On Thu, 22 Nov 2012 18:20:27 +0200, Jani Nikula <jani.nikula at intel.com> wrote:
> On Wed, 21 Nov 2012, Chris Wilson <chris at chris-wilson.co.uk> wrote:
> >
> > mode = drm_mode_duplicate(dev, lvds_connector->base.panel.fixed_mode);
> > if (mode == NULL)
> > @@ -1065,8 +1060,6 @@ bool intel_lvds_init(struct drm_device *dev)
> > kfree(edid);
> > edid = ERR_PTR(-EINVAL);
> > }
> > - } else {
> > - edid = ERR_PTR(-ENOENT);
>
> I'd like to keep this though. I was thinking a more generic, future EDID
> caching feature might use the difference between -EINVAL and -ENOENT,
> NULL being the "we just haven't read it yet" flag, to prevent trying to
> re-read the EDID when it isn't there. *shrug*.
Okay, that's a reasonable argument. Daniel feel free to drop this hunk.
-Chris
--
Chris Wilson, Intel Open Source Technology Centre
