Re: [RFC][PATCH] Make cryptoapi non-optional?

[Jamie Lokier]

David Wagner wrote:
> If you see a 64-bit values repeat twice in a row when querying a
> cryptographic pseudorandom generator, the crypto-PRNG is almost surely
> broken.

True.  However if you generate lots of 64-bit values and see a
collision between any two, that is much more likely.  See "birthday paradox".

I know you understand this, David.  Just so everyone else is clear,
because part of this thread is about locking problems (same result
within a short time), and part of this thread is about uniqueness
(collisions between any pair).

Enjoy,
-- Jamie
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/