https://sourceware.org/bugzilla/show_bug.cgi?id=33549
--- Comment #1 from Nick Alcock <nick.alcock at oracle dot com> --- Why would an invalid size be input? Feeding hand-written non-CTF garbage to libctf is not in its threat model (at least, not yet). CTFv4, in preparation (but huge), will add a lot more verification in this area (and add a bunch of range-checking to the archive code), but right now *entirely syntactically valid* but semantically invalid CTF (e.g. containing loops in the type graph) can cause the deduplicator to recurse infinitely or loop forever. -- You are receiving this mail because: You are on the CC list for the bug.
