> It's not that hard. You should recompile only applications with IPC usage.
Indeed, I just wonder about the possibility of a process getting a
pid of 40K or thereabouts from an IPC structure -- then sign
extending it and using it for kill(2)
Ouch.
> No. It will not happen. If default will not be >32767 then even 10
> years later we'll have "developers" who will be asking "How so ?
> PIDs > 32767 are suddenly allowed ? We did not hear anything about
> this :-(". Latest version of Adobe Acrobat are still libc5-based,
> for example :-/
Hmm... how about we fix the pid_t width in the IPC structure if it
has not been done so already, then in 2.5.x we can raise the limit;
that way -- 2.5.x testers will find these problems and hopefully fix
them, whether or not authors do. This is what I was alluding too.
Commercial vendors will need more prodding, and sadly because even
the glibc-2.1.2 IPC structures need updating, things like Oracle
_will_ break...
-cw
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Jan 23 2000 - 21:00:14 EST