-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi! :)

On Thursday, October 03, 2002 13:56, Wei Dai wrote:
> On Wed, Oct 02, 2002 at 10:09:11PM -0400, Hezekiah wrote:
> >     Notice the HUGE difference in the number of bytes generated. Is this a
> > bug in Crypto++, or is there some weird condition that makes the
> > operating system random number generator go weird when the CPU maxes out?
> > Either way, should I be worried about this affecting the quality of the
> > cryptography?
>
> That is wierd. The test that failed tries to make sure that the blocking
> RNG's (in this case /dev/random) entropy estimation isn't totally off. It
> seems unlikely that /dev/random was actually able to gather 14588
> bytes of entropy in 2 seconds, so I'd say there might be a bug in the
> /dev/random driver.

        Yeah. This IS weird! :) There might be a bug in the /dev/random driver, but 
is it possible that the amount /dev/random returned got mangled someplace 
along the way? (Just a thought. I could be [and probably am] way off!)

> Can you look in the source code for your OS, find out
> the person who wrote the /dev/random driver, and ask him to look into
> this?

        Well, I looked at the source for /dev/random for Linux. I've reproduced the 
information on the author below:

/*
 * random.c -- A strong random number generator
 *
 * Version 1.89, last modified 19-Sep-99
 *
 * Copyright Theodore Ts'o, 1994, 1995, 1996, 1997, 1998, 1999.  All
 * rights reserved.

        That's all that I saw. The only other names in the begining comment block are 
just credits. Since Mr. Ts'o didn't give us an email address, I don't know 
how to contact him (though perhaps it could be dug up with a search through 
the kernel archives.)
        Even if we could contact him, I would suggest that you talked to him. I will 
be the first to admit that I am NO crypto genius! IF there really is a bug in 
/dev/random, then I think the two of you could hunt it down a lot faster than 
I could! If this helps any, I'm running a patched version of the Linux 2.4.19 
kernel. I've also noticed this same thing under 2.4.18 as well.

        If there is anything I else I can do to help, just ask. I can't promise that 
I'll be able to do it (as I said, I'm no crypto genius), but I'm more than 
willing to try! :)

        Sincerely,
          Hezekiah

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9nPcdicjSr4uiPQERApRlAJ9yUpj+7NhKP8F10UZSmrGKti489QCfZjqv
Z1MLesxBSZQ+jAdg5ARdXkQ=
=Ex4B
-----END PGP SIGNATURE-----

Reply via email to