On Tue, Mar 26, 2013 at 04:24:48PM -0700, Jesse Barnes wrote:
> On Wed, 20 Mar 2013 14:36:25 +0200
> Imre Deak <imre.d...@intel.com> wrote:
>
> > > + pipe = -1;
> > >
> > > if (encoder->get_hw_state(encoder, &pipe)) {
> > > - encoder->base.crtc =
> > > - dev_priv->pipe_to_crtc_mapping[pipe];
> > > - } else {
> > > + crtc =
> > > to_intel_crtc(dev_priv->pipe_to_crtc_mapping[pipe]);
> >
> > get_hw_state() can return true and still leave pipe at -1, causing an
> > invalid access here.
>
> Fixed, thanks.
That shouldn't actually be possible - if encoder->get_hw_state returns
true but doesn't set the pipe, that's a bug in the callback. So maybe yell
with a WARN?
-Daniel
--
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx
