Mailing List Archive

Why not use rtc or clk for randomness?
Forgive me if this is a stupid question.

One of the most-asked questions is ends up being about how to get the
kernel to generate more entropy for key generation (from a remote
terminal). I asked it myself and have seen it since a few times.

I got my answer, which was to configure /dev/random using rndcontrol -s
-[irq], first finding a good irq by looking at vmstat -i. I was
specifically cautioned against using the irqs associated with clk and
rtc. Why is that? They have such wonderfully high rates. I'm sure using
them would speed up key generation. Are they poor sources of true
randomness, or what?

Thanks,
John G.

===============================
John Goodleaf
goodleaf@goodleaf.net

PGP key: finger John@clyde.goodleaf.net
===============================

--
Archive is at http://lists.gnupg.org - Unsubscribe by sending mail
with a subject of "unsubscribe" to gnupg-users-request@gnupg.org
Re: Why not use rtc or clk for randomness? [ In reply to ]
On Tue, 22 Aug 2000, John Goodleaf wrote:

> -[irq], first finding a good irq by looking at vmstat -i. I was
> specifically cautioned against using the irqs associated with clk and
> rtc. Why is that? They have such wonderfully high rates. I'm sure using

The output of a clock is not very random. Well there are some
exception like the one in our kitchen :-)

--
Werner Koch GnuPG key: 621CC013
OpenIT GmbH http://www.OpenIT.de

--
Archive is at http://lists.gnupg.org - Unsubscribe by sending mail
with a subject of "unsubscribe" to gnupg-users-request@gnupg.org