From owner-freebsd-current Sun Dec 7 17:52:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id RAA14359 for current-outgoing; Sun, 7 Dec 1997 17:52:12 -0800 (PST) (envelope-from owner-freebsd-current) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id RAA14290; Sun, 7 Dec 1997 17:51:50 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id UAA00245; Sun, 7 Dec 1997 20:51:43 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199712080151.UAA00245@dyson.iquest.net> Subject: Re: VM system info In-Reply-To: <867m9gpsts.fsf@bitbox.follo.net> from Eivind Eklund at "Dec 8, 97 01:43:59 am" To: perhaps@yes.no (Eivind Eklund) Date: Sun, 7 Dec 1997 20:51:43 -0500 (EST) Cc: dyson@FreeBSD.ORG, current@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL31 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-current@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Eivind Eklund said: > > I'm going to be extremely politically incorrect here: > That is perfectly okay :-). I think we are all supposed to discuss, and not just rubber-stamp :-). > > I want this info in the kernel. At the very least, I want > documentation as a part of the SYSCTL_*() macro parameters, unused but > available as a (mandatory) part of the source - better would be as a > part of the kernel that can be compiled away by setting a kernel > option (e.g. NO_SYSCTL_DOCS). > The biggest problem with that is the size of the "documentation." I agree that it would be a good idea to document everything. Internally would be nice (I guess), because it would tend to stay better in sync. Some kind of literate programming scheme would be interesting also. > > I don't care about the amount of hardwired memory it wastes - memory > is extremely cheap, people's time aren't. > I understand, since it is likely that all of the internal strings wouldn't be larger than say 64K, that is a drop in the bucket for the system, and could be disabled anyway. I just don't know what to think about your notion, it *might* be a good idea for the kernel. I know that the sysctls can be tedious to wade through, and some of the features might not be clear. I don't know what the rest of the other FreeBSD developers think about this idea, and it would be a good idea to get more input. All I know is that now when people run out of swap space, I can tell them the sysctl to turn paging off :-). Then they won't try to use any more :-). -- John dyson@freebsd.org jdyson@nc.com