https://bugzilla.mindrot.org/show_bug.cgi?id=3269
--- Comment #5 from Darren Tucker <[email protected]> --- Created attachment 3475 --> https://bugzilla.mindrot.org/attachment.cgi?id=3475&action=edit explicitly use uint32 for enum in monitor (In reply to Stephen Goetze from comment #3) > When I switch to c89, the problem persists, so it appears that IBM > doesn't downshift their treatment of enumerations when running in > c89 mode. That sounds like a compiler bug. > You may want to make a note of this new behavior in c99 if you ever > decide to specify a newer standard for building; the potential > exists for a nasty bug. I think we should still do something about this. We're slowly using C99 features (eg recently, variadic macros) and I don't think it's strictly correct for C89 either: the same problem could occur on a platform where int<32bits (although I know of no such platform that it'll currently run on) or int>32 bits (which is plausible, and I'd imagine whether or not it would work properly would depend on the endianness). Does this patch fix it for you? -- You are receiving this mail because: You are watching someone on the CC list of the bug. You are watching the assignee of the bug. _______________________________________________ openssh-bugs mailing list [email protected] https://lists.mindrot.org/mailman/listinfo/openssh-bugs
