------- Additional Comments From rguenth at gcc dot gnu dot org 2005-05-19 20:20 ------- None. But I have to deal with the possibilities of these during folding of &a[2] == &a[1]. At least until the middle-end makes a promise that they cannot exist (which they do), or the C frontend rejects this. It's really a question for advice on details of our own extension to the C language.
Richard. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21664
