Messages in this thread | | | From | (H. Peter Anvin) | Subject | Re: /dev/one - why not /dev/repeat? | Date | 24 Dec 1998 21:18:28 GMT |
| |
Followup to: <199812241606.NAA01198@sleipnir.valparaiso.cl> By author: Horst von Brand <vonbrand@sleipnir.valparaiso.cl> In newsgroup: linux.dev.kernel > > > That's legitimate, if you have an application that really would be > > substantially faster with a /dev/random. What is this app, and can > > you show benchmarks? > > That's right! Show benchmarks, and we'll beat the blue sky out of them by > careful use of an inlined RNG that _must_ beat the cost of doing a system > call + assorted data copies + the actual RNG work. Besides, a program where > the RNG is a significant fraction of the cost strikes me as quite odd... > > The _only_ reason for /dev/{,u}random is as a (limited?) source of truly > random bits (AFAIU, the distribution isn't even known to be nice at > all!)
"Truly random" implies a flat distribution. /dev/random is the limited source of as-close-as-possible-to-guaranteed-to-be-random bits, /dev/urandom degrades to a cryptographically strong PRNG when it runs out of enthropy.
-hpa
-- PGP: 2047/2A960705 BA 03 D3 2C 14 A8 A8 BD 1E DF FE 69 EE 35 BD 74 See http://www.zytor.com/~hpa/ for web page and full PGP public key I am Bahá'í -- ask me about it or see http://www.bahai.org/ "To love another person is to see the face of God." -- Les Misérables
- 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/
| |