Re: Intel announcements at RSA '99

New Message Reply About this list Date view Thread view Subject view Author view

David R. Conrad (drc@adni.net)
Thu, 21 Jan 1999 19:47:02 -0500 (EST)


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

On Wed, 20 Jan 1999, Steve Bellovin wrote:

> Intel has announced a number of interesting things at the RSA conference.
> The most important, to me, is the inclusion of a hardware random number
> generator (based on thermal noise) in the Pentium III instruction set.
> They also announced hardware support for IPSEC.

Doesn't seem to me that the new features are of much use to anyone. As
others have pointed out, it's quite difficult to assure oneself that the
RNG is true and not a fair PRNG in disguise. The code for, for instance,
the linux /dev/random driver will probably change slightly:

initialization {
        if (cpu flags AND rng flag) {
                set rng_opt
        }
        ...
}

adding new entropy {
        if (rng_opt) {
                get 32-bit allegedly true random value
                mix it into the pool
        }
        ...
}

A nice, but not excessively exciting feature. (I've been wondering why
intel didn't do this for some time now, but now that it's coming, I just
can't seem to work up any enthusiasm over it.)

On the other hand, the serial number is going to stir things up a bit.
But in the end, it's just as irrelevent. Theft prevention? What software
will publish the number to the net? It can be overridden, and/or the CPU
can be replaced.

Copy protection? Will the instruction to read the serial number be
privileged, or not? If it is, then the OS can lie to the application
about the number. Remember SETVER for MS-DOS? But even if this worked,
it would be just as good to have each installation of the OS make up a
random 128-bit number at install time and use that as its serial number.
Perhaps using the hardware RNG!

On the other hand, if the MOV EAX, SERNO instruction *isn't* privileged
then any process running on the machine can read it. So you can forget
about using it for anything even remotely secure. (And even if it isn't
privileged, there may be ways to 'make' it privileged. Witness the hacks
to solve the F00F bug.)

P.S. I hope the serial number doesn't fit in EAX. Does anyone have
details on how long it is? And what the interface to the RNG is like?
In my pseudocode above I assumed one would grab 32 random bits at a time,
but that's all that was: an assumption.

David R. Conrad <drc@adni.net>
This is why I love America -- that any kid can dream "I'm going to get
naked with the President" ... and that dream can actually come true.
What a great country! -- Michael Moore

-----BEGIN PGP SIGNATURE-----
Version: PGPfreeware 5.0i for non-commercial use
Charset: noconv

iQA/AwUBNqfKloPOYu8Zk+GuEQJ30ACbB0gDv42YEdHNTlvOWXTzc2smxUoAoO9z
0ymXe6+zZ3/u7uHTFZr5b5yt
=0cHK
-----END PGP SIGNATURE-----


New Message Reply About this list Date view Thread view Subject view Author view

 
All trademarks and copyrights are the property of their respective owners.

Other Directory Sites: SeekWonder | Directory Owners Forum

The following archive was created by hippie-mail 7.98617-22 on Sat Apr 10 1999 - 01:18:04