Re: Intel 810 Random Number Generator

From: Khimenko Victor (khim@sch57.msk.ru)
Date: Tue Jan 25 2000 - 00:07:15 EST


In <388D2152.A8C0E990@mandrakesoft.com> Jeff Garzik (jgarzik@mandrakesoft.com) wrote:
> David Whysong wrote:
>> Taking the least significant bits of a fast timer between keypresses is a
>> very good way of generating entropy. I'd trust it more than some unknown
>> hardware generator. The same is true of network packets.

> Unfortunately none of the non-RNG sources are very fast. Try to
> generate a 4096-bit key in GPG with and without an RNG...

If you are using /dev/random (that is: BLOCKING version of random generator)
you are doing it on purpose and you SHOULD be ready to wait. When you can not
wait you must use /dev/urandom. So since we know VERY little about RNG it
should be used exactly this way: as source of random data with unknown
amount of enthropy (perhaps 1bit out of every word can be considered as
guaranteed but even this is not clear). If it'll give you good random
numbers - great: /dev/urandom will work more like /dev/random. If not - no
problem: it was /dev/urandom after all so you should not expect "true"
random numbers...

-
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 : Mon Jan 31 2000 - 21:00:14 EST