From owner-cvs-etc Sun Jun 1 16:33:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA13542 for cvs-etc-outgoing; Sun, 1 Jun 1997 16:33:51 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA13520; Sun, 1 Jun 1997 16:33:34 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA05955; Sun, 1 Jun 1997 16:33:01 -0700 (PDT) Date: Sun, 1 Jun 1997 16:33:01 -0700 (PDT) Message-Id: <199706012333.QAA05955@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/01 16:33:01 PDT Modified files: etc rc Log: Add rm -f /var/run/*.pid to prevent random process killing due to picking old .pid file Revision Changes Path 1.125 +2 -1 src/etc/rc From owner-cvs-etc Sun Jun 1 19:59:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id TAA21397 for cvs-etc-outgoing; Sun, 1 Jun 1997 19:59:21 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA21330; Sun, 1 Jun 1997 19:58:43 -0700 (PDT) From: "Jordan K. Hubbard" Received: (from jkh@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA06453; Sun, 1 Jun 1997 19:58:09 -0700 (PDT) Date: Sun, 1 Jun 1997 19:58:09 -0700 (PDT) Message-Id: <199706020258.TAA06453@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc.conf Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk jkh 1997/06/01 19:58:09 PDT Modified files: etc rc.conf Log: timed_enabled -> timed_enable Submitted by: "James D. Stewart" Revision Changes Path 1.16 +2 -2 src/etc/rc.conf From owner-cvs-etc Sun Jun 1 19:59:23 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id TAA21411 for cvs-etc-outgoing; Sun, 1 Jun 1997 19:59:23 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA21345; Sun, 1 Jun 1997 19:58:54 -0700 (PDT) From: "Jordan K. Hubbard" Received: (from jkh@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id TAA06486; Sun, 1 Jun 1997 19:58:20 -0700 (PDT) Date: Sun, 1 Jun 1997 19:58:20 -0700 (PDT) Message-Id: <199706020258.TAA06486@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc.conf Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk jkh 1997/06/01 19:58:20 PDT Modified files: (Branch: RELENG_2_2) etc rc.conf Log: YAMFC Revision Changes Path 1.1.2.15 +2 -2 src/etc/rc.conf From owner-cvs-etc Sun Jun 1 23:47:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA01953 for cvs-etc-outgoing; Sun, 1 Jun 1997 23:47:40 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA01843; Sun, 1 Jun 1997 23:44:28 -0700 (PDT) From: Mark Murray Received: (from markm@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id XAA08091; Sun, 1 Jun 1997 23:43:53 -0700 (PDT) Date: Sun, 1 Jun 1997 23:43:53 -0700 (PDT) Message-Id: <199706020643.XAA08091@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc.conf src/etc/etc.i386 rc.i386 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk markm 1997/06/01 23:43:53 PDT Modified files: etc rc.conf etc/etc.i386 rc.i386 Log: Correct the way IRQs are passed to rndcontrol. If there are more than one IRQ being used, the -s options must be specified for each of them. In this case rc.conf would only allow 1, as the -s was coded into rc.i386. Revision Changes Path 1.17 +2 -2 src/etc/rc.conf 1.28 +2 -2 src/etc/etc.i386/rc.i386 From owner-cvs-etc Mon Jun 2 03:01:43 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA10478 for cvs-etc-outgoing; Mon, 2 Jun 1997 03:01:43 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA10429; Mon, 2 Jun 1997 03:00:55 -0700 (PDT) From: Mark Murray Received: (from markm@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA09230; Mon, 2 Jun 1997 03:00:18 -0700 (PDT) Date: Mon, 2 Jun 1997 03:00:18 -0700 (PDT) Message-Id: <199706021000.DAA09230@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc.conf src/etc/etc.i386 rc.i386 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk markm 1997/06/02 03:00:18 PDT Modified files: (Branch: RELENG_2_2) etc rc.conf etc/etc.i386 rc.i386 Log: YAMFC: fix the handling of multiple IRQs for rndcontrol Revision Changes Path 1.1.2.16 +2 -2 src/etc/rc.conf 1.18.2.6 +2 -2 src/etc/etc.i386/rc.i386 From owner-cvs-etc Mon Jun 2 03:22:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA11330 for cvs-etc-outgoing; Mon, 2 Jun 1997 03:22:10 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA11304; Mon, 2 Jun 1997 03:21:50 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA09347; Mon, 2 Jun 1997 03:21:13 -0700 (PDT) Date: Mon, 2 Jun 1997 03:21:13 -0700 (PDT) Message-Id: <199706021021.DAA09347@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/02 03:21:13 PDT Modified files: etc rc Log: Back out /var/run/*.pid cleaning since some programms (like innd) use .pid to detect crash Revision Changes Path 1.126 +1 -2 src/etc/rc From owner-cvs-etc Mon Jun 2 09:41:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA27703 for cvs-etc-outgoing; Mon, 2 Jun 1997 09:41:26 -0700 (PDT) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA27470; Mon, 2 Jun 1997 09:37:49 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by agora.rdrop.com (8.8.5/8.8.5) with ESMTP id JAA15709; Mon, 2 Jun 1997 09:37:36 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id RAA10191; Mon, 2 Jun 1997 17:02:37 GMT From: Adam David Message-Id: <199706021702.RAA10191@veda.is> Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706012333.QAA05955@freefall.freebsd.org> from "Andrey A. Chernov" at "Jun 1, 97 04:33:01 pm" To: ache@FreeBSD.ORG (Andrey A. Chernov) Date: Mon, 2 Jun 1997 17:02:35 +0000 (GMT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > ache 1997/06/01 16:33:01 PDT > > Modified files: > etc rc > Log: > Add rm -f /var/run/*.pid to prevent random process killing due to > picking old .pid file > > Revision Changes Path > 1.125 +2 -1 src/etc/rc > PIDs gathered from /var/run should always be checked for validity before killing anyway. The old process could have left a stale pidfile without a new process being started. This can happen at any time, not only at reboot. -- Adam David From owner-cvs-etc Mon Jun 2 10:49:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA01509 for cvs-etc-outgoing; Mon, 2 Jun 1997 10:49:08 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA01502; Mon, 2 Jun 1997 10:48:59 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id VAA22466; Mon, 2 Jun 1997 21:48:38 +0400 (MSD) Date: Mon, 2 Jun 1997 21:48:38 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Adam David cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706021702.RAA10191@veda.is> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Mon, 2 Jun 1997, Adam David wrote: > PIDs gathered from /var/run should always be checked for validity before > killing anyway. The old process could have left a stale pidfile without a > new process being started. This can happen at any time, not only at reboot. It is impossible. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Mon Jun 2 11:05:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA02659 for cvs-etc-outgoing; Mon, 2 Jun 1997 11:05:03 -0700 (PDT) Received: from spinner.dialix.com.au (spinner.dialix.com.au [192.203.228.67]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA02639; Mon, 2 Jun 1997 11:04:56 -0700 (PDT) Received: from spinner.dialix.com.au (localhost.dialix.com.au [127.0.0.1]) by spinner.dialix.com.au with ESMTP id CAA27545; Tue, 3 Jun 1997 02:03:38 +0800 (WST) Message-Id: <199706021803.CAA27545@spinner.dialix.com.au> X-Mailer: exmh version 2.0gamma 1/27/96 To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: Adam David , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc In-reply-to: Your message of "Mon, 02 Jun 1997 21:48:38 +0400." Date: Tue, 03 Jun 1997 02:03:35 +0800 From: Peter Wemm Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= wrote: > On Mon, 2 Jun 1997, Adam David wrote: > > > PIDs gathered from /var/run should always be checked for validity before > > killing anyway. The old process could have left a stale pidfile without a > > new process being started. This can happen at any time, not only at reboot. > > It is impossible. Well, not quite "impossible", but it's not exactly simple for scripts to deal with.. Gated uses a flock() lock on it's pid file as the "lock", the existance of the file does not imply that it's valid. So, things like gdc etc try and do a conflicting lock request which should succeed if gated is not runni ng, and should fail if it is there. But doing this for other daemons requires code changes in each one. The problem with gated in particular was that if it was started with the network code, ie: before NFS mounts, it would put it's pid file in /var/ run. However, after the NFS mounts took place, /var/run used to be cleared.. (in case /var was nfs mounted I guess). So, gdc was not able to talk to the running gated. Anyway, for innd, I always put the pid file in the newslib dir, right next to the active file since it was important to innd that it not "go away" on boot. > -- > Andrey A. Chernov > > http://www.nagual.pp.ru/~ache/ Cheers, -Peter From owner-cvs-etc Mon Jun 2 11:10:41 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA02977 for cvs-etc-outgoing; Mon, 2 Jun 1997 11:10:41 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA02972; Mon, 2 Jun 1997 11:10:37 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id WAA22778; Mon, 2 Jun 1997 22:09:44 +0400 (MSD) Date: Mon, 2 Jun 1997 22:09:44 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Peter Wemm cc: Adam David , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706021803.CAA27545@spinner.dialix.com.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Tue, 3 Jun 1997, Peter Wemm wrote: > > It is impossible. > > Well, not quite "impossible", but it's not exactly simple for scripts to > deal with.. I mean impossible without modifying application in case it not cares :-) > Anyway, for innd, I always put the pid file in the newslib dir, right next > to the active file since it was important to innd that it not "go away" on > boot. We should check what our INND port does (could anybody tell me what, I not run INND) and probably fix it. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Mon Jun 2 11:16:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA03462 for cvs-etc-outgoing; Mon, 2 Jun 1997 11:16:11 -0700 (PDT) Received: from alpo.whistle.com (alpo.whistle.com [207.76.204.38]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA03453; Mon, 2 Jun 1997 11:16:07 -0700 (PDT) Received: (from daemon@localhost) by alpo.whistle.com (8.8.5/8.8.5) id LAA29556; Mon, 2 Jun 1997 11:13:07 -0700 (PDT) Received: from current1.whistle.com(207.76.205.22) via SMTP by alpo.whistle.com, id smtpd029549; Mon Jun 2 18:12:59 1997 Message-ID: <33930CFD.794BDF32@whistle.com> Date: Mon, 02 Jun 1997 11:12:13 -0700 From: Julian Elischer Organization: Whistle Communications X-Mailer: Mozilla 3.0Gold (X11; I; FreeBSD 2.2-CURRENT i386) MIME-Version: 1.0 To: Peter Wemm CC: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Adam David , cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc References: <199706021803.CAA27545@spinner.dialix.com.au> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Peter Wemm wrote: > Gated uses a flock() lock on it's pid file as the "lock", the existance of > the file does not imply that it's valid. So, things like gdc etc try and > do a conflicting lock request which should succeed if gated is not runni > ng, and should fail if it is there. But doing this for other daemons > requires code changes in each one. > I seem to remember old UUCP (syV) did an open in O_EXCL mode or something.. (you could always make a PIDfile that was a fifo..) From owner-cvs-etc Mon Jun 2 11:29:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id LAA04223 for cvs-etc-outgoing; Mon, 2 Jun 1997 11:29:40 -0700 (PDT) Received: from scds.ziplink.net (scds.ziplink.net [206.15.128.34]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id LAA04195; Mon, 2 Jun 1997 11:29:01 -0700 (PDT) Received: (from jseger@localhost) by scds.ziplink.net (8.8.5/8.8.4) id NAA14167; Mon, 2 Jun 1997 13:38:26 -0400 (EDT) Date: Mon, 2 Jun 1997 13:38:26 -0400 (EDT) From: "Justin M. Seger" Message-Id: <199706021738.NAA14167@scds.ziplink.net> To: ache@nagual.pp.ru, peter@spinner.dialix.com.au Subject: Re: cvs commit: src/etc rc Cc: adam@veda.is, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Just a suggestion... If we want the pids to exist in /var/run for all programs that keep track of them, and programs that need them after reboot to keep them in /var/db, why not create a symlink between the ones in /var/run to the ones in /var/db where neccessary? -Justin Seger-. From owner-cvs-etc Mon Jun 2 12:56:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA08231 for cvs-etc-outgoing; Mon, 2 Jun 1997 12:56:16 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA08157; Mon, 2 Jun 1997 12:54:52 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id UAA10482; Mon, 2 Jun 1997 20:20:31 GMT From: Adam David Message-Id: <199706022020.UAA10482@veda.is> Subject: Re: cvs commit: src/etc rc In-Reply-To: from "[______ ______]" at "Jun 2, 97 09:48:38 pm" To: ache@nagual.pp.ru (=?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?=) Date: Mon, 2 Jun 1997 20:20:30 +0000 (GMT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > On Mon, 2 Jun 1997, Adam David wrote: > > > PIDs gathered from /var/run should always be checked for validity before > > killing anyway. The old process could have left a stale pidfile without a > > new process being started. This can happen at any time, not only at reboot. > > It is impossible. > > -- > Andrey A. Chernov > > http://www.nagual.pp.ru/~ache/ > That sentence is too terse to make clear sense. Do you mean that processes don't die and leave pidfiles behind, or that it is impossible to check whether there is a program running with the same name and PID as the pidfile claims? I am witness to the former, and have used the latter with success in the case of individual programs. It might not be possible to automatically double-check in a generic context because there is no guaranteed naming convention. However, given the name of the program and its pidfile, it is a simple matter to check first, because two different programs cannot have the same PID at the same time. If more than one program file creates the same pidfile when it runs, there is a deeper problem than this, of an administrative nature not programmatical. -- Adam David From owner-cvs-etc Mon Jun 2 13:03:03 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA08685 for cvs-etc-outgoing; Mon, 2 Jun 1997 13:03:03 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA08626; Mon, 2 Jun 1997 13:02:22 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id AAA24275; Tue, 3 Jun 1997 00:01:43 +0400 (MSD) Date: Tue, 3 Jun 1997 00:01:42 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Adam David cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706022020.UAA10482@veda.is> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Mon, 2 Jun 1997, Adam David wrote: > Do you mean that processes don't die and leave pidfiles behind, or that > it is impossible to check whether there is a program running with the same > name and PID as the pidfile claims? > I am witness to the former, and have used the latter with success in the > case of individual programs. It might not be possible to automatically > double-check in a generic context because there is no guaranteed naming > convention. However, given the name of the program and its pidfile, it > is a simple matter to check first, because two different programs cannot > have the same PID at the same time. Extracting program name from 'ps'-like info isn't always possible, it is small probability that garbadge will be returned, another name, or program just wipe their own name with something else. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Mon Jun 2 13:33:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA10641 for cvs-etc-outgoing; Mon, 2 Jun 1997 13:33:19 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA10634; Mon, 2 Jun 1997 13:33:10 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id UAA10554; Mon, 2 Jun 1997 20:59:24 GMT From: Adam David Message-Id: <199706022059.UAA10554@veda.is> Subject: Re: cvs commit: src/etc rc In-Reply-To: from "[______ ______]" at "Jun 3, 97 00:01:42 am" To: ache@nagual.pp.ru (=?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?=) Date: Mon, 2 Jun 1997 20:59:22 +0000 (GMT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > given the name of the program and its pidfile, it > > is a simple matter to check first, because two different programs cannot > > have the same PID at the same time. > > Extracting program name from 'ps'-like info isn't always possible, > it is small probability that garbadge will be returned, another name, > or program just wipe their own name with something else. > > -- > Andrey A. Chernov > > http://www.nagual.pp.ru/~ache/ > Compliant programs, and those that use the library version of setproctitle all retain the program name within parentheses at the end of the line. IMHO, if the program wipes its name out altogether it deserves whatever crap it gets. -- Adam David From owner-cvs-etc Mon Jun 2 13:45:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA11270 for cvs-etc-outgoing; Mon, 2 Jun 1997 13:45:04 -0700 (PDT) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA11146; Mon, 2 Jun 1997 13:43:26 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by agora.rdrop.com (8.8.5/8.8.5) with ESMTP id NAA09371; Mon, 2 Jun 1997 13:43:20 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id AAA24676; Tue, 3 Jun 1997 00:42:00 +0400 (MSD) Date: Tue, 3 Jun 1997 00:42:00 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Adam David cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706022059.UAA10554@veda.is> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Mon, 2 Jun 1997, Adam David wrote: > IMHO, if the program wipes its name out altogether it deserves whatever crap > it gets. Program name not related to its pid file, you try to introduce unnecesarry restriction. If program not test pid file via flock(), it deserves ... I.e. all programs except gated and few others got what they want. So, they require modifications to check locking, so it is impossible to validate pid files without applications modifications (as I already say in previous letter). Modification of each such application is not acceptable solution. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Mon Jun 2 15:04:18 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA15584 for cvs-etc-outgoing; Mon, 2 Jun 1997 15:04:18 -0700 (PDT) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA15423; Mon, 2 Jun 1997 15:02:29 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by agora.rdrop.com (8.8.5/8.8.5) with ESMTP id PAA15362; Mon, 2 Jun 1997 15:02:16 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id WAA10666; Mon, 2 Jun 1997 22:27:08 GMT From: Adam David Message-Id: <199706022227.WAA10666@veda.is> Subject: Re: cvs commit: src/etc rc In-Reply-To: from "[______ ______]" at "Jun 3, 97 00:42:00 am" To: ache@nagual.pp.ru (=?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?=) Date: Mon, 2 Jun 1997 22:27:07 +0000 (GMT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > IMHO, if the program wipes its name out altogether it deserves whatever crap > > it gets. > > Program name not related to its pid file, you try to introduce unnecesarry > restriction. I already stated that it is necessary to know both names in order to check. It is only when the program clobbers itself that it gets into trouble. If a program is so paranoid as to erase its tracks entirely, it has no business maintaining pidfiles. > If program not test pid file via flock(), it deserves ... I.e. all > programs except gated and few others got what they want. So, they require > modifications to check locking, so it is impossible to validate pid files > without applications modifications (as I already say in previous letter). #!/bin/sh kill `cat /var/run/progname.pid` (yes, there are programs that make the assumption that pidfiles exist for the duration of a programs execution, and not a moment longer). Most of the time when the program has expired, it will produce an unexpected and often confusing error, but occasionally it hits the wrong program altogether. Even once is too often. It is not the offender that suffers but an innocent bystander. (randomly directed violence) I have also seen stale locks, but not on a healthy system. (file claims to be locked, but no process holds the lock). > Modification of each such application is not acceptable solution. Each such application already uses its own methods, which are demonstrably deficient in many cases. -- Adam David From owner-cvs-etc Mon Jun 2 18:33:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA26049 for cvs-etc-outgoing; Mon, 2 Jun 1997 18:33:42 -0700 (PDT) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA26038; Mon, 2 Jun 1997 18:33:34 -0700 (PDT) Received: from awfulhak.demon.co.uk (localhost [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id CAA19996; Tue, 3 Jun 1997 02:05:05 +0100 (BST) Message-Id: <199706030105.CAA19996@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Adam David cc: ache@nagual.pp.ru (=?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?=), cvs-committers@freebsd.org, cvs-all@freebsd.org, cvs-etc@freebsd.org Subject: Re: cvs commit: src/etc rc In-reply-to: Your message of "Mon, 02 Jun 1997 22:27:07 -0000." <199706022227.WAA10666@veda.is> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 03 Jun 1997 02:05:05 +0100 From: Brian Somers Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk [.....] > I have also seen stale locks, but not on a healthy system. > (file claims to be locked, but no process holds the lock). Sounds like a *very* sick system. One that's likely to be doing other stupid things like SEGVing things for no reason 'cos the descriptor tables are shagged. > > Modification of each such application is not acceptable solution. > > Each such application already uses its own methods, which are demonstrably > deficient in many cases. What we need is a set of calls in libutil: int pfile_create(const char *prog); /* create a pid file, return fd */ void pfile_destroy(int fd); /* remove a pid file */ pid_t pfile_read(const char *prog); /* Get pid from file */ And maybe a program that uses pfile_read for use in scripts. As has already been said, the existence of a lock on an existing file says whether the process is running or not. Of course to cater for all eventualities, we may need a "max" arg in pfile_create and an "n" arg in pfile_read, allowing more than one of a given process. Or maybe this is over- complicating things. If programs want to use this mechanism, they can. Killers like newsyslog should extract the pid to kill using pfile_read, therefore disabling it from killing un-cooperating processes (pfile_read will come up with nothing from a stale file). > -- > Adam David > -- Brian , Don't _EVER_ lose your sense of humour.... From owner-cvs-etc Mon Jun 2 23:31:18 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA09961 for cvs-etc-outgoing; Mon, 2 Jun 1997 23:31:18 -0700 (PDT) Received: from androcles.com (androcles.com [204.57.240.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id XAA09864; Mon, 2 Jun 1997 23:29:48 -0700 (PDT) Received: (from dhh@localhost) by androcles.com (8.8.5/8.6.12) id XAA19614; Mon, 2 Jun 1997 23:30:20 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.0 [p0] on FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit MIME-Version: 1.0 In-Reply-To: <33930CFD.794BDF32@whistle.com> Date: Mon, 02 Jun 1997 22:56:04 -0700 (PDT) From: "Duane H. Hesser" To: Julian Elischer Subject: Re: cvs commit: src/etc rc Cc: cvs-etc@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, Adam David , =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Peter Wemm Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Umm...sorry to intrude on your thread, guys (throw me out if you like), but I think I'm a little confused about WHICH problem you're talking about (maybe I missed something?). There are two possible problems with lock files that I know about. One is easy, the other a little harder, and at least the easy one can be handled pretty handily from shell scripts (as well as C). 1) dead or 'stale' locks: if test kill -0 `cat PIDFILE` then do what you do if the process LIVES else RIP fi ...or the C equivalent. This supposes that the pid file contains the PID and only the pid 2) wraparound--the demon dies and sometime later another process re-uses the pid. Problem (1) is actually fairly frequent...some dummy (not me, no NEVER me...) does a 'kill -9', or the SIGSEGV demon randomly kills a perfectly good program... (2) is less likely. In 18 years of watching closely I've never seen one demon assume another's process identity. It COULD happen, and it could HAVE happened, but I'd like to think I would've noticed. No matter, it could happen. Still, 'ps -c' (if I remember) prints from p_comm, rather than the arg string, and I don't think a process is able to mangle that easily, is it? I suppose that you could say that if it does, it deserves it's fate. Forgive me, that is unkind... So what IS the problem? What am I missing? I agree that taking an flock on a pid file is a nice idea, so long as you can afford an open fd, but it doesn't seem to me that revising code (except maybe some shell scripts) is necessary to check for dead locks. I (for one) would be willing to take my chances on the wraparound problem. But please, NO MORE symbolic lunks! Please? I don't think I could STAND another symbolic lunk. Duane H. Hesser dhh@androcles.com From owner-cvs-etc Tue Jun 3 08:33:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id IAA04376 for cvs-etc-outgoing; Tue, 3 Jun 1997 08:33:06 -0700 (PDT) Received: from alpha.xerox.com (alpha.Xerox.COM [13.1.64.93]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id IAA04371; Tue, 3 Jun 1997 08:33:04 -0700 (PDT) Received: from crevenia.parc.xerox.com ([13.2.116.11]) by alpha.xerox.com with SMTP id <18085(6)>; Tue, 3 Jun 1997 08:32:23 PDT Received: from localhost by crevenia.parc.xerox.com with SMTP id <177489>; Tue, 3 Jun 1997 08:32:13 -0700 To: cvs-etc@freebsd.org, cvs-all@freebsd.org, cvs-committers@freebsd.org Subject: Re: cvs commit: src/etc rc Date: Tue, 3 Jun 1997 08:32:05 PDT From: Bill Fenner Message-Id: <97Jun3.083213pdt.177489@crevenia.parc.xerox.com> Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Don't forget that if you use a flock(), there is a race condition when two seperate processes are checking to see if the daemon is running where one process may think it's not running (its attempt to lock the pidfile succeeds) and the other process thinks it's running (since its attempt fails since the first process hadn't released its lock yet). Bill From owner-cvs-etc Tue Jun 3 14:24:24 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA23694 for cvs-etc-outgoing; Tue, 3 Jun 1997 14:24:24 -0700 (PDT) Received: from awfulhak.demon.co.uk (awfulhak.demon.co.uk [158.152.17.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA23487; Tue, 3 Jun 1997 14:22:19 -0700 (PDT) Received: from awfulhak.demon.co.uk (localhost [127.0.0.1]) by awfulhak.demon.co.uk (8.8.5/8.8.5) with ESMTP id VAA00608; Tue, 3 Jun 1997 21:01:14 +0100 (BST) Message-Id: <199706032001.VAA00608@awfulhak.demon.co.uk> X-Mailer: exmh version 1.6.9 8/22/96 To: Bill Fenner cc: cvs-etc@freebsd.org, cvs-all@freebsd.org, cvs-committers@freebsd.org Subject: Re: cvs commit: src/etc rc In-reply-to: Your message of "Tue, 03 Jun 1997 08:32:05 PDT." <97Jun3.083213pdt.177489@crevenia.parc.xerox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 03 Jun 1997 21:01:14 +0100 From: Brian Somers Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Don't forget that if you use a flock(), there is a race condition when > two seperate processes are checking to see if the daemon is running > where one process may think it's not running (its attempt to lock the > pidfile succeeds) and the other process thinks it's running (since its > attempt fails since the first process hadn't released its lock yet). Not true. The lock is a write lock. The check is a read lock. A read lock won't affect another read lock. Remember, you can do just about anything, given the ability to both read lock and write lock. > Bill > -- Brian , Don't _EVER_ lose your sense of humour.... From owner-cvs-etc Tue Jun 3 17:05:45 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA03010 for cvs-etc-outgoing; Tue, 3 Jun 1997 17:05:45 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA02778; Tue, 3 Jun 1997 17:02:51 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA03981; Tue, 3 Jun 1997 17:02:01 -0700 (PDT) Date: Tue, 3 Jun 1997 17:02:01 -0700 (PDT) Message-Id: <199706040002.RAA03981@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.x11.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/03 17:02:01 PDT Added files: etc/mtree BSD.x11.dist Log: New mtree file for /usr/X11R6. Approved (a long time ago) by: David Dawes From owner-cvs-etc Tue Jun 3 17:06:44 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA03077 for cvs-etc-outgoing; Tue, 3 Jun 1997 17:06:44 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA02846; Tue, 3 Jun 1997 17:03:44 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA04015; Tue, 3 Jun 1997 17:02:54 -0700 (PDT) Date: Tue, 3 Jun 1997 17:02:54 -0700 (PDT) Message-Id: <199706040002.RAA04015@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.x11.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/03 17:02:54 PDT Added files: (Branch: RELENG_2_2) etc/mtree BSD.x11.dist Log: Add this file to 2.2 branch too. From owner-cvs-etc Tue Jun 3 17:07:22 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA03095 for cvs-etc-outgoing; Tue, 3 Jun 1997 17:07:22 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA02925; Tue, 3 Jun 1997 17:04:27 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id RAA04048; Tue, 3 Jun 1997 17:03:36 -0700 (PDT) Date: Tue, 3 Jun 1997 17:03:36 -0700 (PDT) Message-Id: <199706040003.RAA04048@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.x11.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/03 17:03:36 PDT Added files: (Branch: RELENG_2_1_0) etc/mtree BSD.x11.dist Log: Add this file to 2.1 branch too (just in case). From owner-cvs-etc Tue Jun 3 21:01:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id VAA18193 for cvs-etc-outgoing; Tue, 3 Jun 1997 21:01:30 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA18043; Tue, 3 Jun 1997 20:59:44 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA06057; Tue, 3 Jun 1997 20:58:53 -0700 (PDT) Date: Tue, 3 Jun 1997 20:58:53 -0700 (PDT) Message-Id: <199706040358.UAA06057@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc Makefile Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/03 20:58:53 PDT Modified files: etc Makefile Log: Remove /usr/local meddling. Approved by: jkh Revision Changes Path 1.151 +1 -16 src/etc/Makefile From owner-cvs-etc Tue Jun 3 21:02:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id VAA18301 for cvs-etc-outgoing; Tue, 3 Jun 1997 21:02:11 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA18121; Tue, 3 Jun 1997 21:00:37 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id UAA06108; Tue, 3 Jun 1997 20:59:46 -0700 (PDT) Date: Tue, 3 Jun 1997 20:59:46 -0700 (PDT) Message-Id: <199706040359.UAA06108@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc Makefile Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/03 20:59:45 PDT Modified files: (Branch: RELENG_2_2) etc Makefile Log: Merge 1.151 (remove /usr/local meddling). Revision Changes Path 1.143.2.5 +1 -16 src/etc/Makefile From owner-cvs-etc Wed Jun 4 03:51:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA16963 for cvs-etc-outgoing; Wed, 4 Jun 1997 03:51:33 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA16939; Wed, 4 Jun 1997 03:51:11 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA04248; Wed, 4 Jun 1997 03:51:10 -0700 (PDT) Date: Wed, 4 Jun 1997 03:51:10 -0700 (PDT) Message-Id: <199706041051.DAA04248@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.usr.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/04 03:51:10 PDT Modified files: etc/mtree BSD.usr.dist Log: Make local "root:wheel" so it will be consistent with BSD.local.dist Reviewed by: bde Revision Changes Path 1.91 +2 -2 src/etc/mtree/BSD.usr.dist From owner-cvs-etc Wed Jun 4 03:51:59 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA17014 for cvs-etc-outgoing; Wed, 4 Jun 1997 03:51:59 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA16982; Wed, 4 Jun 1997 03:51:45 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA04291; Wed, 4 Jun 1997 03:51:44 -0700 (PDT) Date: Wed, 4 Jun 1997 03:51:44 -0700 (PDT) Message-Id: <199706041051.DAA04291@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.usr.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/04 03:51:44 PDT Modified files: (Branch: RELENG_2_2) etc/mtree BSD.usr.dist Log: Merge 1.91. Revision Changes Path 1.74.2.9 +2 -2 src/etc/mtree/BSD.usr.dist From owner-cvs-etc Wed Jun 4 03:52:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id DAA17095 for cvs-etc-outgoing; Wed, 4 Jun 1997 03:52:15 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id DAA17020; Wed, 4 Jun 1997 03:52:00 -0700 (PDT) From: Satoshi Asami Received: (from asami@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id DAA04328; Wed, 4 Jun 1997 03:51:58 -0700 (PDT) Date: Wed, 4 Jun 1997 03:51:58 -0700 (PDT) Message-Id: <199706041051.DAA04328@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.usr.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk asami 1997/06/04 03:51:58 PDT Modified files: (Branch: RELENG_2_1_0) etc/mtree BSD.usr.dist Log: Merge 1.91. Revision Changes Path 1.40.4.13 +2 -2 src/etc/mtree/BSD.usr.dist From owner-cvs-etc Wed Jun 4 12:23:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA23666 for cvs-etc-outgoing; Wed, 4 Jun 1997 12:23:15 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA23333; Wed, 4 Jun 1997 12:20:09 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id MAA07515; Wed, 4 Jun 1997 12:20:05 -0700 (PDT) Date: Wed, 4 Jun 1997 12:20:05 -0700 (PDT) Message-Id: <199706041920.MAA07515@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/04 12:20:04 PDT Modified files: etc rc Log: Restore /var/run/*.pid cleaning, gated already moved off the way and innd store its .pid inside its own directories Revision Changes Path 1.127 +1 -0 src/etc/rc From owner-cvs-etc Wed Jun 4 12:39:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA26025 for cvs-etc-outgoing; Wed, 4 Jun 1997 12:39:51 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id MAA25979; Wed, 4 Jun 1997 12:39:33 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id UAA01180; Wed, 4 Jun 1997 20:01:48 GMT From: Adam David Message-Id: <199706042001.UAA01180@veda.is> Subject: Re: cvs commit: src/etc rc In-Reply-To: <199706041920.MAA07515@freefall.freebsd.org> from "Andrey A. Chernov" at "Jun 4, 97 12:20:05 pm" To: ache@FreeBSD.ORG (Andrey A. Chernov) Date: Wed, 4 Jun 1997 20:01:46 +0000 (GMT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > ache 1997/06/04 12:20:04 PDT > > Modified files: > etc rc > Log: > Restore /var/run/*.pid cleaning, gated already moved off the way > and innd store its .pid inside its own directories > > Revision Changes Path > 1.127 +1 -0 src/etc/rc > Why not clean /var/run/* completely as we used to? Is there anything in /var/run that needs to survive reboot, or should there be? -- Adam David From owner-cvs-etc Wed Jun 4 13:04:05 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA00327 for cvs-etc-outgoing; Wed, 4 Jun 1997 13:04:05 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA29953; Wed, 4 Jun 1997 13:03:04 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id NAA07616; Wed, 4 Jun 1997 13:02:59 -0700 (PDT) Date: Wed, 4 Jun 1997 13:02:59 -0700 (PDT) Message-Id: <199706042002.NAA07616@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc rc Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/04 13:02:59 PDT Modified files: etc rc Log: Remove /var/run/* now instead of *.pid Revision Changes Path 1.128 +3 -3 src/etc/rc From owner-cvs-etc Wed Jun 4 15:26:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA19605 for cvs-etc-outgoing; Wed, 4 Jun 1997 15:26:11 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA19526; Wed, 4 Jun 1997 15:25:37 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id PAA07945; Wed, 4 Jun 1997 15:25:31 -0700 (PDT) Date: Wed, 4 Jun 1997 15:25:31 -0700 (PDT) Message-Id: <199706042225.PAA07945@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.include.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/04 15:25:31 PDT Modified files: etc/mtree BSD.include.dist Log: Remove directories belongs to /usr/src to stop mtree to modify /usr/src permissions. Having them here is wrong from several other poins too: they are never be a directories (simlinks only), so why give a chance to mtree to make them as directories? Since they never be a directories, permissions of them will never be modified by old mtree too. Revision Changes Path 1.12 +1 -39 src/etc/mtree/BSD.include.dist From owner-cvs-etc Wed Jun 4 15:46:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA23577 for cvs-etc-outgoing; Wed, 4 Jun 1997 15:46:39 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA23547; Wed, 4 Jun 1997 15:46:28 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id IAA20236; Thu, 5 Jun 1997 08:40:52 +1000 Date: Thu, 5 Jun 1997 08:40:52 +1000 From: Bruce Evans Message-Id: <199706042240.IAA20236@godzilla.zeta.org.au> To: ache@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Modified files: > etc/mtree BSD.include.dist > Log: > Remove directories belongs to /usr/src to stop mtree to modify /usr/src > permissions. > Having them here is wrong from several other poins too: > they are never be a directories (simlinks only), so why give a chance to mtree to make > them as directories? > Since they never be a directories, permissions of them will never be > modified by old mtree too. > > Revision Changes Path > 1.12 +1 -39 src/etc/mtree/BSD.include.dist Erm, they are always directories in releases. I'm not sure where this is configured for releases, but it is option SHARED=copies in include/Makefile. Bruce From owner-cvs-etc Wed Jun 4 15:49:34 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA24539 for cvs-etc-outgoing; Wed, 4 Jun 1997 15:49:34 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA24533; Wed, 4 Jun 1997 15:49:29 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id CAA03154; Thu, 5 Jun 1997 02:49:12 +0400 (MSD) Date: Thu, 5 Jun 1997 02:49:12 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Bruce Evans cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706042240.IAA20236@godzilla.zeta.org.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, 5 Jun 1997, Bruce Evans wrote: > Erm, they are always directories in releases. I'm not sure where this is > configured for releases, but it is option SHARED=copies in include/Makefile. Do we really need them as directories? In any case some deeper level directories not listed here, so I don't understand, how it is possible for release. Why just not make symlinks _always_? -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 15:58:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA25931 for cvs-etc-outgoing; Wed, 4 Jun 1997 15:58:48 -0700 (PDT) Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA25852; Wed, 4 Jun 1997 15:58:21 -0700 (PDT) Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.5/8.7.3) id PAA01401; Wed, 4 Jun 1997 15:58:11 -0700 (PDT) Date: Wed, 4 Jun 1997 15:58:11 -0700 (PDT) Message-Id: <199706042258.PAA01401@vader.cs.berkeley.edu> To: ache@nagual.pp.ru CC: bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: (message from =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= on Thu, 5 Jun 1997 02:49:12 +0400 (MSD)) Subject: Re: cvs commit: src/etc/mtree BSD.include.dist From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * Do we really need them as directories? I guess so, unless we mandate the source dist. You know where the symlinks are pointing at, right? * In any case some deeper level * directories not listed here, so I don't understand, how it is possible for * release. Why just not make symlinks _always_? Well, here on a release system (2.1.5 actually): === >> du 33 ./arpa 459 ./g++ 1 ./kerberosIV 358 ./machine 110 ./net 70 ./netccitt 136 ./netinet 189 ./netiso 48 ./netns 71 ./nfs 1 ./objc 22 ./protocols 30 ./readline 63 ./rpc 110 ./rpcsvc 5 ./ss 626 ./sys 27 ./ufs/ffs 21 ./ufs/lfs 12 ./ufs/mfs 38 ./ufs/ufs 99 ./ufs 91 ./vm 2928 . === These are all that is listed in BSD.include.dist. At any rate, I suggest you back out the change and ask Jordan for review. This is clearly his region as it concerns the release and not only "make world". Satoshi From owner-cvs-etc Wed Jun 4 16:06:42 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA27787 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:06:42 -0700 (PDT) Received: from freefall.freebsd.org (freefall.cdrom.com [204.216.27.21]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA27362; Wed, 4 Jun 1997 16:05:39 -0700 (PDT) From: "Andrey A. Chernov" Received: (from ache@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id QAA08096; Wed, 4 Jun 1997 16:05:32 -0700 (PDT) Date: Wed, 4 Jun 1997 16:05:32 -0700 (PDT) Message-Id: <199706042305.QAA08096@freefall.freebsd.org> To: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: cvs commit: src/etc/mtree BSD.include.dist Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk ache 1997/06/04 16:05:32 PDT Modified files: etc/mtree BSD.include.dist Log: Back out symlinks/directories removing until solution will be found since they are really directories in the release (for unknown reason for me) Revision Changes Path 1.13 +38 -0 src/etc/mtree/BSD.include.dist From owner-cvs-etc Wed Jun 4 16:09:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA28106 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:09:10 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA28068; Wed, 4 Jun 1997 16:08:56 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id DAA03322; Thu, 5 Jun 1997 03:08:45 +0400 (MSD) Date: Thu, 5 Jun 1997 03:08:45 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Satoshi Asami cc: bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706042258.PAA01401@vader.cs.berkeley.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Wed, 4 Jun 1997, Satoshi Asami wrote: > * Do we really need them as directories? > > I guess so, unless we mandate the source dist. You know where the > symlinks are pointing at, right? What bad happen in keeping the same scheme for release and for source distribution and avoid such special cases? I see some troubles happens if somebody will try to extract source distribution over release (symlinks which overwrites directories) > > * In any case some deeper level > * directories not listed here, so I don't understand, how it is possible for > * release. Why just not make symlinks _always_? > > Well, here on a release system (2.1.5 actually): What about deeper level directories I ask? > At any rate, I suggest you back out the change and ask Jordan for > review. This is clearly his region as it concerns the release and not > only "make world". Ok. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 16:16:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA29345 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:16:55 -0700 (PDT) Received: from labs.usn.blaze.net.au (labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA29337; Wed, 4 Jun 1997 16:16:47 -0700 (PDT) Received: from labs.usn.blaze.net.au (local [127.0.0.1]) by labs.usn.blaze.net.au (8.8.5/8.8.5) with ESMTP id JAA16937; Thu, 5 Jun 1997 09:16:17 +1000 (EST) Message-Id: <199706042316.JAA16937@labs.usn.blaze.net.au> X-Mailer: exmh version 2.0gamma 1/27/96 To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 02:49:12 +0400." X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 05 Jun 1997 09:16:17 +1000 From: David Nugent Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > Erm, they are always directories in releases. I'm not sure > > where this is configured for releases, but it is option > > SHARED=copies in include/Makefile. > > Do we really need them as directories? Yes. it's kind of nice to be able to build your hello world program and ports on a newly installed FreeBSD box without having to also install kernel sources. Otherwise, we should leave installation of the entire compiler kit out unless /sys is also installed (which is actually pretty typical of any Linux installation, and I hated it there too). None of this may be obvious to the end-user, which is why it should remain as it is. Personally, I don't much like mtree following symlinks, but I guess I don't really get a vote. :-) This is one case where you really don't want to do so, so I guess some extension to the mtree format or a command line switch (less flexible here) would be necessary to take care of these special cases where you don't want to follow them. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia Voice +61-3-9791-9547 Data/BBS +61-3-9792-3507 3:632/348@fidonet davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-cvs-etc Wed Jun 4 16:19:39 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA29723 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:19:39 -0700 (PDT) Received: from vader.cs.berkeley.edu (vader.CS.Berkeley.EDU [128.32.38.234]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA29618; Wed, 4 Jun 1997 16:19:06 -0700 (PDT) Received: (from asami@localhost) by vader.cs.berkeley.edu (8.8.5/8.7.3) id QAA01463; Wed, 4 Jun 1997 16:19:02 -0700 (PDT) Date: Wed, 4 Jun 1997 16:19:02 -0700 (PDT) Message-Id: <199706042319.QAA01463@vader.cs.berkeley.edu> To: ache@nagual.pp.ru CC: bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: (message from =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= on Thu, 5 Jun 1997 03:08:45 +0400 (MSD)) Subject: Re: cvs commit: src/etc/mtree BSD.include.dist From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * What bad happen in keeping the same scheme for release and for source * distribution and avoid such special cases? So are you suggesting that we make users install the source distribution just because they want to compile something? Or are you saying we do it the other way, having them always as directories? * I see some troubles happens if somebody will try to extract source * distribution over release (symlinks which overwrites directories) I think you got this backwards. ;) If you extract a source distribution, it will all go under /usr/src. The /usr/include/* will stay as directories. Only when you do a make world (or say the magic words in /usr/src/include), the directories in /usr/include will be replaced with symlinks to the source dist. * What about deeper level directories I ask? All I said was that there doesn't appear to be any directories not specified in BSD.include.dist (it was an answer to your question to Peter about if there is anything special for the release). Are there any other directories needed under /usr/include for compilation of userland code? Satoshi From owner-cvs-etc Wed Jun 4 16:20:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA29880 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:20:10 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA29767; Wed, 4 Jun 1997 16:19:55 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.8.5/8.7.3) id QAA19758; Wed, 4 Jun 1997 16:19:41 -0700 (PDT) From: "Rodney W. Grimes" Message-Id: <199706042319.QAA19758@GndRsh.aac.dev.com> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706042225.PAA07945@freefall.freebsd.org> from "Andrey A. Chernov" at "Jun 4, 97 03:25:31 pm" To: ache@FreeBSD.ORG (Andrey A. Chernov) Date: Wed, 4 Jun 1997 16:19:41 -0700 (PDT) Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > ache 1997/06/04 15:25:31 PDT > > Modified files: > etc/mtree BSD.include.dist > Log: > Remove directories belongs to /usr/src to stop mtree to modify /usr/src > permissions. > Having them here is wrong from several other poins too: > they are never be a directories (simlinks only), so why give a chance to mtree to make > them as directories? They are not _always_ symlinks, esp. for systems without /usr/src that have been built with SHARED=copies (see /usr/src/include/Makefile for what that does). -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation, Inc. Reliable computers for FreeBSD From owner-cvs-etc Wed Jun 4 16:20:19 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA29906 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:20:19 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA29900; Wed, 4 Jun 1997 16:20:15 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id DAA03384; Thu, 5 Jun 1997 03:19:54 +0400 (MSD) Date: Thu, 5 Jun 1997 03:19:54 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: David Nugent cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706042316.JAA16937@labs.usn.blaze.net.au> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, 5 Jun 1997, David Nugent wrote: > Yes. it's kind of nice to be able to build your hello world program > and ports on a newly installed FreeBSD box without having to also > install kernel sources. Otherwise, we should leave installation of > the entire compiler kit out unless /sys is also installed (which is > actually pretty typical of any Linux installation, and I hated it > there too). None of this may be obvious to the end-user, which is > why it should remain as it is. This method make troubles if you plan to install kernel sources over installed release such as kernel & usr/include files mismatch. Why not to install only needed part of /sys, not whole kernel? -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 16:20:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA29989 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:20:40 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA29965; Wed, 4 Jun 1997 16:20:34 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09460; Wed, 4 Jun 1997 16:20:25 -0700 (PDT) To: asami@cs.berkeley.edu (Satoshi Asami) cc: ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 15:58:11 PDT." <199706042258.PAA01401@vader.cs.berkeley.edu> Date: Wed, 04 Jun 1997 16:20:24 -0700 Message-ID: <9456.865466424@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > At any rate, I suggest you back out the change and ask Jordan for > review. This is clearly his region as it concerns the release and not > only "make world". I'd like to argue for the introduction of SHARED=copies as the hard-wired behavior. If we're ever to create a more compartmentalized system, as everyone is ranting for, then we need to decouple things like "bin" and "src". Jordan From owner-cvs-etc Wed Jun 4 16:22:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA00390 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:22:33 -0700 (PDT) Received: from labs.usn.blaze.net.au (labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA00382; Wed, 4 Jun 1997 16:22:27 -0700 (PDT) Received: from labs.usn.blaze.net.au (local [127.0.0.1]) by labs.usn.blaze.net.au (8.8.5/8.8.5) with ESMTP id JAA16970; Thu, 5 Jun 1997 09:22:21 +1000 (EST) Message-Id: <199706042322.JAA16970@labs.usn.blaze.net.au> X-Mailer: exmh version 2.0gamma 1/27/96 To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 03:08:45 +0400." X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 05 Jun 1997 09:22:20 +1000 From: David Nugent Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > I guess so, unless we mandate the source dist. You know where the > > symlinks are pointing at, right? > > What bad happen in keeping the same scheme for release and for source > distribution and avoid such special cases? > I see some troubles happens if somebody will try to extract source > distribution over release (symlinks which overwrites directories) No, the symlinks go the other way (from /usr into /usr/src), and since all of the source distribution is in /usr/src, there's no danger in overwriting when extracting the source distribution. make world which includes make includes clobbers the directories in /usr/include and (re)makes the symlinks, kernel builds don't use /usr/include anyway, so there's no problem there either. I believe mtree being able to selectively not follow symlinks would be a possible solution here. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia Voice +61-3-9791-9547 Data/BBS +61-3-9792-3507 3:632/348@fidonet davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-cvs-etc Wed Jun 4 16:25:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA00929 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:25:51 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA00878; Wed, 4 Jun 1997 16:25:43 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id DAA03464; Thu, 5 Jun 1997 03:25:22 +0400 (MSD) Date: Thu, 5 Jun 1997 03:25:22 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: Satoshi Asami cc: bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706042319.QAA01463@vader.cs.berkeley.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Wed, 4 Jun 1997, Satoshi Asami wrote: > * What bad happen in keeping the same scheme for release and for source > * distribution and avoid such special cases? > > So are you suggesting that we make users install the source > distribution just because they want to compile something? Or are you > saying we do it the other way, having them always as directories? No, I suggest to install only needed /sys part, i.e. *.h files from it, and keep symlinks as in source case. Current variant will cause /usr/include and /sys files mismatch if somebody extract kernel sources over release. > I think you got this backwards. ;) If you extract a source > distribution, it will all go under /usr/src. The /usr/include/* will > stay as directories. It is just the case I fear: if you change something in kernel include file (i.e. FD_SETSIZE), you need not forget to copy it to /usr/include too or mismatch happens. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 16:27:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA01198 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:27:16 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA01163; Wed, 4 Jun 1997 16:27:12 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09528; Wed, 4 Jun 1997 16:26:58 -0700 (PDT) To: David Nugent cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 09:22:20 +1000." <199706042322.JAA16970@labs.usn.blaze.net.au> Date: Wed, 04 Jun 1997 16:26:58 -0700 Message-ID: <9524.865466818@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I believe mtree being able to selectively not follow symlinks would be > a possible solution here. I think that's also a good idea but, as I said in my other mail, I think the true evil here that we have SHARED=copies on systems which have simply installed a bin dist and we have SHARED=symlinks as the default in /usr/src/include/Makefile so that anyone building the world suddenly diverges from the path. I can also see where the original designers may have thought this a very clever, space-saving idea but it's also Deeply Wrong and should die now. :-) Jordan From owner-cvs-etc Wed Jun 4 16:27:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA01376 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:27:54 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA01355; Wed, 4 Jun 1997 16:27:49 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09547; Wed, 4 Jun 1997 16:27:42 -0700 (PDT) To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 03:25:22 +0400." Date: Wed, 04 Jun 1997 16:27:42 -0700 Message-ID: <9544.865466862@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > No, I suggest to install only needed /sys part, i.e. *.h files from it, > and keep symlinks as in source case. Current variant will cause [Now starts jumping up and down] "No! No symlinks! Death to the symlinks!" Jordan From owner-cvs-etc Wed Jun 4 16:30:21 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA01902 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:30:21 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA01873; Wed, 4 Jun 1997 16:30:13 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id DAA03500; Thu, 5 Jun 1997 03:29:35 +0400 (MSD) Date: Thu, 5 Jun 1997 03:29:35 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: "Jordan K. Hubbard" cc: Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <9456.865466424@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Wed, 4 Jun 1997, Jordan K. Hubbard wrote: > > At any rate, I suggest you back out the change and ask Jordan for > > review. This is clearly his region as it concerns the release and not > > only "make world". > > I'd like to argue for the introduction of SHARED=copies as the > hard-wired behavior. If we're ever to create a more compartmentalized > system, as everyone is ranting for, then we need to decouple things > like "bin" and "src". I see most unpleasant case with current method is potential /sys & /usr/include files mispatch (in case someone bump kernel file limit f.e.) and possible missing availability of some kernel deeper level directories from user program, I suggest to keep symlinks as they are in source case and extract only *.h files from kernel source to /sys to prevent bloating. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 16:31:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA02095 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:31:01 -0700 (PDT) Received: from labs.usn.blaze.net.au (labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA02058; Wed, 4 Jun 1997 16:30:54 -0700 (PDT) Received: from labs.usn.blaze.net.au (local [127.0.0.1]) by labs.usn.blaze.net.au (8.8.5/8.8.5) with ESMTP id JAA17021; Thu, 5 Jun 1997 09:30:46 +1000 (EST) Message-Id: <199706042330.JAA17021@labs.usn.blaze.net.au> X-Mailer: exmh version 2.0gamma 1/27/96 To: "Jordan K. Hubbard" cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 16:20:24 MST." <9456.865466424@time.cdrom.com> X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 05 Jun 1997 09:30:45 +1000 From: David Nugent Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I'd like to argue for the introduction of SHARED=copies as the > hard-wired behavior. If we're ever to create a more compartmentalized > system, as everyone is ranting for, then we need to decouple things > like "bin" and "src". I've got a similar idea, except I'd prefer to see it as optional to avoid the duplication. Err, hey, that's what we've got now! :-) If you change the default, then fine, but add a comment in /etc/make.conf? Regards, David From owner-cvs-etc Wed Jun 4 16:32:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA02337 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:32:11 -0700 (PDT) Received: from lsd.relcom.eu.net (lsd.relcom.eu.net [193.124.23.23]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA02279; Wed, 4 Jun 1997 16:32:02 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by lsd.relcom.eu.net (8.8.5/8.8.5) with SMTP id DAA03521; Thu, 5 Jun 1997 03:31:46 +0400 (MSD) Date: Thu, 5 Jun 1997 03:31:45 +0400 (MSD) From: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= X-Sender: ache@lsd.relcom.eu.net To: "Jordan K. Hubbard" cc: David Nugent , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <9524.865466818@time.cdrom.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Wed, 4 Jun 1997, Jordan K. Hubbard wrote: > I can also see where the original designers may have thought this a > very clever, space-saving idea but it's also Deeply Wrong and should > die now. :-) Extracting only *.h files from kernel dist for release (not whole thing) and keeping symlinks as they are looks as final solution for me. -- Andrey A. Chernov http://www.nagual.pp.ru/~ache/ From owner-cvs-etc Wed Jun 4 16:41:25 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA03744 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:41:25 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA03734; Wed, 4 Jun 1997 16:41:21 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09651; Wed, 4 Jun 1997 16:41:04 -0700 (PDT) To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 03:29:35 +0400." Date: Wed, 04 Jun 1997 16:41:04 -0700 Message-ID: <9647.865467664@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I see most unpleasant case with current method is potential /sys & > /usr/include files mispatch (in case someone bump kernel file limit f.e.) > and possible missing availability of some kernel deeper level directories > from user program, I suggest to keep symlinks as they are in source case I knew that you'd make this exact argument, and it's unfortunately bogus. :-) Updating your include files, both directly in /usr/include and in its subdirectories, is just plain something you should do if you're dinking around with sources and building your own kernels and such. I feel very strongly about that, so strongly that I'll go even further in saying that people who mismatch their /sys and /usr/include directories deserve to lose and that adding an architectural wart like this to try and save them from themselves only bites us in far worse ways than it helps us. It is the WRONG solution now, it was the WRONG solution when we inheirited it and now it needs to DIE DIE DIE. Sorry for the extra dose of vehemence, Andrey, but this crock has been so much like a low-grade toothache for me throughout the years that I basically just got used to it and was ignoring it until you guys called attention to it again. Now I'm wondering WTF is wrong with me that I've lived with it for so long without just going to see a dentist to have that sucker yanked out. ;-) Jordan From owner-cvs-etc Wed Jun 4 16:51:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA05323 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:51:06 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA05311; Wed, 4 Jun 1997 16:51:03 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09707; Wed, 4 Jun 1997 16:50:56 -0700 (PDT) To: David Nugent cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 09:30:45 +1000." <199706042330.JAA17021@labs.usn.blaze.net.au> Date: Wed, 04 Jun 1997 16:50:56 -0700 Message-ID: <9703.865468256@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > I've got a similar idea, except I'd prefer to see it as optional to avoid > the duplication. > > Err, hey, that's what we've got now! :-) > > If you change the default, then fine, but add a comment in /etc/make.conf? I actually want to rip the support out completely, and let me explain why I'm after such a seemingly drastic castration of this mechanism: If I'm to give the FreeBSD community a consolidated set of distribution packages for which full MD5 checksumming and the intelligence to use it for all subcomponents is supported in future "hands off" upgrades that actually work, then the less files in a distribution that they end up tweaking en-masse the better. Make a lot of gratuitous tweaks and you'll only cause the "preserve the user's changes" mechanism to start working against you. Sure, it saves a little space to have the symlinks, but it violates the idea of having includes be self-contained, something which buys us far more in being able to update them as a unit at any time and without odd inter-dependencies on another package which can only cause additional confusion. When the symlinks were invented, disk space was also probably 5X as expensive as it is now. I'd much rather trade the disk space for the increased compartmentalization at this point since C18N is something which FreeBSD is singularly bad at and needs improving. :( Jordan From owner-cvs-etc Wed Jun 4 16:52:45 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id QAA05770 for cvs-etc-outgoing; Wed, 4 Jun 1997 16:52:45 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id QAA05764; Wed, 4 Jun 1997 16:52:42 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id QAA09728; Wed, 4 Jun 1997 16:52:30 -0700 (PDT) To: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= cc: David Nugent , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 03:31:45 +0400." Date: Wed, 04 Jun 1997 16:52:30 -0700 Message-ID: <9725.865468350@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Extracting only *.h files from kernel dist for release (not whole thing) > and keeping symlinks as they are looks as final solution for me. No, please read my other messages - I really want to eliminate *all* the symlinks in /usr/include. It's not useful to me to eliminate only some of them, I need all of them to go if the evil include->source dependency is to be severed (and completely severed it really need to be). Jordan From owner-cvs-etc Wed Jun 4 17:06:36 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA08700 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:06:36 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA08614; Wed, 4 Jun 1997 17:06:27 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id KAA23320; Thu, 5 Jun 1997 10:02:11 +1000 Date: Thu, 5 Jun 1997 10:02:11 +1000 From: Bruce Evans Message-Id: <199706050002.KAA23320@godzilla.zeta.org.au> To: asami@cs.berkeley.edu, jkh@time.cdrom.com Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >I'd like to argue for the introduction of SHARED=copies as the >hard-wired behavior. If we're ever to create a more compartmentalized >system, as everyone is ranting for, then we need to decouple things >like "bin" and "src". First decouple building src from /usr/include. Bruce From owner-cvs-etc Wed Jun 4 17:17:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA12218 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:17:48 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA12198; Wed, 4 Jun 1997 17:17:42 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id KAA23780; Thu, 5 Jun 1997 10:13:56 +1000 Date: Thu, 5 Jun 1997 10:13:56 +1000 From: Bruce Evans Message-Id: <199706050013.KAA23780@godzilla.zeta.org.au> To: davidn@labs.usn.blaze.net.au, jkh@time.cdrom.com Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: ache@nagual.pp.ru, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >I think that's also a good idea but, as I said in my other mail, I >think the true evil here that we have SHARED=copies on systems which >have simply installed a bin dist and we have SHARED=symlinks as the >default in /usr/src/include/Makefile so that anyone building the world >suddenly diverges from the path. > >I can also see where the original designers may have thought this a >very clever, space-saving idea but it's also Deeply Wrong and should >die now. :-) You have this backwards :-). SHARED=copies is a hack to avoid dependencies on /usr/src/*sys. This no longer saves a significant amount of space. Symlinks are fundamentally broken when there are multiple source trees installed, but less so than /usr/include itself. Bruce From owner-cvs-etc Wed Jun 4 17:33:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA17272 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:33:48 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA17245; Wed, 4 Jun 1997 17:33:44 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id RAA09998; Wed, 4 Jun 1997 17:33:22 -0700 (PDT) To: Bruce Evans cc: asami@cs.berkeley.edu, ache@nagual.pp.ru, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 10:02:11 +1000." <199706050002.KAA23320@godzilla.zeta.org.au> Date: Wed, 04 Jun 1997 17:33:22 -0700 Message-ID: <9994.865470802@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > >I'd like to argue for the introduction of SHARED=copies as the > >hard-wired behavior. If we're ever to create a more compartmentalized > >system, as everyone is ranting for, then we need to decouple things > >like "bin" and "src". > > First decouple building src from /usr/include. That would be nice, but that still doesn't stop me from building /usr/include as copies and not symlinks right now. :-) Asking for Rome before removing any of the rubble surrounding it is unrealistic. I agree that what you request should be done, but not as a prerequisite. Jordan From owner-cvs-etc Wed Jun 4 17:38:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA19063 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:38:16 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA19048; Wed, 4 Jun 1997 17:38:12 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id RAA10050; Wed, 4 Jun 1997 17:37:44 -0700 (PDT) To: Bruce Evans cc: davidn@labs.usn.blaze.net.au, ache@nagual.pp.ru, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 10:13:56 +1000." <199706050013.KAA23780@godzilla.zeta.org.au> Date: Wed, 04 Jun 1997 17:37:44 -0700 Message-ID: <10047.865471064@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > >I can also see where the original designers may have thought this a > >very clever, space-saving idea but it's also Deeply Wrong and should > >die now. :-) > > You have this backwards :-). SHARED=copies is a hack to avoid > dependencies on /usr/src/*sys. This no longer saves a significant My mistake. Those designers are due my thanks rather than my bricks in that case. :-) Jordan From owner-cvs-etc Wed Jun 4 17:40:26 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA19375 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:40:26 -0700 (PDT) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA19350; Wed, 4 Jun 1997 17:40:23 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id KAA19374; Thu, 5 Jun 1997 10:09:44 +0930 (CST) From: Michael Smith Message-Id: <199706050039.KAA19374@genesis.atrad.adelaide.edu.au> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <9456.865466424@time.cdrom.com> from "Jordan K. Hubbard" at "Jun 4, 97 04:20:24 pm" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Thu, 5 Jun 1997 10:09:44 +0930 (CST) Cc: asami@cs.berkeley.edu, ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Jordan K. Hubbard stands accused of saying: > > At any rate, I suggest you back out the change and ask Jordan for > > review. This is clearly his region as it concerns the release and not > > only "make world". > > I'd like to argue for the introduction of SHARED=copies as the > hard-wired behavior. If we're ever to create a more compartmentalized > system, as everyone is ranting for, then we need to decouple things > like "bin" and "src". Wasn't the whole idea behind this linking stuff to keep the etc files theoretically in sync with the current kernel so that rebuilding kernel-dependant tools was relatively straightforward? Is this still substantively an issue? > Jordan -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-cvs-etc Wed Jun 4 17:41:49 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA19544 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:41:49 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA19538; Wed, 4 Jun 1997 17:41:47 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id RAA10134; Wed, 4 Jun 1997 17:41:21 -0700 (PDT) To: Michael Smith cc: asami@cs.berkeley.edu, ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 10:09:44 +0930." <199706050039.KAA19374@genesis.atrad.adelaide.edu.au> Date: Wed, 04 Jun 1997 17:41:21 -0700 Message-ID: <10130.865471281@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Wasn't the whole idea behind this linking stuff to keep the > etc files theoretically in sync with the current kernel so that > rebuilding kernel-dependant tools was relatively straightforward? Yes, it was. However, I think that syncronization is an admirable goal which would nonetheless be better achieved in other ways. :) Jordan From owner-cvs-etc Wed Jun 4 17:44:08 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id RAA19811 for cvs-etc-outgoing; Wed, 4 Jun 1997 17:44:08 -0700 (PDT) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id RAA19722; Wed, 4 Jun 1997 17:43:59 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id KAA19423; Thu, 5 Jun 1997 10:13:36 +0930 (CST) From: Michael Smith Message-Id: <199706050043.KAA19423@genesis.atrad.adelaide.edu.au> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <9647.865467664@time.cdrom.com> from "Jordan K. Hubbard" at "Jun 4, 97 04:41:04 pm" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Thu, 5 Jun 1997 10:13:36 +0930 (CST) Cc: ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Jordan K. Hubbard stands accused of saying: > > Updating your include files, both directly in /usr/include and in its > subdirectories, is just plain something you should do if you're > dinking around with sources and building your own kernels and such. I OK, so we expect to see a new makefile rule at the top level of the kernel source tree that will copy all the includes to the relevant places under /usr/include? -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-cvs-etc Wed Jun 4 18:01:58 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA22125 for cvs-etc-outgoing; Wed, 4 Jun 1997 18:01:58 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA22118; Wed, 4 Jun 1997 18:01:53 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id RAA10592; Wed, 4 Jun 1997 17:58:12 -0700 (PDT) To: Michael Smith cc: ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 10:13:36 +0930." <199706050043.KAA19423@genesis.atrad.adelaide.edu.au> Date: Wed, 04 Jun 1997 17:58:11 -0700 Message-ID: <10588.865472291@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > OK, so we expect to see a new makefile rule at the top level of the > kernel source tree that will copy all the includes to the relevant > places under /usr/include? Yeah. It's called "make includes" :-) Seriously, with SHARED=copies as the default and a couple of anomalies like /usr/include/ufs cleaned up, it works pretty well right now. Jordan From owner-cvs-etc Wed Jun 4 18:13:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA24202 for cvs-etc-outgoing; Wed, 4 Jun 1997 18:13:40 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA24192; Wed, 4 Jun 1997 18:13:37 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id SAA28344; Wed, 4 Jun 1997 18:14:34 -0700 (PDT) Message-Id: <199706050114.SAA28344@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: "Jordan K. Hubbard" cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 16:41:04 PDT." <9647.865467664@time.cdrom.com> From: David Greenman Reply-To: dg@root.com Date: Wed, 04 Jun 1997 18:14:34 -0700 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> I see most unpleasant case with current method is potential /sys & >> /usr/include files mispatch (in case someone bump kernel file limit f.e.) >> and possible missing availability of some kernel deeper level directories >> from user program, I suggest to keep symlinks as they are in source case > >I knew that you'd make this exact argument, and it's unfortunately >bogus. :-) > >Updating your include files, both directly in /usr/include and in its >subdirectories, is just plain something you should do if you're >dinking around with sources and building your own kernels and such. I >feel very strongly about that, so strongly that I'll go even further >in saying that people who mismatch their /sys and /usr/include >directories deserve to lose and that adding an architectural wart like >this to try and save them from themselves only bites us in far worse >ways than it helps us. It is the WRONG solution now, it was the WRONG >solution when we inheirited it and now it needs to DIE DIE DIE. > >Sorry for the extra dose of vehemence, Andrey, but this crock has been >so much like a low-grade toothache for me throughout the years that I >basically just got used to it and was ignoring it until you guys >called attention to it again. Now I'm wondering WTF is wrong with me >that I've lived with it for so long without just going to see a >dentist to have that sucker yanked out. ;-) Well, I disagree with you, Jordan. I think we should make the kernel sources a mandatory part of the system and /usr/include/sys should be a symlink. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-cvs-etc Wed Jun 4 18:18:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA24486 for cvs-etc-outgoing; Wed, 4 Jun 1997 18:18:51 -0700 (PDT) Received: from implode.root.com (implode.root.com [198.145.90.17]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA24478; Wed, 4 Jun 1997 18:18:47 -0700 (PDT) Received: from localhost (localhost [127.0.0.1]) by implode.root.com (8.8.5/8.8.5) with SMTP id SAA28368; Wed, 4 Jun 1997 18:19:38 -0700 (PDT) Message-Id: <199706050119.SAA28368@implode.root.com> X-Authentication-Warning: implode.root.com: localhost [127.0.0.1] didn't use HELO protocol To: "Jordan K. Hubbard" cc: Michael Smith , asami@cs.berkeley.edu, ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 17:41:21 PDT." <10130.865471281@time.cdrom.com> From: David Greenman Reply-To: dg@root.com Date: Wed, 04 Jun 1997 18:19:38 -0700 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> Wasn't the whole idea behind this linking stuff to keep the >> etc files theoretically in sync with the current kernel so that >> rebuilding kernel-dependant tools was relatively straightforward? > >Yes, it was. However, I think that syncronization is an admirable >goal which would nonetheless be better achieved in other ways. :) The fact that the sys include files are duplicated in /usr/include is the problem. They shouldn't be there - there probably shouldn't be any symlinks, either, but those are probably better than trying to fix it up in cpp. The kernel include files belong in the /sys tree and nowhere else. -DG David Greenman Core-team/Principal Architect, The FreeBSD Project From owner-cvs-etc Wed Jun 4 18:48:52 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id SAA26463 for cvs-etc-outgoing; Wed, 4 Jun 1997 18:48:52 -0700 (PDT) Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA26456; Wed, 4 Jun 1997 18:48:39 -0700 (PDT) Received: (from bde@localhost) by godzilla.zeta.org.au (8.8.5/8.6.9) id LAA26835; Thu, 5 Jun 1997 11:34:36 +1000 Date: Thu, 5 Jun 1997 11:34:36 +1000 From: Bruce Evans Message-Id: <199706050134.LAA26835@godzilla.zeta.org.au> To: jkh@time.cdrom.com, msmith@atrad.adelaide.edu.au Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk >> Updating your include files, both directly in /usr/include and in its >> subdirectories, is just plain something you should do if you're >> dinking around with sources and building your own kernels and such. I > >OK, so we expect to see a new makefile rule at the top level of the >kernel source tree that will copy all the includes to the relevant >places under /usr/include? No. Putting it at the top level doesn't help at low levels. Otherwise, the standard rules in /usr/src/include/Makefile are sufficient (cd /usr/src/include; make; make install). Isn't this how everyone keeps their include files up to date? :-) Bruce From owner-cvs-etc Wed Jun 4 20:17:04 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA03510 for cvs-etc-outgoing; Wed, 4 Jun 1997 20:17:04 -0700 (PDT) Received: from androcles.com (androcles.com [204.57.240.10]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA03347; Wed, 4 Jun 1997 20:16:08 -0700 (PDT) Received: (from dhh@localhost) by androcles.com (8.8.5/8.6.12) id UAA00292; Wed, 4 Jun 1997 20:16:32 -0700 (PDT) Message-ID: X-Mailer: XFMail 1.0 [p0] on FreeBSD Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <9725.865468350@time.cdrom.com> Date: Wed, 04 Jun 1997 20:14:56 -0700 (PDT) From: "Duane H. Hesser" To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: cvs-etc@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG, David Nugent , =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Yaaaay!, Jordan... On 04-Jun-97 Jordan K. Hubbard wrote: >> Extracting only *.h files from kernel dist for release (not whole thing) >> and keeping symlinks as they are looks as final solution for me. > >No, please read my other messages - I really want to eliminate *all* >the symlinks in /usr/include. It's not useful to me to eliminate only >some of them, I need all of them to go if the evil include->source >dependency is to be severed (and completely severed it really need to >be). > > Jordan > Duane H. Hesser dhh@androcles.com From owner-cvs-etc Wed Jun 4 20:47:31 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA10765 for cvs-etc-outgoing; Wed, 4 Jun 1997 20:47:31 -0700 (PDT) Received: from mail.webspan.net (root@mail.webspan.net [206.154.70.7]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id UAA10674; Wed, 4 Jun 1997 20:46:59 -0700 (PDT) Received: from orion.webspan.net (orion.webspan.net [206.154.70.5]) by mail.webspan.net (8.8.5/8.8.5) with ESMTP id XAA22149; Wed, 4 Jun 1997 23:46:55 -0400 (EDT) Received: from orion.webspan.net (localhost [127.0.0.1]) by orion.webspan.net (WEBSPN/970116) with ESMTP id XAA10336; Wed, 4 Jun 1997 23:46:55 -0400 (EDT) To: "Jordan K. Hubbard" cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG From: "Gary Palmer" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 16:41:04 PDT." <9647.865467664@time.cdrom.com> Date: Wed, 04 Jun 1997 23:46:54 -0400 Message-ID: <10333.865482414@orion.webspan.net> Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk "Jordan K. Hubbard" wrote in message ID <9647.865467664@time.cdrom.com>: > called attention to it again. Now I'm wondering WTF is wrong with me > that I've lived with it for so long without just going to see a > dentist to have that sucker yanked out. ;-) Satoshi & Bills masochism is spreading? :-) Or maybe they got it from you in the first place? :) Gary -- Gary Palmer FreeBSD Core Team Member FreeBSD: Turning PC's into workstations. See http://www.FreeBSD.ORG/ for info From owner-cvs-etc Wed Jun 4 21:31:51 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id VAA18934 for cvs-etc-outgoing; Wed, 4 Jun 1997 21:31:51 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id VAA18911; Wed, 4 Jun 1997 21:31:46 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id VAA11208; Wed, 4 Jun 1997 21:31:25 -0700 (PDT) To: dg@root.com cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 18:14:34 PDT." <199706050114.SAA28344@implode.root.com> Date: Wed, 04 Jun 1997 21:31:24 -0700 Message-ID: <11204.865485084@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Well, I disagree with you, Jordan. I think we should make the kernel > sources a mandatory part of the system and /usr/include/sys should be a > symlink. Well, that would still get me what I want since they would then be part of "bin" (you're not exactly _disagreeing_ with me, you're simply proposing another way of getting to the same place), but I still think that this is an area where you'd get a LOT of negative feedback. Jordan From owner-cvs-etc Wed Jun 4 23:00:33 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA27276 for cvs-etc-outgoing; Wed, 4 Jun 1997 23:00:33 -0700 (PDT) Received: from alpha.xerox.com (alpha.Xerox.COM [13.1.64.93]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id XAA27249; Wed, 4 Jun 1997 23:00:12 -0700 (PDT) Received: from crevenia.parc.xerox.com ([13.2.116.11]) by alpha.xerox.com with SMTP id <17608(7)>; Wed, 4 Jun 1997 22:59:41 PDT Received: from localhost by crevenia.parc.xerox.com with SMTP id <177489>; Wed, 4 Jun 1997 22:59:28 -0700 To: cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 97 16:50:56 PDT." <9703.865468256@time.cdrom.com> Date: Wed, 4 Jun 1997 22:59:25 PDT From: Bill Fenner Message-Id: <97Jun4.225928pdt.177489@crevenia.parc.xerox.com> Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Not to mention that the symlinks make it easy to have a newer /usr/include/* than rest of system -- what happens when you "make world", then cvs update /usr/src but don't "make world"? You're going to end up with includes in /usr/include that don't agree with your world. (I admit that it's a little weird to update /usr/src and not make world, but it's so easy to make it not be a problem...) Bill From owner-cvs-etc Thu Jun 5 00:10:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA29672 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:10:15 -0700 (PDT) Received: from labs.usn.blaze.net.au (labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA29665; Thu, 5 Jun 1997 00:10:07 -0700 (PDT) Received: from labs.usn.blaze.net.au (local [127.0.0.1]) by labs.usn.blaze.net.au (8.8.5/8.8.5) with ESMTP id RAA00482; Thu, 5 Jun 1997 17:09:42 +1000 (EST) Message-Id: <199706050709.RAA00482@labs.usn.blaze.net.au> X-Mailer: exmh version 2.0gamma 1/27/96 To: "Jordan K. Hubbard" cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Wed, 04 Jun 1997 16:50:56 MST." <9703.865468256@time.cdrom.com> X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Thu, 05 Jun 1997 17:09:40 +1000 From: David Nugent Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Sure, it saves a little space to have the symlinks, but it violates > the idea of having includes be self-contained, something which buys us > far more in being able to update them as a unit at any time and > without odd inter-dependencies on another package which can only cause > additional confusion. Ok. If we go this route - and I have no major objection - can we completely detach builds in /usr/src from hardwired /usr/include and /usr/lib? Possibly others too, like /usr/share/mk. I realise this is more extreme than what you're saying, but if you see where I'm pointing, it'll ultimately make it possible to build a completely different version of the OS on any other version without the runtime dependencies in the installed system, and it should not have any effect on the installed system either. This necessarily means using the compiler, binaries, libs and includes from $DESTDIR consistently throughout the source tree. >From my last attempt to do this, it wasn't really possible without first building the base system into $DESTDIR, chroot and build again; and even then there were no guarantees if the installed OS was markedly different from what you were attempting to build. Only the the basic tools required to build the bootstrap should come from the installed system. In other words, "make release", apart from the distributions building process, would come closer to a one-pass build rather than two full passes as is currently the case. I'm not sure how possible this is or how close we are to being able to achieve it now. All I know is that every time I've tested using alternate $DESTDIR's from "/" I've seen all sorts of subtle breakages. Bruce fixed some gcc related ones recently which were the main culprits (namely special casing old style declarations which typically caused compilation of yacc output to fail here and there), but I haven't tried since then for temporary lack of disk space. I also attempted a few weeks ago to build a 2.2.1 release on a 2.1.7 system. This was less than successful as you can imagine, but after a little fudging to get to the chroot/build stage I was able to get it done. I was under pressure at the time to do it quickly, so unfortunately I wasn't taking notes. :-) I did think that the time that it was theoretically possible to do if only the build process was completely independant from the rest of the system, which seems to be the direction you want to head anyway. Regards, David David Nugent - Unique Computing Pty Ltd - Melbourne, Australia Voice +61-3-9791-9547 Data/BBS +61-3-9792-3507 3:632/348@fidonet davidn@freebsd.org davidn@blaze.net.au http://www.blaze.net.au/~davidn/ From owner-cvs-etc Thu Jun 5 00:22:38 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA00238 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:22:38 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA00231; Thu, 5 Jun 1997 00:22:36 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id AAA12202; Thu, 5 Jun 1997 00:22:12 -0700 (PDT) To: David Nugent cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 17:09:40 +1000." <199706050709.RAA00482@labs.usn.blaze.net.au> Date: Thu, 05 Jun 1997 00:22:12 -0700 Message-ID: <12199.865495332@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Ok. If we go this route - and I have no major objection - can we > completely detach builds in /usr/src from hardwired /usr/include and > /usr/lib? Possibly others too, like /usr/share/mk. I realise this is > more extreme than what you're saying, but if you see where I'm pointing, > it'll ultimately make it possible to build a completely different version > of the OS on any other version without the runtime dependencies in > the installed system, and it should not have any effect on the > installed system either. I'm all for it. Are you really clear on what's entailed in this, however? :-) Jordan From owner-cvs-etc Thu Jun 5 00:24:50 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA00466 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:24:50 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (wck-ca7-22.ix.netcom.com [204.31.231.54]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA00436; Thu, 5 Jun 1997 00:24:39 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id AAA16343; Thu, 5 Jun 1997 00:24:16 -0700 (PDT) Date: Thu, 5 Jun 1997 00:24:16 -0700 (PDT) Message-Id: <199706050724.AAA16343@silvia.HIP.Berkeley.EDU> To: davidn@labs.usn.blaze.net.au CC: jkh@time.cdrom.com, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: <199706050709.RAA00482@labs.usn.blaze.net.au> (message from David Nugent on Thu, 05 Jun 1997 17:09:40 +1000) Subject: Re: cvs commit: src/etc/mtree BSD.include.dist From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * it'll ultimately make it possible to build a completely different version * of the OS on any other version without the runtime dependencies in * the installed system, and it should not have any effect on the * installed system either. Actually I have been working on this for a while with Bruce and John P. (You may have seen some of the DESTDIR commits lately.) If you are interested, please send me a note. * This necessarily means using the compiler, binaries, libs and * includes from $DESTDIR consistently throughout the source tree. * From my last attempt to do this, it wasn't really possible * without first building the base system into $DESTDIR, chroot * and build again; and even then there were no guarantees if the * installed OS was markedly different from what you were attempting * to build. Only the the basic tools required to build the bootstrap * should come from the installed system. Basicaly what I do is build only necessary stuff in $DESTDIR. No chroot is necessary. At least if you are building an Intel FreeBSD system on an Intel FreeBSD system, you have a fairly good luck. I could build 3.0 on 2.1.5. (Well almost, I'm still trying to figure out how to fix "make".) Satoshi From owner-cvs-etc Thu Jun 5 00:27:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA00605 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:27:07 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA00600; Thu, 5 Jun 1997 00:27:04 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id AAA12240; Thu, 5 Jun 1997 00:26:48 -0700 (PDT) To: asami@cs.berkeley.edu (Satoshi Asami) cc: davidn@labs.usn.blaze.net.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 00:24:16 PDT." <199706050724.AAA16343@silvia.HIP.Berkeley.EDU> Date: Thu, 05 Jun 1997 00:26:48 -0700 Message-ID: <12236.865495608@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Basicaly what I do is build only necessary stuff in $DESTDIR. No > chroot is necessary. At least if you are building an Intel FreeBSD > system on an Intel FreeBSD system, you have a fairly good luck. I > could build 3.0 on 2.1.5. (Well almost, I'm still trying to figure > out how to fix "make".) You mean this problem? ... cc -O2 -pipe -I/usr/src/usr.bin/make -c /usr/src/usr.bin/make/arch.c In file included from /usr/include/sys/time.h:118, from /usr/include/sys/stat.h:50, from /usr/src/usr.bin/make/arch.c:94: /usr/include/time.h:92: redefinition of `struct timespec' /usr/include/time.h:144: parse error before `struct' /usr/include/time.h:145: parse error before `struct' /usr/include/time.h:146: parse error before `const' :-) I've been investigating this too, trying to make a single SNAP server produce releases on multiple branches. Jordan From owner-cvs-etc Thu Jun 5 00:32:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA00834 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:32:40 -0700 (PDT) Received: from silvia.HIP.Berkeley.EDU (wck-ca7-22.ix.netcom.com [204.31.231.54]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA00812; Thu, 5 Jun 1997 00:32:26 -0700 (PDT) Received: (from asami@localhost) by silvia.HIP.Berkeley.EDU (8.8.5/8.6.9) id AAA16569; Thu, 5 Jun 1997 00:32:19 -0700 (PDT) Date: Thu, 5 Jun 1997 00:32:19 -0700 (PDT) Message-Id: <199706050732.AAA16569@silvia.HIP.Berkeley.EDU> To: jkh@time.cdrom.com CC: davidn@labs.usn.blaze.net.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: <12236.865495608@time.cdrom.com> (jkh@time.cdrom.com) Subject: Re: cvs commit: src/etc/mtree BSD.include.dist From: asami@cs.berkeley.edu (Satoshi Asami) Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk * You mean this problem? * * ..... * cc -O2 -pipe -I/usr/src/usr.bin/make -c /usr/src/usr.bin/make/arch.c * In file included from /usr/include/sys/time.h:118, * from /usr/include/sys/stat.h:50, * from /usr/src/usr.bin/make/arch.c:94: * /usr/include/time.h:92: redefinition of `struct timespec' * /usr/include/time.h:144: parse error before `struct' * /usr/include/time.h:145: parse error before `struct' * /usr/include/time.h:146: parse error before `const' Huh? I haven't seen that one. I'm talking about this: === >> make ===> share/info make: illegal option -- m usage: make [-eiknqrst] [-D variable] [-d flags] [-f makefile ] [-I directory] [-j max_jobs] [variable=value] *** Error code 2 Stop. === The old make didn't have the "-m" option. ;) I got a suggestion from Bruce, that I will try out. Satoshi From owner-cvs-etc Thu Jun 5 00:39:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id AAA01036 for cvs-etc-outgoing; Thu, 5 Jun 1997 00:39:30 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id AAA01031; Thu, 5 Jun 1997 00:39:27 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id AAA12316; Thu, 5 Jun 1997 00:39:24 -0700 (PDT) To: asami@cs.berkeley.edu (Satoshi Asami) cc: davidn@labs.usn.blaze.net.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 00:32:19 PDT." <199706050732.AAA16569@silvia.HIP.Berkeley.EDU> Date: Thu, 05 Jun 1997 00:39:24 -0700 Message-ID: <12312.865496364@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Huh? I haven't seen that one. I'm talking about this: Ah, OK, yours is somewhat different - I've got an include race over here. :-) Jordan From owner-cvs-etc Thu Jun 5 04:18:29 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id EAA08666 for cvs-etc-outgoing; Thu, 5 Jun 1997 04:18:29 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id EAA08660; Thu, 5 Jun 1997 04:18:24 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id LAA02530; Thu, 5 Jun 1997 11:40:33 GMT From: Adam David Message-Id: <199706051140.LAA02530@veda.is> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706050114.SAA28344@implode.root.com> from David Greenman at "Jun 4, 97 06:14:34 pm" To: dg@root.com Date: Thu, 5 Jun 1997 11:40:33 +0000 (GMT) Cc: jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Well, I disagree with you, Jordan. I think we should make the kernel > sources a mandatory part of the system and /usr/include/sys should be a > symlink. > > -DG > > David Greenman > Core-team/Principal Architect, The FreeBSD Project > Let me second this, and add the rationale that most people don't want to be running a kernel.GENERIC bloated with all kinds of drivers for devices they probably don't have (and not all drivers that they do need). Of course, this all changes when kernel modules are better integrated. However, building a custom kernel will likely always be a typical operation during post-installation machine configuration. -- Adam David From owner-cvs-etc Thu Jun 5 05:08:30 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id FAA10060 for cvs-etc-outgoing; Thu, 5 Jun 1997 05:08:30 -0700 (PDT) Received: from mushi.colo.neosoft.com (qmailr@mushi.colo.neosoft.com [206.109.6.82]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id FAA10039 for ; Thu, 5 Jun 1997 05:08:24 -0700 (PDT) Received: (qmail 24915 invoked from network); 5 Jun 1997 12:08:11 -0000 Received: from bonkers.neosoft.com (HELO bonkers.taronga.com) (root@206.109.2.48) by mushi.colo.neosoft.com with SMTP; 5 Jun 1997 12:08:11 -0000 Received: (from peter@localhost) by bonkers.taronga.com (8.6.11/8.6.9) id HAA25458; Thu, 5 Jun 1997 07:10:30 -0500 From: peter@taronga.com (Peter da Silva) Message-Id: <199706051210.HAA25458@bonkers.taronga.com> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist To: adam@veda.is (Adam David) Date: Thu, 5 Jun 1997 07:10:29 -0500 (CDT) Cc: dg@root.com, jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG In-Reply-To: <199706051140.LAA02530@veda.is> from "Adam David" at Jun 5, 97 11:40:33 am X-Mailer: ELM [version 2.4 PL24] Content-Type: text Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Of course, this all changes when kernel modules are better integrated. > However, building a custom kernel will likely always be a typical operation > during post-installation machine configuration. The only reason I've built a kernel at all since the boot -c came in has been to add the bpfilter and ppp devices, or because I was hacking on something. Of course once I get in I generally trim things down a bunch, but I don't think we can assume that a production machine is going to be running a custom kernel. From owner-cvs-etc Thu Jun 5 05:16:07 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id FAA10282 for cvs-etc-outgoing; Thu, 5 Jun 1997 05:16:07 -0700 (PDT) Received: from veda.is (ubiq.veda.is [193.4.230.60]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA10274; Thu, 5 Jun 1997 05:16:04 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.7.3) id MAA02627; Thu, 5 Jun 1997 12:37:46 GMT From: Adam David Message-Id: <199706051237.MAA02627@veda.is> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706051210.HAA25458@bonkers.taronga.com> from Peter da Silva at "Jun 5, 97 07:10:29 am" To: peter@taronga.com (Peter da Silva) Date: Thu, 5 Jun 1997 12:37:45 +0000 (GMT) Cc: dg@root.com, jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > The only reason I've built a kernel at all since the boot -c came in has been > to add the bpfilter and ppp devices, or because I was hacking on something. > > Of course once I get in I generally trim things down a bunch, but I don't think > we can assume that a production machine is going to be running a custom kernel. I would still say custom kernels are in use in typical installations, while not universally so. -- Adam David From owner-cvs-etc Thu Jun 5 05:42:02 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id FAA11231 for cvs-etc-outgoing; Thu, 5 Jun 1997 05:42:02 -0700 (PDT) Received: from pillar.elsevier.co.uk (root@pillar.elsevier.co.uk [193.131.222.35]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id FAA11204; Thu, 5 Jun 1997 05:41:33 -0700 (PDT) Received: from snowdon.elsevier.co.uk (snowdon.elsevier.co.uk [193.131.197.164]) by pillar.elsevier.co.uk (8.8.5/8.8.5) with ESMTP id NAA25520; Thu, 5 Jun 1997 13:38:08 +0100 (BST) Received: from cadair.elsevier.co.uk by snowdon.elsevier.co.uk with SMTP (PP); Thu, 5 Jun 1997 13:41:01 +0100 Received: from tees.elsevier.co.uk (tees.elsevier.co.uk [193.131.197.60]) by cadair.elsevier.co.uk (8.8.5/8.8.5) with ESMTP id NAA02709; Thu, 5 Jun 1997 13:40:55 +0100 (BST) Received: (from dpr@localhost) by tees.elsevier.co.uk (8.8.5/8.8.5) id NAA02128; Thu, 5 Jun 1997 13:40:54 +0100 (BST) To: dg@root.com Cc: "Jordan K. Hubbard" , =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist References: <199706050114.SAA28344@implode.root.com> From: Paul Richards Date: 05 Jun 1997 13:40:53 +0100 In-Reply-To: David Greenman's message of Wed, 04 Jun 1997 18:14:34 -0700 Message-ID: <57bu5li5nu.fsf@tees.elsevier.co.uk> Lines: 14 X-Mailer: Gnus v5.4.37/Emacs 19.30 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk David Greenman writes: > Well, I disagree with you, Jordan. I think we should make the kernel > sources a mandatory part of the system and /usr/include/sys should be a > symlink. Why ? I don't see any reason why kernel sources (or any sources) should be mandatory. I agree with Jordan that the sources (or includes in this specific case) should be detached from the installed system. -- Dr Paul Richards. [p.richards@elsevier.co.uk] Originative Solutions Ltd. [paul@originat.demon.co.uk] Phone: 0370 462071 (Mobile), +44 (0)1865 843155 (Elsevier) From owner-cvs-etc Thu Jun 5 07:25:00 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id HAA15713 for cvs-etc-outgoing; Thu, 5 Jun 1997 07:25:00 -0700 (PDT) Received: from sumatra.americantv.com (sumatra.americantv.com [207.170.17.37]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA15705; Thu, 5 Jun 1997 07:24:51 -0700 (PDT) Received: from right.PCS (right.PCS [148.105.10.31]) by sumatra.americantv.com (8.8.5/8.8.5) with ESMTP id JAA06079; Thu, 5 Jun 1997 09:58:39 -0500 (CDT) Received: (jlemon@localhost) by right.PCS (8.6.13/8.6.4) id JAA18077; Thu, 5 Jun 1997 09:25:54 -0500 Message-ID: <19970605092554.57000@right.PCS> Date: Thu, 5 Jun 1997 09:25:54 -0500 From: Jonathan Lemon To: Bill Fenner Cc: cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist References: <9703.865468256@time.cdrom.com> <97Jun4.225928pdt.177489@crevenia.parc.xerox.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.61.1 In-Reply-To: <97Jun4.225928pdt.177489@crevenia.parc.xerox.com>; from Bill Fenner on Jun 06, 1997 at 10:59:25PM -0800 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Jun 06, 1997 at 10:59:25PM -0800, Bill Fenner wrote: > Not to mention that the symlinks make it easy to have a newer > /usr/include/* than rest of system -- what happens when you > "make world", then cvs update /usr/src but don't "make world"? > You're going to end up with includes in /usr/include that don't > agree with your world. > > (I admit that it's a little weird to update /usr/src and not > make world, but it's so easy to make it not be a problem...) Not at all. 'make world' now takes about 12 hours on my poor little 486, up from 8 hours a few months ago. I might do repeated runs of 'cvsup' during the week, but put off rebuilding until later. -- Jonathan From owner-cvs-etc Thu Jun 5 09:10:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA21775 for cvs-etc-outgoing; Thu, 5 Jun 1997 09:10:27 -0700 (PDT) Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA21764; Thu, 5 Jun 1997 09:10:20 -0700 (PDT) Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.8.5/8.7.3) id JAA20298; Thu, 5 Jun 1997 09:08:38 -0700 (PDT) From: "Rodney W. Grimes" Message-Id: <199706051608.JAA20298@GndRsh.aac.dev.com> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <10588.865472291@time.cdrom.com> from "Jordan K. Hubbard" at "Jun 4, 97 05:58:11 pm" To: jkh@time.cdrom.com (Jordan K. Hubbard) Date: Thu, 5 Jun 1997 09:08:38 -0700 (PDT) Cc: msmith@atrad.adelaide.edu.au, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL25 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > OK, so we expect to see a new makefile rule at the top level of the > > kernel source tree that will copy all the includes to the relevant > > places under /usr/include? > > Yeah. It's called "make includes" :-) > > Seriously, with SHARED=copies as the default and a couple of anomalies > like /usr/include/ufs cleaned up, it works pretty well right now. > > Jordan Since I was the one who turned on SHARED=copies and fixed it up to do the right things way back in the 1.X days for releases I feel I need to say a few things about why it _is_ a good idea in support of Jordan here, though I won't go so far as to support ripping out the SHARED knob. 1) The reason it was turned on at all was so that a binary distribution would include a complete /usr/include tree so that the compiler(s) could find them without needing a /sys tree. In the 386BSD days Bill shipped the system with SHARED=symlinks (the Berkeley default as far back as I can remember) and it was a FAQ on the list that you ``need to install the src distribution to be able to compile programs'' (at that time the sources where all in 1 distribution pack, some 15 odd floppies if I recall correctly.) 2) /usr/include containes the interface to the currently running /usr/lib and /kernel, _not_ to the current contents of /usr/src. /usr/include should be updated when things are installed from /usr/src into /usr/lib or a new kernel is installed. Not one second before, as doing so causes a potential miss match in API's. 3) Some day /usr/src is not going to depend on /usr/include, the opposite should be true, and can easily be done today with the correct setting of SHARED=copies. 4) I've been running my *BSD systems for 5 years with SHARED=copies and have had no problems. Also all of the systems I build and ship have been setup that way and I have yet to here a customer have a problem with it. 5) On a multideveloper machine you do not want one developers current edits to /usr/src/sys/param.h that are not in /kernel to effect the compilation of some user land program. (ooppss.. guess this is the same things as 2 above). 6) There are still many actual files installed in /usr/include when SHARED=symlinks, not all /usr/src changes are instantly propogated to /usr/include. This is an inconsistent nightmare IMHO. Now, the reason to leave the SHARED knob in (ie, I am just advocating changing the default from symlinks to copies, until such time as /usr/src stops depending on /usr/include, at which time the knob can be ripped out since it would then be pointless.) 1) A developer working on a kernel/userland interface issue can easily save some time buy running with the symlinks. A change in a kernel header file is instantly avaliable to test compile the userland utility. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation, Inc. Reliable computers for FreeBSD From owner-cvs-etc Thu Jun 5 09:55:53 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id JAA24120 for cvs-etc-outgoing; Thu, 5 Jun 1997 09:55:53 -0700 (PDT) Received: from bitbox.follo.net (bitbox.follo.net [194.198.43.36]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA24076; Thu, 5 Jun 1997 09:55:12 -0700 (PDT) Received: (from eivind@localhost) by bitbox.follo.net (8.8.5/8.7.3) id SAA06705; Thu, 5 Jun 1997 18:54:52 +0200 (MET DST) Date: Thu, 5 Jun 1997 18:54:52 +0200 (MET DST) Message-Id: <199706051654.SAA06705@bitbox.follo.net> From: Eivind Eklund To: "Jordan K. Hubbard" CC: ache@nagual.pp.ru, davidn@labs.usn.blaze.net.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG In-reply-to: "Jordan K. Hubbard"'s message of Wed, 04 Jun 1997 16:52:30 -0700 Subject: Re: cvs commit: src/etc/mtree BSD.include.dist References: <9725.865468350@time.cdrom.com> Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > > > Extracting only *.h files from kernel dist for release (not whole thing) > > and keeping symlinks as they are looks as final solution for me. > > No, please read my other messages - I really want to eliminate *all* > the symlinks in /usr/include. It's not useful to me to eliminate only > some of them, I need all of them to go if the evil include->source > dependency is to be severed (and completely severed it really need to > be). I'm definitely in favor. I don't get bitten _that_ often, but I still consider /usr/include depending directly on /usr/src a very dirty wart, especially when not _all_ of /usr/include is created this way. Disk space is cheap and getting cheaper; time is expensive and getting more so. Eivind. From owner-cvs-etc Thu Jun 5 12:12:00 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01209 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:12:00 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01161; Thu, 5 Jun 1997 12:11:30 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZhwR-0002yw-00; Thu, 5 Jun 1997 13:11:23 -0600 To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Wed, 04 Jun 1997 16:20:24 PDT." <9456.865466424@time.cdrom.com> References: <9456.865466424@time.cdrom.com> Date: Thu, 05 Jun 1997 13:11:23 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <9456.865466424@time.cdrom.com> "Jordan K. Hubbard" writes: : I'd like to argue for the introduction of SHARED=copies as the : hard-wired behavior. If we're ever to create a more compartmentalized : system, as everyone is ranting for, then we need to decouple things : like "bin" and "src". I'd argue for that for other reasons. I don't want my "installed" include files to change out from under me as I do a cvs update in my /usr/src/sys area. This is just wrong, because that's what make install is for. Some people like it, I know, but in general, for most people it is confusing and tends to cause lots of errors. Oh, I tend to have /usr/src on a jaz drive, and if I unmount it, I can't compile *ANYTHING*. Warner From owner-cvs-etc Thu Jun 5 12:14:28 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01412 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:14:28 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01365; Thu, 5 Jun 1997 12:13:51 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZhyA-0002zH-00; Thu, 5 Jun 1997 13:13:10 -0600 To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Wed, 04 Jun 1997 16:27:42 PDT." <9544.865466862@time.cdrom.com> References: <9544.865466862@time.cdrom.com> Date: Thu, 05 Jun 1997 13:13:10 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <9544.865466862@time.cdrom.com> "Jordan K. Hubbard" writes: : > No, I suggest to install only needed /sys part, i.e. *.h files from it, : > and keep symlinks as in source case. Current variant will cause : : [Now starts jumping up and down] "No! No symlinks! Death to the symlinks!" "Shrink, I want to kill. I want to see veins in my teeth. Eat Dead babies. I mean Kill Kill Kill" and he started jumping up and down yelling "Kill kill kill" and we were both jumping up and down yelling "kill kill kill" and a sargent came by and pinned a metal on me.... DEATH TO THIS STUPID SYMLINK BY DEFAULT. Warner From owner-cvs-etc Thu Jun 5 12:15:12 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01477 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:15:12 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01434; Thu, 5 Jun 1997 12:14:54 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZhzQ-0002zV-00; Thu, 5 Jun 1997 13:14:28 -0600 To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Wed, 04 Jun 1997 16:41:04 PDT." <9647.865467664@time.cdrom.com> References: <9647.865467664@time.cdrom.com> Date: Thu, 05 Jun 1997 13:14:28 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <9647.865467664@time.cdrom.com> "Jordan K. Hubbard" writes: : Sorry for the extra dose of vehemence, Andrey, but this crock has been : so much like a low-grade toothache for me throughout the years that I : basically just got used to it and was ignoring it until you guys : called attention to it again. Now I'm wondering WTF is wrong with me : that I've lived with it for so long without just going to see a : dentist to have that sucker yanked out. ;-) Let me dog pile on this one. Let's get FreeBSD to a Dentist! Warner From owner-cvs-etc Thu Jun 5 12:16:48 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01553 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:16:48 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01529; Thu, 5 Jun 1997 12:16:36 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZi11-0002zo-00; Thu, 5 Jun 1997 13:16:07 -0600 To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , David Nugent , cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Wed, 04 Jun 1997 16:52:30 PDT." <9725.865468350@time.cdrom.com> References: <9725.865468350@time.cdrom.com> Date: Thu, 05 Jun 1997 13:16:07 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <9725.865468350@time.cdrom.com> "Jordan K. Hubbard" writes: : > Extracting only *.h files from kernel dist for release (not whole thing) : > and keeping symlinks as they are looks as final solution for me. : : No, please read my other messages - I really want to eliminate *all* : the symlinks in /usr/include. It's not useful to me to eliminate only : some of them, I need all of them to go if the evil include->source : dependency is to be severed (and completely severed it really need to : be). cd /usr/src ; cvs update -PAd Should *NOT* update anything in /usr/include, directly or indirectly. Eliminating the symlinks completely solves this problem. The small amount of disk space savings just isn't worth the hassles. Warner "Dog Pile" Losh From owner-cvs-etc Thu Jun 5 12:22:16 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01831 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:22:16 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01791; Thu, 5 Jun 1997 12:21:36 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZi5l-00030o-00; Thu, 5 Jun 1997 13:21:01 -0600 To: "Jordan K. Hubbard" Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: Michael Smith , asami@cs.berkeley.edu, ache@nagual.pp.ru, bde@zeta.org.au, cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Wed, 04 Jun 1997 17:41:21 PDT." <10130.865471281@time.cdrom.com> References: <10130.865471281@time.cdrom.com> Date: Thu, 05 Jun 1997 13:21:01 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <10130.865471281@time.cdrom.com> "Jordan K. Hubbard" writes: : > Wasn't the whole idea behind this linking stuff to keep the : > etc files theoretically in sync with the current kernel so that : > rebuilding kernel-dependant tools was relatively straightforward? : : Yes, it was. However, I think that syncronization is an admirable : goal which would nonetheless be better achieved in other ways. :) That's what make install is for. Or the make includes variant of make install. Besides, it introduces some interesting security problems. If my account is compromized, and I own the header files, the intruder can then insert arbitrary code into all or nearly programs on the system. If the include files are installed properly, then this can't happen. I always do a 'cvs diff -u -r HEAD' before doing a make install, before you say what about... Warner From owner-cvs-etc Thu Jun 5 12:25:11 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id MAA01972 for cvs-etc-outgoing; Thu, 5 Jun 1997 12:25:11 -0700 (PDT) Received: from rover.village.org (rover.village.org [204.144.255.49]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id MAA01942; Thu, 5 Jun 1997 12:24:58 -0700 (PDT) Received: from rover.village.org [127.0.0.1] by rover.village.org with esmtp (Exim 1.60 #1) id 0wZi8y-000316-00; Thu, 5 Jun 1997 13:24:20 -0600 To: David Nugent Subject: Re: cvs commit: src/etc/mtree BSD.include.dist Cc: "Jordan K. Hubbard" , cvs-all@freebsd.org, cvs-committers@freebsd.org, cvs-etc@freebsd.org In-reply-to: Your message of "Thu, 05 Jun 1997 17:09:40 +1000." <199706050709.RAA00482@labs.usn.blaze.net.au> References: <199706050709.RAA00482@labs.usn.blaze.net.au> Date: Thu, 05 Jun 1997 13:24:20 -0600 From: Warner Losh Message-Id: Sender: owner-cvs-etc@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In message <199706050709.RAA00482@labs.usn.blaze.net.au> David Nugent writes: : Ok. If we go this route - and I have no major objection - can we : completely detach builds in /usr/src from hardwired /usr/include and : /usr/lib? Possibly others too, like /usr/share/mk. I realise this is : more extreme than what you're saying, but if you see where I'm pointing, : it'll ultimately make it possible to build a completely different version : of the OS on any other version without the runtime dependencies in : the installed system, and it should not have any effect on the : installed system either. I'd love to see this for cross compilation purposes. It will take some doing, however. I don't think it should be required to move forward on the symlink stuff, however, that the whole tree build w/o hard wired, directly or indirectly, /usr/lib and /usr/include. Warner From owner-cvs-etc Thu Jun 5 14:23:54 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id OAA07871 for cvs-etc-outgoing; Thu, 5 Jun 1997 14:23:54 -0700 (PDT) Received: from time.cdrom.com (root@time.cdrom.com [204.216.27.226]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id OAA07861; Thu, 5 Jun 1997 14:23:51 -0700 (PDT) Received: from time.cdrom.com (jkh@localhost [127.0.0.1]) by time.cdrom.com (8.8.5/8.6.9) with ESMTP id OAA16874; Thu, 5 Jun 1997 14:23:12 -0700 (PDT) To: Warner Losh cc: =?KOI8-R?B?4c7E0sXKIP7F0s7P1w==?= , Satoshi Asami , bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-reply-to: Your message of "Thu, 05 Jun 1997 13:13:10 MDT." Date: Thu, 05 Jun 1997 14:23:12 -0700 Message-ID: <16871.865545792@time.cdrom.com> From: "Jordan K. Hubbard" Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > DEATH TO THIS STUPID SYMLINK BY DEFAULT. Done!! :-) Jordan From owner-cvs-etc Thu Jun 5 15:09:40 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id PAA10001 for cvs-etc-outgoing; Thu, 5 Jun 1997 15:09:40 -0700 (PDT) Received: from agora.rdrop.com (root@agora.rdrop.com [199.2.210.241]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id PAA09995; Thu, 5 Jun 1997 15:09:31 -0700 (PDT) Received: from florence.pavilion.net (mailrelay1.pavilion.net [194.242.128.25]) by agora.rdrop.com (8.8.5/8.8.5) with ESMTP id PAA01181; Thu, 5 Jun 1997 15:09:26 -0700 (PDT) Received: (from joe@localhost) by florence.pavilion.net (8.8.5/8.8.5) id XAA10405; Thu, 5 Jun 1997 23:07:26 +0100 (BST) Message-ID: <19970605230726.06023@pavilion.net> Date: Thu, 5 Jun 1997 23:07:26 +0100 From: Josef Karthauser To: Warner Losh Cc: "Jordan K. Hubbard" , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist References: <9456.865466424@time.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.74 In-Reply-To: ; from Warner Losh on Thu, Jun 05, 1997 at 01:11:23PM -0600 Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, Jun 05, 1997 at 01:11:23PM -0600, Warner Losh wrote: > I'd argue for that for other reasons. I don't want my "installed" > include files to change out from under me as I do a cvs update in my > /usr/src/sys area. This is just wrong, because that's what make > install is for. > > Some people like it, I know, but in general, for most people it is > confusing and tends to cause lots of errors. > > Oh, I tend to have /usr/src on a jaz drive, and if I unmount it, I > can't compile *ANYTHING*. I agree also (if I'm allowed to mail here.) I needed more space on one of my servers and so had to destroy /usr/src. This inadvertently removed my capability to compile a lot of things. (There was much unhappiness ;( ) Joe -- Josef Karthauser Technical Manager Email: joe@pavilion.net Pavilion Internet plc. [Tel: +44 1273 607072 Fax: +44 1273 607073] From owner-cvs-etc Thu Jun 5 19:32:44 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id TAA21645 for cvs-etc-outgoing; Thu, 5 Jun 1997 19:32:44 -0700 (PDT) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id TAA21640; Thu, 5 Jun 1997 19:32:40 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id MAA29428; Fri, 6 Jun 1997 12:00:56 +0930 (CST) From: Michael Smith Message-Id: <199706060230.MAA29428@genesis.atrad.adelaide.edu.au> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706051140.LAA02530@veda.is> from Adam David at "Jun 5, 97 11:40:33 am" To: adam@veda.is (Adam David) Date: Fri, 6 Jun 1997 12:00:56 +0930 (CST) Cc: dg@root.com, jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Adam David stands accused of saying: > > Well, I disagree with you, Jordan. I think we should make the kernel > > sources a mandatory part of the system and /usr/include/sys should be a > > symlink. Count this as a resounding NO vote to the above. > > David Greenman > > Let me second this, and add the rationale that most people don't want to > be running a kernel.GENERIC bloated with all kinds of drivers for devices > they probably don't have (and not all drivers that they do need). This is contradicted by my experience and that of numerous other posters over the years. Most users don't give a damn whether there are extra drivers in their kernel; OTOH they generally get pretty pissed if a driver they need _isn't_. > However, building a custom kernel will likely always be a typical operation > during post-installation machine configuration. Don't make me laugh. It _might_ be, if the user/admin is particularly paranoid, but believe me when I say that GENERIC is a lot more popular than you seem to think. > Adam David -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[ From owner-cvs-etc Fri Jun 6 10:47:27 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id KAA01418 for cvs-etc-outgoing; Fri, 6 Jun 1997 10:47:27 -0700 (PDT) Received: from news1.gtn.com (news1.gtn.com [194.77.0.15]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id KAA01365; Fri, 6 Jun 1997 10:45:39 -0700 (PDT) Received: (from uucp@localhost) by news1.gtn.com (8.7.2/8.7.2) with UUCP id TAA15325; Fri, 6 Jun 1997 19:30:19 +0200 (MET DST) Received: (from andreas@localhost) by klemm.gtn.com (8.8.5/8.8.5) id TAA17134; Fri, 6 Jun 1997 19:15:36 +0200 (CEST) Message-ID: <19970606191535.06310@klemm.gtn.com> Date: Fri, 6 Jun 1997 19:15:35 +0200 From: Andreas Klemm To: Warner Losh Cc: "Jordan K. Hubbard" , =?iso-8859-1?Q?=E1=CE=C4=D2=C5=CA_=FE=C5=D2=CE=CF=D7?= , David Nugent , cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG Subject: Re: cvs commit: src/etc/mtree BSD.include.dist References: <9725.865468350@time.cdrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 0.75 In-Reply-To: ; from Warner Losh on Thu, Jun 05, 1997 at 01:16:07PM -0600 X-Disclaimer: A free society is one where it is safe to be unpopular X-Operating-System: FreeBSD 3.0-CURRENT SMP Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk On Thu, Jun 05, 1997 at 01:16:07PM -0600, Warner Losh wrote: > cd /usr/src ; cvs update -PAd > > Should *NOT* update anything in /usr/include, directly or indirectly. > Eliminating the symlinks completely solves this problem. The small > amount of disk space savings just isn't worth the hassles. Tr ||| ue ... ;-) -- Andreas Klemm | klemm.gtn.com - powered by Symmetric MultiProcessor FreeBSD http://www.freebsd.org/~fsmp/SMP/SMP.html http://www.freebsd.org/~fsmp/SMP/benches.html From owner-cvs-etc Fri Jun 6 13:44:10 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id NAA10975 for cvs-etc-outgoing; Fri, 6 Jun 1997 13:44:10 -0700 (PDT) Received: from veda.is (veda.is [193.4.230.1]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id NAA10911; Fri, 6 Jun 1997 13:43:46 -0700 (PDT) Received: (from adam@localhost) by veda.is (8.8.5/8.8.5) id UAA00265; Fri, 6 Jun 1997 20:45:08 GMT From: Adam David Message-Id: <199706062045.UAA00265@veda.is> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706060230.MAA29428@genesis.atrad.adelaide.edu.au> from Michael Smith at "Jun 6, 97 12:00:56 pm" To: msmith@atrad.adelaide.edu.au (Michael Smith) Date: Fri, 6 Jun 1997 20:45:07 +0000 (GMT) Cc: dg@root.com, jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@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-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk > Adam David stands accused of saying: > > > Well, I disagree with you, Jordan. I think we should make the kernel > > > sources a mandatory part of the system and /usr/include/sys should be a > > > symlink. Please check your attribution better, it belongs further down. I guess you can blame blind/dumb software. > Count this as a resounding NO vote to the above. > > > > David Greenman > > > > Let me second this, and add the rationale that most people don't want to > > be running a kernel.GENERIC bloated with all kinds of drivers for devices > > they probably don't have (and not all drivers that they do need). > > This is contradicted by my experience and that of numerous other > posters over the years. Most users don't give a damn whether there > are extra drivers in their kernel; OTOH they generally get pretty pissed > if a driver they need _isn't_. According to this, most users are wasting precious memory. > > However, building a custom kernel will likely always be a typical operation > > during post-installation machine configuration. > > Don't make me laugh. It _might_ be, if the user/admin is particularly > paranoid, but believe me when I say that GENERIC is a lot more popular > than you seem to think. It is not a contest, and I think most would agree that both camps are significant. -- Adam David From owner-cvs-etc Fri Jun 6 22:16:01 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id WAA00496 for cvs-etc-outgoing; Fri, 6 Jun 1997 22:16:01 -0700 (PDT) Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id WAA00476; Fri, 6 Jun 1997 22:15:57 -0700 (PDT) Received: (from msmith@localhost) by genesis.atrad.adelaide.edu.au (8.8.5/8.7.3) id OAA10682; Sat, 7 Jun 1997 14:45:12 +0930 (CST) From: Michael Smith Message-Id: <199706070515.OAA10682@genesis.atrad.adelaide.edu.au> Subject: Re: cvs commit: src/etc/mtree BSD.include.dist In-Reply-To: <199706062045.UAA00265@veda.is> from Adam David at "Jun 6, 97 08:45:07 pm" To: adam@veda.is (Adam David) Date: Sat, 7 Jun 1997 14:45:12 +0930 (CST) Cc: msmith@atrad.adelaide.edu.au, dg@root.com, jkh@time.cdrom.com, ache@nagual.pp.ru, asami@cs.berkeley.edu, bde@zeta.org.au, cvs-all@FreeBSD.ORG, cvs-committers@FreeBSD.ORG, cvs-etc@FreeBSD.ORG X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-etc@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Adam David stands accused of saying: > > Adam David stands accused of saying: > > > > Well, I disagree with you, Jordan. I think we should make the kernel > > > > sources a mandatory part of the system and /usr/include/sys should be a > > > > symlink. > > Please check your attribution better, it belongs further down. > I guess you can blame blind/dumb software. You can blame me trimming bulk out of the message. I think it's fair to read three levels of indentation as "possibly not this individual". > > This is contradicted by my experience and that of numerous other > > posters over the years. Most users don't give a damn whether there > > are extra drivers in their kernel; OTOH they generally get pretty pissed > > if a driver they need _isn't_. > > According to this, most users are wasting precious memory. Er, yes. I would have thought that this was a fairly obvious fact, hardly deserving of comment. > > Don't make me laugh. It _might_ be, if the user/admin is particularly > > paranoid, but believe me when I say that GENERIC is a lot more popular > > than you seem to think. > > It is not a contest, and I think most would agree that both camps are > significant. Significant, yes. However this is effectively a variation on the disagreement I had with PST; those who care can fix it. Those that can't, can't. So, if you care, fix it. If not, or if being able to throw a single kernel at dozens of different sites without having to rummage through your archives to find their superoptimised 10k-smaller kernel config, then GENERIC wins. > Adam David -- ]] Mike Smith, Software Engineer msmith@gsoft.com.au [[ ]] Genesis Software genesis@gsoft.com.au [[ ]] High-speed data acquisition and (GSM mobile) 0411-222-496 [[ ]] realtime instrument control. (ph) +61-8-8267-3493 [[ ]] Unix hardware collector. "Where are your PEZ?" The Tick [[