From owner-freebsd-hackers Tue Oct 8 02:36:09 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id CAA12181 for hackers-outgoing; Tue, 8 Oct 1996 02:36:09 -0700 (PDT) Received: from haldjas.folklore.ee (Haldjas.folklore.ee [193.40.6.121]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id CAA12039; Tue, 8 Oct 1996 02:32:04 -0700 (PDT) Received: (from narvi@localhost) by haldjas.folklore.ee (8.7.5/8.6.12) id MAA12042; Tue, 8 Oct 1996 12:26:45 +0300 (EET DST) Date: Tue, 8 Oct 1996 12:26:44 +0300 (EET DST) From: Narvi To: Harlan Stenn cc: hackers@freebsd.org, current@freebsd.org Subject: Re: random() fix explanation to avoid misunderstanding In-Reply-To: <8061.844717430@mumps.pfcs.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Mon, 7 Oct 1996, Harlan Stenn wrote: > Terry is right when he talk about needing to keep the PSEUDO-random code > "golden". > > Jordan is right when he says that if you want repeatable results, it's > best to carry your pseudo-random number code with you. > > If folks are going to change random.c anyway, I'd ask that either the > old code be moved to a "compat" library, and/or find a way to have a > per-process switch that provides a way to choose which version of the > generator you're getting. Wouldn't just an environment variable (USE_OLD_RANDOM) do the job? Most people will not have it defined. For those who will need it, define it in the manual page for random. Yes - it does increase the code size (=bloats it) but is there any other way - as noted already, majority does not (or does not in some time in the near future) need the old code. The minority could just have it set in .login? Sander > > H >