ b/drivers/gpu/drm/i915/intel_display.c
> > @@ -14059,6 +14059,7 @@ intel_check_cursor_plane(struct drm_plane *plane,
> >  	struct drm_crtc *crtc = crtc_state->base.crtc;
> >  	struct drm_framebuffer *fb = state->base.fb;
> >  	struct drm_i915_gem_object *obj = intel_fb_obj(fb);
> > +	enum pipe pipe = to_intel_plane(plane)->pipe;
> >  	unsigned stride;
> >  	int ret;
> >  
> > @@ -14092,6 +14093,22 @@ intel_check_cursor_plane(struct drm_plane *plane,
> >  		return -EINVAL;
> >  	}
> >  
> > +	/*
> > +	 * There's something wrong with the cursor on CHV pipe C.
> > +	 * If it straddles the left edge of the screen then
> > +	 * moving it away from the edge or disabling it often
> > +	 * results in a pipe underrun, and often that can lead to
> > +	 * dead pipe (constant underrun reported, and it scans
> > +	 * out just a solid color). To recover from that, the
> > +	 * display power well must be turned off and on again.
> > +	 * Refuse the put the cursor into that compromised position.
> > +	 */
> > +	if (IS_CHERRYVIEW(plane->dev) && pipe == PIPE_C &&
> > +	    state->visible && state->base.crtc_x < 0) {
> > +		DRM_DEBUG_KMS("CHV cursor C not allowed to straddle the left screen edge\n");
> > +		return -EINVAL;
> > +	}
> > +
> >  	return 0;
> >  }
> >  
> > -- 
> > 2.4.10
> > 
> > _______________________________________________
> > Intel-gfx mailing list
> > Intel-gfx at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/intel-gfx
> 
> -- 
> Daniel Vetter
> Software Engineer, Intel Corporation
> http://blog.ffwll.ch

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch