This mostly stems from

 https://bugs.freedesktop.org/show_bug.cgi?id=76554

which is a super-tricky bug even Intel has given up on. The problem is that noone knows what behavior is expected by HW, and it's mostly a guess game. The only thing which is for sure is that old kernels (3.7?) have absolutely positively never ever exposed the problem. Still, it's not bisectable at all due to the unreliable nature of the bug triggering.

I'd like to dedicate hackweek to gain better understanding of the problem and perhaps ultimately fix it at the end of the day, as it's annoying a LOT of people with GM45 mobile i915 chipset (including myself).