From #2608 (comment):
Failing debuginfo/nested_cdb.d on Win64 (most likely Win32 too) also appears to be a CodeView debuginfo regression (CodeView support is supposed to be greatly improved by v6); once again related to the GEPs we use for nested variables (inner functions only, i.e., where the GEP stems from the bitcast context pointer argument; no problem in the outer function, where it's a GEP from the frame alloca).
The issue is a wrong number of indirections IIRC, so the values are garbage.
Revert f1712b2 when it's fixed.