The stringprep_utf8_nfkc_normalize function in lib/nfkc.c in libidn before 1.33 allows context-dependent attackers to cause a denial of service (out-of-bounds read and crash) via crafted UTF-8 data.
https://security.gentoo.org/glsa/201908-06
https://lists.gnu.org/archive/html/help-libidn/2016-07/msg00009.html
http://www.ubuntu.com/usn/USN-3068-1
http://www.securityfocus.com/bid/92070
http://www.openwall.com/lists/oss-security/2016/07/21/4
http://www.openwall.com/lists/oss-security/2016/07/20/6
http://www.debian.org/security/2016/dsa-3658
http://lists.opensuse.org/opensuse-updates/2016-08/msg00098.html
http://lists.opensuse.org/opensuse-updates/2016-08/msg00005.html
http://git.savannah.gnu.org/cgit/libidn.git/commit/?id=1fbee57ef3c72db2206dd87e4162108b2f425555