https://gcc.gnu.org/bugzilla/show_bug.cgi?id=19430
--- Comment #32 from Manuel López-Ibáñez <manu at gcc dot gnu.org> --- (In reply to Vincent Lefèvre from comment #31) > In any case, no warnings are generated. So, the problem here is not related > to whether the address of j is taken, but to something else. With a constant, you hit PR18501. If you use "j = i", it warns as it should.
