PATCH 21.5 This fixes the only other complaint I'm getting from valgrind about something of the form memcpy(&x, &x, sizeof(x));. Twice in the redisplay code, we try to optimize by building a dynarray entry in place, then throw it all away by doing Dynarr_add, which copies the bytes from where they are to ... um, where they are. The fix is to use Dynarr_increment in that case instead to bump up the size of the dynarray to include the new entry. I have run with this for a couple of hours today, but have not given it extensive testing. -- Jerry James http://loganjerry.googlepages.com/ _______________________________________________ XEmacs-Patches mailing list XEmacs-Patches(a)xemacs.org http://calypso.tux.org/cgi-bin/mailman/listinfo/xemacs-patches