Subject: Re: replace kernel random number function
To: Johan Danielsson <>
From: Michael Graff <>
List: tech-kern
Date: 10/22/2000 10:49:54 (Johan Danielsson) writes:

> Shouldn't this use rnd(9), or is that too slow?

We could build arc4random, export it to the world as /dev/arandom, and
have it in the kernel as well.

However, random(3) is fast, and sometimes all you really need is weak
randomness for non-security related uses.