Re: CONFIG_RANDOM option for 1.99.2

Martin.Dalecki (
Tue, 14 May 1996 18:40:59 +0200 (MET DST)

On Tue, 14 May 1996, Ingo Molnar wrote:

> the random driver is cool and well implemented, and it has a >minimal<
> interface. Why not use it? We are not talking about a 10000 functions API.

The interface is in fact minimal but the impementation isn't.

> note that good random numbers are/will be crutial for future networking.
> Why not a bit of testing. Or do you want to send your credit card number
> mangled with standard libc srand() ? =)

Nope I wouldn't send my credit card number over the net anyway. I simply
prefer cach, as most of the people in europe. Both for political and
technical reasons. And IMHO not only is the libc interface standardized, but
also the pseudo random number generater used there! And remember: think about

When You are speaking about testing I agree: Let's make it an *option* for
testing. In fact I sended as a coincidence yestoday an corresponding
patch to Linus. It was IMHO somehow cleaner then the one seen here before.
I fixed the problem with BOOTP without random. For removal of the
random.h-hooks I used defines like this:

void foo(int)
#define foo(x) do {} while (0)

> btw, libc should use this interface IMHO, in some way, thus applications
> using libc *rand* functions could be enhanced.

Yes!! there is a strong need for secure/slow random numbers in xkobo ;-).

> -- mingo
> ps. the Linux entry from the NSA bible:
> Linux: the OS with blocking /dev/random
> =B-)

Why do You think the /dev/random is nowhere used (at least appriopriatly) at
current time? Or why didn't anybody of the other major players implement
such an animal before? The idea isn't new in any way.
I didn't intend to offend anybody. It's simply the feeling of growing
featueritis in Linux overcomming me when seeing something like this.

And just an idea. When You are really feeling the strong need for strong
random numbers, simpy connect a radio reciver to Yor's soundcard... You got
the idea.