On Mon, Jan 10, 2000 at 08:25:54AM +0100, Ph. Marek wrote:
> Hi Linus, hi Alan, hi everybody else not explicitly mentioned,
>
> here's a
> patch for testing (comments please cc: to me) and
> inclusion for the 2.3
> series (if believed to be ok).
>
> Short summary:
> More and more programs (eg
> CGI-Scripts) use the PID as pseudo-random number.
> As this isn't really
> random and easily guessed from the outside (esp. with
> daemons loaded while
> booting) it's a little bit troublesome.
> My patch allows some kind of
> modification to the current routing which just
> increments the last pid
> until a free pid is found.
>
> Possible are
> - last_pid+1 (as before);
> -
> complete random (entropy pool);
> - low and/or high byte randomness added;
> -
> multiply last_pid with prime, add another prime;
> - pseudo-random.
>
I did a patch to do this a while ago (it was submitted by Matthew
Kirkwood) and it was rejected. I believe that Alan didn't feel that
it contirbuted much other than hiding /tmp races in badly witten
programs.
Sean
-
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 : Sat Jan 15 2000 - 21:00:17 EST