On 04/11/2014 06:58 PM, Jani Nikula wrote: > On Fri, 11 Apr 2014, Aaron Lu <aaron.lu@xxxxxxxxx> wrote: >> On 04/11/2014 04:13 AM, Matthew Garrett wrote: >>> The list of machines in the "Use native backlight" table is getting longer >>> and longer, which is a solid indication that we're doing something wrong. >>> Disable the ACPI backlight interface if the system claims to be Windows 8 >>> or later. >> >> And has a native backlight control interface. > > This. One thing we were doing wrong in i915 was unconditionally > registering a native backlight interface even when it was not > functional. We can't really know whether that is the case except by > trusting the VBT, and although we've been let down in the past, trusting > the VBT on this one is probably a better approximation of the reality > than assuming it always works. Thus we've queued [1] and [2] for 3.15 > which should help. > > [1] http://cgit.freedesktop.org/drm-intel/commit/?id=39fbc9c8f6765959b55e0b127dd5c57df5a47d67 > [2] http://cgit.freedesktop.org/drm-intel/commit/?id=c675949ec58ca50d5a3ae3c757892f1560f6e896 That's pretty cool :-) Regards, Aaron _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
